DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
Applicant's arguments filed 11/25/2025 regarding the rejection of Claim 21 under 35 USC 112(a) have been fully considered and are persuasive. Claim 21 has been cancelled via the instant amendment. Accordingly, the rejection is moot. The rejection of Claim 21 under 35 USC 112(a) is withdrawn.
Applicant’s arguments regarding the rejection in Independent Claim 1 under 35 USC 103 as unpatentable over U.S. Patent Publication No. 2019/0180879 A1 to Jain et al. (“Jain”) in view of U.S. Patent Publication No. 2021/0098093 A1 to Shadid et al. (“Shadid”), Non-Patent Literature Anja Thieme, Danielle Belgrave, and Gavin Doherty. 2020. Machine Learning in Mental Health: A Systematic Review of the HCI Literature to Support the Development of Effective and Implementable ML Systems. ACM Trans. Comput.-Hum. Interact. 27, 5, Article 34 (October 2020), 53 pages (“Thieme”) and Alexander Yanovski, Reuben E. Kron, Raymond R. Townsend, Virginia Ford, The Clinical Utility of Ambulatory Blood Pressure and Heart Rate Monitoring in Psychiatric Inpatients, American Journal of Hypertension, Volume 11, Issue 3, March 1998, Pages 309–315 (“Yanowski”) have been fully considered but they are not persuasive.
Applicant argues that Shadid (the pertinent reference) does not disclose the amended Claim 1 limitations "determining whether the second risk score is within a second predetermined range corresponding to adequate therapeutic response to the medication" and "in response to the second risk score not being within the second predetermined range corresponding to adequate therapeutic response to the medication and determining that the second risk score is indicative of an inadequate therapeutic response to the medication: automatically scheduling a follow up" because “Shadid's system monitors for disease progression or emergence of new conditions, whereas the amended claim monitors therapeutic efficacy of medication administered for the diagnosed condition before automatically scheduling a follow up.” Applicant cites Shadid’s Para. [0119], which Applicant argues “makes clear that Shadid's appointment scheduling occurs when a new disease pattern is detected, not when medication proves inadequate for treating the existing condition.”
The Examiner does not disagree with Applicant’s characterization of Shadid, but disagrees that the breadth of Claim 1 precludes Shadid’s applicability. That is to say, the limitation “in response to the second risk score not being within the second predetermined range corresponding to adequate therapeutic response to the medication and determining that the second risk score is indicative of an inadequate therapeutic response to the medication: automatically scheduling a follow up” is so broad that it encompasses both “scheduling … when a new disease pattern is detected” and “when medication proves inadequate for treating the existing condition” (Applicant’s Remarks at Pg. 13). Applicant appears to attribute to the purported distinction to the language “corresponding to adequate therapeutic response to the medication and determining that the second risk score is indicative of an inadequate therapeutic response.” However, the broad term “corresponding to” is at odds with such a narrow interpretation. Shadid does not schedule a follow-up if the administered medication is effective: by Applicant’s own characterization, Shadid instead schedules a follow up when a new disease pattern is detected. Shadid’s “corresponding” goes the opposite direction of that of Claim 1, but is still such “corresponding” as claimed. As such, Shadid’s follow-up is indeed scheduled “in response to the second risk score not being within the second predetermined range corresponding to adequate therapeutic response to the medication and determining that the second risk score is indicative of an inadequate therapeutic response to the medication.”
Applicant’s arguments regarding the rejection of Independent Claims 12 and 20 under 35 USC 103 are similar to Applicant’s arguments regarding Independent Claim 1. Applicant’s arguments have been fully considered but are not persuasive for the same reasons as explained above.
Applicant’s arguments regarding the rejection of Claims 1-5 and 7-21 under 35 USC 101 have been fully considered but are not persuasive. Applicant’s arguments cite two pieces of precedent whose conclusions regarding eligibility diverge (Ex parte Desjardins, Appeal 2024-000567, Application 16/319,040 (Decision on Director Review) and Recentive Analytics, Inc. v. Fox Corp., No. 2023-2437 (Fed. Cir. Apr. 18, 2025)), assert that the claims at issue are more akin to the former than the latter, and conclude patent eligibility accordingly. The Examiner respectfully disagrees. In the Examiner’s view, the Present Claims are more similar to those at issue in Recentive than Desjardins. By the logic of Recentive, and consistent with that of Desjardins, the Examiner maintains that Claims 1-5 and 7-21 are directed to patent ineligible subject matter.
Both Recentive and Desjardins abide by the same test for patent eligibility (see Recentive at Pg. 10 of 18; Desjardins at Pg. 4 of 10). Using this test, the claims of both Recentive and Desjardins were found to recite abstract ideas (see Recentive at Pg. 16 of 18; Desjardins at Pg. 6 of 10). Patent eligibility in both Recentive and Desjardins turned on whether the claims at issue amount to significantly more than the recited abstract ideas (see Recentive at Pg. 16 of 18; Desjardins at Pg. 6 of 10).
Recentive and Desjardins diverge in this respect. In Recentive, the Federal Circuit found the claims at issue to be ineligible for failing to satisfy step two of the Alice inquiry (Recentive at Pg. 17 of 18). In contrast, the Appeals Review Panel of Desjardins found the claims eligible because they satisfied Step 2A, Prong Two of the Alice test (see Desjardins at Pg. 9 of 10, referencing MPEP 2106). The Panel’s holding in Desjardins turned on its finding that the claims there-at-issue constituted “an improvement to how the machine learning model itself operates, and not, for example, the identified mathematical calculation” (see Desjardins at Pg. 9 of 10, referencing MPEP 2106.05(a) and Intellectual Ventures I LLC v. Symantec Corp., 838 F.3d 1307, 1316 (Fed. Cir. 2016)). The Federal Circuit held the opposite in Recentive, finding the claims there-at-issue to be mere application of generic machine learning to a new data environment (see Recentive at Pg. 18 of 18).
The instant claims are more akin to those at issue in Recentive than in Desjardins. In contrast to the claims of Desjardins, Independent Claim 1 does not constitute “an improvement to how the machine learning model itself operates.” Independent Claim 1 is more akin to the claim set of Recentive, which amount to “the application of generic machine learning to new data environments, without disclosing improvements to the machine learning models to be applied…” (see Recentive at Pg. 18 of 18).
In contrast to Desjardins, Independent Claim 1 does not recite an improvement to the functioning of a computer. MPEP 2106.05(a) states “If it is asserted that the invention improves upon conventional functioning of a computer, or upon conventional technology or technological processes, a technical explanation as to how to implement the invention should be present in the specification. That is, the disclosure must provide sufficient details such that one of ordinary skill in the art would recognize the claimed invention as providing an improvement.” No such detail is provided in the Specification. Applicant argues on Pg 15-16 of Applicant’s Remarks that “the key improvement” is the limitation “retraining the predictive machine learning model based on additional training data sets associated with information corresponding to the wearable device users.” This purported improvement does not meet the criteria of MPEP 2106.05(a) for two reasons:
The purported improvement is to the abstract idea itself. The “retraining” limitation highlighted by Applicant was asserted at Para. 13(b)(i)(1) of the Non-Final Office Action dated 8/27/205 to recite an abstract idea. Applicant has not refuted that assertion. A similar assertion is made herein. In accordance with MPEP 2106.05(a), “…the judicial exception alone cannot provide the improvement. The improvement can be provided by one or more additional elements.” As such, and consistent with the Panel’s holding in Desjardins, Claim 1 does not recite such an improvement as would render its subject matter patent eligible.
The above finding is similar to the Federal Circuit’s in Recentive, where the purportedly inventive concept of “using machine learning to dynamically generate optimized maps and schedules based on real-time data and update them based on changing conditions” was held to be “no more than claiming the abstract idea itself” (see Recentive at Pg. 16 of 18).
The purported improvement is not an improvement in the sense contemplated by MPEP 2106.05(a) because it is not sufficiently described as such by the Present Specification. The “retraining” step highlighted by Applicant is described only at a high level of generality, and without reference to the specifics of any particular computer algorithm whose functionality Applicant argues is improved. Accordingly, it would not be apparent to a person of ordinary skill in the art that the “retraining” of Claim 1 constitutes an improvement to the functioning of a computer.
This fact meaningfully differentiates Independent Claim 1 from Desjardins. In Desjardins, the Panel cites Paragraph 21 of Desjardins’s Specification as identifying improvements in training the machine learning model itself (see Desjardins at Pg. 8 through Pg. 9). The Present Specification is lacking in this regard.
Applicant’s arguments regarding Independent Claims 12 and 20 are similar to Applicant’s arguments regarding Independent Claim 1. Applicant’s arguments have been fully considered but are not persuasive for the same reasons as explained above.
Applicant’s arguments regarding dependent Claims 2-5, 7-11, 13-20 and 21 are based on Applicant’s arguments regarding the independent Claims from which each respectively depends. Applicant’s arguments have been fully considered but are not persuasive for the same reasons explained above.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-5 and 7-20 and 22 are rejected under 35 U.S.C. § 101 because the claimed invention is directed to a judicial exception (i.e., an abstract idea) without significantly more.
Regarding Independent Claim 1, Claim 1 is ineligible.
Eligibility Step 1: Claim 1 is directed to “a method” (i.e., a process) and thus falls within one of the four statutory categories.
Eligibility Step 2A, Prong One: Claim 1 recites an abstract idea.
“training a predictive machine learning model to predict risk values for mental or neurological disorders based on training data sets…” recites an abstract idea (specifically, a mathematical calculation) when afforded its broadest reasonable interpretation. Under the broadest reasonable interpretation, the terms of the claim are presumed to have their plain meaning consistent with the specification as it would be interpreted by one of ordinary skill in the art. See MPEP 2111. The Present Specification states at Para. [0037] that “…the predictive model may comprise a machine learning model or regression model (e.g. a multiple logistic regression model)….” While the term “a predictive machine learning model” as recited in Claim 1 is broader than (and therefore not limited solely to) such a “multiple logistic regression model” as described at Para. [0037], one of ordinary skill in the art would understand from the Specification the claimed “predictive machine learning model” to mean such a model that calculates the probability of something happening based on multiple sets of variables. Accordingly, a mathematical calculation is recited. See Example 47 of July 2024 Subject Matter Eligibility Examples from the 2024 Guidance Update on Patent Subject Matter Eligibility, Including on Artificial Intelligence.
The limitation “retraining the predictive machine learning model” recites a mathematical calculation for the same reasons.
“determine a first risk score corresponding to the type of mental disorder or neurological disorder” recites an abstract idea (specifically, a mental process) when afforded its broadest reasonable interpretation. Such determining as claimed could be performed in the human mind. For example, a human could observe data detected by the sensor (e.g., via reading a computer printout of the information) and exercise judgment to assign (i.e., “determine”) a risk score which corresponds to the type of mental disorder based on that information.
The similar limitation “determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder” recites a mental process for the same reasons.
“determining whether the first risk score is within a first predetermined range” recites an abstract idea (specifically, a mental process) when afforded its broadest reasonable interpretation. The claimed “determining” is practically performable in the human mind. For example, a human could observe both the first score and the first predetermined range, and exercise judgment regarding whether the first risk score is within the first predetermined range.
The similar limitation “determining whether the second risk score is within a second predetermined range” recites a mental process for the same reasons.
Eligibility Step 2A, Prong Two: Claim 1 does not recite additional elements that integrate the judicial exception into a practical application.
“storing real-time patient information, corresponding to a user, and detected by a sensor of a wearable device, into a particular Electronic Medical Record (EMR) associated with the user” amounts to generally linking the use of a judicial exception to a particular technological environment or field of use. Storing data that is subsequently analyzed does not add a meaningful limitation to the analysis of such data.
The similar limitation “storing the additional real-time patient information into the particular EMR associated with the user” does not integrate the recited judicial exceptions into a practical application for the same reasons.
“applying the predictive machine learning model to the real-time patient information, from the particular EMR,” amounts to merely reciting the words “apply it” with the judicial exception, and as such is insufficient to integrate the recited abstract idea into a practical application. See MPEP 2106.04(d)(I). MPEP 2106.05(f) provides the following considerations for determining whether a claim simply recites a judicial exception with the words “apply it” (or an equivalent), such as mere instructions to implement an abstract idea on a computer: (1) whether the claim recites only the idea of a solution or outcome i.e., the claim fails to recite details of how a solution to a problem is accomplished; (2) whether the claim invokes computers or other machinery merely as a tool to perform an existing process; and (3) the particularity or generality of the application of the judicial exception. The judicial exception “determine a first risk score” is performed by “applying the predictive machine learning model….” The predictive machine learning model is used generally to apply the abstract idea without placing any limits on how the predictive machine learning model functions, and does not provide any details regarding how the determining is accomplished. Instead, only the outcome of a first risk score being determined are recited.
The similar recitation “applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR” does not integrate the recited judicial exceptions into a practical application for the same reasons.
“automatically providing a first notification, wherein the user is administered a medication based on the first notification” amounts to necessary data outputting in conjunction with the recited mental process, and is insignificant extra-solution activity. The provided notification is the result of the mental process “determining whether the first risk score is within a first predetermined range,” and is necessary for use of the claimed “determining” as without it the determination would remain unknown. The claimed “…providing a notification…” thus does not (either on its own or in combination with the remainder of the claim) add a meaningful limitation, and is insufficient to integrate the claimed mental process into a practical application.
The portion of the limitation which states “wherein the patient is administered a medication based on the notification” does not modify the claimed “providing a notification.” Administration of a medication is not affirmatively recited. Instead, the patient “is” administered a medication “based on” the notification. Per the broadest reasonable interpretation of the claim, such administration could take place days later, and could be necessitated by factors other than that upon which the notification was based, provided the notification was considered in the decision to administer medication.
“receiving feedback from the wearable device during a defined monitoring period subsequent to the medication being administered to the user based on the first notification, the feedback comprising additional real-time patient information detected by the sensor of the wearable device, wherein the feedback indicated physiological or behavioral changes in the user following administration of the medication” amounts to necessary data gathering in conjunction with the recited mental process, and is insignificant extra-solution activity. The mental process of determining a second risk score relies on the “receiv[ed] feedback.” Receipt (i.e., “gathering”) of the feedback (i.e., “data”) is therefore required for use of the claimed abstract idea. The claimed “receiving feedback…” thus does not (either on its own or in combination with the remainder of the claim) add a meaningful limitation, and is insufficient to integrate the claimed mental process into a practical application.
Neither the fact that the feedback is received during a particular time nor what the feedback indicates change that receipt of such feedback amounts to necessary data gathering in conjunction with the recited mental process in the manner explained above.
“automatically scheduling a follow up…” amounts to necessary data outputting in conjunction with the recited mental process, and is insignificant extra-solution activity. The claimed “scheduling” is the result of the mental process “determining whether the second risk score is within the second predetermined range.” Contextually, the claimed “scheduling” is outputting the result of determining whether the administered medication was effective. Such outputting is necessary for use of the claimed “determining.” as without it the determination would remain unknown. The claimed “…scheduling a follow up…” thus does not add a meaningful limitation, and is insufficient to integrate the claimed mental process into a practical application.
That the recited “automatically scheduling a follow up” is done “in response to the second risk score not being within the second predetermined range…” does not change the fact that the claimed “scheduling” is the result of the above-noted mental process.
Eligibility Step 2B: Claim 1 does not amount to significantly more than the abstract ideas recited therein.
“storing real-time patient information, corresponding to a user, and detected by a sensor of a wearable device, into a particular Electronic Medical Record (EMR) associated with the user” does not contribute an inventive concept. Such storing is recited at a high level of generality, and is well-understood, routine and conventional. See MPEP 2106.05(d)(II).
The similar limitation “storing the additional real-time patient information into the particular EMR associated with the user” does not contribute an inventive concept for the same reasons.
“applying the predictive machine learning model to the real-time patient information, from the particular EMR,” does not contribute an inventive concept. Such applying is a mere instruction to apply the judicial exception. Mere instructions to apply a judicial exception cannot provide an inventive concept. MPEP 2106.05(f).
The similar recitation “applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR” does not contribute an inventive concept for the same reasons.
“automatically providing a first notification, wherein the user is administered a medication based on the first notification” does not contribute an inventive concept. Providing a notification to a wearable-device-user upon sensing a problematic parameter is well-understood, routine and conventional in the art. For example, Non-Patent Literature S. C. Mukhopadhyay, “Wearable Sensors for Human Activity Monitoring: A Review,” in IEEE Sensors Journal, vol. 15, no. 3, pp. 1321-1330, March 2015 states that “wearable sensors have become very popular in many applications such as medical … fields,” (Pg. 1321, Left Column, First Paragraph after Abstract) and describes as part of their “basic architecture” the ability to “generate a warning message” based on the processed data (Pg. 1322, Right Column, First Paragraph).
“receiving feedback from the wearable device during a defined monitoring period subsequent to the medication being administered to the user based on the first notification, the feedback comprising additional real-time patient information detected by the sensor of the wearable device, wherein the feedback indicated physiological or behavioral changes in the user following administration of the medication” does not contribute an inventive concept. Receiving user information from a sensor of a wearable device is well-understood, routine and conventional. For example, Non-Patent Literature Sara E. Schaefer et al.; "A Feasibility Study of Wearable Activity Monitors for Pre-Adolescent School-Age Children;" May 22, 2014; Centers for Disease Control and Prevention; Preventing Chronic Disease, Volume 11 describes such sensors as “increasingly used in health research to provide more complete, accurate and objective information,” “becoming increasingly available commercially” (Pg. 1 of 8, Fifth and Sixth paragraphs)
“automatically scheduling a follow up…” does not contribute an inventive concept. As explained above, such “scheduling” is in effect a notification. Providing a notification to a wearable-device-user upon sensing a problematic parameter is well-understood, routine and conventional in the art. For example, Non-Patent Literature S. C. Mukhopadhyay, “Wearable Sensors for Human Activity Monitoring: A Review,” in IEEE Sensors Journal, vol. 15, no. 3, pp. 1321-1330, March 2015 states that “wearable sensors have become very popular in many applications such as medical … fields,” (Pg. 1321, Left Column, First Paragraph after Abstract) and describes as part of their “basic architecture” the ability to “generate a warning message” based on the processed data (Pg. 1322, Right Column, First Paragraph).
Regarding Claims 2-5 and 7-9, Claims 2-5 and 7-9 are ineligible. Claims 2-5 and 7-9 fall within one of the four statutory categories. Claims 2-5 and 7-9 further limit the abstract ideas recited in Claim 1. Claims 2-5 and 7-9 do not contain any additional elements, and therefore do not integrate the recited abstract ideas into a practical application and do not contribute an inventive concept.
Regarding Claim 10, Claim 10 is ineligible.
Eligibility Step 1: Claim 10 is directed to “a method” (i.e., a process) and thus falls within one of the four statutory categories.
Eligibility Step 2A, Prong One: Claim 10 recites an abstract idea.
“determining a plurality of significant influencing factors corresponding to the type of mental disorder or neurological disorder using the predictive machine learning model” recites an abstract idea, and more specifically a mental process. Such “determining” as claimed could be performed in the human mind. For example, a human could create a simple predictive model using responses from the validating questionnaire (e.g., a model that looks at whether the entire questionnaire was completed to assign an agreeability score to the user) and use that simple model to decide upon the presence of such significant influencing factors as claimed (e.g., “low agreeability is a significant influencing factor for x”).
The limitation “determining the first risk score further using the plurality of significant influencing factors” modifies the mental process “determine a first risk score…” recited in Claim 1.
Eligibility Step 2A, Prong Two: Claim 10 does not recite additional elements that integrate the judicial exception into a practical application.
“Receiving a response from the user to a validating questionnaire for the type of mental disorder or neurological disorder” amounts to necessary data gathering in conjunction with the recited mental process, and is insignificant extra-solution activity. The mental process of “determine a first risk score” recited in Claim 1 relies on (i.e., via the further limitation imposed on it by Claim 10) the “receiv[ed] response.” Receipt (i.e., “gathering”) of the information (i.e., “data”) is therefore required for use of the claimed abstract idea. The claimed “receiving a response…” thus does not (either on its own or in combination with the remainder of the claim) add a meaningful limitation, and is insufficient to integrate the claimed mental process into a practical application
Eligibility Step 2B: Claim 10 does not amount to significantly more than the abstract ideas recited therein.
“receiving a response from the user to a validating questionnaire for the type of mental disorder or neurological disorder” does not contribute an inventive concept. Such receipt is well-understood, routine and conventional. For example, Non-Patent Literature D. McMillan et al.; "Defining successful treatment outcome in depression using the PHQ-9: A comparison of methods;" December 2010; Journal of Affective Disorders, Vol. 127, Issues 1-3, Pgs. 122-129 describes such a questionnaire, which it states “is widely used in primary care” (Pg. 122, Background Section of Abstract).
Regarding Claim 11, Claim 11 is ineligible. Claim 11 falls within one of the four statutory categories. Claim 11 further limit the abstract ideas recited in Claim 10. Claim 11 does not contain any additional elements, and therefore do not integrate the recited abstract ideas into a practical application and do not contribute an inventive concept.
Regarding Independent Claim 12, Claim 12 is ineligible.
Eligibility Step 1: Claim 12 is directed to “a non-transitory computer-readable medium” (i.e., a machine) and thus falls within one of the four statutory categories.
Eligibility Step 2A, Prong One: Claim 12 recites an abstract idea.
“training a predictive machine learning model to predict risk values for mental or neurological disorders based on training data sets…” recites an abstract idea (specifically, a mathematical calculation) when afforded its broadest reasonable interpretation for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
The limitation “retraining the predictive machine learning model” recites a mathematical calculation for the same reasons.
“determine a first risk score corresponding to the type of mental disorder or neurological disorder” recites an abstract idea (specifically, a mental process) when afforded its broadest reasonable interpretation for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
The similar limitation “determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder” recites a mental process for the same reasons.
“determining whether the first risk score is within a first predetermined range” recites an abstract idea (specifically, a mental process) when afforded its broadest reasonable interpretation for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
The similar limitation “determining whether the second risk score is within a second predetermined range” recites a mental process for the same reasons.
Eligibility Step 2A, Prong Two: Claim 12 does not recite additional elements that integrate the judicial exception into a practical application.
“A non-transitory computer-readable storage medium having instructions embodied thereon” is (1) insignificant extra-solution activity insufficient to integrate the judicial exception into a practical application and (2) a generic computer structure for performing a generic computer function, and thus simply amounts to using a computer as a tool to implement the abstract idea.
“storing real-time patient information, corresponding to a user, and detected by a sensor of a wearable device, into a particular Electronic Medical Record (EMR) associated with the user” amounts to generally linking the use of a judicial exception to a particular technological environment or field of use. Storing data that is subsequently analyzed does not add a meaningful limitation to the analysis of such data.
The similar limitation “storing the additional real-time patient information into the particular EMR associated with the user” does not integrate the recited judicial exceptions into a practical application for the same reasons.
“applying the predictive machine learning model to the real-time patient information, from the particular EMR,” amounts to merely reciting the words “apply it” with the judicial exception, and as such is insufficient to integrate the recited abstract idea into a practical application for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
The similar recitation “applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR” does not integrate the recited judicial exceptions into a practical application for the same reasons.
“automatically providing a first notification, wherein the user is administered a medication based on the first notification” amounts to necessary data outputting in conjunction with the recited mental process, and is insignificant extra-solution activity for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
“receiving feedback from the wearable device during a defined monitoring period subsequent to the medication being administered to the user based on the first notification, the feedback comprising additional real-time patient information detected by the sensor of the wearable device wherein the feedback indicates physiological or behavioral changes in the user following administration of the medication” amounts to necessary data gathering in conjunction with the recited mental process, and is insignificant extra-solution activity insufficient to integrate the judicial exception into a practical application for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
“automatically scheduling a follow up…” amounts to necessary data outputting in conjunction with the recited mental process, and is insignificant extra-solution activity insufficient to integrate the claimed mental process into a practical application for the same reasons as explained above with respect to the similar limitation recited in Claim 1
Eligibility Step 2B: Claim 12 does not amount to significantly more than the abstract ideas recited therein.
“A non-transitory computer-readable storage medium having instructions embodied thereon” does not contribute an inventive concept. The claimed “non-transitory computer-readable storage medium” is recited at a high level of generality, and use of such a storage medium as claimed is well-understood, routine and conventional in the art. For example, Non-Patent Literature S. C. Mukhopadhyay, "Wearable Sensors for Human Activity Monitoring: A Review," in IEEE Sensors Journal, vol. 15, no. 3, pp. 1321-1330, March 2015 states that “wearable sensors have become very popular in many applications such as medical … fields,” (Pg. 1321, Left Column, First Paragraph after Abstract) and describes as such a storage as being part of their “basic architecture” (Pg. 1322, Left Column, Third Paragraph; Pg. 1322, Fig. 3, “Microcontroller”).
“storing real-time patient information, corresponding to a user, and detected by a sensor of a wearable device, into a particular Electronic Medical Record (EMR) associated with the user” does not contribute an inventive concept for the same reasons explained above with respect to the similar limitation recited in Claim 1.
The similar limitation “storing the additional real-time patient information into the particular EMR associated with the user” does not contribute an inventive concept for the same reasons.
“applying the predictive machine learning model to the real-time patient information, from the particular EMR,” does not contribute an inventive concept for the same reasons explained above with respect to the similar limitation recited in Claim 1.
The similar recitation “applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR” does not contribute an inventive concept for the same reasons.
“automatically providing a first notification, wherein the user is administered a medication based on the first notification” does not contribute an inventive concept for the same reasons explained above with respect to the similar limitation recited in Claim 1.
“receiving feedback from the wearable device during a defined monitoring period subsequent to the medication being administered to the user based on the first notification, the feedback comprising additional real-time patient information detected by the sensor of the wearable device wherein the feedback indicates physiological or behavioral changes in the user following administration of the medication” does not contribute an inventive concept for the same reasons explained above with respect to the similar limitation recited in Claim 1.
“automatically scheduling a follow up…” does not contribute an inventive concept for the same reasons explained above with respect to the similar limitation recited in Claim 1.
Regarding Claim 13, Claim 13 is ineligible. Claim 13 falls within one of the four statutory categories. Claim 13 further limits the abstract ideas recited in Claim 12. Claim 13 does not contain any additional elements, and therefore does not integrate the recited abstract ideas into a practical application and does not contribute an inventive concept.
Regarding Claim 14, Claim 14 is ineligible.
Eligibility Step 1: Claim 14 is directed to a “media” (i.e., a machine) and thus falls within one of the four statutory categories.
Eligibility Step 2A, Prong One: Claim 14 recites an abstract idea.
“determining a third risk score of the user” recites an abstract idea, and more specifically a mental process for the same reasons as explained above regarding the similar limitation of Claim 1.
Eligibility Step 2A, Prong Two: Claim 14 does not recite additional elements that integrate the judicial exception into a practical application.
“receiving further real-time patient information from the wearable device” amounts to necessary data gathering in conjunction with the recited mental process, and is insignificant extra-solution activity. The mental process of “determining a third risk score” relies on the received information. Receipt (i.e., “gathering”) of the information (i.e., “data”) is therefore required for use of the claimed abstract idea. The claimed “receiving further real-time patient information” thus does not (either on its own or in combination with the remainder of the claim) add a meaningful limitation, and is insufficient to integrate the claimed mental process into a practical application.
“automatically providing a third notification” amounts to necessary data outputting in conjunction with the recited mental process, and is insignificant extra-solution activity for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
Eligibility Step 2B: Claim 14 does not amount to significantly more than the abstract ideas recited therein.
“receiving further real-time patient information from the wearable device” does not contribute an inventive concept. Such receipt is well-understood, routine and conventional. For example, Non-Patent Literature D. McMillan et al.; "Defining successful treatment outcome in depression using the PHQ-9: A comparison of methods;" December 2010; Journal of Affective Disorders, Vol. 127, Issues 1-3, Pgs. 122-129 describes such a questionnaire, which it states “is widely used in primary care” (Pg. 122, Background Section of Abstract).
“automatically providing a third notification” does not contribute an inventive concept for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
Regarding Claim 15, Claim 15 is ineligible.
Eligibility Step 1: Claim 15 is directed to a “media” (i.e., a machine) and thus falls within one of the four statutory categories.
Eligibility Step 2A, Prong One: Claim 15 recites an abstract idea.
“determining a plurality of influencing factors corresponding to the type of mental disorder or neurological disorder using wearable device information from the wearable device users and their responses to the validating questionnaire” recites an abstract idea, and more specifically a mental process for the same reasons as explained above regarding the similar limitation of Claim 10.
“determining confidence scores for each of the plurality of influencing factors” recites an abstract idea, and more specifically a mental process. Such “determining” as claimed could be performed in the human mind. For example, a human could glance at each of the two influencing factors he had previously determined and arbitrarily assign each a confidence score. No specific manner of determination is recited, and the particulars of the how the determination is made are not claimed.
“determining the first risk score by additionally using at least one of the plurality of influencing factors having a confidence score above a threshold” recites an abstract idea, and more specifically a mental process for the same reasons as explained above regarding the similar limitation of Claim 1.
Eligibility Step 2A, Prong Two: Claim 15 does not recite additional elements that integrate the judicial exception into a practical application.
“receiving a response from the user to the validating questionnaire for the type of mental disorder or neurological disorder” amounts to necessary data gathering in conjunction with the recited mental process, and is insignificant extra-solution activity. The mental process of “determining the first risk score” relies on the received information. Receipt (i.e., “gathering”) of the information (i.e., “data”) is therefore required for use of the claimed abstract idea. The claimed “receiving further real-time patient information” thus does not (either on its own or in combination with the remainder of the claim) add a meaningful limitation, and is insufficient to integrate the claimed mental process into a practical application.
Eligibility Step 2B: Claim 15 does not amount to significantly more than the abstract ideas recited therein.
“receiving a response from the user to the validating questionnaire for the type of mental disorder or neurological disorder” does not contribute an inventive concept. Such receipt is well-understood, routine and conventional. For example, Non-Patent Literature D. McMillan et al.; "Defining successful treatment outcome in depression using the PHQ-9: A comparison of methods;" December 2010; Journal of Affective Disorders, Vol. 127, Issues 1-3, Pgs. 122-129 describes such a questionnaire, which it states “is widely used in primary care” (Pg. 122, Background Section of Abstract).
Regarding Claims 16-19, Claims 16-19 are ineligible. Claims 16-19 fall within one of the four statutory categories. Claims 16-19 further limit the abstract ideas recited in Claim 12. Claims 16-19 do not contain any additional elements, and therefore do not integrate the recited abstract ideas into a practical application and do not contribute an inventive concept.
Regarding Independent Claim 20, Claim 20 is ineligible.
Eligibility Step 1: Claim 20 is directed to “a system” (i.e., a machine) and thus falls within one of the four statutory categories.
Eligibility Step 2A, Prong One: Claim 20 recites an abstract idea.
“training a predictive machine learning model to predict risk values for mental or neurological disorders based on training data sets…” recites an abstract idea (specifically, a mathematical calculation) when afforded its broadest reasonable interpretation for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
The limitation “retraining the predictive machine learning model” recites a mathematical calculation for the same reasons.
“determine a first risk score corresponding to the type of mental disorder or neurological disorder” recites an abstract idea (specifically, a mental process) when afforded its broadest reasonable interpretation for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
The similar limitation “determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder” recites a mental process for the same reasons.
“determining whether the first risk score is within a first predetermined range” recites an abstract idea (specifically, a mental process) when afforded its broadest reasonable interpretation for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
The similar limitation “determining whether the second risk score is within a second predetermined range…” recites a mental process for the same reasons.
Eligibility Step 2A, Prong Two: Claim 20 does not recite additional elements that integrate the judicial exception into a practical application.
“one or more processors;” is (1) insignificant extra-solution activity insufficient to integrate the judicial exception into a practical application and (2) a generic computer structure for performing a generic computer function, and thus simply amounts to using a computer as a tool to implement the abstract idea.
“one or more computer storage media storing computer-useable instructions” is (1) insignificant extra-solution activity insufficient to integrate the judicial exception into a practical application and (2) a generic computer structure for performing a generic computer function, and thus simply amounts to using a computer as a tool to implement the abstract idea.
“storing real-time patient information, corresponding to a user, and detected by a sensor of a wearable device, into a particular Electronic Medical Record (EMR) associated with the user” amounts to generally linking the use of a judicial exception to a particular technological environment or field of use. Storing data that is subsequently analyzed does not add a meaningful limitation to the analysis of such data.
The similar limitation “storing the additional real-time patient information into the particular EMR associated with the user” does not integrate the recited judicial exceptions into a practical application for the same reasons.
“applying the predictive machine learning model to the real-time patient information, from the particular EMR,” amounts to merely reciting the words “apply it” with the judicial exception, and as such is insufficient to integrate the recited abstract idea into a practical application for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
The similar recitation “applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR” does not integrate the recited judicial exceptions into a practical application for the same reasons.
“automatically providing a first notification, wherein the user is administered a medication based on the first notification” amounts to necessary data outputting in conjunction with the recited mental process, and is insignificant extra-solution activity for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
“receiving feedback from the wearable device during a defined monitoring period subsequent to the medication being administered to the user based on the first notification, the feedback comprising additional real-time patient information detected by the sensor of the wearable device wherein the feedback indicates physiological or behavioral changes in the user following administration of the modification” amounts to necessary data gathering in conjunction with the recited mental process, and is insignificant extra-solution activity insufficient to integrate the judicial exception into a practical application for the same reasons as explained above with respect to the similar limitation recited in Claim 1.
“automatically scheduling a follow up…” amounts to necessary data outputting in conjunction with the recited mental process, and is insignificant extra-solution activity insufficient to integrate the claimed mental process into a practical application for the same reasons as explained above with respect to the similar limitation recited in Claim 1
Eligibility Step 2B: Claim 20 does not amount to significantly more than the abstract ideas recited therein.
“one or more processors;” does not contribute an inventive concept. The claimed processors are recited at a high level of generality, and use of such processors as claimed is well-understood, routine and conventional in the art. For example, Non-Patent Literature S. C. Mukhopadhyay, "Wearable Sensors for Human Activity Monitoring: A Review," in IEEE Sensors Journal, vol. 15, no. 3, pp. 1321-1330, March 2015 states that “wearable sensors have become very popular in many applications such as medical … fields,” (Pg. 1321, Left Column, First Paragraph after Abstract) and describes as such a processor as being part of their “basic architecture” (Pg. 1322, Left Column, Third Paragraph; Pf. 1322, Fig. 1).
“one or more computer storage media storing computer-useable instructions” does not contribute an inventive concept. The claimed “computer storage medium” is recited at a high level of generality, and use of such a storage medium as claimed is well-understood, routine and conventional in the art. For example, Non-Patent Literature S. C. Mukhopadhyay, "Wearable Sensors for Human Activity Monitoring: A Review," in IEEE Sensors Journal, vol. 15, no. 3, pp. 1321-1330, March 2015 states that “wearable sensors have become very popular in many applications such as medical … fields,” (Pg. 1321, Left Column, First Paragraph after Abstract) and describes as such a storage as being part of their “basic architecture” (Pg. 1322, Left Column, Third Paragraph; Pg. 1322, Fig. 3, “Microcontroller”).
“storing real-time patient information, corresponding to a user, and detected by a sensor of a wearable device, into a particular Electronic Medical Record (EMR) associated with the user” does not contribute an inventive concept for the same reasons explained above with respect to the similar limitation recited in Claim 1.
The similar limitation “storing the additional real-time patient information into the particular EMR associated with the user” does not contribute an inventive concept for the same reasons.
“applying the predictive machine learning model to the real-time patient information, from the particular EMR,” does not contribute an inventive concept for the same reasons explained above with respect to the similar limitation recited in Claim 1.
The similar recitation “applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR” does not contribute an inventive concept for the same reasons.
“automatically providing a first notification, wherein the user is administered a medication based on the first notification” does not contribute an inventive concept for the same reasons explained above with respect to the similar limitation recited in Claim 1.
“receiving feedback from the wearable device during a defined monitoring period subsequent to the medication being administered to the user based on the first notification, the feedback comprising additional real-time patient information detected by the sensor of the wearable device wherein the feedback indicates physiological or behavioral changes in the user following administration of the modification” does not contribute an inventive concept for the same reasons explained above with respect to the similar limitation recited in Claim 1.
“automatically scheduling a follow up…” does not contribute an inventive concept for the same reasons explained above with respect to the similar limitation recited in Claim 1.
Regarding Claim 22, Claim 22 is ineligible.
Eligibility Step 1: Claim 22 is directed to “a method” (i.e., a process) and thus falls within one of the four statutory categories.
Eligibility Step 2A, Prong One: Claim 22 recites an abstract idea.
“combining the first risk score with the predisposition probability value to calculate a probability of mental health” recites a mental process and a mathematical calculation. The claimed computing is practically performable in the human mind. No particular manner of computing is required, and the specifics of how the claimed computing is done are not recited. Additionally, the claimed computing is a mathematical calculation.
Eligibility Step 2A, Prong Two: Claim 22 does not recite any additional elements.
Eligibility Step 2B: Claim 22 does not amount to significantly more than the abstract ideas recited therein.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim 1, 2, 4, 7-14, 16, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over previously cited U.S. Patent Publication No. 2019/0180879 A1 to Jain et al. (“Jain”) in view of previously cited U.S. Patent Publication No. 2021/0098093 A1 to Shadid et al. (“Shadid”), previously cited Non-Patent Literature Anja Thieme, Danielle Belgrave, and Gavin Doherty. 2020. Machine Learning in Mental Health: A Systematic Review of the HCI Literature to Support the Development of Effective and Implementable ML Systems. ACM Trans. Comput.-Hum. Interact. 27, 5, Article 34 (October 2020), 53 pages (“Thieme”) and Non-Patent Literature Alexander Yanovski, Reuben E. Kron, Raymond R. Townsend, Virginia Ford, The Clinical Utility of Ambulatory Blood Pressure and Heart Rate Monitoring in Psychiatric Inpatients, American Journal of Hypertension, Volume 11, Issue 3, March 1998, Pages 309–315 (”Yanowski”).
Regarding Independent Claim 1, Jain discloses:
A method for predictive, diagnostic, and therapeutic applications of wearables for mental health, the method comprising: (Abstract, “A method may include collecting sensor data related to health of a patient and receiving input that provides quantification of health of the patient.”);
training a predictive … model to predict risk values for mental or neurological disorders based on training data sets, (Para. [0168], “The total health module 116 may be configured to generate a health risk score of the patient. The health risk score may be based on responses by the patient to questionnaires provided by the questionnaire module 104, sensor data collected by the sensors 120 and 122, the SHC score determined by the graded escalation module 110, the CCB score determined by the chronic burden module 108, the LCC score determined by the lifestyle choice module 114, and/or the EHR data included in the EHR database 128. In some embodiments, the health risk score may also be based on the chronic data included in the chronic disease database 126;” see also Paras. [0169] and [0170]);
Jain describes teaching a predictive model in the manner claimed, but does not disclose the predictive model being a “predictive machine learning model.” This deficiency is addressed below.
the training data sets being (a) associated with information corresponding to wearable device users (Para. [0168], “The health risk score may be based on responses by the patient to questionnaires provided by the questionnaire module 104, sensor data collected by the sensors 120 and 122…;” Para. [0048], “In some embodiments, one or more of the sensors 120 and 122 may include on-body (e.g., wearable) devices and/or off-body (e.g., non-wearable) devices.”);
and (b) comprising one or more of: responses from a validating questionnaire for a type of mental disorder or neurological disorder; sleep information; activity level information; Electronic Medical Record (EMR) data; heart rate information; or validation data; (Para. [0168], “The health risk score may be based on responses by the patient to questionnaires provided by the questionnaire module 104, sensor data collected by the sensors 120 and 122…”);
storing real-time patient information, corresponding to a user and detected by a sensor of a wearable device, into a particular Electronic Medical Record (EMR) associated with the user, (Para. [0061], “The memory 117 may store various data in any data structure, such as a relational database structure. For example, the memory 117 may include collected data obtained from one or more of the sensors 120 and 122, the user device 124, the chronic disease database 126, and/or the EHR database 128.”);
wherein the sensor comprises at least one of: a photoplethysmogram sensor, a skin and ambient temperature sensor, a brain wave sensor, a photodetector, a photodiode, a photoresistor, a phototransistor, a charge-coupled-device, an active pixel sensor, a light sensor, an IR sensor, an electroencephalography sensor, an electromyography sensor, an electrooculography sensor, a heart rate monitor, an electrocardiogram, an electroencephalogram, a pedometer, a thermometer, a transdermal transmitter sensor, one or more front-facing cameras, a camera, a microphone, an accelerometer, a gyroscope, a blood pressure sensor, a pulse oximeter, a respiration rate sensor, a blood alcohol concentration sensor, an accelerometer sensor, a force sensor or a biometric sensor; (Para. [0123], “For example, the first sensor 120 and/or the second sensor 122 may include an accelerometer configured to determine movement of the patient…;” Para. [0048], “ For example, one or more of the sensors 120 and 122 may include a global positioning system (GPS) sensor, an accelerometer sensor, a pedometer sensor, a heart rate (HR) sensor, a blood pressure (BP) sensor, a blood glucose sensor, an electromyography (EMG) sensor, an electrocardiogram (ECG) sensor, an electroencephalography EEG sensor, a Galvanic Skin Response (GSR) sensor, a photoplethysmography (PPG) sensor, a temperature sensor, a sleep sensor, a posture sensor, a respiration sensor, a cardiac output sensor, a ballistocardiography (BCG) sensor, a stress sensor, an emotion sensing system, or any other sensor to detect and/or gather data about a physical state of the patient.”);
applying the predictive … model to the real-time patient information, from the particular EMR, to determine a first risk score corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the real-time patient information detected by the sensor of the wearable device; Para. [0168], “The total health module 116 may be configured to generate a health risk score of the patient. The health risk score may be based on responses by the patient to questionnaires provided by the questionnaire module 104, sensor data collected by the sensors 120 and 122, the SHC score determined by the graded escalation module 110, the CCB score determined by the chronic burden module 108, the LCC score determined by the lifestyle choice module 114, and/or the EHR data included in the EHR database 128. In some embodiments, the health risk score may also be based on the chronic data included in the chronic disease database 126.”);
determining whether the first risk score is within a first predetermined range; (Para. [0184], “In some embodiments, the total health module 116 may perform graded escalation on the health score for each of the three health dimensions where a patient may be considered to be at risk as their score falls below the threshold for the given category.”);
and retraining the predictive machine learning model based on additional training data sets associated with information corresponding to the wearable device users. (Para. [0307], “In some embodiments, the physical predictive model may be determined daily for a period of time;” see also Paras. [0316] through [0337] detailing Jain’s “physical predictive model”).
Jain does not disclose:
training a predictive machine learning model to predict risk values for mental or neurological disorders based on training data sets, (i.e., Jain discloses training a model to predict such risk values as claimed, but does not disclose “wherein the model is a predictive machine learning model”);
applying the predictive machine learning model to the real-time patient information, from the particular EMR, to determine a first risk score corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the real- time patient information detected by the sensor of the wearable device; (i.e., Jain discloses applying the model in the manner claimed, but does not disclose “wherein the model is a predictive machine learning model”);
in response to the first risk score not being within the first predetermined range, automatically providing a first notification, wherein the user is administered a medication based on the first notification;
receiving feedback from the wearable device during a defined monitoring period subsequent to the medication being administered to the user based on the first notification, the feedback comprising additional real-time patient information detected by the sensor of the wearable device wherein the feedback indicates physiological or behavioral changes in the user following administration of the medication;
storing the additional real-time patient information into the particular EMR associated with the user;
applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR, to determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the additional real-time patient information detected by the sensor of the wearable device; determining whether the second risk score is within a second predetermined range corresponding to adequate therapeutic response to the medication;
in response to the second risk score not being within the second predetermined range corresponding to adequate therapeutic response to the medication and determining that the second risk score is indicative of an inadequate therapeutic response to the medication, automatically scheduling a follow up;
Shadid describes “Systems and methods are provided for integrated healthcare management. The systems and methods include establishing a medical record data, accessing information about a treatment related to the medical record data, obtaining biometric data related to the treatment, processing the biometric data, determining whether the treatment is appropriate with respect to the biometric data, in response to the treatment being appropriate with respect to the biometric data, administering medication, and in response the treatment not being appropriate with respect to the biometric data, determining a new treatment, and administering new medication based on the new treatment.” (Abstract). Shadid is thus analogous art.
Shadid discloses:
in response to the first risk score not being within the first predetermined range, automatically providing a first notification, wherein the user is administered a medication based on the first notification; (Para. [0118], “In example embodiments, when the IHM module 302, 302 a, via the monitoring module 510 illustrated in FIG. 5, triggers an alert during patient monitoring, the IHM module 302, 302 a triages the alert based upon existing patterns, current patient prescriptions and patient information. If the alert involves a known condition for which the treating medication exists in the patient's treatment database 306(3), the IHM module 302, 302 a controls the dispensation of the medication, at the required dosage that is defined in the treatment database 306(3), to the patient.”);
receiving feedback from the wearable device … subsequent to the medication being administered to the user based on the first notification, the feedback comprising additional real-time patient information detected by the sensor of the wearable device wherein the feedback indicates physiological or behavioral changes in the user following administration of the medication; (Para. [0084], “According to exemplary embodiments, the receiving module 318 may be configured to receive real-time feedback data corresponding to an efficacy of the first medication, the determining module 324 may further determine a modified dosage level of the first medication based on the received real-time feedback data, and the modifying module 328 may modify the first information related to the first administration of the first medication based on the determined modified dosage level of the first medication.”);
Shadid’s feedback indicates physiological changes in the user following administration of the medication. Shadid states at Para. [0084] that modification is made based on “real-time feedback data corresponding to an efficacy of the first medication.” Shadid determines efficacy based in part on sensed “biometric data” (Para. [0082]). Shadid’s “biometric data” is monitored in real-time (see e.g., Shadid at Para. [0089]), and thus changes. Shadid’s feedback thus indicated physiological changes in the user following administration of the medication.
storing the additional real-time patient information into the particular EMR associated with the user; (Para. [0090], “In example embodiments, at step S408, the patient health database 306(1), the medication database 306(2) and the treatment database 306(3) receive the patient biometric data gathered at step S406. For example, with reference to FIG. 3. the IHM module 302 transmits and stores the biometric data at, e.g., the patient health database 306(1), the medication database 306(2) and the treatment database 306(3).”);
applying the predictive … model at least to the additional real-time patient information, from the particular EMR, to determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the additional real-time patient information detected by the sensor of the wearable device; (Para. [0084], “According to exemplary embodiments, the receiving module 318 may be configured to receive real-time feedback data corresponding to an efficacy of the first medication, the determining module 324 may further determine a modified dosage level of the first medication based on the received real-time feedback data, and the modifying module 328 may modify the first information related to the first administration of the first medication based on the determined modified dosage level of the first medication.”);
determining whether the second risk score is within a second predetermined range corresponding to adequate therapeutic response to the medication; (Fig. 14, after “Administer Medication S416,” the process repeats; Para. [0008]; Para. [0019]; Claim 1; Claim 13);
Para. [0008], Para. [0019], Claim 1, Claim 13 describe Shadid’s determination as being based on whether treatment is appropriate with respect to biometric data related to treatment. Post-administration, Shadid’s biometric data is reflective of therapeutic response. Shadid’s determination thus corresponds to adequate therapeutic response as claimed.
in response to the second risk score not being within the second predetermined range corresponding to adequate therapeutic response to the medication and determining that the second risk score is indicative of an inadequate therapeutic response to the medication, automatically scheduling a follow up; (Para. [0119], “If the IHM module 302, 302 a determines that the severity of the alert is routine and not urgent or critical, the IHM module 302, 302 a notifies the healthcare provider. In example embodiments, the IHM module 302, 302 a schedules an appointment, submits the prescription to the healthcare provider and alerts the patient to fill a prescription;” Para. [0008]; Para. [0019]; Claim 1; Claim 13).
Shadid’s follow-up is scheduled when inadequate therapeutic response to the medication is indicated. Shadid’s follow-up is scheduled when a detected response does not match a previously detected response (Para. [0119]). Shadid expects that the response will match a predefined pattern (Para. [0116]). If the response does not match the predefined pattern, Shadid schedules a follow-up (Para. [0119]). Such failure to match Shadid’s predefined pattern indicates inadequate therapeutic response to the administered medication. See also Shadid at Paras. [0025] and [0026] outlining Shadid’s process.
It would have been obvious for a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Jain with the teachings of Shadid (i.e., to modify the method of Jain such that medication is administered, the impact of the administered medication is monitored, and the process is repeated as taught by Shadid) in order to “deliver treatments and medicine directly to patients at the time they need it, as well as to track the medication and the real time biometric data of the patients” (Shadid at Para. [0004]).
The combination of Jain and Shadid does not disclose:
“training a predictive machine learning model to predict risk values for mental or neurological disorders based on training data sets,” (i.e., Jain discloses training a model to predict such risk values as claimed, but does not disclose “wherein the model is a predictive machine learning model”);
“applying the predictive machine learning model to the real-time patient information, from the particular EMR, to determine a first risk score corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the real- time patient information detected by the sensor of the wearable device;” (i.e., Jain discloses applying the model in the manner claimed, but does not disclose “wherein the model is a predictive machine learning model”);
“applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR, to determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the additional real-time patient information detected by the sensor of the wearable device;” (i.e., Shadid discloses applying the model in the manner claimed, but does not disclose “wherein the model is a predictive machine learning model”);
receiving feedback “during a defined monitoring period” (i.e., Shadid discloses receipt of such feedback as claimed, but Shadid’s feedback is not limited to “a defined monitoring period”).
Thieme describes “Machine Learning in Mental Health: A Systematic Review of the HCI Literature to Support the Development of Effective and Implementable ML Systems” (Title). Thieme is thus analogous art.
Thieme discloses:
“training a predictive machine learning model to predict risk values for mental or neurological disorders based on training data sets,” (i.e., Thieme discloses “wherein the model is a predictive machine learning model”); (Pg. 34:15, Fourth Paragraph, “3.3.1 Source and Scale of Mental Health Data. ML algorithms build mathematical models based on training data to make predictions or decisions without being explicitly programmed [93]. The papers in our corpus are split between those that collect data for this purpose (n = 29) and those that make use of existing data (n = 23). Existing data is provided through previously generated datasets (n = 14, plus 1 hybrid) and health records (n = 6, plus 2 hybrids).”);
“applying the predictive machine learning model to the real-time patient information, from the particular EMR, to determine a first risk score corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the real- time patient information detected by the sensor of the wearable device;” i.e., Thieme discloses “wherein the model is a predictive machine learning model”); (Pg. 34:15, Fourth Paragraph, “3.3.1 Source and Scale of Mental Health Data. ML algorithms build mathematical models based on training data to make predictions or decisions without being explicitly programmed [93]. The papers in our corpus are split between those that collect data for this purpose (n = 29) and those that make use of existing data (n = 23). Existing data is provided through previously generated datasets (n = 14, plus 1 hybrid) and health records (n = 6, plus 2 hybrids).”);
“applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR, to determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the additional real-time patient information detected by the sensor of the wearable device;” (i.e., Thieme discloses “wherein the model is a predictive machine learning model”); (Pg. 34:15, Fourth Paragraph, “3.3.1 Source and Scale of Mental Health Data. ML algorithms build mathematical models based on training data to make predictions or decisions without being explicitly programmed [93]. The papers in our corpus are split between those that collect data for this purpose (n = 29) and those that make use of existing data (n = 23). Existing data is provided through previously generated datasets (n = 14, plus 1 hybrid) and health records (n = 6, plus 2 hybrids).”);
It would have been obvious for a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of combined Jain and Shadid with the teachings of Thieme (i.e., to use such a predictive machine learning model as taught by Thieme as the predictive model of combined Jain and Shadid) in order to “assist in the detection, diagnosis and treatment of mental health problem,” as “[machine learning] techniques can potentially offer new routes for learning patterns of human behavior; identifying mental health symptoms and risk factors; developing predictions about disease progression; and personalizing and optimizing therapies” (Thieme at Pg. 34:1, Abstract).
The combination of Jain, Shadid and Thieme does not disclose:
receiving feedback “during a defined monitoring period” (i.e., Shadid discloses receipt of such feedback as claimed, but Shadid’s feedback is not limited to “a defined monitoring period”).
Yanowski describes “Alterations in heart rate and blood pressure (BP) … in patients receiving psychiatric medication” (Abstract). Yanowski is reasonably pertinent to the problem faced by the inventor, and is thus analogous art. See MPEP 2141.01(a)(I).
Yanowkski teaches:
receiving feedback “during a defined monitoring period” (i.e., Yanowski teaches “a defined monitoring period”) (Pg. 310, Right Column, Second Paragraph, “The monitor was programmed to obtain readings every 20 min during the day (6:00 am to 10:00 pm) and every 60 min during the night (10:00 pm to 6:00 am), in accordance with standard recommendations for use of the device. … The nurses followed the usual hospital procedures for care of patients on drugs with known potential for side effects.”).
It would have been obvious for a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the device of combined Jain, Shadid and Thieme with the teachings of Yanowski (i.e., to use such a defined time period as taught by Yanowski for receiving feedback in the device of combined Jain, Shadid and Thieme) in order to facilitate “early management of cardiovascular side effects” when administering psychiatric medications (Yanowski at Pg. 314, Left Column, Third Paragraph).
Regarding Claim 2, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 1 as explained above.
Jain additionally discloses:
“wherein the real-time patient information comprises sleep information, diet information, and activity information, and wherein the method further comprises:” (Para. [0158], “The LRSQ response may indicate habits of the patient that may be harmful. For example, smoking, alcohol, dietary harm, sedentary lifestyle, schedule irregularity, quality of social, and/or stress management of the patient.”);
“receiving EMR data for the user comprising … and demographic information” (Para. [0038], “The electronic device may receive electronic health data record (EHR) data from the physician of the patient;” Fig. 5, “Distal Factors 530;”
Shadid additionally discloses:
“consumed neuropsychiatric drug … information” (Para. [0087], “…the medical record data may include data related to a medical history of the patient, past ailments and diseases, past treatments and procedures performed on the patient, current ailments and diseases, current treatments and procedures, diagnoses for the patient's ailments, types and amounts of medication that five patient has been prescribed in the past and is currently being prescribed…”).
Regarding Claim 4, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 1 as explained above.
Jain additionally discloses:
“wherein the first risk score is further determined based on social information, family medical history, and medication history” (Fig. 2, “Obtain A Full Medical Profile Of A Patient 202;” Fig. 5, “Social Connections 530e”).
Regarding Claim 7, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 1 as explained above.
Jain additionally discloses:
“wherein the type of mental disorder comprises at least one of clinical depression, anxiety disorder, bipolar disorder, dementia, attention-deficit disorder, hyperactivity disorder, schizophrenia, obsessive compulsive disorder, and post-traumatic stress disorder” (Para. [0111], “The patient PHQ2 input received in response to the PHQ2 questionnaire may indicate a state of depression of the patient. In some embodiments, the PHQ2 questionnaire may be provided to the patient via the user device 124 before scheduling an in-person examination or at any other suitable time. In some embodiments, the module can escalate the depression detection by having the PHQ2 questionnaire be followed by the PHQ9 questionnaire if the PHQ2 questionnaire reflects that the patient may likely be depressed. Additionally, any other similar questionnaire that assesses depression can be employed as well.”);
Regarding Claim 8, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 1 as explained above.
Jain additionally discloses:
“wherein the real time patient information comprises calorie intake during a predetermined range of time” (Para. [0066] describes calories of meals being questioned).
Regarding Claim 9, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 1 as explained above.
Jain additionally discloses:
“determining the first risk score using an average value of a heart rate, an average value of a sleep level, and an average value of an activity level over a consecutive period of at least one month and comparing the average values to heart rate values, sleep level values, and activity level values over a consecutive period of at least two more recent days” (Para. [0048], “One or more of the sensors 120 and 122 may include any type of sensor to gather sensor data related to a physical state of a patient. For example, one or more of the sensors 120 and 122 may include a global positioning system (GPS) sensor, an accelerometer sensor, a pedometer sensor, a heart rate (HR) sensor, a blood pressure (BP) sensor, a blood glucose sensor, an electromyography (EMG) sensor, an electrocardiogram (ECG) sensor, an electroencephalography EEG sensor, a Galvanic Skin Response (GSR) sensor, a photoplethysmography (PPG) sensor, a temperature sensor, a sleep sensor, a posture sensor, a respiration sensor, a cardiac output sensor, a ballistocardiography (BCG) sensor, a stress sensor, an emotion sensing system, or any other sensor to detect and/or gather data about a physical state of the patient. Alternatively or additionally, one or more of the sensors 120 and 122 may include any type of sensor to gather sensor data related to a mental state of the patient. For example, one or more of the sensors 120 and 122 may detect emotional resilience, tiredness, mood, or any other mental state of the patient;” Para. [0049], “Additionally, the interpretation of sensor data may be adjusted based on when the sensor data is gathered such as for different months, days, seasons, or any other appropriate time based factor that may affect the health of the patient. The sensor data may permit quantification of health habits in terms of activity, sleep, stress, posture, outdoor time, regularity of routine, number of cigarettes per day, number of times fast food is consumed, or number of times alcohol is consumed or a restaurant that serves alcohol is visited, what type of food is being consumed (e.g., amounts of salt, sugar, trans-fat, or alcohol) of the patient. Alternatively or additionally, the sensor data may permit quantification of blood pressure, heart rate, heart rate variability, cardiac output, oxygen saturation, emotion markers, or pain markers of the patient;” Para. [0073], “In some embodiment, the patient response is taken as the weighted average of sensor derived input and the patient derived input.”).
Regarding Claim 10, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 1 as explained above.
Jain additionally discloses:
“Receiving a response from the user to a validating questionnaire for the type of mental disorder or neurological disorder;” (Para. [0063], “The questionnaire module 104 may be configured to provide one or more health related questionnaires to the patient via user device 124;” see also Paras. [0064] through [0069]);
“determining a plurality of significant influencing factors corresponding to the type of mental disorder or neurological disorder…” (Para. [0040], “In an embodiment, risk prediction can be made for an impact that the lifestyle choices of the patient may have on their chronic diseases. The electronic device may receive patient lifestyle input in response to a healthy lifestyle and personal control questionnaire (HLPCQ). The patient lifestyle input may include data related to dietary health choices, dietary harm avoidance, daily routine, organized physical exercise, and/or social and mental balance of the patient. The electronic device may store a database of statistically significant number of similar and comparable patients and using that it may determine the impact the various lifestyle choices of the patient may have on the chronic diseases of the patient.”);
“and determining the risk score using the plurality of significant influencing factors;” (Para. [0168], “The total health module 116 may be configured to generate a health risk score of the patient. The health risk score may be based on responses by the patient to questionnaires provided by the questionnaire module 104, sensor data collected by the sensors 120 and 122, the SHC score determined by the graded escalation module 110, the CCB score determined by the chronic burden module 108, the LCC score determined by the lifestyle choice module 114, and/or the EHR data included in the EHR database 128. In some embodiments, the health risk score may also be based on the chronic data included in the chronic disease database 126.”);
Thieme additionally discloses:
“using the predictive machine learning model” (Pg. 34:15, Fourth Paragraph, “3.3.1 Source and Scale of Mental Health Data. ML algorithms build mathematical models based on training data to make predictions or decisions without being explicitly programmed [93]. The papers in our corpus are split between those that collect data for this purpose (n = 29) and those that make use of existing data (n = 23). Existing data is provided through previously generated datasets (n = 14, plus 1 hybrid) and health records (n = 6, plus 2 hybrids).”);
Regarding Claim 11, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 10 as explained above.
Jain additionally discloses:
“wherein the plurality of significant influencing factors comprises age and family history” (Para. [0032], “Some of these factors may vary depending on the age, gender, race, ethnicity, geographic location, and/or other demographic factors of the patient.”).
Regarding Independent Claim 12, Jain discloses:
“A non-transitory computer-readable storage medium having instructions embodied thereon, the instructions being executable by one or more processors to perform a method for predictive, diagnostic, and therapeutic applications of wearables for mental health, the method comprising:” (Abstract, “A method may include collecting sensor data related to health of a patient and receiving input that provides quantification of health of the patient;” Para. [0060], “The memory 117 may include computer-readable storage media for collecting or storing data thereon. For example, the memory 117 may include computer-readable storage media that may be tangible or non-transitory computer-readable storage media such as Random Access Memory (RAM), Read-Only Memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Compact Disc Read-Only Memory (CD-ROM) or other optical disk storage, magnetic disk storage or other magnetic storage devices, flash memory devices (e.g., solid state memory devices), or any other tangible and non-transitory storage medium which may be used to store data that may be accessed by a general-purpose or special-purpose computer.”);
“training a predictive … model to predict risk values for mental or neurological disorders based on training data sets,” (Para. [0168], “The total health module 116 may be configured to generate a health risk score of the patient. The health risk score may be based on responses by the patient to questionnaires provided by the questionnaire module 104, sensor data collected by the sensors 120 and 122, the SHC score determined by the graded escalation module 110, the CCB score determined by the chronic burden module 108, the LCC score determined by the lifestyle choice module 114, and/or the EHR data included in the EHR database 128. In some embodiments, the health risk score may also be based on the chronic data included in the chronic disease database 126;” see also Paras. [0169] and [0170]; see Paras. [0306], [307] [336] and [337] detailing Jain’s “physical predictive model”);
“the training data sets being (a) associated with information corresponding to wearable device users” (Para. [0168], “The health risk score may be based on responses by the patient to questionnaires provided by the questionnaire module 104, sensor data collected by the sensors 120 and 122…;” Para. [0048], “In some embodiments, one or more of the sensors 120 and 122 may include on-body (e.g., wearable) devices and/or off-body (e.g., non-wearable) devices.”);
“and (b) comprising one or more of: responses from a validating questionnaire for a type of mental disorder or neurological disorder; sleep information; activity level information; Electronic Medical Record (EMR) data; heart rate information; or validation data;” (Para. [0168], “The health risk score may be based on responses by the patient to questionnaires provided by the questionnaire module 104, sensor data collected by the sensors 120 and 122…”);
“storing real-time patient information, corresponding to a user and detected by a sensor of a wearable device, into a particular Electronic Medical Record (EMR) associated with the user,” (Para. [0061]);
“wherein the sensor comprises at least one of: a photoplethysmogram sensor, a skin and ambient temperature sensor, a brain wave sensor, a photodetector, a photodiode, a photoresistor, a phototransistor, a charge-coupled-device, an active pixel sensor, a light sensor, an IR sensor, an electroencephalography sensor, an electromyography sensor, an electrooculography sensor, a heart rate monitor, an electrocardiogram, an electroencephalogram, a pedometer, a thermometer, a transdermal transmitter sensor, one or more front-facing cameras, a camera, a microphone, an accelerometer, a gyroscope, a blood pressure sensor, a pulse oximeter, a respiration rate sensor, a blood alcohol concentration sensor, an accelerometer sensor, a force sensor or a biometric sensor;” (Para. [0123]; Para. [0048], “ For example, one or more of the sensors 120 and 122 may include a global positioning system (GPS) sensor, an accelerometer sensor, a pedometer sensor, a heart rate (HR) sensor, a blood pressure (BP) sensor, a blood glucose sensor, an electromyography (EMG) sensor, an electrocardiogram (ECG) sensor, an electroencephalography EEG sensor, a Galvanic Skin Response (GSR) sensor, a photoplethysmography (PPG) sensor, a temperature sensor, a sleep sensor, a posture sensor, a respiration sensor, a cardiac output sensor, a ballistocardiography (BCG) sensor, a stress sensor, an emotion sensing system, or any other sensor to detect and/or gather data about a physical state of the patient.”););
“applying the predictive … model to the real-time patient information, from the particular EMR, to determine a first risk score corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the real- time patient information detected by the sensor of the wearable device;” (Para. [0168]);
“determining whether the first risk score is within a first predetermined range;” (Para. [0184]);
“and retraining the predictive … model based on additional training data sets associated with information corresponding to the wearable device users.” (Para. [0307], “In some embodiments, the physical predictive model may be determined daily for a period of time;” see also Paras. [0316] through [0337] detailing Jain’s “physical predictive model”)
Jain does not disclose:
“training a predictive machine learning model to predict risk values for mental or neurological disorders based on training data sets,” (i.e., Jain discloses training a model to predict such risk values as claimed, but does not disclose “wherein the model is a predictive machine learning model”);
“applying the predictive machine learning model to the real-time patient information, from the particular EMR, to determine a first risk score corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the real- time patient information detected by the sensor of the wearable device;” (i.e., Jain discloses applying the model in the manner claimed, but does not disclose “wherein the model is a predictive machine learning model”);
“in response to the first risk score not being within the first predetermined range, automatically providing a first notification, wherein the user is administered a medication based on the first notification;”
“receiving feedback from the wearable device during a defined monitoring period subsequent to the medication being administered to the user based on the first notification, the feedback comprising additional real-time patient information detected by the sensor of the wearable device wherein the feedback indicates physiological or behavioral changes in the user following administration of the medication;”
“storing the additional real-time patient information into the particular EMR associated with the user:”
“applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR, to determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the additional real-time patient information detected by the sensor of the wearable device;”
“determining whether the second risk score is within a second predetermined range corresponding to adequate therapeutic response to the medication;”
“in response to the second risk score not being within the second predetermined range corresponding to adequate therapeutic response to the medication and determining that the second risk score is indicative of an inadequate therapeutic response to the medication, automatically scheduling a follow up;”
Shadid describes “Systems and methods are provided for integrated healthcare management. The systems and methods include establishing a medical record data, accessing information about a treatment related to the medical record data, obtaining biometric data related to the treatment, processing the biometric data, determining whether the treatment is appropriate with respect to the biometric data, in response to the treatment being appropriate with respect to the biometric data, administering medication, and in response the treatment not being appropriate with respect to the biometric data, determining a new treatment, and administering new medication based on the new treatment.” (Abstract). Shadid is thus analogous art.
Shadid discloses:
“in response to the first risk score not being within the first predetermined range, automatically providing a first notification, wherein the user is administered a medication based on the first notification;” (Para. [0118]);
“receiving feedback from the wearable device … subsequent to the medication being administered to the user based on the first notification, the feedback comprising additional real-time patient information detected by the sensor of the wearable device wherein the feedback indicates physiological or behavioral changes in the user following administration of the medication;” (Para. [0084]);
This limitation is being interpreted similarly to the similar Claim 1 limitation elaborated upon above.
“storing the additional real-time patient information into the particular EMR associated with the user:” (Para. [0090]);
“applying the predictive … model at least to the additional real-time patient information, from the particular EMR, to determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the additional real-time patient information detected by the sensor of the wearable device;” (Para. [0084]);
“determining whether the second risk score is within a second predetermined range corresponding to adequate therapeutic response to the medication;” ((Fig. 14, after “Administer Medication S416,” the process repeats; Para. [0008]; Para. [0019]; Claim 1; Claim 13);
“in response to the second risk score not being within the second predetermined range corresponding to adequate therapeutic response to the medication and determining that the second risk score is indicative of an inadequate therapeutic response to the medication, automatically scheduling a follow up; (Para. [0119]; Para. [0008]; Para. [0019]; Claim 1; Claim 13).
This limitation is being interpreted similarly to the similar Claim 1 limitation elaborated upon above.
It would have been obvious for a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the device of Jain with the teachings of Shadid (i.e., to modify the device of Jain such that medication is administered, the impact of the administered medication is monitored, and the process is repeated as taught by Shadid) in order to “deliver treatments and medicine directly to patients at the time they need it, as well as to track the medication and the real time biometric data of the patients” (Shadid at Para. [0004]).
The combination of Jain and Shadid does not disclose:
“training a predictive machine learning model to predict risk values for mental or neurological disorders based on training data sets,” (i.e., Jain discloses training a model to predict such risk values as claimed, but does not disclose “wherein the model is a predictive machine learning model”);
“applying the predictive machine learning model to the real-time patient information, from the particular EMR, to determine a first risk score corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the real- time patient information detected by the sensor of the wearable device;” (i.e., Jain discloses applying the model in the manner claimed, but does not disclose “wherein the model is a predictive machine learning model”);
“applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR, to determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the additional real-time patient information detected by the sensor of the wearable device;” (i.e., Shadid discloses applying the model in the manner claimed, but does not disclose “wherein the model is a predictive machine learning model”);
receiving feedback “during a defined monitoring period” (i.e., Shadid discloses receipt of such feedback as claimed, but Shadid’s feedback is not limited to “a defined monitoring period”).
Thieme describes “Machine Learning in Mental Health: A Systematic Review of the HCI Literature to Support the Development of Effective and Implementable ML Systems” (Title). Thieme is thus analogous art.
Thieme discloses:
“training a predictive machine learning model to predict risk values for mental or neurological disorders based on training data sets,” (i.e., Thieme discloses “wherein the model is a predictive machine learning model”) (Pg. 34:15, Fourth Paragraph);
“applying the predictive machine learning model to the real-time patient information, from the particular EMR, to determine a first risk score corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the real- time patient information detected by the sensor of the wearable device;” (i.e., Thieme discloses “wherein the model is a predictive machine learning model”) (Pg. 34:15, Fourth Paragraph,);
“applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR, to determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the additional real-time patient information detected by the sensor of the wearable device;” (i.e., Thieme discloses “wherein the model is a predictive machine learning model”) (Pg. 34:15, Fourth Paragraph,);
It would have been obvious for a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the device of combined Jain and Shadid with the teachings of Thieme (i.e., to use such a predictive machine learning model as taught by Thieme as the predictive model of combined Jain and Shadid) in order to “assist in the detection, diagnosis and treatment of mental health problem,” as “[machine learning] techniques can potentially offer new routes for learning patterns of human behavior; identifying mental health symptoms and risk factors; developing predictions about disease progression; and personalizing and optimizing therapies” (Thieme at Pg. 34:1, Abstract).
The combination of Jain, Shadid and Thieme does not disclose:
receiving feedback “during a defined monitoring period” (i.e., Shadid discloses receipt of such feedback as claimed, but Shadid’s feedback is not limited to “a defined monitoring period”).
Yanowski describes “Alterations in heart rate and blood pressure (BP) … in patients receiving psychiatric medication” (Abstract). Yanowski is reasonably pertinent to the problem faced by the inventor, and is thus analogous art. See MPEP 2141.01(a)(I).
Yanowkski teaches:
receiving feedback “during a defined monitoring period” (i.e., Yanowski teaches “a defined monitoring period”) (Pg. 310, Right Column, Second Paragraph, “The monitor was programmed to obtain readings every 20 min during the day (6:00 am to 10:00 pm) and every 60 min during the night (10:00 pm to 6:00 am), in accordance with standard recommendations for use of the device. … The nurses followed the usual hospital procedures for care of patients on drugs with known potential for side effects.”).
It would have been obvious for a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the device of combined Jain, Shadid and Thieme with the teachings of Yanowski (i.e., to use such a defined time period as taught by Yanowski for receiving feedback in the device of combined Jain, Shadid and Thieme) in order to facilitate “early management of cardiovascular side effects” when administering psychiatric medications (Yanowski at Pg. 314, Left Column, Third Paragraph).
Regarding Claim 13, the combination of Jain and Shadid renders obvious the entirety of Claim 12 as explained above.
Jain additionally discloses:
“further comprising selecting the wearable device users for the training of the predictive … model based on age and gender” (Para. [0032], “Some of these factors may vary depending on the age, gender, race, ethnicity, geographic location, and/or other demographic factors of the patient.”).
Thieme additionally discloses:
“the predictive machine learning model” (Pg. 34:15, Fourth Paragraph);
Regarding Claim 14, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 12 as explained above.
Shadid additionally discloses:
“after retraining the predictive … model, receiving further real-time patient information from the wearable device; in response to retraining the predictive machine learning model and receiving the additional wearable device information, determining a third risk score for the user;” (Fig. 14, after “Administer Medication S416,” the process repeats);
“and in response to the third risk score not being within the first predetermined range, automatically providing a third notification” (Para. [0118], “In example embodiments, when the IHM module 302, 302 a, via the monitoring module 510 illustrated in FIG. 5, triggers an alert during patient monitoring, the IHM module 302, 302 a triages the alert based upon existing patterns, current patient prescriptions and patient information. If the alert involves a known condition for which the treating medication exists in the patient's treatment database 306(3), the IHM module 302, 302 a controls the dispensation of the medication, at the required dosage that is defined in the treatment database 306(3), to the patient.”);
Thieme additionally discloses:
“the predictive machine learning model” (Pg. 34:15, Fourth Paragraph);
Regarding Claim 16, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 12 as explained above.
Jain additionally discloses:
“wherein the first predetermined range is determined based on a probability of required medical intervention” (Para. [0105]-[0106]).
Regarding Independent Claim 20, Jain discloses:
Jain discloses:
“A system for predictive, diagnostic, and therapeutic applications of wearables for mental health, the system comprising: one or more processors; and one or more computer storage media storing computer-useable instructions that, when used by the one or more processors, cause the one or more processors to perform a method,” (Abstract, “A method may include collecting sensor data related to health of a patient and receiving input that provides quantification of health of the patient;” Para. [0060], “The memory 117 may include computer-readable storage media for collecting or storing data thereon. For example, the memory 117 may include computer-readable storage media that may be tangible or non-transitory computer-readable storage media such as Random Access Memory (RAM), Read-Only Memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Compact Disc Read-Only Memory (CD-ROM) or other optical disk storage, magnetic disk storage or other magnetic storage devices, flash memory devices (e.g., solid state memory devices), or any other tangible and non-transitory storage medium which may be used to store data that may be accessed by a general-purpose or special-purpose computer.”);
“the method comprising: training a predictive … model to predict risk values for mental or neurological disorders based on training data sets,” (Para. [0168]; see also Paras. [0169] and [0170])
“the training data sets being (a) associated with information corresponding to wearable device users” (Para. [0168]; Para. [0048]);
“and (b) comprising one or more of: responses from a validating questionnaire for a type of mental disorder or neurological disorder; sleep information; activity level information; Electronic Medical Record (EMR) data; heart rate information; or validation data;” (Para. [0168]; Para. [0048]);
“storing real-time patient information, corresponding to a user and detected by a sensor of a wearable device, into a particular Electronic Medical Record (EMR) associated with the user,” (Para. [0061]);
“wherein the sensor comprises at least one of: a photoplethysmogram sensor, a skin and ambient temperature sensor, a brain wave sensor, a photodetector, a photodiode, a photoresistor, a phototransistor, a charge-coupled-device, an active pixel sensor, a light sensor, an IR sensor, an electroencephalography sensor, an electromyography sensor, an electrooculography sensor, a heart rate monitor, an electrocardiogram, an electroencephalogram, a pedometer, a thermometer, a transdermal transmitter sensor, one or more front-facing cameras, a camera, a microphone, an accelerometer, a gyroscope, a blood pressure sensor, a pulse oximeter, a respiration rate sensor, a blood alcohol concentration sensor, an accelerometer sensor, a force sensor or a biometric sensor;” (Para. [0123]);
“applying the predictive … model to the real-time patient information, from the particular EMR, to determine a first risk score corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the real- time patient information detected by the sensor of the wearable device;” (Para. [0168]);
“determining whether the first risk score is within a first predetermined range;” (Para. [0184]);
“and retraining the predictive machine learning model based on additional training data sets associated with information corresponding to the wearable device users” (Para. [0307]; see also Paras. [0316] through [0337] detailing Jain’s “physical predictive model”).
Jain does not disclose:
“training a predictive machine learning model to predict risk values for mental or neurological disorders based on training data sets,” (i.e., Jain discloses training a model to predict such risk values as claimed, but does not disclose “wherein the model is a predictive machine learning model”);
“applying the predictive machine learning model to the real-time patient information, from the particular EMR, to determine a first risk score corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the real- time patient information detected by the sensor of the wearable device;” (i.e., Jain discloses applying the model in the manner claimed, but does not disclose “wherein the model is a predictive machine learning model”);
“in response to the first risk score not being within the first predetermined range, automatically providing a first notification, wherein the user is administered a medication based on the first notification;”
“receiving feedback from the wearable device during a defined monitoring period subsequent to the medication being administered to the user based on the first notification, the feedback comprising additional real-time patient information detected by the sensor of the wearable device wherein the feedback indicates physiological or behavioral changes in the user following administration of the medication”
“storing the additional real-time patient information into the particular EMR associated with the user:”
“applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR, to determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the additional real-time patient information detected by the sensor of the wearable device;”
“determining whether the second risk score is within a second predetermined range corresponding to adequate therapeutic response to the medication;”
“in response to the second risk score not being within the second predetermined range corresponding to adequate therapeutic response to the medication and determining that the second risk score is indicative of an inadequate therapeutic response to the medication, automatically scheduling a follow up;”
Shadid describes “Systems and methods are provided for integrated healthcare management. The systems and methods include establishing a medical record data, accessing information about a treatment related to the medical record data, obtaining biometric data related to the treatment, processing the biometric data, determining whether the treatment is appropriate with respect to the biometric data, in response to the treatment being appropriate with respect to the biometric data, administering medication, and in response the treatment not being appropriate with respect to the biometric data, determining a new treatment, and administering new medication based on the new treatment.” (Abstract). Shadid is thus analogous art.
Shadid discloses:
“in response to the first risk score not being within the first predetermined range, automatically providing a first notification, wherein the user is administered a medication based on the first notification;” (Para. [0118]);
“receiving feedback from the wearable device … subsequent to the medication being administered to the user based on the first notification, the feedback comprising additional real-time patient information detected by the sensor of the wearable device wherein the feedback indicates physiological or behavioral changes in the user following administration of the medication;” (Para. [0084]);
This limitation is being interpreted similarly to the similar limitation of Claim 1,which explanation is above.
“storing the additional real-time patient information into the particular EMR associated with the user:” (Para. [0090]);
“applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR, to determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the additional real-time patient information detected by the sensor of the wearable device;” (Para. [0084]);
“determining whether the second risk score is within a second predetermined range corresponding to adequate therapeutic response to the medication;” (Fig. 14, after “Administer Medication S416,” the process repeats; Para. [0008]; Para. [0019]; Claim 1; Claim 13; see rejection of Claim 1, above);
“in response to the second risk score not being within the second predetermined range corresponding to adequate therapeutic response to the medication and determining that the second risk score is indicative of an inadequate therapeutic response to the medication, automatically scheduling a follow up;” (Para. [0119]; Para. [0008]; Para. [0019]; Claim 1; Claim 13);
This limitation is being interpreted similarly to the similar limitation of Claim 1, which explanation is above
It would have been obvious for a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Jain with the teachings of Shadid (i.e., to modify the method of Jain such that medication is administered, the impact of the administered medication is monitored, and the process is repeated as taught by Shadid) in order to “deliver treatments and medicine directly to patients at the time they need it, as well as to track the medication and the real time biometric data of the patients” (Shadid at Para. [0004]).
The combination of Jain and Shadid does not disclose:
“training a predictive machine learning model to predict risk values for mental or neurological disorders based on training data sets,” (i.e., Jain discloses training a model to predict such risk values as claimed, but does not disclose “wherein the model is a predictive machine learning model”);
“applying the predictive machine learning model to the real-time patient information, from the particular EMR, to determine a first risk score corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the real- time patient information detected by the sensor of the wearable device;” (i.e., Jain discloses applying the model in the manner claimed, but does not disclose “wherein the model is a predictive machine learning model”);
“applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR, to determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the additional real-time patient information detected by the sensor of the wearable device;” (i.e., Shadid discloses applying the model in the manner claimed, but does not disclose “wherein the model is a predictive machine learning model”);
receiving feedback “during a defined monitoring period” (i.e., Shadid discloses receipt of such feedback as claimed, but Shadid’s feedback is not limited to “a defined monitoring period”)
Thieme describes “Machine Learning in Mental Health: A Systematic Review of the HCI Literature to Support the Development of Effective and Implementable ML Systems” (Title). Thieme is thus analogous art.
Thieme discloses:
“training a predictive machine learning model to predict risk values for mental or neurological disorders based on training data sets,” (i.e., Thieme discloses “wherein the model is a predictive machine learning model”); (Pg. 34:15, Fourth Paragraph, “3.3.1 Source and Scale of Mental Health Data. ML algorithms build mathematical models based on training data to make predictions or decisions without being explicitly programmed [93]. The papers in our corpus are split between those that collect data for this purpose (n = 29) and those that make use of existing data (n = 23). Existing data is provided through previously generated datasets (n = 14, plus 1 hybrid) and health records (n = 6, plus 2 hybrids).”);
“applying the predictive machine learning model to the real-time patient information, from the particular EMR, to determine a first risk score corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the real- time patient information detected by the sensor of the wearable device;” i.e., Thieme discloses “wherein the model is a predictive machine learning model”); (Pg. 34:15, Fourth Paragraph, “3.3.1 Source and Scale of Mental Health Data. ML algorithms build mathematical models based on training data to make predictions or decisions without being explicitly programmed [93]. The papers in our corpus are split between those that collect data for this purpose (n = 29) and those that make use of existing data (n = 23). Existing data is provided through previously generated datasets (n = 14, plus 1 hybrid) and health records (n = 6, plus 2 hybrids).”);
“applying the predictive machine learning model at least to the additional real-time patient information, from the particular EMR, to determine a second risk score (a) corresponding to the type of mental disorder or neurological disorder, and (b) based at least in part on the additional real-time patient information detected by the sensor of the wearable device;” (i.e., Thieme discloses “wherein the model is a predictive machine learning model”); (Pg. 34:15, Fourth Paragraph, “3.3.1 Source and Scale of Mental Health Data. ML algorithms build mathematical models based on training data to make predictions or decisions without being explicitly programmed [93]. The papers in our corpus are split between those that collect data for this purpose (n = 29) and those that make use of existing data (n = 23). Existing data is provided through previously generated datasets (n = 14, plus 1 hybrid) and health records (n = 6, plus 2 hybrids).”);
It would have been obvious for a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of combined Jain and Shadid with the teachings of Thieme (i.e., to use such a predictive machine learning model as taught by Thieme as the predictive model of combined Jain and Shadid) in order to “assist in the detection, diagnosis and treatment of mental health problem,” as “[machine learning] techniques can potentially offer new routes for learning patterns of human behavior; identifying mental health symptoms and risk factors; developing predictions about disease progression; and personalizing and optimizing therapies” (Thieme at Pg. 34:1, Abstract).
Yanowski describes “Alterations in heart rate and blood pressure (BP) … in patients receiving psychiatric medication” (Abstract). Yanowski is reasonably pertinent to the problem faced by the inventor, and is thus analogous art. See MPEP 2141.01(a)(I).
Yanowkski teaches:
receiving feedback “during a defined monitoring period” (i.e., Yanowski teaches “a defined monitoring period”) (Pg. 310, Right Column, Second Paragraph, “The monitor was programmed to obtain readings every 20 min during the day (6:00 am to 10:00 pm) and every 60 min during the night (10:00 pm to 6:00 am), in accordance with standard recommendations for use of the device. … The nurses followed the usual hospital procedures for care of patients on drugs with known potential for side effects.”).
It would have been obvious for a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the device of combined Jain, Shadid and Thieme with the teachings of Yanowski (i.e., to use such a defined time period as taught by Yanowski for receiving feedback in the device of combined Jain, Shadid and Thieme) in order to facilitate “early management of cardiovascular side effects” when administering psychiatric medications (Yanowski at Pg. 314, Left Column, Third Paragraph).
Claims 3, 18 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over previously cited U.S. Patent Publication No. 2019/0180879 A1 to Jain et al. (“Jain”) in view of previously cited U.S. Patent Publication No. 2021/0098093 A1 to Shadid et al. (“Shadid”), previously cited Non-Patent Literature Anja Thieme, Danielle Belgrave, and Gavin Doherty. 2020. Machine Learning in Mental Health: A Systematic Review of the HCI Literature to Support the Development of Effective and Implementable ML Systems. ACM Trans. Comput.-Hum. Interact. 27, 5, Article 34 (October 2020), 53 pages (“Thieme”) and Non-Patent Literature Alexander Yanovski, Reuben E. Kron, Raymond R. Townsend, Virginia Ford, The Clinical Utility of Ambulatory Blood Pressure and Heart Rate Monitoring in Psychiatric Inpatients, American Journal of Hypertension, Volume 11, Issue 3, March 1998, Pages 309–315 (”Yanowski”) as respectively applied to Claims 2 and 12 above, and further in view of previously cited U.S. Patent No. 10,940,286 B2 to Kansgara (“Kansgara”).
Regarding Claim 3, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 2 as explained above.
Jain additionally discloses:
“wherein the sleep information comprises … body posture information” (Para. [0048], “one or more of the sensors 120 and 122 may include … a posture sensor…”).
Jain does not disclose:
“a quantity of sleep without a disturbance for at least two consecutive days.”
Kansgara is in the same field of endeavor, and discloses:
“a quantity of sleep without a disturbance for at least two consecutive days” (Claim 1).
It would have been obvious before the effective filing date of the claimed invention to one having ordinary skill in the art to modify the Mental Health Assessment method as taught by Jain, Shadid, Thieme and Yanowski to comprise a quantity of sleep without a disturbance for at least two consecutive days as taught by Kansagra, since such a modification would provide the predictable result of providing adequate sleep information to assess patient’s mental state.
Regarding Claim 18, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 12 as explained above.
Jain additionally discloses:
“wherein the real-time patient information comprises sleep information” (Para. [0049], “The sensor data may permit quantification of health habits in terms of activity, sleep…”).
The combination of Jain, Shadid, Thieme and Yanowski does not disclose:
“comprising a quantity of sleep without a disturbance for a period of time.”
Kansgara is in the same field of endeavor, and discloses:
“comprising a quantity of sleep without a disturbance for a period of time” (Claim 1).
It would have been obvious before the effective filing date of the claimed invention to one having ordinary skill in the art to modify the Mental Health Assessment method as taught by Jain, Shadid, Thieme and Yanowski to comprise a quantity of sleep without a disturbance for at least two consecutive days as taught by Kansagra, since such a modification would provide the predictable result of providing adequate sleep information to assess patient’s mental state.
Regarding Claim 19, the combination of Jain, Shadid, Thieme. Yanowski and Kangsgara renders obvious the entirety of Claim 18 as explained above
Jain additionally discloses:
“wherein the sleep information comprises a second quantity of sleep without a disturbance for a second period of time that is a longer duration than the period of time” (Para. [0049] – sleep study based on variation of time).
Claims 5 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over previously cited U.S. Patent Publication No. 2019/0180879 A1 to Jain et al. (“Jain”) in view of previously cited U.S. Patent Publication No. 2021/0098093 A1 to Shadid et al. (“Shadid”), previously cited Non-Patent Literature Anja Thieme, Danielle Belgrave, and Gavin Doherty. 2020. Machine Learning in Mental Health: A Systematic Review of the HCI Literature to Support the Development of Effective and Implementable ML Systems. ACM Trans. Comput.-Hum. Interact. 27, 5, Article 34 (October 2020), 53 pages (“Thieme”) and Non-Patent Literature Alexander Yanovski, Reuben E. Kron, Raymond R. Townsend, Virginia Ford, The Clinical Utility of Ambulatory Blood Pressure and Heart Rate Monitoring in Psychiatric Inpatients, American Journal of Hypertension, Volume 11, Issue 3, March 1998, Pages 309–315 (”Yanowski”) as respectively applied to Claims 1 and 12 above, and further in view of previously cited U.S. Patent No. 11,410,777 to Edelson et al. (“Edelson”).
Regarding Claim 5, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 1 as explained above.
Jain additionally discloses:
“wherein the first risk score is further determined using … a stratified analysis” (Para [0075], [0128], statistical analysis, Para [0197] – Stratified analysis).
The combination of Jain, Shadid, Thieme and Yanowski does not disclose:
“multivariate logistic regression.”
Edelson pertains to “Patient risk evaluation” (Title). Edelson is thus analogous art.
Edelson discloses:
“multivariate logistic regression” (Col 24 Lines 15 -Col 26 Line 15).
It would have been obvious before the effective filing date of the claimed invention to one having ordinary skill in the art to modify the Mental Health Assessment method as taught by the combination of Jain, Shadid, Thieme and Yanowski with using multivariate logistic regression as taught by Edelson, since such a modification would provide the predictable result of providing statistical analysis of a risk score to help assess patient health.
Regarding Claim 17, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 12 as explained above.
The combination of Jain, Shadid and Thiem does not disclose:
“wherein the first predetermined range is determined using multivariate logistic regression.”
Edelson pertains to “Patient risk evaluation” (Title). Edelson is thus analogous art.
Edelson discloses:
“wherein the first predetermined range is determined using multivariate logistic regression” (Col 24 Line 15 through Col 26 Line 15).
It would have been obvious before the effective filing date of the claimed invention to one having ordinary skill in the art to modify the Mental Health Assessment method as taught by the combination of Jain, Shadid and Thiem with using multivariate logistic regression as taught by Edelson, since such a modification would provide the predictable result of providing statistical analysis of a risk score to help assess patient health.
Claims 15 is rejected under 35 U.S.C. 103 as being unpatentable over previously cited U.S. Patent Publication No. 2019/0180879 A1 to Jain et al. (“Jain”) in view of previously cited U.S. Patent Publication No. 2021/0098093 A1 to Shadid et al. (“Shadid”), previously cited Non-Patent Literature Anja Thieme, Danielle Belgrave, and Gavin Doherty. 2020. Machine Learning in Mental Health: A Systematic Review of the HCI Literature to Support the Development of Effective and Implementable ML Systems. ACM Trans. Comput.-Hum. Interact. 27, 5, Article 34 (October 2020), 53 pages (“Thieme”) and Non-Patent Literature Alexander Yanovski, Reuben E. Kron, Raymond R. Townsend, Virginia Ford, The Clinical Utility of Ambulatory Blood Pressure and Heart Rate Monitoring in Psychiatric Inpatients, American Journal of Hypertension, Volume 11, Issue 3, March 1998, Pages 309–315 (”Yanowski”) as applied to Claim 12 above, and further in view of previously cited U.S. Patent No. US 11,103,171 B2 to Gao et al. (“Gao”).
Regarding Claim 15, the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 12 as explained above.
Jain additionally discloses:
“further comprising: receiving a response from the user to the validating questionnaire for the type of mental disorder or neurological disorder;” (Para. [0063], “The questionnaire module 104 may be configured to provide one or more health related questionnaires to the patient via user device 124;” see also Paras. [0064] through [0069]);
“determining a plurality of influencing factors corresponding to the type of mental disorder or neurological disorder using wearable device information from the wearable device users and their responses to the validating questionnaire;” (Para. [0040], “In an embodiment, risk prediction can be made for an impact that the lifestyle choices of the patient may have on their chronic diseases. The electronic device may receive patient lifestyle input in response to a healthy lifestyle and personal control questionnaire (HLPCQ). The patient lifestyle input may include data related to dietary health choices, dietary harm avoidance, daily routine, organized physical exercise, and/or social and mental balance of the patient. The electronic device may store a database of statistically significant number of similar and comparable patients and using that it may determine the impact the various lifestyle choices of the patient may have on the chronic diseases of the patient.”);
“determining the first risk score by additionally using at least one of the plurality of influencing factors…” (Para. [0168], “The total health module 116 may be configured to generate a health risk score of the patient. The health risk score may be based on responses by the patient to questionnaires provided by the questionnaire module 104, sensor data collected by the sensors 120 and 122, the SHC score determined by the graded escalation module 110, the CCB score determined by the chronic burden module 108, the LCC score determined by the lifestyle choice module 114, and/or the EHR data included in the EHR database 128. In some embodiments, the health risk score may also be based on the chronic data included in the chronic disease database 126.”);
The combination of Jain, Shadid, Thieme and Yanowski does not disclose:
“determining confidence scores for each of the plurality of influencing factors;”
“and determining the first risk score by additionally using at least one of the plurality of influencing factors having a confidence score above a threshold.”
Gao describes “Systems and methods for screening, diagnosing, and stratifying patients” (Title). Gao is thus analogous art.
Gao discloses:
“determining confidence scores for each of the plurality of influencing factors;” (Claim 1, “calculating a confidence score for each of the generated plurality of rules, the confidence score being representative of a capacity to predict the category label;
“determining the first risk score by additionally using at least one of the plurality of influencing factors having a confidence score above a threshold.” (Claim 1, “A system for evaluating a patient for mental health issues, the system comprising … using a Bayesian Decision List … wherein the Bayesian Decision List was generated by … calculating a confidence score for each of the generated plurality of rules….”).
It would have been obvious for a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the device of combined Jain, Shadid, Thieme and Yanowski with the teachings of Goa (i.e., to include such confidence scores as taught by Gao) in order to facilitate data classification (Gao at Col. 2, Lns. 9-16).
Claim 22 is rejected under 35 U.S.C. 103 as being unpatentable over previously cited U.S. Patent Publication No. 2019/0180879 A1 to Jain et al. (“Jain”) in view of previously cited U.S. Patent Publication No. 2021/0098093 A1 to Shadid et al. (“Shadid”), previously cited Non-Patent Literature Anja Thieme, Danielle Belgrave, and Gavin Doherty. 2020. Machine Learning in Mental Health: A Systematic Review of the HCI Literature to Support the Development of Effective and Implementable ML Systems. ACM Trans. Comput.-Hum. Interact. 27, 5, Article 34 (October 2020), 53 pages (“Thieme”) and Non-Patent Literature Alexander Yanovski, Reuben E. Kron, Raymond R. Townsend, Virginia Ford, The Clinical Utility of Ambulatory Blood Pressure and Heart Rate Monitoring in Psychiatric Inpatients, American Journal of Hypertension, Volume 11, Issue 3, March 1998, Pages 309–315 (”Yanowski”) as applied to Claim 1 above, and further in view of previously cited WO 2019/155071 A1 to Tuytten et al. (“Tuytten”).
Regarding Claim 22, the combination of the combination of Jain, Shadid, Thieme and Yanowski renders obvious the entirety of Claim 1 as explained above.
The combination of Jain, Shadid, Thieme and Yanowski does not disclose:
wherein a predisposition probability value representing a predisposition of the user to the type of mental disorder or neurological disorder is determined using at least information from the wearable device;
and further comprising combining the first risk score with the predisposition probability value to calculate a probability of mental health
Tuytten describes “a method of generating a model M to detect, or predict an outcome, in particular the risk of a health condition in a subject” (Pg. 1, Ln. 9-11). Tuytenn is thus analogous art.
Tuytenn discloses:
wherein a predisposition probability value representing a predisposition of the user to the type of mental disorder or neurological disorder is determined using at least information from the wearable device; (Pg. 5, Ln. 18-23, “The method employs iterative population segregation methodology, to obtain enriched sub populations of subjects having (a predisposition for) an outcome, or /and enriched sub populations of subjects not having (a predisposition for) an outcome. The method comprises the generation of PPV-defined models with high detection rates of subjects with the (predisposition for) the outcome, or/and NPV-defined models with high detections rates of subjects not having (a predisposition for) the outcome;” Pg. 8, Ln. 5-9, “Measurement data for a plurality of diagnostic or prognostic variables are obtained from each of the subjects. The variables may be selected from biometric, life-style and physiological characteristics, and in one particular embodiment, the variables will include biological molecules such as proteins, metabolites or combinations thereof.”);
and further comprising combining the first risk score with the predisposition probability value to calculate a probability of mental health (Claim 1, “…wherein the model M comprises the first model M1 and the second model M2; and outputting a value indicative of a detection, or a prediction of risk, of the health condition in the subject based on said model M.”)
It would have been obvious for a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the device of combined Jain, Shadid, Thieme and Yanowski with the teachings of Tuytten (i.e., to train the predictive machine learning model of combined Jain, Shadid, Thieme and Yanowski to predict predisposition probability values for the type of mental disorder or neurological disorder as taught by Tuytten, and to combine such predisposition probability values with risk score in the manner of Tuyten) in order to create a “more robust and accurate way of detecting, or predict[ing] risk of, an outcome” (Tuytten at Abstract).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CHRISTOPHER J MUTCHLER whose telephone number is (571)272-8012. The examiner can normally be reached M-F 7:00 am - 4:00 pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jennifer McDonald can be reached on 571-270-3061. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/C.J.M./Examiner, Art Unit 3796
/Jennifer Pitrak McDonald/Supervisory Patent Examiner, Art Unit 3796