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 .
Priority
Claim of priority to provisional patent application 63/634659 is acknowledged.
Status of the Claims
The status of the claims as of the response filed 6/23/2026 is as follows: Claims 1, 4, 7-8, 10, 12, 15, and 18-19 are currently amended. Claims 2-3, 5-6, 9, 11, 13-14, 16-17, and 20 are original. Claims 1-20 are currently pending in the application and have been considered below.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 5/26/2026 is in compliance with the provisions of 37 CFR 1.97 and is being considered by the examiner.
Response to Amendment
Rejection Under 35 USC 101
The claims have been amended but the 35 USC 101 rejections for claims 1-20 are upheld.
Rejection Under 35 USC 102/103
The amendments made to the claims introduce limitations that are not fully addressed in the previous office action, and thus the corresponding 35 USC 102/103 rejections are withdrawn. However, Examiner will consider the amended claims in light of an updated prior art search and address their patentability with respect to prior art below.
Response to Arguments
Rejection Under 35 USC 101
On pages 9-10 of the response filed 6/23/2026 Applicant argues that “even if the present claims can be performed (at least partially) by a human, the claims are clearly not ‘directed to’ managing any personal behavior or interactions,” and are instead “‘directed to’ a complex and technical process for activating and deactivating sensors based on performance of an assistive action, as well as using machine learning to predict outcomes of such actions.” Applicant’s arguments are fully considered, but are not persuasive. The claims recite no specific complex or technical processes for activating and deactivating the sensors, and the activation/deactivation of the sensors merely functions within the claims as a means of obtaining the data necessary for the abstract patient positioning analysis and outcome determination steps of the invention such that they amount to insignificant extra-solution activity in the form of data gathering. The presence of additional elements in the claims beyond the abstract idea itself does not preclude the claims from reciting a judicial exception under Step 2A – Prong 1; in the instant case, Examiner maintains that the crux of the invention is the evaluation of patient data to make determinations about positioning, characteristics, and clinical outcomes so that appropriate intervention may be initiated between human actors, which fits into the “mental process” and “certain methods of organizing human activity” groupings of abstract idea (as explained in more detail in the updated 35 USC 101 rejections below). Accordingly, Examiner maintains that the claims are directed to a judicial exception and do not merely “involve” an exception as Applicant asserts.
On pages 10-11 Applicant argues that “dynamic sensor activation and deactivation reflects a clear technical improvement in the functioning of the computer system” as outlined in at least [0066] of the specification. Applicant’s arguments are fully considered, but are not persuasive. There is no technical detail about how the computing system achieves the recited dynamic activation and deactivation of the sensors, and, when considered in the context of the claims as a whole, such activation/deactivation of the sensors merely functions as a means of obtaining the data necessary for the abstract patient positioning analysis and outcome determination steps of the invention such that they amount to insignificant extra-solution activity in the form of data gathering under Step 2A – Prong 2 (see MPEP 2106.05(g)). When reevaluated under Step 2B, this dynamic activation/deactivation of the sensors again amounts to insignificant extra-solution activity, and is also found to be a well-understood, routine, and conventional function in the field of clinical monitoring, as evidenced by at least [0061] of Mariottini et al. (US 20160147959 A1); [0033] of Dsouza et al. (US 20230107394 A1); [0045] of Seukhtipyaroge (US 20230214704 A1); abstract, [0086], & [0092] of Knickerbocker et al. (US 20220199235 A1); and [0040] & [0043] of Greiner (US 20190059725 A1). Accordingly, this feature does not amount to a practical application or inventive concept under Steps 2A – Prong 2 or 2B.
For the reasons outlined above, the 35 USC 101 rejections are upheld for claims 1-20.
Rejection Under 35 USC 102/103
Applicant’s arguments on pages 12-13 with respect to alleged deficiencies of the previously cited prior art references have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 4 and 15 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Claims 4 and 15 each recite “evaluating image data using a convolutional neural network to infer, based on movement of the user, that performance of the action has been completed.” Applicant’s original specification does not provide sufficient explanation of how the CNN acts to evaluate image data to infer that performance of an action has been completed based on movement of a user that would convince one of ordinary skill in the art that Applicant had possession of this feature at the time of filing. A convolutional neural network is only mentioned in the following portions of Applicant’s specification:
[0063]: “In some embodiments, the system may infer or determine action performance indirectly. For example, the system may evaluate video or image data (e.g., using convolutional neural networks) to detect particular patterns or movements indicative of performing given actions.”
[0065]: “In some embodiments, the system may infer or determine action completion indirectly. For example, the system may evaluate video or image data (e.g., using convolutional neural networks) to detect particular patterns or movements indicative of completion of given actions.”
[0091]: “For example, for image data, the monitoring system 105 may use one r more convolutional neural networks (CNNs) to identify objects (e.g., the user’s hands), track movement through physical space (e.g., to monitor how the user’s hands move in three-dimensional space), and the like.”
Such disclosures repeat the functional language of the claims that merely recite the result of the CNN without explaining how the CNN would specifically be trained or operated to transform inputs of image data into the desired output of inferred completion of an assistive action between a user and a patient. Because Applicant has not sufficiently shown that they were in possession of this feature at the time of filing, this limitation is rejected under 35 U.S.C. 112(a).
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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1
In the instant case, claims 1-12 are directed to methods (i.e. processes) and claims 13-20 are directed to a system (i.e. a machine). Thus, each of the claims falls within one of the four statutory categories. Nevertheless, the claims fall within the judicial exception of an abstract idea.
Step 2A – Prong 1
Independent claims 1, 10, and 12 recite steps that, under their broadest reasonable interpretations, cover mental processes as well as certain methods of organizing human activity, e.g. managing personal behavior, relationships, or interactions between people. Specifically, claim 12 (as representative) recites:
A system, comprising: one or more processors; and one or more memories storing a program, which, when executed on any combination of the one or more processors, performs operations, the operations comprising:
determining that a user is performing an action to assist a patient in a physical environment;
in response to the determining that the user is performing the action, activating a set of sensors in the physical environment;
accessing first sensor data collected by the set of sensors;
determining that the user has completed performance of the action;
in response to the determining that the user has completed performance of the action, deactivating the set of sensors in the physical environment;
determining, based on the first sensor data:
a first positioning of the patient in the physical environment prior to performance of the action; and
a second positioning of the patient in the physical environment subsequent to performance of the action;
determining a set of patient characteristics for the patient;
generating an outcome score for the patient, using a first trained machine learning model, based on the first positioning, the second positioning, and the set of patient characteristics; and
in response to determining that the outcome score does not satisfy one or more criteria, initiating one or more interventions for the patient.
But for the recitation of generic computer components like processors, memories, and machine learning, the italicized functions, when considered as a whole, describe a patient positioning analysis operation that could be achieved by a human actor (e.g. a clinician or other medical professional) mentally and/or via managing their personal behavior and interactions with others. For example, a person could observe a user interacting with a patient to mentally determine whether the person is performing an assistance action for the patient (e.g. repositioning the patient in a bed, providing medication or other treatment, providing new/additional bedding or pillows, assisting with toileting, etc.) as well as determine when the assistance action has been completed. The person could then access sensor data from the time of the assistance action (e.g. by looking at sensor readouts from positioning/pressure sensors, seeing recorded videos or images from a camera, listening to recorded audio, etc.) and mentally make determinations about a first positioning of the patient prior to performance of the action and a second positioning of the patient subsequent to performance of the action (e.g. by noting that a patient was in a supine position, but that the user adjusted the bed configuration such that the patient is now sitting up). The person could also mentally determine a set of patient characteristics by using observation and judgment, for example noting the age, race, sex, medical conditions, or other characteristics of the patient. The person could then use their expertise to mentally (or with aid of pen and paper) generate an outcome score based on the first and second positionings as well as the patient characteristics, e.g. by estimating that the patient is at higher risk for a pressure injury due to the new position compared to the old position and based on characteristics like age and medical condition. The person could finally mentally determine that the outcome score does not satisfy a criterion and manage their personal behavior and interactions with the patient to initiate an appropriate intervention, e.g. physically repositioning the patient, providing the patient with different bedding, speaking with the patient about their risk, etc. Accordingly, claim 12 recites an abstract idea in the form of a mental process as well as a certain method of organizing human activity. Claim 1 recites substantially similar subject matter as claim 12 and is also found to recite an abstract idea under the same analysis.
Similarly, claim 10 recites:
A method, comprising:
determining that a user is performing an action to assist a patient in a physical environment;
in response to the determining that the user is performing the action, activating a set of sensors in the physical environment;
accessing first sensor data collected by the set of sensors;
determining that the user has completed performance of the action;
in response to the determining that the user has completed performance of the action, deactivating the set of sensors in the physical environment;
determining, based on the first sensor data:
a first positioning of the patient in the physical environment prior to performance of the action; and
a second positioning of the patient in the physical environment subsequent to performance of the action;
determining a set of patient characteristics for the patient;
training a first machine learning model to generate outcome scores for patient positioning based on processing the first positioning, the second positioning, and the set of patient characteristics using the first machine learning model; and
deploying the first machine learning model to generate outcome scores.
But for the recitation of generic computer components like machine learning, the italicized functions, when considered as a whole, describe a patient positioning analysis operation that could be achieved by a human actor (e.g. a clinician or other medical professional) mentally or with the aid of pen and paper. For example, a person could observe a user interacting with a patient to mentally determine whether the person is performing an assistance action for the patient (e.g. repositioning the patient in a bed, providing medication or other treatment, providing new/additional bedding or pillows, assisting with toileting, etc.) as well as determine when the assistance action has been completed. The person could then access sensor data from the time of the assistance action (e.g. by looking at sensor readouts from positioning/pressure sensors, seeing recorded videos or images from a camera, listening to recorded audio, etc.) and mentally make determinations about a first positioning of the patient prior to performance of the action and a second positioning of the patient subsequent to performance of the action (e.g. by noting that a patient was in a supine position, but that the user adjusted the bed configuration such that the patient is now sitting up). The person could also mentally determine a set of patient characteristics by using observation and judgment, for example noting the age, race, sex, medical conditions, or other characteristics of the patient. The person could then use their expertise to mentally (or with aid of pen and paper) generate an outcome score based on learned correlations among the first and second positionings as well as the patient characteristics, e.g. by estimating that the patient is at higher risk for a pressure injury due to the new position compared to the old position and based on characteristics like age and medical condition. Accordingly, claim 10 recites an abstract idea in the form of a mental process.
Dependent claims 2-9, 11, and 13-20 inherit the limitations that recite an abstract idea from their dependence on claims 1, 10, and 12, respectively, and thus these claims also recite an abstract idea under the Step 2A – Prong 1 analysis. In addition, claims 2-5, 7-9, 11, 13-16, and 18-20 recite additional limitations that further describe the abstract idea identified in the independent claims.
Specifically, claims 2 and 13 describe types of sensor data, each of which are types of sensor data that a person would be capable of observing readouts from and evaluating to make determinations about patient positioning and outcomes.
Claims 3 and 14 specify that the first sensor data is accessed in response to receiving an indication from a user that the user is performing an action to reposition the patient, which could be accomplished as part of a certain method of organizing human activity by the person taking special care to access sensor data (e.g. by observing readouts) right after a colleague (e.g. a nurse or other aide) has informed them that they are imminently repositioning the patient.
Claims 4 and 15 recite evaluating image data to infer, based on movement of the user, that performance of the action has been completed, which a person would be capable of achieving by looking at recorded images/video and mentally making an inference that the action has been completed based on movements of the user in the images/video.
Claims 5 and 16 describe the type of patient characteristics, each of which are types of information that a person would be capable of mentally determining about a patient and taking into consideration when making positioning and outcome evaluations.
Claims 7 and 18 specify that generating the outcome score comprises predicting the second positioning of the patient based on processing the first sensor data, and processing the second positioning of the patient. A person would be capable of mentally predicting the second positioning of the patient to make determinations about corresponding outcomes.
Claims 8 and 19 recite that the outcome score indicates at least one of a probability that the second positioning of the patient will be comfortable, or a prediction of whether the second positioning will improve or worsen one or more medical conditions of the patient, which are each types of predictions that a person would be capable of mentally generating when observing positioning data in light of other known patient characteristics.
Claims 9 and 20 specify that initiating the one or more interventions comprises transmitting a notification to a user assisting the patient, which a person could achieve by managing their personal behavior during an interaction with an aide or colleague to inform them that they should assist the patient.
Claim 11 describes generating predictions for patient positioning based on the first sensor data. A person would be capable of learning correlations between patient positioning and outcome scores such that they are capable of determining outcome scores by looking at patient positioning gleaned from sensor data as an input.
However, recitation of an abstract idea is not the end of the analysis. Each of the claims must be analyzed for additional elements that indicate the abstract idea is integrated into a practical application to determine whether the claim is considered to be “directed to” an abstract idea.
Step 2A – Prong 2
The judicial exception is not integrated into a practical application. In particular, independent claims 1, 10, and 12 do not include additional elements that integrate the abstract idea into a practical application. The additional elements of claims 1, 10, and 12 include a set of sensors that collect the first sensor data, as well as specifying that the set of sensors are activated in response to determining that the user is performing the action and deactivated in response to determining that the user has completed performance of the action. Each of the independent claims also recites using/deploying a first trained machine learning model to generate the outcome score, while claim 10 further recites training the first machine learning model. Claim 12 also recites the additional elements of one or more processors and one or more memories storing a program to perform the recited functions.
The collection of data by a set of sensors that are activated and deactivated responsive to abstract determinations amounts to insignificant extra-solution activity in the form of mere data gathering, because the set of sensors is merely invoked as a means of obtaining the input data necessary for the main analysis steps of the invention (see MPEP 2106.05(g)). The use of a trained machine learning model as well as the processor and memory for performing the functions of the invention amount to mere instructions to “apply” the abstract idea using generic computer components because they are merely invoked as tools with which to digitize and/or automate the otherwise-abstract functions of making determinations, accessing and evaluating sensor data to make further determinations, determining patient characteristics, generating an outcome score, comparing the outcome score to one or more criteria, and initiating interventions when the criteria are not met (see MPEP 2106.05(f)). The trained machine learning model is used to generally apply the abstract idea without placing any limits on how the trained model functions; rather, this limitation only recites the outcome of “generating an outcome score” based on various data without any details about how these specific types of data are actually evaluated to generate the outcome score. Similarly, the training of the model as in claim 10 is recited at a very high level of generality with no details about any specific training methods or architectures, and also amounts to instructions to “apply” the abstract idea because it merely digitizes/automates the learning of correlations between clinical inputs and outputs that a person could achieve mentally (e.g. via targeted training or experience over time) such that the model can later be deployed to digitize/automate the otherwise-abstract determinations that a person could make. There are no improvements to the underlying model training methods or machine learning structure or functioning that would amount to an improvement to a technology; rather, high-level machine learning techniques are merely being applied to the field of outcome prediction to digitize and/or automate the otherwise-abstract process of generating outcome scores based on known patient positionings and characteristics. Accordingly, claims 1, 10, and 12 as a whole are each directed to an abstract idea without integration into a practical application.
The judicial exception recited in dependent claims 2-9, 11, and 13-20 is also not integrated into a practical application under a similar analysis as above. Claims 2-3, 5, 8-9, 13-14, 16, and 19-20 do not introduce any new additional elements of their own, merely invoking the same sensor and machine learning features addressed for the independent claims above, and accordingly also do not provide integration into a practical application.
Claims 4 and 15 introduce a convolutional neural network that is used to perform the inference step, which amounts to mere instructions to “apply” the exception in a similar manner as explained for the first machine learning model of the independent claims. That is, the CNN is invoked at a high level of generality and merely functions to digitize and/or automate the otherwise-abstract step of evaluating image data to infer that performance of an action has been completed based on movement of the user. The CNN is used to generally apply the abstract idea without placing any limits on how the CNN functions; rather, this limitation only recites the outcome of “inferring that performance of the action has been completed” based on movement of the user in image data without any details about how the image movement data is actually evaluated to infer that the action has been completed.
Claims 6 and 17 specify that the first trained machine learning model was trained based on a set of position exemplars, each respective position exemplar of the set of position exemplars comprising respective sensor data and respective outcome data for a corresponding patient. This limitation again amounts to mere instructions to “apply” the judicial exception, because it merely describes how the model was previously trained via unspecified, high-level supervised learning to learn correlations between desired inputs and outputs, which merely digitizes/automates the learning of correlations between clinical inputs and outputs that a person could achieve mentally (e.g. via targeted training or experience over time) such that the model can later be deployed to digitize/automate the otherwise-abstract determinations that a person could make. There are no improvements to the underlying model training methods or machine learning structure or functioning that would amount to an improvement to a technology; rather, high-level machine learning techniques are merely being applied to the field of outcome prediction to digitize and/or automate the otherwise-abstract process of generating outcome scores based on known patient positionings and characteristics.
Claims 7 and 18 introduce a second trained machine learning model that predicts the second position of the patient based on processing the first sensor data, so that the second positioning can be used as an input to the first trained machine learning model. The second trained machine learning model is similar to the first trained machine learning model, in that it is recited at a high level of generality and is merely invoked as a tool with which to digitize and/or automate an otherwise-abstract function (i.e. prediction the second positioning of the patient based on sensor data). The trained machine learning model is used to generally apply the abstract idea without placing any limits on how the trained model functions; rather, this limitation only recites the outcome of “predicting the second position of the patient” based on unspecified processing of the first sensor data, without any details about how the sensor data are actually processed to predict the second positioning.
Claim 11 similarly recites a step for training a machine learning model to generate predictions for patient positioning based on the first sensor data, which is then used as input for the first machine learning model, which amounts to mere instructions to “apply” the judicial exception as explained for similar limitations of claims 7, 10, and 18 above. The training of the model is recited at a very high level of generality with no details about any specific training methods or architectures, and amounts to instructions to “apply” the abstract idea because it merely digitizes/automates the learning of correlations between clinical inputs and outputs that a person could achieve mentally (e.g. via targeted training or experience over time) such that the model can later be deployed to digitize/automate the otherwise-abstract positioning predictions that a person could make. There are no improvements to the underlying model training methods or machine learning structure or functioning that would amount to an improvement to a technology; rather, high-level machine learning techniques are merely being applied to the field of positioning prediction to digitize and/or automate the otherwise-abstract process of predicting a patient’s positioning based on known sensor information.
Accordingly, the additional elements of claims 1-20 do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. Claims 1-20 are directed to an abstract idea.
Step 2B
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of trained machine learning models, processors, and memories for performing the determining, accessing, generating, initiating, deploying, etc. steps of the invention amount to mere instructions to apply the exception using generic computer components. As evidence of the generic nature of the above recited additional elements, Examiner notes that Applicant’s specification is silent to the particulars of the machine learning models and computer elements; paras. [0440]-[0442] provide a broad overview of generic examples of computing system components like processors and memories, while para. [0140] provides one example of the machine learning model as being a neural network, but notes that “the particular operations and techniques used to train the outcome machine learning model 525 may vary depending in the particular model architecture and implementation,” indicating to one of ordinary skill in the art that other known machine learning model architectures are intended to be contemplated. Similarly, a CNN is broadly described as an example means of determining when an action has been completed in para. [0065] and further described as an example implementation of a type of machine learning technique that would vary based on implementation in [0091], with no explanation of any specific training or operation of the CNN that show it is anything other than a known, existing type of model that one of ordinary skill in the art would recognize.
As indicated above, the use of a set of sensors to collect the sensor data amounts to insignificant extra-solution activity in the form of mere data gathering. Examiner further notes that it is well-understood, routine, and conventional to utilize sensors to obtain sensor data as input for clinical prediction and outcome scoring purposes, as evidenced by at least [0070] & [0073] of Areias et al. (US 20250210176 A1); [0036] of Syal et al. (US 20230074628 A1); and [0035]-[0038] of Kraal et al. (US 20220223255 A1). The selective activation/deactivation of these sensors responsive to meeting a trigger condition is also well-understood, routine, and conventional in patient monitoring, as evidenced by at least [0061] of Mariottini et al. (US 20160147959 A1); [0033] of Dsouza et al. (US 20230107394 A1); [0045] of Seukhtipyaroge (US 20230214704 A1); abstract, [0086], & [0092] of Knickerbocker et al. (US 20220199235 A1); and [0040] & [0043] of Greiner (US 20190059725 A1).
Analyzing these additional elements as an ordered combination adds nothing that is not already present when considering the elements individually; the overall effect of the activated/deactivated set of sensors, trained machine learning models, and computer implementation in combination is to digitize and/or automate patient positioning analysis and outcome prediction operations that could otherwise be achieved mentally and/or as certain methods of organizing human activity. Examiner further notes that the combination of a set of sensors to collect data and trained machine learning models (e.g. CNNs) executed by processors and memory to generate outcome scores and make clinical predictions is well-understood, routine, and conventional, as evidenced by at least abstract, Fig. 1, [0034]-[0039], [0052], & [0152] of Kayser et al. (US 20230013233 A1); abstract, Figs. 1-4, & [0184] of Areias; abstract & Figs. 1-4 of Syal; and abstract & Figs. 2-6 of Kraal et al. Thus, when considered as a whole and in combination, claims 1-20 are not patent eligible.
Claim Rejections - 35 USC § 103
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.
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
Claims 1-3, 5, 8-10, 12-14, 16, and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Thompson et al. (US 20250111948 A1) in view of Sukhtipyaroge (US 20230214704 A1).
Claims 1 and 12
Thompson teaches a method, comprising:
determining that a user is performing an action to assist a patient in a physical environment (Thompson [0030], noting the system monitors the location of caregivers to determine which actions a caregiver performs and how much time is spent on each action in a clinical environment, e.g. helping a patient);
accessing first sensor data collected by the set of sensors (Thompson [0023], [0034], [0036]-[0037], noting the system gathers and analyzes sensor data collected by a set of sensors in the physical environment of a patient);
determining that the user has completed performance of the action (Thompson [0030], noting the system monitors the location of caregivers to determine which actions a caregiver performs and how much time is spent on each action; determining the amount of time spent on each action indicates that both a start and a completion of each action is determined. See also [0028], noting the system can update its outcome score predictions “when one or more actions are completed by one or more caregivers,” indicating that the system is capable of determining that the user has completed performance of an action);
determining, based on the first sensor data: a first positioning of the patient in the physical environment (Thompson [0063], noting sensor data (e.g. from pressure sensors, a camera, etc.) can be evaluated to indicate “one or more body postures of the patient” (i.e. including at least two positionings of the patient in the physical environment));
determining a set of patient characteristics for the patient (Thompson [0032], [0054], noting EMR system data may be used to identify patient characteristics like medical history, diagnoses, clinical interventions, vital signs, allergies, etc.; see also [0063], noting ADT system data can be used to identify patient characteristics like demographic information);
generating an outcome score for the patient, using a first trained machine learning model, based on the first positioning, the second positioning, and the set of patient characteristics (Thompson [0041], [0047], [0062]-[0063], noting the system can evaluate the collected sensor data (e.g. including multiple patient postures as in [0063]) in concert with the patient characteristics via a trained machine learning model to generate a post-operative score, e.g. a risk of pressure injury as an outcome); and
in response to determining that the outcome score does not satisfy one or more criteria, initiating one or more interventions for the patient (Thompson [0048], [0050]-[0051], [0069]-[0070], noting an alert and/or recommendation is issued when the score is above a threshold value (i.e. when the score does not satisfy a criteria of being below the threshold value) which may include recommended post-operative care interventions, e.g. turning the patient bed to reduce the pressure injury risk score).
In summary, Thompson teaches a method for monitoring clinician-patient interactions as well as patient postures via a variety of sensor devices to generate outcome scores and recommend appropriate interventions. Though Thompson teaches determining when a clinician user is performing an action related to patient care (e.g. an interaction with a patient), it fails to explicitly disclose activating the set of sensors in response to determining that the user is performing the action and deactivating the set of sensors in response to determining that the user has completed performance of the action. Additionally, though Thompson contemplates determining and evaluating at least two postures of the patient to determine the outcome score (see [0063]), it fails to explicitly disclose that the first posture represents a first positioning of the patient in the physical environment prior to performance of the action, and that the second posture represents a second positioning of the patient in the physical environment subsequent to performance of the action.
However, [0028] of Thompson further teaches that the system outputs can be continuously updated, e.g. by updating calculation of a post-operative score (which includes determination of body postures of the patient as in [0063]) “when one or more actions are completed by one or more caregivers C to follow recommendations that are designed to reduce the post-operative score below the threshold value.” It therefore would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the “one or more postures” of [0063] to include at least a first posture representing a first positioning of the patient prior to performance of the action and a second posture representing a second positioning of the patient subsequent to performance of the action in order to effectively evaluate the impact of the clinician action/intervention on the patient’s score (as suggested by Thompson [0028]).
Additionally, Sukhtipyaroge teaches an analogous patient and caregiver interaction monitoring system that includes functionality for activating sensor data recording when a caregiver is determined to begin an assistive action and deactivating sensor data recording when the caregiver is determined to have completed the assistive action (Sukhtipyaroge [0045], noting a user can trigger and stop motion data recording by sensors by pressing a button to indicate the beginning and end of a clinical action). It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the collection of sensor data that may be evaluated to determine the impact of a user’s performance of an assistive action as in Thompson such that the sensors are only activated during performance of the action as in Sukhtipyaroge in order to restrict data gathering to the relevant time period associated with the clinical action, thereby reducing the computational burden of data gathering and analysis (as suggested by Sukhtipyaroge [0045]).
Regarding claim 12, Thompson in view of Sukhtipyaroge teaches a system, comprising: one or more processors; and one or more memories storing a program, which, when executed on any combination of the one or more processors, performs operations (Thompson [0041], noting the system includes a processor that receives and executes instructions from a memory to perform the functions of the invention) substantially similar to the steps of claim 1, as explained above.
Claims 2 and 13
Thompson in view of Sukhtipyaroge teaches the method of claim 1, and the combination further teaches wherein the first sensor data comprises at least one of: accelerometer data, orientation data, pressure data, video data, image data, or audio data (Thompson [0034], [0036]-[0037], [0063], [0076], noting sensor data includes pressure data from load cells, audio data from microphones, images and/or videos capturing posture/orientation data and other image or video data, etc.).
Claim 13 recites substantially similar subject matter as claim 2, and is also rejected as above.
Claims 3 and 14
Thompson in view of Sukhtipyaroge teaches the method of claim 1, and the combination further teaches wherein the first sensor data is accessed in response to receiving an indication, from a user, that the user is performing an action to reposition the patient (Sukhtipvaroge [0045], noting sensors (i.e. the sensors of Thompson when considered in the context of the combination) can be activated and accessed for processing data based on receiving an indication from a user that they are performing an action, e.g. to reposition a patient in a bed or into a sitting, standing, or other position as in [0036]-[0037]).
Claim 14 recites substantially similar subject matter as claim 3, and is also rejected as above.
Claims 5 and 16
Thompson in view of Sukhtipyaroge teaches the method of claim 1, and the combination further teaches wherein the set of patient characteristics comprise at least one of demographics of the patient, or one or more medical conditions of the patient (Thompson [0032], [0054], [0063], noting patient characteristics like medical history, diagnoses, allergies (i.e. one or more medical conditions), as well as demographic information).
Claim 16 recites substantially similar subject matter as claim 5, and is also rejected as above.
Claims 8 and 19
Thompson in view of Sukhtipyaroge teaches the method of claim 1, and the combination further teaches wherein the outcome score indicates at least one of: (i) a probability that the second positioning of the patient will be comfortable for the patient, or (ii) a prediction of whether the second positioning of the patient will improve or worsen one or more medical conditions of the patient (Thompson [0062]-[0065], noting the outcome score includes a pressure injury score and/or an infection score, considered equivalent to a prediction of whether the second positioning of the patient will improve or worsen one or more medical conditions of the patient because developing pressure injuries and/or infections following surgery would necessarily worsen the medical condition of the patient).
Claim 19 recites substantially similar subject matter as claim 8, and is also rejected as above.
Claims 9 and 20
Thompson in view of Sukhtipyaroge teaches the method of claim 1, and the combination further teaches wherein initiating the one or more interventions comprises transmitting a notification to a user assisting the patient (Thompson [0048], [0050]-[0051], [0069]-[0070], noting an alert and/or recommendation is issued to a caregiver which may include recommended post-operative care interventions to perform, e.g. turning the patient bed to reduce the pressure injury risk score).
Claim 20 recites substantially similar subject matter as claim 9, and is also rejected as above.
Claim 10
Thompson teaches a method, comprising:
determining that a user is performing an action to assist a patient in a physical environment (Thompson [0030], noting the system monitors the location of caregivers to determine which actions a caregiver performs and how much time is spent on each action in a clinical environment, e.g. helping a patient);
accessing first sensor data collected by the set of sensors (Thompson [0023], [0034], [0036]-[0037], noting the system gathers and analyzes sensor data collected by a set of sensors in the physical environment of a patient);
determining that the user has completed performance of the action (Thompson [0030], noting the system monitors the location of caregivers to determine which actions a caregiver performs and how much time is spent on each action; determining the amount of time spent on each action indicates that both a start and a completion of each action is determined. See also [0028], noting the system can update its outcome score predictions “when one or more actions are completed by one or more caregivers,” indicating that the system is capable of determining that the user has completed performance of an action);
determining, based on the first sensor data: a first positioning of the patient in the physical environment (Thompson [0063], noting sensor data (e.g. from pressure sensors, a camera, etc.) can be evaluated to indicate “one or more body postures of the patient” (i.e. including at least two positionings of the patient in the physical environment));
determining a set of patient characteristics for the patient (Thompson [0032], [0054], noting EMR system data may be used to identify patient characteristics like medical history, diagnoses, clinical interventions, vital signs, allergies, etc.; see also [0063], noting ADT system data can be used to identify patient characteristics like demographic information);
training a first machine learning model to generate outcome scores for patient positioning based on processing the first positioning, the second positioning, and the set of patient characteristics using the first machine learning model; and deploying the first machine learning model to generate outcome scores (Thompson [0041], [0047], [0062]-[0063], noting the system can evaluate the collected sensor data (e.g. including multiple patient postures as in [0063]) in concert with the patient characteristics via a machine learning model that is trained to generate a post-operative score, e.g. a risk of pressure injury as an outcome).
In summary, Thompson teaches a method for monitoring clinician-patient interactions as well as patient postures via a variety of sensor devices to generate outcome scores via training and using a machine learning model. Though Thompson teaches determining when a clinician user is performing an action related to patient care (e.g. an interaction with a patient), it fails to explicitly disclose activating the set of sensors in response to determining that the user is performing the action and deactivating the set of sensors in response to determining that the user has completed performance of the action. Additionally, though Thompson contemplates determining and evaluating at least two postures of the patient to determine the outcome score (see [0063]), it fails to explicitly disclose that the first posture represents a first positioning of the patient in the physical environment prior to performance of the action, and that the second posture represents a second positioning of the patient in the physical environment subsequent to performance of the action.
However, [0028] of Thompson further teaches that the system outputs can be continuously updated, e.g. by updating calculation of a post-operative score (which includes determination of body postures of the patient as in [0063]) “when one or more actions are completed by one or more caregivers C to follow recommendations that are designed to reduce the post-operative score below the threshold value.” It therefore would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the “one or more postures” of [0063] to include at least a first posture representing a first positioning of the patient prior to performance of the action and a second posture representing a second positioning of the patient subsequent to performance of the action in order to effectively evaluate the impact of the clinician action/intervention on the patient’s score (as suggested by Thompson [0028]).
Additionally, Sukhtipyaroge teaches an analogous patient and caregiver interaction monitoring system that includes functionality for activating sensor data recording when a caregiver is determined to begin an assistive action and deactivating sensor data recording when the caregiver is determined to have completed the assistive action (Sukhtipyaroge [0045], noting a user can trigger and stop motion data recording by sensors by pressing a button to indicate the beginning and end of a clinical action). It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the collection of sensor data that may be evaluated to determine the impact of a user’s performance of an assistive action as in Thompson such that the sensors are only activated during performance of the action as in Sukhtipyaroge in order to restrict data gathering to the relevant time period associated with the clinical action, thereby reducing the computational burden of data gathering and analysis (as suggested by Sukhtipyaroge [0045]).
Claims 4, 7, 11, 15, and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Thompson and Sukhtipyaroge as applied to claims 1, 10, or 12 above, and further in view of Kayser et al. (US 20230013233 A1).
Claims 4 and 15
Thompson in view of Sukhtipyaroge teaches the method of claim 1, and the combination further teaches wherein determining that the user has completed performance of the action comprises: evaluating image data (Thompson [0037], noting image data from cameras may be analyzed to monitor movements of patients and caregivers to determine actions performed by and interactions between those users, e.g. to determine completion of actions as in [0028]). However, the present combination fails to explicitly disclose that the determination is made using a convolutional neural network. However, Kayser teaches an analogous method of monitoring patient and caregiver poses, motions, interactions, etc. in image data that utilizes a convolutional neural network (Kayser [0046]-[0053], [0090]). It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the determination of movements and actions of users in images as in the combination such that it is accomplished via a CNN as in Kayser in order to utilize an existing machine learning image analysis approach that is known to be useful for detecting and tracking users in clinical images for the purpose of making positioning-based outcome predictions (as suggested by Kayser [0046] & [0052]-[0053]).
Claim 15 recites substantially similar subject matter as claim 4, and is also rejected as above.
Claims 7 and 18
Thompson in view of Sukhtipyaroge teaches the method of claim 1, and the combination further teaches wherein generating the outcome score comprises: predicting the second positioning of the patient based on processing the first sensor data (Thompson [0063], noting sensor data (e.g. from pressure sensors, a camera, etc.) can be evaluated to indicate/predict “one or more body postures of the patient” (i.e. including the second positioning when considered in the context of the combination), which is then used as input to generate the post-operative score (e.g. via a first trained machine learning model as in [0041])). Though Thompson contemplates that “processing the input data 400 includes utilizing one or more artificial intelligence models to generate system outputs 204,” (see [0041]), the present combination fails to explicitly disclose that the second positioning is predicted using a second trained machine learning model.
However, Kayser teaches an analogous clinical outcome prediction method that includes use of a trained machine learning model to identify and track the pose of an individual in sensor data (Kayser [0051]-[0055]) that is then used as a basis for predicting outcomes like risk of pressure injury (Kayser [0064], [0067]-[0068], [0104]-[0112]). It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the determination of posture from sensor data as in the combination such that it is achieved via a second trained machine learning model as in Kayser in order to utilize positioning prediction methods that incorporate expert-verified examples and may improve over time, thereby improving performance of the prediction (as suggested by Kayser [0054]).
Claim 18 recites substantially similar subject matter as claim 7, and is also rejected as above.
Claim 11
Thompson in view of Sukhtipyaroge teaches the method of claim 10, and further teaches (Thompson [0063], noting sensor data (e.g. from pressure sensors, a camera, etc.) can be evaluated to indicate/predict “one or more body postures of the patient” (i.e. positionings), which is then used as input to generate the post-operative score (e.g. via a first trained machine learning model as in [0041])). Though Thompson contemplates that “processing the input data 400 includes utilizing one or more artificial intelligence models to generate system outputs 204,” (see [0041]), the present combination fails to explicitly disclose training a machine learning model to generate the positioning predictions.
However, Kayser teaches an analogous clinical outcome prediction method that includes training and use of a machine learning model to identify and track the pose of an individual in sensor data (Kayser [0051]-[0055]) that is then used as a basis for predicting outcomes like risk of pressure injury (Kayser [0064], [0067]-[0068], [0104]-[0112]). It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the determination of posture from sensor data as in the combination such that it is achieved via training and using a machine learning model as in Kayser in order to utilize positioning prediction methods that incorporate expert-verified examples and may improve over time, thereby improving performance of the prediction (as suggested by Kayser [0054]).
Claims 6 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Thompson and Sukhtipyaroge as applied to claims 1 or 12 above, and further in view of Sethi et al. (US 20250143632 A1).
Claims 6 and 17
Thompson in view of Sukhtipyaroge teaches the method of claim 1, showing a machine learning model that is trained to generate a post-operative score based on sensor data and other inputs collected about a patient and/or caregiver (Thompson [0041], [0062]-[0065]). However, the present combination does not provide specific details about the training of the model, and thus fails to explicitly disclose wherein the first trained machine learning model was trained based on a set of position exemplars, each respective position exemplar of the set of position exemplars comprising respective sensor data and respective outcome data for a corresponding patient. However, Sethi teaches an analogous outcome prediction (e.g. pressure injury scores) machine learning model that is trained via supervised learning based on training data correlating sensor data inputs with ground truth outcome labels (Sethi [0094]-[0102]). It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the training of an outcome prediction machine learning model as in the combination to include supervised learning via labeled training data as in Sethi in order to utilize known supervised learning techniques that ensure the model will learn the associations between input sensor data and ground truth outcome labels (as suggested by Sethi [0094]-[0095]).
Claim 17 recites substantially similar subject matter as claim 6, and is also rejected as above.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Main et al. (US 20220087617 A1) describes systems for utilizing machine learning models to predict patient pressure injury and/or fall risk outcomes based on collected sensor data and presenting notifications of recommended interventions to mitigate the risks.
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 KAREN A HRANEK whose telephone number is (571)272-1679. The examiner can normally be reached M-F 8:00-4:00 ET.
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, Shahid Merchant can be reached at 571-270-1360. 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.
/KAREN A HRANEK/ Primary Examiner, Art Unit 3684
Prosecution Insights