DETAILED ACTION
Status of Claims
This is a non-final office action on the merits in response to the arguments and/or amendments filed on 19 December 2025 and the request for continued examination filed on 19 December 2025.
Claim(s) 1, 19, and 20 is/are amended.
Claim(s) 1-6 and 8-21 is/are currently pending and have been examined.
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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 19 December 2025 has been entered.
Claim Interpretation Notes
The claims recite generate a gradient boosting machine learning model for the member by assigning a plurality of model features an importance value using Shapley Additive Explanations (SHAP) values. Upon review of the amendments filed 19 December 2025, Examiner noticed that the limitation might be incorrected understood as generating a ML model through the use of SHAP values, e.g., using SHAP values themselves in order to set the coefficients within the model. For clarity of the record, Examiner notes that this is not how the claim is interpreted. SHAP values come from Shapley Additive exPlanations which is a technique used to interpret models (See Lundberg “A Unified Approach to Interpreting Model Predictions”) rather than create models. Because of this background, the generation limitation must be understood to refer to the generation of a model with characterization values. In other words, the limitation requires 1) generating a gradient boosting machine learning model, and 2) assigning a plurality of model features an importance value using Shapley Additive Explanations (SHAP) values, where the model actually produced by the full limitation is the gradient boosting machine learning model with the importance values. This interpretation is consistent with the specification, which does not actually suggest or support generating a ML model through the use of SHAP values.
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 1-6 and 8-21 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claim 19, which is representative of claims 1 and 20, recites: a system for observing member behavior and managing a care gap associated with the member, comprising:
determine a current gap-in-care for the member by: receiving a
determine, for the current gap-in-care for the member, one or more actions for the member that, if followed, are capable of at least partially closing the current gap-in-care within a clinically-defined period of time for the member;
provide, based on the probability of at least partially closing the current gap-in-care, a communication to the member that describes the one or more actions for the member via a communication channel.
The preceding recitation of the claim has had strikethroughs applied to the features beyond one abstract idea to more clearly demonstrate the limitations setting forth that abstract idea. The remaining limitations describe a concept of analyzing information regarding a gap in care and providing a recommendation to close that gap. This concept describes a mental process that a provider should follow to induce a patient to close a gap-in-care, similar to the “mental process that a neurologist should follow when testing a patient for nervous system malfunctions” given in MPEP 2106.04(a)(2)(II)(C) as an example of managing personal behavior in the methods of organizing human activity sub-grouping. As such, these limitations manage personal behavior and thus set forth a method of organizing human activity. Alternatively, the identified concept is analogous to the examples of “observations”, “evaluation”, “judgement”, and “opinion” given in MPEP2106.04(a)(2)(III) and can be practically performed in the human mind. As such, these limitations set forth a mental process.
The claims also recite generate a gradient boosting machine learning model for the member by assigning a plurality of model features an importance value using Shapely Additive Explanations (SHAP) values. The broadest reasonable interpretation of “gradient boosting” refers to a specific class of machine learning algorithms. The specification does not define the term but uses it consistent with the identified BRI. The broadest reasonable interpretation of “using Shapley Additive Explanations” refers to a specific mathematical technique for quantifying variable importance. The specification does not define the term but uses it consistent with the identified BRI. Thus when given their broadest reasonable interpretation in light of the specification, generate a gradient boosting machine learning model for the member by assigning a plurality of model features an importance value using Shapely Additive Explanations (SHAP) values describe mathematical calculations. As such, these limitations set forth a mathematical concept.
While the above limitations set forth concepts that fall within different groupings of abstract ideas, they all set forth abstract ideas. As such, per MPEP 2106.04(II)(B), these concepts are considered together as a single abstract idea for further analysis. Therefore the claims are determined to recite an abstract idea.
MPEP 2106, reflecting the 2019 PEG, directs examiners at Step 2A Prong Two to consider whether the additional elements of the claims integrate a recited abstract idea into a practical application.
Claim 19 recites a system comprising: a processor; and a memory. Claim 20 recites a non-transitory computer-readable medium. These additional elements are recited at an extremely high level of generality, and are interpreted as generic computing devices used to implement the abstract idea. Per MPEP 2106.05(f), implementing an abstract idea on a generic computing device does not integrate an abstract idea into a practical application in Step 2A Prong Two, similar to how the recitation of the computer in the claim in Alice amounted to mere instructions to apply the abstract idea on a generic computer. As such, these additional elements do not integrate the abstract idea into a practical application.
The claims further recite using the machine learning model. This limitations provides nothing more than mere instructions to implement the abstract idea on a generic computer. As such, these additional elements do not integrate the abstract idea into a practical application.
The claims further recite an additional element of receiving electronic records. This additional element does not reflect any improvement to any technology or technical field. Additionally, this additional element does not implement the judicial exception with or using a particular machine. Further, this additional element does not effect a transformation or reduction of a particular article. Finally, this additional element does not apply or use the abstract idea in some other meaningful way. Instead, this addition element only generally links the abstract idea to a technological environment of computing devices. As such, this additional element does not integrate the abstract idea into a practical application.
If the features of generate a gradient boosting machine learning model for the member by assigning a plurality of model features an importance value using Shapely Additive Explanations (SHAP) values were considered as additional elements, these additional elements would only amount to instructions to implement the abstract idea with a generic computing device. As such, this limitation would not integrate the abstract idea into a practical application.
There are no further additional elements. When considered as a combination, the additional elements only generally link the abstract idea to a technological environment of computing devices. As such, the combination of additional elements does not integrate the abstract idea into a practical application. Therefore the claims are determined to be directed to an abstract idea.
At Step 2B of the Mayo/Alice analysis, examiners are to consider whether the additional elements amount to significantly more than the abstract idea.
As previously noted, the claims recite additional elements which may be interpreted as generic computing devices used to implement the abstract idea. However, per MPEP 2106.05(f), implementing an abstract idea on a generic computing does not add significantly more in Step 2B, similar to how the recitation of the computer in the claim in Alice amounted to mere instructions to apply the abstract idea on a generic computer. As such, these additional elements do not amount to significantly more.
As previously noted, the claims recite an additional element of receiving electronic records. Per MPEP 2106.05(d), the courts have recognizing receiving data and electronic recordkeeping as well understood, routine, and conventional computer functions. As such, this additional element, individually and in combination with the prior computing devices, does not amount to significantly more than the abstract idea.
If the feature of generate a gradient boosting machine learning model for the member by assigning a plurality of model features an importance value using Shapely Additive Explanations (SHAP) values were considered as additional elements, Pantel et al. (US 2015/01005224 A1) demonstrates (a conventional gradient-boosted decision trees method. See at least [0025]) that generating a gradient boosting machine learning model was conventional long before the priority date of the claimed invention, and Fryer et al. (Shapley values for feature selection: The good, the bad, and the axioms) demonstrates (“The Shapley value has, over recent years, become a popular method for interpretable feature attribution in fitted ML models … The ML methods that stand out in terms of popularity are SHapley Additive exPlanations (SHAP).” Page 1, “The simple and intuitive nature of the Shapley value axioms encourages an assessment that the Shapley value is an explainable or interpretable approach to computing the importance of features. “ Page 2) that assigning a plurality of model features an importance value using Shapley Additive Explanations (SHAP) values was conventional before the priority date of the claimed invention. As such these additional elements would not amount to significantly more than the abstract idea.
There are no further additional elements. When considered as a combination at Step 2B, the additional elements only generally link the abstract idea to a technological environment of computing devices. As such, the combination of additional elements does not amount to significantly more than the abstract idea. Therefore, when considered individually and as an ordered combination, the additional elements of the independent claims do not amount to significantly more than the judicial exception. Thus the independent claims are not patent eligible.
Dependent claims 2-6, 8-18, and 21, further describe the abstract idea, but the claims continue to recite an abstract idea, albeit a narrowed one. Dependent claims 2-6, 9-18, and 21 do not recite any further additional elements. The previously identified additional elements, individually and as a combination, fail to integrate the narrowed abstract idea into a practical application or amount to significantly more than the narrowed abstract idea for the same reasons articulated above. Dependent claim 8 recites the additional element of a channel comprising email. This additional element, when considered individually and in combination with the previously identified additional elements, only generally links the abstract idea to a technological environment involving computers. As such, this additional element does not integrate the abstract idea into a practical application. Further, Baji et al. (US 5027400) demonstrates (“conventional electronic mail” See at least Column 24, Line 24) that this additional element was a conventional computer function long before the priority date of the claimed invention. As such, this additional element does not amount to significantly more than the abstract idea. Further, this additional element, when considered individually and in combination with the previously identified additional elements, only generally links the abstract idea to a technological environment involving computers. As such, the combination of additional elements does not amount to significantly more. Thus as the dependent claims remain directed to a judicial exception, and as the additional elements of the claims do not amount to significantly more, the dependent claims are not patent eligible.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1-6, 8-13, and 15-21 is/are rejected under 35 U.S.C. 103 as being unpatentable over Reisman et al. (US 2009/0216558 A1) in view of Holub et al. (US 2021/0313063 A1) and Rho et al. (US 2022/0067580 A1).
Regarding Claim 1 and 20: Reisman discloses a method of observing member behavior and managing a care gap associated with the member, the method comprising:
determining a current gap-in-care for the member by: receiving an electronic record associated with the member that describes a current health-related behavior of the member (In steps 200-202, the health care organization 100 collects a wide spectrum of medical care information 114, 122, 124, 128 and aggregates it in the medical database 118 for subsequent analysis. See at least [0046]. Also: Clinical data 114 originates from medical services claims, pharmacy data, as well as from lab results, and includes information associated with the patient-health care provider interaction, including information related to the patient's diagnosis and treatment, medical procedures, drug prescription information, in-patient information and health care provider notes. See at least [0038]).
receiving guideline behavior for the member supported by a professional clinical recommendation (In step 204, the health care organization 100 establishes a set of clinical rules 120 for a plurality of conditions, such as by having an on-site medical professional team continuously review collected health reference information 122, including evidence-based medical literature. See at least [0046]. Also: An on-staff team of medical professionals within the health care organization 100 consults various sources of health reference information 122, including evidence-based preventive health data, to establish and continuously or periodically revise a set of clinical rules 120 that reflect best evidence-based medical standards of care for a plurality of conditions. The clinical rules 120 are stored in the medical database 118. See at least [0039]); and
determining a difference between the current health-related behavior of the member and the guideline behavior for the member, wherein the difference defines, at least in part, the current gap-in-care for the member (In steps 210 and 212, the rules engine module 126 applies the latest evidence-based medical standards of care included within the clinical rules 120 to the patient's actual care, as evidenced from the claims, pharmacy, lab and patient-entered clinical data, to identify at least one instance where the patient's actual care is inconsistent with the expected care embodied by the clinical rules 120. See at least [0046]. Also: the rules engine module 126 instantiates a patient-specific rule processing session 772 (FIG. 18) and generates one or more clinical alerts 104 when the incoming data, as well as previously stored patient data, indicates a deviation from the best evidence-based best medical standards of care in light of the clinical rules 120. See at least [0075]).
determining, for the current gap-in-care for the member, one or more actions for the member that, if followed, are capable of at least partially closing the current gap-in-care within a clinically-defined period of time for the member (when the rules engine module identifies an instance of actual care inconsistent with the established, best evidence-based medical standards of care, the patient is presented with a clinical alert via the PHR. In embodiments, the clinical alerts include notifications to contact the health care provider in order to start or stop a specific medication and/or to undergo a specific examination or test procedure associated with one or more conditions and co-morbidities specific to the patient. See at least [0009]. Also: RTRecommendationList--a list of real-time alerts 104, 106 generated by the rules engine module 126, including an alert number, alert name, instructional text, severity code, creation date, and a completion status indicator (e.g., open, completed, ignore) for each generated alert. See at least [0065]).
providing a communication to the member that describes the one or more actions for the member via a communication channel (in steps 214-216, the rules engine module 126 stores an alert indicator in the patient's 102 medical data file within the medical database 118, including the associated alert detail, and presents the patient with one or more clinical alerts 104 and/or personalized wellness alerts 106 via the appropriate interface of the PHR 108. See at least [0046]. Also: the alerts list 402 includes a number of clinical alerts 104 suggesting specific tests related to patient's diabetes and recommending use of statins (e.g., to lower cholesterol levels). In one embodiment, the list 402 includes one or more personalized wellness alerts 106, such a recommendation to undergo periodic breast cancer screenings for female patients of predetermined age range that have not had a recent screening. See at least [0048]).
Reisman does not expressly disclose generating a gradient boosting machine learning model for the member, or using the machine learning model to determine a probability of at least partially closing the current gap-in-care for each of the one or more actions, or providing, based on the probability of at least partially closing the current gap-in-care, a communication.
However, Holub teaches generating a gradient boosting machine learning model for the member (the process 400 may include generating one or more machine learning models. See at least [0052]. Also: the machine learning models may also comprise, for example, … gradient-boosting machines (GBM). See at least [0041]), and using the machine learning model to determine a probability of at least partially closing the current gap-in-care for each of the one or more actions (At block 420, the process 400 may include generating actions for addressing one or more gaps in care and generating a probability of successfully addressing the one or more gaps in care. See at least [0061]. Also: FIG. 2 also depicts output data 220 that may be generated from the machine learning models 218. The output data 220 may include data regarding gap(s) in care (e.g., colonoscopy, mammogram, and/or untreated diabetes), as well as a probability. See at least [0042]), and providing, based on the probability of at least partially closing the current gap-in-care, a communication (sending the notification comprises sending the notification based at least in part on the probability satisfying a threshold probability for successfully addressing the at least one of the haps in medical care or the one or more actions. See at least [0023]).
Reisman provides a system which identifies gaps in care, identifies actions to address a gap in care, and provides recommendations to perform identified actions, upon which the claimed invention’s determination of a probability of an action to address a gap in care can be seen as an improvement. However, Holub demonstrates that the prior art already knew of identifying probabilities of an action addressing a gap in care and using that probability to determine whether to send a notification. One of ordinary skill in the art could have trivially applied the techniques of Holub to the system of Reisman. Further, one of ordinary skill in the art would have recognized that the application of Holub would have resulted in an improved system which would limit recommended actions to those sufficiently likely to address gaps in care. As such, the application of Holub would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention in view of the disclosures of Reisman and the teachings of Holub.
Additionally, Reisman does not disclose assigning a plurality of model features an importance value using Shapley Additive Explanations.
Rho teaches generating a gradient boosting machine learning model by assigning a plurality of model features an importance value using Shapley Additive Explanations (in some examples, executed analytical engine 156 may also perform operations that generate, for each of the analyst-specified features, a feature value contribution that characterizes a contribution of the analyst-specified features to the outcome of the selected machine learning or artificial intelligence process. For example, executed analytical engine 156 may compute a Shapley value feature contribution for each of the analyst-specified features based on the elements of predicted output data 324A, 324B, and 324C and corresponding ones of the modified feature vectors 314. In some instances, executed analytical engine 156 may calculate one or more of the Shapley value feature contributions in accordance with a Shapley Additive exPlanations (SNAP) algorithm (e.g., when the selected machine learning or artificial intelligence process corresponds to a gradient-boosted decision tree algorithm). See at least [0100]).
Reisman and Holub suggests a systems which uses a gradient boosted machine learning model, upon which the claimed invention’s use of Shapley Additive Explanations to generate an interpreted model can be seen as an improvement. However, Rho demonstrates that the prior art already knew of applying SHAP values to gradient boosted machine learning models to generate interpreted models. One of ordinary skill in the art could have easily applied the SHAP techniques of Rho to generating an interpreted model of Reisman and Holub, and one of ordinary skill in the art would have recognized that such an application of Rho would have resulted in a system using a more trustworthy machine learning model. As such, the application of Rho, and the claimed invention, would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention in view of the disclosures of Reisman and the teachings of Holub and Rho.
Regarding Claim 2 and 21: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses determining an impact associated with at least partially closing the current gap-in-care, wherein the impact comprises a clinical impact (Step 212 further includes identifying whether the patient 102 should be notified about newly available evidence-based standards of preventive health care 122 via a personalized wellness alert, such as when the preventive health care information is beneficial to the patient's actual care (e.g., notifications regarding the benefits of breast cancer screening). If the rules engine module 126 does not detect a discrepancy between the actual care given by the caregiver and the best evidence-based medical standards of care, or when the newly received health reference is not beneficial (e.g., cumulative in light of existing information), the method returns to step 200. See at least [0046]. Also: During processing, the rules engine module 126 records alert justification information in the medical database 118. In one embodiment, the alert justification information specifies which rules have been triggered/processed by the incoming data (e.g., by rule number), which alerts have been generated (e.g., by alert number), a time/date stamp for each alert 104, 106, the specific exclusionary and inclusionary information for a given patient that caused the rule to trigger (e.g., known drug allergies are used to exclude alerts recommending a drug regimen that may cause an allergic reaction), as well as the patient-entered and claim information associated with the incoming real-time data that triggered a given rule. See at least [0058]).
Regarding Claim 3: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses wherein the clinical impact is measured by a health biomarker (Specifically, the rules engine module 126 processes the newly-received data point 744 in light of the previously stored health profile (e.g., prior health indicator readings, patient's chronic conditions, age, and sex) and the best evidence-based medical standards of care 120 to generate in real-time a normal or target range 748, as well as a high risk indicator 750, which provide context for the updated readings. For health indicators, such as blood pressure, which need to stay within a given target range 748, the high risk indicator 750 is demarcated via a high range and a low range. In addition to providing the target range and the health risk indicator, the rules engine provides specific messaging to the member to alert them if the health indicator like blood pressure is critically high to seek urgent medical care. In embodiments, the health indicator includes cholesterol levels, blood pressure readings, HbA1c test results, and body mass index (BMI) readings. See at least [0080]).
Regarding Claim 4: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses wherein the health biomarker comprises at least one of HbAlc, blood pressure, and health complications (Specifically, the rules engine module 126 processes the newly-received data point 744 in light of the previously stored health profile (e.g., prior health indicator readings, patient's chronic conditions, age, and sex) and the best evidence-based medical standards of care 120 to generate in real-time a normal or target range 748, as well as a high risk indicator 750, which provide context for the updated readings. For health indicators, such as blood pressure, which need to stay within a given target range 748, the high risk indicator 750 is demarcated via a high range and a low range. In addition to providing the target range and the health risk indicator, the rules engine provides specific messaging to the member to alert them if the health indicator like blood pressure is critically high to seek urgent medical care. In embodiments, the health indicator includes cholesterol levels, blood pressure readings, HbA1c test results, and body mass index (BMI) readings. See at least [0080]).
Regarding Claim 5: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses wherein the health complications comprise at least one of a stroke, myocardial infraction, in-member admission, and emergency room admission (the rules engine module 126 applies clinical data 114 and clinical components of the patient-entered data 128 to generate a real-time risk score 105 for various medical conditions (e.g., points are assigned to various clinical factors that increase the risk for heart disease and based on the member's conditions and lifestyle behaviors, a percentage score is calculated to identify the member's risk for future heart disease). The risk score 105 quantifies the severity of existing medical conditions and assesses the risk for future conditions in light of evaluating multiple risk factors in accordance with the clinical rules 120. For example, the risk score 105 may identify high risk diabetics or patients subject to a risk of future stroke. See at least [0067]. Also: The rules engine module 126 identifies relevant health reference information 122 and medical news 124 based on a real-time analysis of the clinical data 114, patient-entered data 128, risk score 105. See at least [0077]).
Regarding Claim 6: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses determining an impact associated with at least partially closing the current gap-in-care, wherein the impact comprises a cost impact (Step 212 further includes identifying whether the patient 102 should be notified about newly available evidence-based standards of preventive health care 122 via a personalized wellness alert, such as when the preventive health care information is beneficial to the patient's actual care (e.g., notifications regarding the benefits of breast cancer screening). If the rules engine module 126 does not detect a discrepancy between the actual care given by the caregiver and the best evidence-based medical standards of care, or when the newly received health reference is not beneficial (e.g., cumulative in light of existing information), the method returns to step 200. See at least [0046]. Also: During processing, the rules engine module 126 records alert justification information in the medical database 118. In one embodiment, the alert justification information specifies which rules have been triggered/processed by the incoming data (e.g., by rule number), which alerts have been generated (e.g., by alert number), a time/date stamp for each alert 104, 106, the specific exclusionary and inclusionary information for a given patient that caused the rule to trigger (e.g., known drug allergies are used to exclude alerts recommending a drug regimen that may cause an allergic reaction), as well as the patient-entered and claim information associated with the incoming real-time data that triggered a given rule. See at least [0058]).
Regarding Claim 8: Reisman in view of Holub and Rho makes obvious the above limitations. Reisman separately discloses a communication channel comprises at least one of email, direct mail, SMS, and an automated outbound calling campaign (via phone, mail, email or other communications. See at least [0044]).
Reisman, Holub, and Rho suggests a systems which provides a communication to a user regarding a gap-in-care, which differs from the claimed invention by the substitution of an unstated communication channel for one of particular channels. However, Reisman separate demonstrates that those particular channels were already known by the prior art. One of ordinary skill in the art could have trivially substituted email into the system of Reisman, Holub, and Rho to provide a notification. Further, one of ordinary skill in the art would have recognized that such a substitution would have predictably resulted in a system which contacts patients regarding gaps in car via email. As such, the identified substitution and the claimed invention would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention in view of the disclosures of Reisman and the teachings of Holub and Rho.
Regarding Claim 9: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses wherein the electronic record associated with the member comprises claims-based electronic data (The medical care information collected by the health care organization comprises patient-specific clinical data (e.g., based on claims, health care provider, and patient-entered input), as well as health reference information, including evidence-based literature relating to a plurality of medical conditions. See at least [0004]. Also: A medical insurance carrier collects clinical information originating from medical services claims, performed procedures, pharmacy data, lab results, and provides it to the health care organization for storage in a medical database. See at least [0005]).
Regarding Claim 10: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses wherein the electronic record further comprises electronic medical record (EMR) data (In steps 200-202, the health care organization 100 collects a wide spectrum of medical care information 114, 122, 124, 128 and aggregates it in the medical database 118 for subsequent analysis. See at least [0046]. Also: Clinical data 114 originates from medical services claims, pharmacy data, as well as from lab results, and includes information associated with the patient-health care provider interaction, including information related to the patient's diagnosis and treatment, medical procedures, drug prescription information, in-patient information and health care provider notes. See at least [0038]).
Regarding Claim 11: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses wherein the claims-based electronic data comprises data describing at least one insurance medical and/or insurance claim made by at least one of the member and a provider (The medical care information collected by the health care organization comprises patient-specific clinical data (e.g., based on claims, health care provider, and patient-entered input), as well as health reference information, including evidence-based literature relating to a plurality of medical conditions. See at least [0004]. Also: A medical insurance carrier collects clinical information originating from medical services claims, performed procedures, pharmacy data, lab results, and provides it to the health care organization for storage in a medical database. See at least [0005]).
Regarding Claim 12: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses wherein the electronic record associated with the member comprises device data obtained from at least one device associated with the member (Exemplary patient-entered data 128 includes additional clinical data, such as patient's family history, use of non-prescription drugs, known allergies, unreported and/or untreated conditions (e.g., chronic low back pain, migraines, etc.), as well as results of self-administered medical tests (e.g., periodic blood pressure and/or blood sugar readings). See at least [0040]).
Regarding Claim 13: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses wherein the device data comprises at least one of gyroscopic data, accelerometer data, beacon data, glucose readings, heart rate data, blood pressure data, blood oxygen data, temperature data, kinetics data, location data, motion data, a device identifier, and a timestamp (Exemplary patient-entered data 128 includes additional clinical data, such as patient's family history, use of non-prescription drugs, known allergies, unreported and/or untreated conditions (e.g., chronic low back pain, migraines, etc.), as well as results of self-administered medical tests (e.g., periodic blood pressure and/or blood sugar readings). See at least [0040]).
Regarding Claim 15: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses wherein the guideline behavior for the member supported by the professional clinical recommendation comprises guidance based on at least one of medical history, demographics, social indices, biomarkers, behavior data, engagement data, historical gap-in-care data, and a machine learning model-derived output (In step 204, the health care organization 100 establishes a set of clinical rules 120 for a plurality of conditions, such as by having an on-site medical professional team continuously review collected health reference information 122, including evidence-based medical literature. See at least [0046]. Also: An on-staff team of medical professionals within the health care organization 100 consults various sources of health reference information 122, including evidence-based preventive health data, to establish and continuously or periodically revise a set of clinical rules 120 that reflect best evidence-based medical standards of care for a plurality of conditions. The clinical rules 120 are stored in the medical database 118. See at least [0039]. Also: based on a history of a heart attack and the patient's drug regimen compliance information (e.g., as entered by a health care provider), the rules engine module 126 presents relevant drug-related educational materials 122, 124 relating to the importance of taking medications for heart attacks. See at least [0077]).
Regarding Claim 16: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses determining an impact associated with at least partially closing the current gap-in-care (Step 212 further includes identifying whether the patient 102 should be notified about newly available evidence-based standards of preventive health care 122 via a personalized wellness alert, such as when the preventive health care information is beneficial to the patient's actual care (e.g., notifications regarding the benefits of breast cancer screening). If the rules engine module 126 does not detect a discrepancy between the actual care given by the caregiver and the best evidence-based medical standards of care, or when the newly received health reference is not beneficial (e.g., cumulative in light of existing information), the method returns to step 200. See at least [0046]. Also: During processing, the rules engine module 126 records alert justification information in the medical database 118. In one embodiment, the alert justification information specifies which rules have been triggered/processed by the incoming data (e.g., by rule number), which alerts have been generated (e.g., by alert number), a time/date stamp for each alert 104, 106, the specific exclusionary and inclusionary information for a given patient that caused the rule to trigger (e.g., known drug allergies are used to exclude alerts recommending a drug regimen that may cause an allergic reaction), as well as the patient-entered and claim information associated with the incoming real-time data that triggered a given rule. See at least [0058]) and adjusting the determined impact by a baseline biomarker of the member (Specifically, the rules engine module 126 processes the newly-received data point 744 in light of the previously stored health profile (e.g., prior health indicator readings, patient's chronic conditions, age, and sex) and the best evidence-based medical standards of care 120 to generate in real-time a normal or target range 748, as well as a high risk indicator 750, which provide context for the updated readings. For health indicators, such as blood pressure, which need to stay within a given target range 748, the high risk indicator 750 is demarcated via a high range and a low range. In addition to providing the target range and the health risk indicator, the rules engine provides specific messaging to the member to alert them if the health indicator like blood pressure is critically high to seek urgent medical care. In embodiments, the health indicator includes cholesterol levels, blood pressure readings, HbA1c test results, and body mass index (BMI) readings. See at least [0080]).
Regarding Claim 17: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses determining an impact associated with at least partially closing the current gap-in-care (Step 212 further includes identifying whether the patient 102 should be notified about newly available evidence-based standards of preventive health care 122 via a personalized wellness alert, such as when the preventive health care information is beneficial to the patient's actual care (e.g., notifications regarding the benefits of breast cancer screening). If the rules engine module 126 does not detect a discrepancy between the actual care given by the caregiver and the best evidence-based medical standards of care, or when the newly received health reference is not beneficial (e.g., cumulative in light of existing information), the method returns to step 200. See at least [0046]. Also: During processing, the rules engine module 126 records alert justification information in the medical database 118. In one embodiment, the alert justification information specifies which rules have been triggered/processed by the incoming data (e.g., by rule number), which alerts have been generated (e.g., by alert number), a time/date stamp for each alert 104, 106, the specific exclusionary and inclusionary information for a given patient that caused the rule to trigger (e.g., known drug allergies are used to exclude alerts recommending a drug regimen that may cause an allergic reaction), as well as the patient-entered and claim information associated with the incoming real-time data that triggered a given rule. See at least [0058]) and adjusting the determined impact by a degree of member management of the member’s condition (Specifically, the rules engine module 126 processes the newly-received data point 744 in light of the previously stored health profile (e.g., prior health indicator readings, patient's chronic conditions, age, and sex) and the best evidence-based medical standards of care 120 to generate in real-time a normal or target range 748, as well as a high risk indicator 750, which provide context for the updated readings. For health indicators, such as blood pressure, which need to stay within a given target range 748, the high risk indicator 750 is demarcated via a high range and a low range. In addition to providing the target range and the health risk indicator, the rules engine provides specific messaging to the member to alert them if the health indicator like blood pressure is critically high to seek urgent medical care. In embodiments, the health indicator includes cholesterol levels, blood pressure readings, HbA1c test results, and body mass index (BMI) readings. See at least [0080]).
Regarding Claim 18: Reisman in view of Holub and Rho makes obvious the above limitations. Additionally, Reisman discloses determining an impact associated with at least partially closing the current gap-in-care (Step 212 further includes identifying whether the patient 102 should be notified about newly available evidence-based standards of preventive health care 122 via a personalized wellness alert, such as when the preventive health care information is beneficial to the patient's actual care (e.g., notifications regarding the benefits of breast cancer screening). If the rules engine module 126 does not detect a discrepancy between the actual care given by the caregiver and the best evidence-based medical standards of care, or when the newly received health reference is not beneficial (e.g., cumulative in light of existing information), the method returns to step 200. See at least [0046]. Also: During processing, the rules engine module 126 records alert justification information in the medical database 118. In one embodiment, the alert justification information specifies which rules have been triggered/processed by the incoming data (e.g., by rule number), which alerts have been generated (e.g., by alert number), a time/date stamp for each alert 104, 106, the specific exclusionary and inclusionary information for a given patient that caused the rule to trigger (e.g., known drug allergies are used to exclude alerts recommending a drug regimen that may cause an allergic reaction), as well as the patient-entered and claim information associated with the incoming real-time data that triggered a given rule. See at least [0058]), wherein the impact associated with at least partially closing the current gap-in-care for the member comprises at least partially closing one or more additional current gaps-in-care for the member (when the rules engine module identifies an instance of actual care inconsistent with the established, best evidence-based medical standards of care, the patient is presented with a clinical alert via the PHR. In embodiments, the clinical alerts include notifications to contact the health care provider in order to start or stop a specific medication and/or to undergo a specific examination or test procedure associated with one or more conditions and co-morbidities specific to the patient. See at least [0009]. Also: RTRecommendationList--a list of real-time alerts 104, 106 generated by the rules engine module 126, including an alert number, alert name, instructional text, severity code, creation date, and a completion status indicator (e.g., open, completed, ignore) for each generated alert. See at least [0065]).
Regarding Claim 19: Reisman discloses a system for observing member behavior and managing a care gap associated with the member, comprising:
a processor; and a memory coupled with the processor (See at least [0056]), wherein the memory stores data that, when executed by the processor, enables the processor to: determine a current gap-in-care for the member by:
receiving an electronic record associated with the member that describes a current health-related behavior of the member (In steps 200-202, the health care organization 100 collects a wide spectrum of medical care information 114, 122, 124, 128 and aggregates it in the medical database 118 for subsequent analysis. See at least [0046]. Also: Clinical data 114 originates from medical services claims, pharmacy data, as well as from lab results, and includes information associated with the patient-health care provider interaction, including information related to the patient's diagnosis and treatment, medical procedures, drug prescription information, in-patient information and health care provider notes. See at least [0038]).
receiving guideline behavior for the member supported by a professional clinical recommendation (In step 204, the health care organization 100 establishes a set of clinical rules 120 for a plurality of conditions, such as by having an on-site medical professional team continuously review collected health reference information 122, including evidence-based medical literature. See at least [0046]. Also: An on-staff team of medical professionals within the health care organization 100 consults various sources of health reference information 122, including evidence-based preventive health data, to establish and continuously or periodically revise a set of clinical rules 120 that reflect best evidence-based medical standards of care for a plurality of conditions. The clinical rules 120 are stored in the medical database 118. See at least [0039]); and
determining a difference between the current health-related behavior of the member and the guideline behavior for the member, wherein the difference defines, at least in part, the current gap-in-care for the member (In steps 210 and 212, the rules engine module 126 applies the latest evidence-based medical standards of care included within the clinical rules 120 to the patient's actual care, as evidenced from the claims, pharmacy, lab and patient-entered clinical data, to identify at least one instance where the patient's actual care is inconsistent with the expected care embodied by the clinical rules 120. See at least [0046]. Also: the rules engine module 126 instantiates a patient-specific rule processing session 772 (FIG. 18) and generates one or more clinical alerts 104 when the incoming data, as well as previously stored patient data, indicates a deviation from the best evidence-based best medical standards of care in light of the clinical rules 120. See at least [0075]).
determine, for the current gap-in-care for the member, one or more actions for the member that, if followed, are capable of at least partially closing the current gap-in-care within a clinically-defined period of time for the member (when the rules engine module identifies an instance of actual care inconsistent with the established, best evidence-based medical standards of care, the patient is presented with a clinical alert via the PHR. In embodiments, the clinical alerts include notifications to contact the health care provider in order to start or stop a specific medication and/or to undergo a specific examination or test procedure associated with one or more conditions and co-morbidities specific to the patient. See at least [0009]. Also: RTRecommendationList--a list of real-time alerts 104, 106 generated by the rules engine module 126, including an alert number, alert name, instructional text, severity code, creation date, and a completion status indicator (e.g., open, completed, ignore) for each generated alert. See at least [0065]).
provide a communication to the member that describes the one or more actions for the member via a communication channel (in steps 214-216, the rules engine module 126 stores an alert indicator in the patient's 102 medical data file within the medical database 118, including the associated alert detail, and presents the patient with one or more clinical alerts 104 and/or personalized wellness alerts 106 via the appropriate interface of the PHR 108. See at least [0046]. Also: the alerts list 402 includes a number of clinical alerts 104 suggesting specific tests related to patient's diabetes and recommending use of statins (e.g., to lower cholesterol levels). In one embodiment, the list 402 includes one or more personalized wellness alerts 106, such a recommendation to undergo periodic breast cancer screenings for female patients of predetermined age range that have not had a recent screening. See at least [0048]).
Reisman does not expressly disclose generate a gradient boosting machine learning model for the member, or use the machine learning model to determine a probability of at least partially closing the current gap-in-care for each of the one or more actions, or prove, based on the probability of at least partially closing the current gap-in-care, a communication.
However, Holub teaches generate a gradient boosting machine learning model for the member (the process 400 may include generating one or more machine learning models. See at least [0052]. Also: the machine learning models may also comprise, for example, … gradient-boosting machines (GBM). See at least [0041]), and use the machine learning model to determine a probability of at least partially closing the current gap-in-care for each of the one or more actions (At block 420, the process 400 may include generating actions for addressing one or more gaps in care and generating a probability of successfully addressing the one or more gaps in care. See at least [0061]. Also: FIG. 2 also depicts output data 220 that may be generated from the machine learning models 218. The output data 220 may include data regarding gap(s) in care (e.g., colonoscopy, mammogram, and/or untreated diabetes), as well as a probability. See at least [0042]), and provide, based on the probability of at least partially closing the current gap-in-care, a communication (sending the notification comprises sending the notification based at least in part on the probability satisfying a threshold probability for successfully addressing the at least one of the haps in medical care or the one or more actions. See at least [0023]).
Reisman provides a system which identifies gaps in care, identifies actions to address a gap in care, and provides recommendations to perform identified actions, upon which the claimed invention’s determination of a probability of an action to address a gap in care can be seen as an improvement. However, Holub demonstrates that the prior art already knew of identifying probabilities of an action addressing a gap in care and using that probability to determine whether to send a notification. One of ordinary skill in the art could have trivially applied the techniques of Holub to the system of Reisman. Further, one of ordinary skill in the art would have recognized that the application of Holub would have resulted in an improved system which would limit recommended actions to those sufficiently likely to address gaps in care. As such, the application of Holub would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention in view of the disclosures of Reisman and the teachings of Holub.
Additionally, Reisman does not disclose assigning a plurality of model features an importance value using Shapley Additive Explanations.
Rho teaches generating a gradient boosting machine learning model by assigning a plurality of model features an importance value using Shapley Additive Explanations (in some examples, executed analytical engine 156 may also perform operations that generate, for each of the analyst-specified features, a feature value contribution that characterizes a contribution of the analyst-specified features to the outcome of the selected machine learning or artificial intelligence process. For example, executed analytical engine 156 may compute a Shapley value feature contribution for each of the analyst-specified features based on the elements of predicted output data 324A, 324B, and 324C and corresponding ones of the modified feature vectors 314. In some instances, executed analytical engine 156 may calculate one or more of the Shapley value feature contributions in accordance with a Shapley Additive exPlanations (SNAP) algorithm (e.g., when the selected machine learning or artificial intelligence process corresponds to a gradient-boosted decision tree algorithm). See at least [0100]).
Reisman and Holub suggests a systems which uses a gradient boosted machine learning model, upon which the claimed invention’s use of Shapley Additive Explanations to generate an interpreted model can be seen as an improvement. However, Rho demonstrates that the prior art already knew of applying SHAP values to gradient boosted machine learning models to generate interpreted models. One of ordinary skill in the art could have easily applied the SHAP techniques of Rho to generating an interpreted model of Reisman and Holub, and one of ordinary skill in the art would have recognized that such an application of Rho would have resulted in a system using a more trustworthy machine learning model. As such, the application of Rho, and the claimed invention, would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention in view of the disclosures of Reisman and the teachings of Holub and Rho.
Claim(s) 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Reisman et al. (US 2009/0216558 A1) in view of Holub et al. (US 2021/0313063 A1) and Rho et al. (US 2022/0067580 A1), and further in view of Irish et al. (US 2020/0281542 A1).
Regarding Claim 14: Reisman in view of Holub and Rho makes obvious the above limitations. Reisman does not appear to disclose wherein the electronic record comprises an image of the member.
However, Irish teaches an electronic record comprises an image of the member (transmitting, by the imaging device, the captured image to the user interface system in response to generating the captured image, wherein the captured image is medically related to a patient; automatically associating, by the user interface system, the captured image to the patient upon receipt; and storing, by the user interface system, the received captured image in an electronic medical record (EMR) of the patient associated with the captured image. See at least [0006]).
Reisman, Holub, and Rho suggest a system which delivers health recommendations to users based on their electronic medical records, which differs from the claimed invention by the substitution of Reisman’s generic electronic medical record for a medical record containing an image of the patient. Irish demonstrates that the prior art already knew of electronic medical records containing images of a patient. One of ordinary skill in the art could have trivially substituted Irish’s EMR in for the EMR of Reisman, Holub, and Rho. Further, one of ordinary skill in the art would have recognized that such a substitution would have predictably resulted in a system which would analyze EMRs including patient images to determine health recommendations. As such, the identified substitution, and the claimed invention, would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention in view of the disclosures of Reisman and the teachings of Holub, Rho, and Irish.
Alternative Claim Rejections - 35 USC § 103
Claim(s) 1 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Reisman et al. (US 2009/0216558 A1) in view of Holub et al. (US 2021/0313063 A1) and Driouich (Building a flight price analysis model with machine learning).
Regarding Claim 1 and 20: Reisman discloses a method of observing member behavior and managing a care gap associated with the member, the method comprising:
determining a current gap-in-care for the member by: receiving an electronic record associated with the member that describes a current health-related behavior of the member (In steps 200-202, the health care organization 100 collects a wide spectrum of medical care information 114, 122, 124, 128 and aggregates it in the medical database 118 for subsequent analysis. See at least [0046]. Also: Clinical data 114 originates from medical services claims, pharmacy data, as well as from lab results, and includes information associated with the patient-health care provider interaction, including information related to the patient's diagnosis and treatment, medical procedures, drug prescription information, in-patient information and health care provider notes. See at least [0038]).
receiving guideline behavior for the member supported by a professional clinical recommendation (In step 204, the health care organization 100 establishes a set of clinical rules 120 for a plurality of conditions, such as by having an on-site medical professional team continuously review collected health reference information 122, including evidence-based medical literature. See at least [0046]. Also: An on-staff team of medical professionals within the health care organization 100 consults various sources of health reference information 122, including evidence-based preventive health data, to establish and continuously or periodically revise a set of clinical rules 120 that reflect best evidence-based medical standards of care for a plurality of conditions. The clinical rules 120 are stored in the medical database 118. See at least [0039]); and
determining a difference between the current health-related behavior of the member and the guideline behavior for the member, wherein the difference defines, at least in part, the current gap-in-care for the member (In steps 210 and 212, the rules engine module 126 applies the latest evidence-based medical standards of care included within the clinical rules 120 to the patient's actual care, as evidenced from the claims, pharmacy, lab and patient-entered clinical data, to identify at least one instance where the patient's actual care is inconsistent with the expected care embodied by the clinical rules 120. See at least [0046]. Also: the rules engine module 126 instantiates a patient-specific rule processing session 772 (FIG. 18) and generates one or more clinical alerts 104 when the incoming data, as well as previously stored patient data, indicates a deviation from the best evidence-based best medical standards of care in light of the clinical rules 120. See at least [0075]).
determining, for the current gap-in-care for the member, one or more actions for the member that, if followed, are capable of at least partially closing the current gap-in-care within a clinically-defined period of time for the member (when the rules engine module identifies an instance of actual care inconsistent with the established, best evidence-based medical standards of care, the patient is presented with a clinical alert via the PHR. In embodiments, the clinical alerts include notifications to contact the health care provider in order to start or stop a specific medication and/or to undergo a specific examination or test procedure associated with one or more conditions and co-morbidities specific to the patient. See at least [0009]. Also: RTRecommendationList--a list of real-time alerts 104, 106 generated by the rules engine module 126, including an alert number, alert name, instructional text, severity code, creation date, and a completion status indicator (e.g., open, completed, ignore) for each generated alert. See at least [0065]).
providing a communication to the member that describes the one or more actions for the member via a communication channel (in steps 214-216, the rules engine module 126 stores an alert indicator in the patient's 102 medical data file within the medical database 118, including the associated alert detail, and presents the patient with one or more clinical alerts 104 and/or personalized wellness alerts 106 via the appropriate interface of the PHR 108. See at least [0046]. Also: the alerts list 402 includes a number of clinical alerts 104 suggesting specific tests related to patient's diabetes and recommending use of statins (e.g., to lower cholesterol levels). In one embodiment, the list 402 includes one or more personalized wellness alerts 106, such a recommendation to undergo periodic breast cancer screenings for female patients of predetermined age range that have not had a recent screening. See at least [0048]).
Reisman does not expressly disclose generating a gradient boosting machine learning model for the member, or using the machine learning model to determine a probability of at least partially closing the current gap-in-care for each of the one or more actions, or providing, based on the probability of at least partially closing the current gap-in-care, a communication.
However, Holub teaches generating a gradient boosting machine learning model for the member (the process 400 may include generating one or more machine learning models. See at least [0052]. Also: the machine learning models may also comprise, for example, … gradient-boosting machines (GBM). See at least [0041]), and using the machine learning model to determine a probability of at least partially closing the current gap-in-care for each of the one or more actions (At block 420, the process 400 may include generating actions for addressing one or more gaps in care and generating a probability of successfully addressing the one or more gaps in care. See at least [0061]. Also: FIG. 2 also depicts output data 220 that may be generated from the machine learning models 218. The output data 220 may include data regarding gap(s) in care (e.g., colonoscopy, mammogram, and/or untreated diabetes), as well as a probability. See at least [0042]), and providing, based on the probability of at least partially closing the current gap-in-care, a communication (sending the notification comprises sending the notification based at least in part on the probability satisfying a threshold probability for successfully addressing the at least one of the haps in medical care or the one or more actions. See at least [0023]).
Reisman provides a system which identifies gaps in care, identifies actions to address a gap in care, and provides recommendations to perform identified actions, upon which the claimed invention’s determination of a probability of an action to address a gap in care can be seen as an improvement. However, Holub demonstrates that the prior art already knew of identifying probabilities of an action addressing a gap in care and using that probability to determine whether to send a notification. One of ordinary skill in the art could have trivially applied the techniques of Holub to the system of Reisman. Further, one of ordinary skill in the art would have recognized that the application of Holub would have resulted in an improved system which would limit recommended actions to those sufficiently likely to address gaps in care. As such, the application of Holub would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention in view of the disclosures of Reisman and the teachings of Holub.
Additionally, Reisman does not disclose assigning a plurality of model features an importance value using Shapley Additive Explanations.
Driouich teaches generating a machine learning model by assigning a plurality of model features an importance value using Shapley Additive Explanations (SHAP) values (“One important aspect of any machine learning project is explainability. A first glance at model performances only gives us a global idea about the predictions, but thanks to state-of-the-art Explainable AI (XAI) techniques, we can interpret the model globally but also take a closer look at data points where the model did or did not perform well. For this purpose, we used the SHAP XAI library to deep dive into our models’ predictions. SHAP (SHapley Additive exPlanations) is ‘a game theoretic approach to explain the output of any machine learning model. It connects optimal credit allocation with local explanations using the classic Shapley values from game theory and their related extensions.’ Concretely, SHAP is a XAI tool used to explain the contribution of each feature value to the forecast. For instance, the feature value round_trip = 1 increased the predicted price from $250 to $296: … As highlighted previously, we could explain why the model was predicting a given price on worst-case scenario forecasts and adjust our training set accordingly to decrease overfitting effect or tweak our algorithm’s hyperparameters to better deal with these cases.” See at least Page 11).
Reisman and Holub suggests a systems which uses a gradient boosted machine learning model, upon which the claimed invention’s use of Shapley Additive Explanations to interpret the model can be seen as an improvement. However, Driouich demonstrates that the prior art already knew of improving gradient boosted machine learning models by applying SHAP values to evaluate and interpret generated models. One of ordinary skill in the art could have easily applied the SHAP techniques of Driouich to generating the model of Reisman and Holub, and one of ordinary skill in the art would have recognized that such an application of Driouich would have resulted in a system using a better understood machine learning model. As such, the application of Driouich, and the claimed invention, would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention in view of the disclosures of Reisman and the teachings of Holub and Driouich.
Response to Arguments
Applicant’s Argument Regarding 101 Rejections of claims 1-6 and 8-21:
The features of “determine a current gap-in-care,” “receiving guideline behavior,” “determining a difference between the current health-related behavior of the member and the guideline behavior,” “determine a probability of at least partially closing the current gap-in-care,” and “provide … a communication to the member” are not equivalent in any way to an observation, evaluation, judgement, or opinion.
The claimed features of using Shapley Additive Explanations (SHAP) values cannot be equated to a mathematical calculation as it is not a mathematical operation (such as multiplication) or an act of calculating using mathematical methods to determine a variable or a number, e.g., performing an arithmetic operation such as exponentiation. See MPEP 2106.04(a)(I2)(I)(C).
The present claims are directed as a whole towards providing a communication to a member that describes one or more actions for the member via a communication.
Applicant submits that alleged subtract ideas of the claims are, when considered as a combination, meaningfully limited and integrated into a practical application.
Examiner’s Response: Applicant's arguments filed 19 December 2025 have been fully considered but they are not persuasive.
Applicant’s basis for asserting that “determine a current gap-in-care” is “not equivalent in any way” to “evaluation” or “judgement”, and that “provide … a communication to the member” is “not equivalent in any way” to “opinion” is not articulated and is not apparent to the examiner.
Applicant is incorrect that using SHAP values is not “an act of calculating using mathematical methods to determine a variable or a number.” Shapley Additive exPlanations is a specific computational method for determining importance values of various features in a model.
Examiner notes that Applicant’s characterization of the claim as directed to “providing a communication to a member that describes one or more actions for the member via a communication” strongly supports the conclusion that the claims describe a method of organizing human activity.
Applicant’s assertion is untethered from any identification of additional elements with which the claim would be meaningfully limited by. When the additional elements identified by the examiner are considered, they do not appear to meaningfully limit the abstract idea. As such, Applicant’s argument is unpersuasive.
Applicant’s Argument Regarding 103 Rejections of claims 1-6 and 8-21:
Applicant submits that the Driouich reference relied upon in the 103 rejection does not qualify as prior art. … A date on a web page printout does not, without more, demonstrate public availability, indexing, or dissemination. … While the copy of the Driouich reference provided in the Office Action mailed June 25, 2025 references the date of December 15, 2020, the date is not indicated as being a publication date.
The website archive.org includes a capture of the URL … dated June 19, 2021. According to archive.org, the Driouich reference website was blank on that date, save for a notice of Cookie’s Policy.
Holub does not disclose, teach, or suggest, “using [a] machine learning model to determine a probability of at least partially closing the current gap-in-care for each of … one or more actions”, where the machine learning model is generated for the member as recited in the independent claims. Instead, Holub teaches that machine learning models may be individually trained for different medications or actions for medication and may be trained to identify potential issues with a medication. Nothing in Holub discloses, teaches, or suggests using the machine learning model that was generated for a member to determine a probability of at least partially closing the current gap-in-care for each of the one or more actions.
Examiner’s Response: Applicant's arguments filed 19 December 2025 have been fully considered but they are not persuasive.
Examiner disagrees that the page does not indicate that the included date is a publication date. The location of the date after a title and next to the author's name for a blog post, is an indication that this date is the publication date of the article.
As noted in the Advisory Action mailed 19 November 2025, “The referenced website when accessed today (17 November 2025), reflects the printout mailed with the 25 June 2025 office action. However, an archive.org capture of the website today initially appears to be blank but for a notice of "page does not exist" and a loading icon, and eventually just displays the loading icon. It appears that archive.org's software consistently does not correctly capture the blog at developers.amadeus.com.”
Applicant appears to be implying that the limitation “generating a gradient boosting machine learning model for the member” requires that the model be member specific. However, the broadest reasonable interpretation of the limitation includes a model generated for members at large including a particular member. As such, Holub does disclose the identified limitation.
Applicant’s Argument Regarding Double Patenting Rejections of claims 1, 19, and 20: A Terminal Disclaimer over such patent has been submitted to cure any basis for the rejection.
Examiner’s Response: Applicant's Terminal Disclaimer filed 19 December 2025 resolves the identified issue.
Additional Considerations
The prior art made of record and not relied upon that is considered pertinent to applicant’s disclosure can be found in the PTO-892 Notice of References Cited.
Segal (US 2020/0373018 A1) is noted for evaluating features that “contribute … to the predicted probability” of a health issue by applying SHAP techniques to Gradient Boosting Decision Trees, which closely parallels specific techniques disclosed in the present application.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Bion A Shelden whose telephone number is (571)270-0515. The examiner can normally be reached M-F, 12pm-10pm EST.
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, Kambiz Abdi can be reached at (571) 272-6702. 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.
/Bion A Shelden/ Primary Examiner, Art Unit 3685 2026-02-06