DETAILED ACTION
This action is a non-final First Office Action.
Claims 1-20 are currently presented for examination.
This action is in response to communications filed on 08/23/2023.
This nonprovisional application is a child of provisional application “Modeling Mental Healthcare User Data To Generate Recommendations For Optimized Outcomes” filed on 08/23/2022 as 63/400,770.
Claims 1-20 are rejected under 35 U.S.C. 101 as being directed to a non-statutory matter, a judicial exception, an abstract idea (mental process and mathematical concept) without significantly more.
Claims 1, 5-9, 11-12, and 15-20 are rejected under 35 U.S.C. 103 as being unpatentable over Goldner et al. “Predictive Guidance Systems For Personalized Health And Self-Care, And Associated Methods” (US20210391081A1) in view of Lefkofsky et al. “Method And Process
For Predicting And Analyzing Patient Cohort Response, Progression, And Survival” (US20210125731A1).
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Goldner in view of Lefkofsky in further view of Solomon et al. “System And Method For Mechanical Failure Classification, Condition Assessment and Remediation Recommendation” (US20210388950A1).
Claims 3 and 4 are rejected under 35 U.S.C. 103 as being unpatentable over Goldner in view of Lefkofsky in further view of Wang et al. “Method For Identifying High-Risk AML Patients” (US20210371937A1).
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Goldner in view of Lefkofsky in further view of Sengupta et al. “Video Based Matching And Tracking By Analyzing One Or More Image Abstractions” (US8600172B2).
Claim 13 is rejected under 35 U.S.C. 103 as being unpatentable over Goldner in view of Lefkofsky in further view of Solomon and Wang.
Claim 14 is rejected under 35 U.S.C. 103 as being unpatentable over Goldner, Lefkofsky, Solomon, and Wang in view of IEEE TechnologyNavigator “Covariance Matrices” and Sarabandi et al. “Automotive Radar Scene Simulator” (US20200111382A1).
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 .
Information Disclosure Statement
The information disclosure statements (IDS) have been considered by the Examiner.
Drawings
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(4) because reference character “130” has been used to designate both “Database” in Figure 1 and “storage system” in specification paragraphs [0024]-[0025]. I would recommend changing the drawings label to match the specification as data could be stored without the need of a database. Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Specification
The disclosure is objected to because of the following informalities: Lack of space between paragraphs [0027] and [0028]. Paragraph [0028] refers to “one or more network interface controllers” as 240, while Figure 2 refers to “network interface controller(s)” as 230. Paragraph [0029] refers to “user data” as 216, while Figure 2 refers to “user data” as 214. I would advise changing either the drawing numbers or the specification numbers to be in line with each other. Appropriate correction is required.
The equations found on page 12 of the specification are blurry and can be misinterpreted when viewed from the paper version of the specification. A clear set of equations should be included so that anybody viewing the specification should be able to easily tell what the equations represent. See MPEP 37 CFR 1.52(a)(1)(v) “Presented in a form having sufficient clarity and contrast between the paper and the writing thereon to permit the direct reproduction of readily legible copies in any number by use of photographic, electrostatic, photo-offset, and microfilming processes and electronic capture by use of digital imaging and optical character recognition.” Appropriate correction is required.
Claim Objections
Claim 7 objected to because of the following informalities: claim 7 states wherein the future state data includes at least one of a predicted state or a potential future treatment process. Specifying “potential” within the claim language suggests that any future treatment process, no matter how unrelated to the actual state data, could be included as a treatment process. Claim language should be amended to clarify that future treatment processes should be at least positively correlated to the future state data. Appropriate correction is required.
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 a judicial exception (i.e. abstract idea) without anything significantly more.
Step 1: Claims 1-10 are directed to a method, which is a process, which is a statutory category of invention. Claims 11-20 are directed to a non-transitory processor-readable medium, which is a manufacture, which is a statutory category of invention. Therefore, claims 1-20 are directed to patent eligible categories of invention.
Step 2A, Prong 1: Claims 1 and 11 recite the abstract idea of generating a future state model based on user and group data, constituting an abstract idea based on Mathematical Concepts based on mathematical relationships, mathematical formulas/equations, and mathematical calculations, and based on Mental Processes based on concepts performed in the human mind, both with the aid of pencil and paper.
In regard to claim 1, the limitation of receiving, at the processor, (1) past state data associated with a user and (2) past activity data associated with the user; covers mental processes including making an observation about the past state data and past activity data.
Additionally, the limitation of generating, via the processor, (1) a covariance matrix based on a state model, an activity dissipation model, and an activity magnitude model, associated with a group including the user, (2) at least one activity impact dissipation parameter associated with the user, and (3) at least one activity magnitude parameter associated with the user; covers mathematical processes including a mathematical relationship regarding the covariance matrix and the models and parameters associated with the user and covers mental processes including making a judgement about the state model, activity dissipation model, and the activity magnitude model.
Additionally, the limitation of generating at least one state impact value based on the at least one activity impact dissipation parameter, and the at least one activity magnitude parameter; covers mental processes including making an observation about the activity impact dissipation parameter and a judgement about how to make at least one state impact value.
Additionally, the limitation of generating a non-linear regression model associated with a future state based on the past state data, the past activity data, and the at least one state impact value; covers mathematical concepts including making mathematical calculations about the past state data, the past activity data, and at least one state impact value.
Additionally, the limitation of generating a non-linear regression model associated with a future state based on the past state data, the past activity data, and the at least one state impact value; covers mathematical concepts including making mathematical calculations about the past state data, the pas activity data, and the at least one state impact value. Additionally, transmitting, via the processor, at least one signal configured to cause display of the future state data to the user covers mental processes including making an opinion of the future state data.
In regard to claim 11, the limitation of receive (1) at least one intensity value associated with a physical state and (2) at least one time value associated with at least one program type; covers mental processes include making an observation about the intensity and time values.
Additionally, the limitation calculate at least one type-specific state reduction value associated with the physical state based on (a) at least one response model for the at least one program type and (b) the at least one time value; covers mathematical concepts including mathematical calculations in the calculation of the state reduction value associated with the physical state.
Additionally, the limitation calculate a total state reduction value based on the at least one type-specific state reduction value; covers mathematical concepts including mathematical calculations used in the total state reduction value based on the state reduction value.
Additionally, the limitation calculate at least one future intensity value based on the total state reduction value; and covers mathematical concepts including mathematical calculations used in the future intensity value based on the total state reduction value.
Additionally, the limitation cause display of the at least one future intensity value covers mental processes including making an observation about the future intensity value.
Thus, the claims recite the abstract idea of generating a future state model based on user and group data.
Dependent claims 2-10 and 12-20 further narrow the abstract ideas, identified in the independent claims.
Step 2A, Prong 2: The judicial exception is not integrated into a practical application. In claim 1, the limitation of receiving, at a processor, (1) past state data associated with a user and (2) past activity data associated with the user; amounts to insignificant extra-solution activity of receiving data, i.e. pre-solution activity for gathering data for use in the claimed process. See MPEP 2106.05(g).
Additionally, the limitation of generating, via the processor, (1) a covariance matrix based on a state model, an activity dissipation model, and an activity magnitude model, associated with a group including the user, (2) at least one activity impact dissipation parameter associated with the user, and (3) at least one activity magnitude parameter associated with the user; generating at least one state impact value based on the at least one activity impact dissipation parameter, and the at least one activity magnitude parameter; amounts to mere instructions to apply an exception, where it recites the idea of a solution. The limitation does not indicate what the activity dissipation model, activity magnitude model, activity impact dissipation parameter, or activity magnitude parameter are or how they are associated with the state impact value. See MPEP 2106.05(f)(1).
Additionally, the limitation generating a non-linear regression model associated with a future state based on the past state data, the past activity data, and the at least one state impact value; amounts to mere instructions to apply an exception, where it invokes computers or other machinery merely as a tool to perform an existing process. The inclusion of a non-linear regression model would fall under this category. See MPEP 2106.05(f)(2).
Additionally, the limitation of calculating, via the processor, future state data based on the non-linear regression model; and amounts to mere instructions to apply an exception, where it recites the idea of a solution. The limitation does not indicate what the future state data or how it is associated with the non-linear regression model. See MPEP 2106.05(f)(1).
Additionally, the limitation of transmitting, via the processor, at least one signal configured to cause display of the future state data to the user amounts to mere instructions to apply an exception, where it is selecting a particular data source or type of data to be manipulated, in this case selecting information for display. See MPEP 2106.05(g).
Further, the claim includes the additional element of a processor. The processor is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a computer and/or a generic computer component. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
In claim 11, the limitation of receive (1) at least one intensity value associated with a physical state and (2) at least one time value associated with at least one program type; amounts to insignificant extra-solution activity of receiving data, i.e. pre-solution activity for gathering data for use in the claimed process. See MPEP 2106.05(g).
Additionally, the limitation of calculate at least one type-specific state reduction value associated with the physical state based on (a) at least one response model for the at least one program type and (b) the at least one time value; amounts to mere instructions to apply an exception, where it recites an idea of a solution. The limitation does not indicate what the type-specific state reduction value is. See MPEP 2106.05(f)(1).
Additionally, the limitation calculate a total state reduction value based on the at least one type-specific state reduction value; amounts to mere instructions to apply an exception, where it recites the idea of a solution. The limitation does not indicate what a total state reduction value is.
Additionally, the limitation calculate at least one future intensity value based on the total state reduction value; and amounts to mere instructions to apply an exception, where it recites the idea of a solution. The limitation does not indicate what the future intensity value is.
Additionally, the limitation cause display of the at least one future intensity value amounts to mere instructions to apply an exception, where it is selecting a particular data source or type of data to be manipulated, in this case selecting information for display. See MPEP 2106.05(g).
Further, the claim includes the additional element of a processor. The processor is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a computer and/or a generic computer component. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claims 2-10 and 12-20 further narrow the abstract ideas, identified in the independent claims, and do not introduce further additional elements for consideration beyond those addressed above.
Step 2B: In claim 1, the limitations of receiving, at a processor, (1) past state data associated with a user and (2) past activity data associated with the user;” and “transmitting, via the processor, at least one signal configured to cause display of the future state data to the user are shown to reflect the court decisions of Versata Dev. Group, Inc. v SAP Am., Inc. iv Storing and retrieving information in memory, shown in MPEP 2106.05(d)(II).
Additionally, the limitation of generating, via the processor, (1) a covariance matrix based on a state model, an activity dissipation model, and an activity magnitude model, associated with a group including the user, (2) at least one activity impact dissipation parameter associated with the user, and (3) at least one activity magnitude parameter associated with the user; generating at least one state impact value based on the at least one activity impact dissipation parameter, and the at least one activity magnitude parameter; amounts to mere instructions to apply an exception, where it recites the idea of a solution. The limitation does not indicate what the activity dissipation model, activity magnitude model, activity impact dissipation parameter, or activity magnitude parameter are or how they are associated with the state impact value. See MPEP 2106.05(f)(1).
Additionally, the limitation generating a non-linear regression model associated with a future state based on the past state data, the past activity data, and the at least one state impact value; amounts to mere instructions to apply an exception, where it invokes computers or other machinery merely as a tool to perform an existing process. The inclusion of a non-linear regression model would fall under this category. See MPEP 2106.05(f)(2).
Additionally, the limitation of calculating, via the processor, future state data based on the non-linear regression model; and amounts to mere instructions to apply an exception, where it recites the idea of a solution. The limitation does not indicate what the future state data or how it is associated with the non-linear regression model. See MPEP 2106.05(f)(1).
Further, the claim includes the additional element of a processor. The processor is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a computer and/or a generic computer component. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
In claim 11, the limitation of receive (1) at least one intensity value associated with a physical state and (2) at least one time value associated with at least one program type;” and “cause display of the at least one future intensity value are shown to reflect the court decisions of Versata Dev. Group, Inc. v SAP Am., Inc. iv Storing and retrieving information in memory, shown in MPEP 2106.05(d)(II).
Additionally, the limitation of calculate at least one type-specific state reduction value associated with the physical state based on (a) at least one response model for the at least one program type and (b) the at least one time value; amounts to mere instructions to apply an exception, where it recites an idea of a solution. The limitation does not indicate what the type-specific state reduction value is. See MPEP 2106.05(f)(1).
Additionally, the limitation calculate a total state reduction value based on the at least one type-specific state reduction value; amounts to mere instructions to apply an exception, where it recites the idea of a solution. The limitation does not indicate what a total state reduction value is. See MPEP 2106.05(f)(1).
Additionally, the limitation calculate at least one future intensity value based on the total state reduction value; and amounts to mere instructions to apply an exception, where it recites the idea of a solution. The limitation does not indicate what the future intensity value is. See MPEP 2106.05(f)(1).
Further, 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 element of the processor amounts to no more than mere instructions to apply the exception using a generic computer component that does not impose any meaningful limits on practicing the abstract idea and therefore cannot provide an inventive concept. See MPEP 2106.05(b).
The dependent claims include the same abstract ideas recited as recited in the independent claims and merely incorporate additional details that narrow the abstract ideas and fail to add significantly more to the claims.
Dependent claim 2 recites wherein the state model is associated with a lognormal distribution. This limitation covers a “Mathematical Concept” grouping of abstract ideas, as the limitation is describing a mathematical relationship with the distribution of the state model. See MPEP 2106.04(a)(2).
Dependent claim 3 recites wherein the activity dissipation model is a multivariate logit model. This limitation covers a “Mathematical Concept” grouping of abstract ideas, as the limitation is describing the model used for the activity dissipation model. See MPEP 2106.04(a)(2).
Dependent claim 4 recites wherein the activity magnitude model is a multivariate logit model. This limitation covers a “Mathematical Concept” grouping of abstract ideas, as the limitation is describing the model used for the activity dissipation model. See MPEP 2106.04(a)(2).
Dependent claim 5 recites wherein at least one of the state model, the activity dissipation model, or the activity magnitude model is associated with at least one of an extensive covariate model, a mixture model, a non-linear mixed effects model, a neural network, or a Bayesian model. This limitation covers a “Mathematical Concept” grouping of abstract ideas, as the limitation is describing a set of possible mathematical models for use in the modeling process. See MPEP 2106.04(a)(2).
Dependent claim 6 recites generating, via the processor, an observed activity impact dissipation parameter and an observed activity magnitude parameter based on the past state data and the past activity data, the observed activity impact dissipation parameter and the observed activity magnitude parameter associated with the user; and. Under the broadest reasonable interpretation, this limitation is a process step that covers performance in the human mind or with the aid of pencil and paper. As such, this limitation falls within the “Mental Process” grouping of abstract ideas.
Also, the claim limitation includes the additional element of a processor. The processor is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a computer and/or a generic computer component. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
Additionally, dependent claim 6 recites generating the covariance matrix based on the observed activity impact dissipation parameter and the observed activity magnitude parameter. This limitation covers a “Mathematical Concept” grouping of abstract ideas, as the limitation is describing the makeup of the covariance matrix. See MPEP 2106.04(a)(2).
Dependent claim 7 recites wherein the future state data includes at least one of a predicted state or a potential future treatment process. This limitation amounts to insignificant extra-solution activity of mere data gathering. See MPEP 2106.05(g).
Dependent claim 8 recites wherein: the past state data includes an indication of a physical state; and. This limitation is shown to reflect the court decisions of Versata Dev. Group, Inc. v SAP Am., Inc. iv Storing and retrieving information in memory, shown in MPEP 2106.05(d)(II).
Additionally, dependent claim 8 recites the past activity data includes at least one of an indication of a program time, an indication of program frequency, or an indication of a program duration. This limitation is shown to reflect the court decisions of Versata Dev. Group, Inc. v SAP Am., Inc. iv Storing and retrieving information in memory, shown in MPEP 2106.05(d)(II).
Dependent claim 9 recites wherein the at least one state impact value is associated with a reduction in a quantitative value associated with the past state data. This limitation covers a “Mathematical Concept” grouping of abstract ideas, as the limitation describes a quantitative reduction in the desired result. See MPEP 2106.04(a)(2).
Dependent claim 10 recites wherein the covariance matrix is associated with the at least one of a bimodal distribution or a multi-modal distribution. This limitation covers a “Mathematical Concept” grouping of abstract ideas, as the limitation describes the distribution of the covariance matrix. See MPEP 2106.04(a)(2).
Dependent claim 12 recites wherein the at least one response model includes longitudinal data. This limitation covers a “Mathematical Concept” grouping of abstract ideas, as the limitation covers the specific type of data used in at least one response model. See MPEP 2106.04(a)(2).
Dependent claim 13 recites wherein the at least one response model includes a lognormal distribution for state dynamics, a first multivariate logit model for dissipation of program effect, and a second multivariate logit model for program magnitude. This limitation covers a “Mathematical Concept” grouping of abstract ideas, as the limitation covers the distribution and model components required for the listed variables. See MPEP 2106.04(a)(2).
Dependent claim 14 recites wherein the code to cause the one or more processors to calculate the at least one type-specific state reduction value includes code to cause the one or more processors to: generate a first covariance matrix portion associated with the first multivariate logit model, a second covariance matrix portion associated with the second multivariate logit model, and a third covariance matrix portion associated with the lognormal distribution, the first covariance matrix portion, the second covariance matrix portion, and the third covariance matrix portion being diagonally arranged; and. This limitation covers a “Mathematical Concept” grouping of abstract ideas, as the limitation covers the use of covariance matrices and their arrangements within the established multivariate logit models. See MPEP 2106.04(a)(2).
Additionally, dependent claim 14 recites parameterize the at least one response model based on the first covariance matrix portion, the second covariance matrix portion, and the third covariance matrix portion. This limitation covers a “Mathematical Concept” grouping of abstract ideas, as the limitations covers the use of parameterization of the covariance matrices. See MPEP 2106.04(a)(2).
Also, the claim limitation includes the additional element of a processor. The processor is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a computer and/or a generic computer component. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 15 recites wherein the instructions further include code to cause the one or more processors to improve a program based on the at least one future intensity value, at least one constraint, and reinforcement learning. This limitation amounts to mere instructions to apply an exception, where it invokes computers or other machinery merely as a tool to perform an existing process. The inclusion of reinforcement learning to improve a program would fall under this category. See MPEP 2106.05(f)(2).
Also, the claim limitation includes the additional element of a processor. The processor is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a computer and/or a generic computer component. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 16 recites wherein the at least one response model includes at least one of a mixture model, a non-linear mixed effects model, a recurrent neural network, a long short-term memory (LSTM) network, a Bayesian network, a Bayesian structural time series, or a graph neural network. This limitation covers a “Mathematical Concept” grouping of abstract ideas, as the limitation covers the use of mathematical models for use of the response model. See MPEP 2106.04(a)(2).
Dependent claim 17 recites wherein the code further comprises code to cause the one or more processors to generate a process based on the at least one future intensity value, the process configured to improve the physical state. Under the broadest reasonable interpretation, this limitation is a process step that covers performance in the human mind or with the aid of pencil and paper. As such, this limitation falls within the “Mental Process” grouping of abstract ideas.
Also, the claim limitation includes the additional element of a processor. The processor is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a computer and/or a generic computer component. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 18 recites wherein the code further comprises code to cause the one or more processors to generate the process based on at least one of a user availability and a cost. Under the broadest reasonable interpretation, this limitation is a process step that covers performance in the human mind or with the aid of pencil and paper. As such, this limitation falls within the “Mental Process” grouping of abstract ideas.
Also, the claim limitation includes the additional element of a processor. The processor is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a computer and/or a generic computer component. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 19 recites wherein the at least one intensity value, the at least one time value, and the at least one future intensity value are each arranged sequentially. Under the broadest reasonable interpretation, this limitation is a process step that covers performance in the human mind or with the aid of pencil and paper. As such, this limitation falls within the “Mental Process” grouping of abstract ideas.
Dependent claim 20 recites wherein the at least one time value includes an indication of at least one of a program date, a program duration, a program frequency, or a program interval. This limitation is shown to reflect the court decisions of Versata Dev. Group, Inc. v SAP Am., Inc. iv Storing and retrieving information in memory, shown in MPEP 2106.05(d)(II).
Accordingly, claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e. an abstract idea) without anything significantly more.
Claim Rejections - 35 USC § 103
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 (i.e., changing from AIA to pre-AIA ) 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.
Claims 1, 5-9, 11-12, and 15-20 are rejected under 35 U.S.C. 103 as being unpatentable over Goldner et al. “Predictive Guidance Systems For Personalized Health And Self-Care, And Associated Methods” (US20210391081A1) in view of Lefkofsky et al. “Method And Process For Predicting And Analyzing Patient Cohort Response, Progression, And Survival” (US20210125731A1).
Regarding claim 1, Goldner teaches receiving, at a processor, (1) past state data associated with a user and (2) past activity data associated with the user; ([0036], “the analyzing devices 102 can analyze the obtained input data, including historical data, current real-time data, continuously supplied data, and/or any other data (e.g., using a statistical analysis, machine learning analysis, etc.), and generate output data. The output data can include predictions of a patient's health state, interpretations, recommendations, notifications, instructions, support, and/or other information related to the obtained input data. The analyzing devices 102 can perform such analyses at any suitable frequency and/or any suitable number of times (e.g., once, multiple times, on a continuous basis, etc.).”)
generating, via the processor, (1) a covariance matrix based on a state model, ([0056], “A covariance matrix of features can be computed for each of the clusters.”)
an activity dissipation model, and ([0049], “For example, if the system recommends that the patient exercise, the system can quickly adapt based on the patient's subsequent actions. If no exercise takes place, the system can use the new information that exercise is not likely to happen at the moment, and recalculate to develop a new estimate of what mode of self-care will be most effective (e.g., a mode that is more likely to be used). If exercise does take place, the monitoring activity can provide data suggesting how much exercise is happening and/or how often, and that information can be used to determine whether the focus of support should stay with exercise, or move on to other modes (e.g., diet or sleep).”
Examiner’s Note: As per the Specification, the activity dissipation model is described as “can be configured to represent a diminished effect that an activity has on a state”. As per the given definition and the cited material above, a POSITA could reasonably link the two together, especially given the fact that both sources are within the healthcare field.)
an activity magnitude model, ([0068], “in some embodiments, expected results are evaluated in such a way that there is a trade-off between achieving a better outcome (e.g. lower blood glucose concentration, lower blood pressure, lower weight, or improved kidney function, improved heart function, etc.) verses how long it takes to achieve. (e.g., a 10 mg/dL reduction in 4 months is preferable to an 11 mg/dL reduction that takes 8 years.)”
Examiner’s Note: As per the Specification, the activity magnitude model is described as “can be configured to represent the magnitude and duration of the impact of an activity”. A POSITA would be able to link this definition with the example listed above, as the cited example deals with the dosage reduction over time in regards to a change in dosage reduction, which could be seen as an activity magnitude.)
associated with a group including the user, ([0025] and [0051], “The health state can be any status, condition, parameter, etc. that is associated with or otherwise related to the patient's health. In some embodiments, the system 100 can be used to identify, manage, monitor, and/or provide recommendations relating to diabetes, hypoglycemia, hyperglycemia, pre-diabetes, hypertension, hyperlipidemia, ketoacidosis, liver failure, congestive heart failure, hypoxia, kidney function, intoxication, dehydration, hyponatremia, shock, sepsis, trauma, water retention, bleeding, endocrine disorders, asthma, lung conditions, muscle breakdown, malnutrition, body function (e.g., lung functions, heart functions, etc.), physical performance (e.g., athletic performance), anaerobic activity, weight loss/gain, nutrition, wellness, mental health, focus, effects of medication, medication levels, health indicators, and/or user compliance. “, “At step 220, the Monitoring System is used to collect a large volume of data. For example, the Monitoring System can collect over 24 billion data points from millions of users in all 195 countries. The Monitoring System can be used to collect various different types of data from the connected devices as described above. The collected data can be processed (e.g., grouped or combined via a predetermined mechanism) and stored in the system 100 for subsequent use.”)
(2) at least one activity impact dissipation parameter associated with the user, and ([0049], “For example, if the system recommends that the patient exercise, the system can quickly adapt based on the patient's subsequent actions. If no exercise takes place, the system can use the new information that exercise is not likely to happen at the moment, and recalculate to develop a new estimate of what mode of self-care will be most effective (e.g., a mode that is more likely to be used). If exercise does take place, the monitoring activity can provide data suggesting how much exercise is happening and/or how often, and that information can be used to determine whether the focus of support should stay with exercise, or move on to other modes (e.g., diet or sleep).”
Examiner’s Note: As seen in the example, based on whether the user exercises, the system adjusts how it affects the ongoing support needed. A POSITA would be able to infer that this could reasonably represent an activity impact dissipation parameter.)
(3) at least one activity magnitude parameter associated with the user; ([0049], “For example, if the system recommends that the patient exercise, the system can quickly adapt based on the patient's subsequent actions. If no exercise takes place, the system can use the new information that exercise is not likely to happen at the moment, and recalculate to develop a new estimate of what mode of self-care will be most effective (e.g., a mode that is more likely to be used). If exercise does take place, the monitoring activity can provide data suggesting how much exercise is happening and/or how often, and that information can be used to determine whether the focus of support should stay with exercise, or move on to other modes (e.g., diet or sleep).”
Examiner’s Note: Similar to the explanation given for the activity impact dissipation parameter, a POSITA could reasonably link the use and frequency of exercise to determine the best mode of self-care in conjunction with the other modes.)
Generating at least one state impact value based on the at least one activity impact dissipation parameter, and the at least one activity magnitude parameter; ([0049], “The adaptation process can include repeating one or more of the prediction, evaluation, support, and/or monitoring processes discussed above to update and/or modify the self-care recommendations for the patient. One or more of these processes can be continually performed, such that the adaptation process is ongoing and responds in real-time to the patient's actions.”)
Goldner does not recite generating a non-linear regression model associated with a future state based on the past state data, the past activity data, and the at least one state impact value; calculating, via the processor, future state data based on the non-linear regression model; and transmitting, via the processor, at least one signal configured to cause display of the future state data to the user.
Lefkofsky teaches generating a non-linear regression model associated with a future state based on the past state data, the past activity data, and the at least one state impact value; ([0280], [0061], [0165], “Machine learning models may apply linear regression algorithms, non-linear regression, logistic regression algorithms, classification models, bootstrap resampling models, subset selection models, dimensionality reduction models, tree-based models (such as bagging, boosting, and random forest), and other supervised or unsupervised models. “, “For example, a user may select for the model to identify which features most closely correlate to patient survival in the cohort, or which features most closely correlate with a positive treatment outcome in the cohort. The user may also select which classification labels from the classification labels of the model that they wish the model to classify.”, “The patient data store 14 may be a pre-existing dataset that includes patient clinical history, such as demographics, comorbidities, diagnoses and recurrences, medications, surgeries, and other treatments along with their response and adverse effects details.”, “In general both temporal and static features may be derived from the patient data but the analysis at this stage is purely backward-looking to avoid leaking future information. Different categories or classes of features include: “time since last/first XXX”; “number of XXX”; or “demographics.” Extracting features may include multiple lookback horizons, for example features may be bounded to the trailing 12 months or may be based on continuous historic analysis.”
Examiner’s Note: Lefkofsky is able to teach everything listed in the previous limitations except for the covariance matrix.)
calculating, via the processor, future state data based on the non-linear regression model; and ([0280], “For example, a user may select for the model to identify which features most closely correlate to patient survival in the cohort, or which features most closely correlate with a positive treatment outcome in the cohort. The user may also select which classification labels from the classification labels of the model that they wish the model to classify.”
Examiner’s Note: While a non-linear regression model was not directly stated, it was established earlier that a non-linear regression model could be a model used in the analysis.)
transmitting, via the processor, at least one signal configured to cause display of the future state data to the user. (Figure 1 and [0060], “Returning to FIG. 1, the back end layer 12 also may include a distributed computing and modeling layer 38 that receives data from the patient cohort timeline data storage 18 to provide inputs to a plurality of modules, including, a time to event modeling module 40 that powers the patient survival analysis user interface 30, an event likelihood module 42 that calculates the likelihood of one or more events received at the patient event likelihood analysis user interface 32 for subsequent display in that user interface, a next event modeling module 44 that generates models of one or more next events for subsequent display at the patient next event analysis user interface 34, and one or more future modeling modules 46 that generate one or more future models for subsequent display at the one or more patient future analysis user interfaces 36.”)
It would have been obvious to one of ordinary skill in the art, before the effective filing date, to combine the teachings of Goldner with Lefkofsky, as both references deal with modeling patient health tracking systems designed to assist with achieving a health-oriented goal. Lefkofsky would modify Goldner by providing the use of non-linear regression models alongside the calculation of future state data and a more thorough display of the future state data to the user. The benefit of doing so is opening up more models to be used in the building of the prediction system as well as the ability to calculate and display future state data and predictions using said models, allowing for greater model choice in the system building and training process.
Regarding claim 5, the combination of Goldner and Lefkofsky teaches the limitations of claim 1. Goldner does not explicitly recite wherein at least one of the state model, the activity dissipation model, or the activity magnitude model is associated with at least one of an extensive covariate model, a mixture model, a non-linear mixed effects model, a neural network, or a Bayesian model.
Lefkofsky teaches wherein at least one of the state model, the activity dissipation model, or the activity magnitude model is associated with at least one of an extensive covariate model, a mixture model, a non-linear mixed effects model, a neural network, or a Bayesian model. ([0085], “Artificial intelligence models referenced herein may be gradient boosting models, random forest models, neural networks (NN), regression models, Naive Bayes models, or machine learning algorithms (MLA).”)
See motivation of claim 1
Regarding claim 6, the combination of Goldner and Lefkofsky teaches the limitations of claim 1. Goldner teaches wherein the generating the covariance matrix includes: generating, via the processor, an observed activity impact dissipation parameter and an observed activity magnitude parameter based on the past state data and the past activity data, the observed activity impact dissipation parameter and the observed activity magnitude parameter associated with the user; and ([0049], “The adaptation process can include repeating one or more of the prediction, evaluation, support, and/or monitoring processes discussed above to update and/or modify the self-care recommendations for the patient. One or more of these processes can be continually performed, such that the adaptation process is ongoing and responds in real-time to the patient's actions. For example, if the system recommends that the patient exercise, the system can quickly adapt based on the patient's subsequent actions. If no exercise takes place, the system can use the new information that exercise is not likely to happen at the moment, and recalculate to develop a new estimate of what mode of self-care will be most effective (e.g., a mode that is more likely to be used). If exercise does take place, the monitoring activity can provide data suggesting how much exercise is happening and/or how often, and that information can be used to determine whether the focus of support should stay with exercise, or move on to other modes (e.g., diet or sleep).”,
Examiner’s Note: As established earlier, a POSITA would be able to ascertain that an activity impact dissipation parameter and activity magnitude parameter would be able to be derived from the health data and history of the user, and since the Specification does not give a concrete explanation to what “observed” does in the scope of this limitation, the examiner has interpreted it to assume a parameter that is presently impacting the system and its ability to operate. The example above utilizing exercise impacting the usefulness of other modes could easily be seen as impacting the system and changing how it operates in predicting the best mode of self-care.)
Goldner also teaches generating the covariance matrix based on the observed activity impact dissipation parameter and the observed activity magnitude parameter. ([0054]-[0055] and [0058], “All behavior and signals can have a multitude of features that go into the outcome models. The Monitoring System can cluster the multitude of features by various categories such as based on A) the features are interdependent and cannot be changed without affecting other features, B) reducing the state space of possible modes that can be changed, or C) specific coherent self-care modes that the user improve in, where those related features reflect the overall changes that occur in user's behavior in that field.”, “To cluster the features, the Monitoring System can compute a distance matrix based on the variation of information of the features.”, “for each cluster representing a self-care mode, change the values of the representative feature, followed by the corresponding change in all other features in the cluster (e.g., reduce logged carbohydrate values for the most recent 30 days by the 5%, or increase recorded physical activity over the past 30 days by 10%) thus, using the covariance matrices for the clusters, re-create the input features for the Outcomes Model hypothetically improved data values, and recalculate the forecast from the revised inputs;”)
Regarding claim 7, the combination of Goldner and Lefkofsky teaches the limitations of claim 1. Goldner teaches wherein the future state data includes at least one of a predicted state or a potential future treatment process ([0041], “In some embodiments, the state estimator 114 can generate activity predictions for a predetermined future duration based on the obtained data. The state estimator 114 can start from a current health state (e.g., blood glucose level) and extrapolate or derive future health states according to the activity predictions. The system 100 can use the future health states to determine and recommend a set of user actions that may be implemented between now and a future time to avoid the thresholding health states and/or to conform to a targeted behavior (e.g., a repeated set of user actions).”)
Regarding claim 8, the combination of Goldner and Lefkofsky teaches the limitations of claim 1. Goldner teaches wherein: the past state data includes an indication of a physical state; and ([0085], “As discussed above, the health data (e.g., the new data and/or the user history 124) can include any data relevant to the user's health state. Examples of health data include, but are not limited to, any of the following: blood pressure data (e.g., current and/or previous measurements of systolic and/or diastolic blood pressure), blood glucose data (e.g., current and/or previous blood glucose measurements, current and/or previous HbA1c data values), heart rate data, food data (e.g., number of meals; timing of meals; number of calories; amount of carbohydrates, fats, sugars, etc.), medical history data (e.g., current and/or previous weight, height, BMI, age, sleeping patterns, medical conditions, cholesterol levels, diabetes type, family history, user health history, diagnoses, blood pressure, etc.)”)
Goldner also teaches “the past activity data includes at least one of an indication of a program time, an indication of program frequency, or an indication of a program duration.” ([0024], “The user devices 104 can also obtain contextual data, such as user calendar (including, e.g., event name or category, location, date, time, participant, etc.), that may be used to categorize other obtained data. For example, the contextual data may be used to categorize the data into user activities or user health states.”)
Regarding claim 9, the combination of Goldner and Lefkofsky teaches the limitations of claim 1. Goldner teaches wherein the at least one state impact value is associated with a reduction in a quantitative value associated with the past state data ([0058], “for each cluster representing a self-care mode, change the values of the representative feature, followed by the corresponding change in all other features in the cluster (e.g., reduce logged carbohydrate values for the most recent 30 days by the 5%, or increase recorded physical activity over the past 30 days by 10%) thus, using the covariance matrices for the clusters, re-create the input features for the Outcomes Model hypothetically improved data values, and recalculate the forecast from the revised inputs;”)
Regarding claim 11, the combination of Goldner and Lefkofsky teaches the limitations of the claim. Goldner teaches receive (1) at least one intensity value associated with a physical state and (2) at least one time value associated with at least one program type; ([0026] and [0047], “Health-related information can also include physiological and/or behavioral parameters, such as vitals (e.g., heart rate, body temperature (such as skin temperature), blood pressure (such as systolic and/or diastolic blood pressure), respiratory rate), cardiovascular data (e.g., pacemaker data, arrhythmia data), body function data, meal or nutrition data (e.g., number of meals; timing of meals; number of calories; amount of carbohydrates, fats, sugars, etc.), physical activity or exercise data (e.g., time and/or duration of activity; activity type such as walking, running, swimming; strenuousness of the activity such as low, moderate, high; etc.), sleep data (e.g., number of hours of sleep, average hours of sleep, variability of hours of sleep, sleep-wake cycle data, data related to sleep apnea events, sleep fragmentation (such as fraction of nighttime hours awake between sleep episodes, etc.), stress level data (e.g., cortisol and/or other chemical indicators of stress levels, perspiration), a1c data, etc.”, “The monitoring process can include continuously collecting information about each user's program engagement, behaviors, biometrics, and/or outcomes. Some examples of the user's program engagement can include application use, interactions with coaches, logging, use of educational resources such as lesson completion, interactions with peer support networks, and others. Some examples of the collected behavior information can include physical activity, sleep, food and mealtimes, consistency of medication use, location, and the like. Some examples of the collected biometrics can include heart rate, blood pressure, skin temperature, blood glucose and other subcutaneous analytes, and others. Some examples of the collected outcomes can include: statistics of blood glucose concentration such as time in range, HbA1c, or average blood glucose; statistics of blood pressure including average and variability; statistics of cholesterol levels; other health statistics including weight, BMI, waist circumference and others; statistics of health markers for conditions including many cancers, heart disease, or others.”)
Calculate at least one type-specific state reduction value associated with the physical state based on (a) at least one response model for the at least one program type and (b) at least one time value ([0080]-[0081], “In some embodiments, recommendations can be provided to the user's coach, who can then explain the importance of the recommendation to the user, and support the user to follow a program emphasizing the recommended mode of self-care. Alternatively or in combination, the system can use the recommendation from the Optimizer to select an automated program designed to promote a particular behavior. For example, if the optimal self-care intervention for a particular user is to work on reducing blood pressure, an Adaptive Support Model can be established for the user, designed to improve consistency of self-monitoring of blood pressure and/or of any other habit related to reduction of blood pressure.”, “Once the system has been built as described above with respect to steps 210-250, the system can be put into practice. For example, upon starting with the system, an individual (“the person”) can communicate which chronic condition(s) are being managed. The system, which may be automated or may be implemented as a human coach, sets an appropriate outcomes goal for the person, and the person begins using the app. Data monitoring and aggregation collects past recorded data shared by the person, ongoing passively collected data, and/or data logged by the person. Periodically, the Outcomes Model is used to forecast the person's outcomes metric, and the Optimizer is used to select the most effective self-care mode to focus on.”)
Goldner does not explicitly recite calculate a total state reduction value based on the at least one type-specific state reduction value;
Calculate at least one future intensity value based on the total state reduction value; and
Cause display of the at least one future intensity value.
Lefkofsky teaches calculate a total state reduction value based on the at least one type-specific state reduction value; ([0167], “Based on the data for each of the patients and for each patient timepoint, an outcome target for an outcome event may be calculated within a horizon time window; a plurality of prior features may be identified; and a state of each of the plurality of prior features at the patient timepoint may be determined.”)
Calculate at least one future intensity value based on the total state reduction value; and ([0167], “An outcome event may include a state of the patient and/or the disease, such as progression or death, and the outcome target may be described with a target label such as a yes or no indication of whether the outcome will occur within a particular horizon time window from the patient timepoint/anchor point, along with a date of the endpoint. The horizon time window may include any suitable periods of time such as 3 months, 6 months, 9 months, 12 months, 24 months, 36 months, 48 months, or 60 months, or other periods of time.”)
Cause display of the at least one future intensity value ([0060], “Returning to FIG. 1, the back end layer 12 also may include a distributed computing and modeling layer 38 that receives data from the patient cohort timeline data storage 18 to provide inputs to a plurality of modules, including, a time to event modeling module 40 that powers the patient survival analysis user interface 30, an event likelihood module 42 that calculates the likelihood of one or more events received at the patient event likelihood analysis user interface 32 for subsequent display in that user interface, a next event modeling module 44 that generates models of one or more next events for subsequent display at the patient next event analysis user interface 34, and one or more future modeling modules 46 that generate one or more future models for subsequent display at the one or more patient future analysis user interfaces 36.”)
See motivation of claim 1
Regarding claim 12, the combination of Goldner and Lefkofsky teaches the limitations of claim 11. Goldner teaches wherein the at least one response model includes longitudinal data ([0042], “The system can continuously update and/or adapt the self-care recommendations, e.g., based on continuously-collected data from the particular patient as well as data from a plurality of other patients. The system can guide individuals toward self-care changes that are likely to improve their chronic health conditions, support them in making those changes, and adapt and/or update continuously over time.”)
Regarding claim 15, the combination of Goldner and Lefkofsky teaches the limitations of claim 11. Goldner teaches wherein the instructions further include code to cause the one or more processors to improve a program based on the at least one future intensity value, at least one constraint, and reinforcement learning ([0070], “In some embodiments, the steps for making a decision using such reinforcement learning system with DQN are: 1) observe the current state of the user (from representative features' values); 2) get the deep-network's assessment of the Q-values for the different actions; 3) select the action that yields the best value. Such selection can be done stochastically with respect to the relative estimated Q-values. In some embodiments, For the model to keep learning the best action over time, and to tie it with the predictions of the Outcome Models, two more steps can be required: 1) record as immediate reward the change in outcome from the last decision time; and 2) add an example for the network to learn from, with the previous state as the input, the estimated Q-values as the output for all actions that were not selected, and Q=r+γ·Qe(s, α) where r is the observed reward, α the action that corresponds to the intervention taken in previous step, and Qe(s,α) is the estimated Q-value for the selected action.”)
Regarding claim 16, the combination of Goldner and Lefkofsky teaches the limitations of claim 11. Goldner teaches wherein the at least one response model includes at least one of a mixture model, a non-linear mixed effects model, a recurrent neural network, a long short-term memory (LSTM) network, a Bayesian network, a Bayesian structural time series, or a graph neural network ([0037], “In some embodiments, the analyzing devices 102 is configured to analyze the input data and generate the output data using one or more machine learning models 122. (…)Bayesian algorithms (e.g., naïve Bayes, Gaussian naïve Bayes, multinomial naïve Bayes, averaged one-dependence estimators, Bayesian belief networks, Bayesian networks), clustering algorithms (e.g., k-means, k-medians, expectation maximization, hierarchical clustering), association rule learning algorithms (e.g., apriori algorithm, ECLAT algorithm), artificial neural networks (e.g., perceptron, multilayer perceptrons, back-propagation, stochastic gradient descent, Hopfield networks, radial basis function networks), deep learning algorithms (e.g., convolutional neural networks, recurrent neural networks, long short-term memory networks, stacked auto-encoders, deep Boltzmann machines, deep belief networks)”)
Regarding claim 17, the combination of Goldner and Lefkofsky teaches the limitations of claim 11. Goldner teaches wherein the code further comprises code to cause the one or more processors to generate a process based on the at least one future intensity value, the process configured to improve the physical state. ([0045], “The Outcomes Model can be used in combination with an optimization algorithm (“Optimizer”) to determine an improvement in outcome (relative to a current forecast of the patient's outcome) that may be achieved by improving and/or changing various modes of self-care. This information can be used to determine the mode of self-care that, at the current time, is likely to provide the best long-term improvement in outcomes.”)
Regarding claim 18, the combination of Goldner and Lefkofsky teaches the limitations of claim 17. Goldner teaches wherein the code further comprises code to cause the one or more processors to generate the process based on at least one of a user availability and a cost. ([0129], “The healthcare guidance system 1410 can collect health user data of the user 1402 a and can identify a health metric based on the health data. The system 1410 then identifies a self-care mode from a plurality of self-care modes by, for example, analyzing a plurality of available self-care modes, each having one or more contributing factors, determining one or more predictions of the health care metric based on the self-care modes, identifying contribution factors of the self-care modes for the predictions, and selecting a self-care mode from the plurality of available self-care modes based on the contribution factors.”, “Self-care modes can be recommended continuously or periodically at regular or irregular intervals or in response to an event (e.g., unwanted predicted event, user inputted event, etc.”
Examiner’s Note: The specification does not elaborate on what is considered user availability and cost, so the examiner has interpreted both as referring to any reason why the system could not select a certain self-care mode, such as a health limitation or outside event.)
Regarding claim 19, the combination of Goldner and Lefkofsky teaches the limitations of claim 11. Goldner does not explicitly recite wherein the at least one intensity value, the at least one time value, and the at least one future intensity value are each arranged sequentially.
Lefkofsky teaches wherein the at least one intensity value, the at least one time value, and the at least one future intensity value are each arranged sequentially. ([0161], “The prediction model may be implemented based on data from a large number of patients, using information about the patients' medical history and treatments along with information about their survival. In order to chronologically align the data from numerous patients, one or more anchor points (also referred to as “patient timepoints”) may be identified within the data (FIG. 26). The anchor points identify points in time that may be common to all or at least many of the patients and which may help to standardize the time course of the data relative to events such as disease progression. The anchor points may include events such as time of first diagnosis, time of first metastasis, or time of first treatment, although other anchor point events are also possible. FIG. 26 shows an alignment of timelines for patients P1, P2, P3, . . . , Pn based on a common anchor event.”)
Regarding claim 20, the combination of Goldner and Lefkofsky teaches the limitations of claim 11. Goldner does not explicitly recite wherein the at least one time value includes an indication of at least one of a program date, a program duration, a program frequency, or a program interval.
Lefkofsky teaches wherein the at least one time value includes an indication of at least one of a program date, a program duration, a program frequency, or a program interval. ([0164], “The data may be processed to identify a plurality of patient timepoints (anchor points) that occur within the period of time covered by each patient's data. As discussed above, the anchor points or patient timepoints may include timepoints associated with any patient interaction with the medical system, including any interaction with an individual or facility that provides medical care or obtains medical information such as a care provider, a genetic sequencing organization, a hospital outpatient or inpatient facility, etc. The patient timepoints may be identified by a date attached to or associated with each piece of data in the received set of patient data.”)
Claim 2 is rejected as being unpatentable over Goldner in view of Lefkofsky in further view of Solomon et al. “System And Method For Mechanical Failure Classification, Condition Assessment And Remediation Recommendation” (US20210388950A1).
Regarding claim 2, the combination of Goldner and Lefkofsky teaches the limitations of claim 1. Both Goldner and Lefkofsky do not teach wherein the state model is associated with a lognormal distribution.
Solomon teaches wherein the state model is associated with a lognormal distribution. ([0193], “The statistical model may be a lognormal model (e.g., a lognormal distribution graph 800)”)
It would have been obvious to one of ordinary skill in the art, before the effective filing date, to combine the teachings of Goldner and Lefkofsky with Solomon as the references deal with the use of models and the properties associated with them. Solomon would modify Goldner and Lefkofsky by explicitly defining the use of lognormal distributions within model usage. The benefit of doing so is that this establishes lognormal distribution within the parameters used within the state model, allowing researchers and analysts to assume that quantitative data should follow a lognormal distribution.
Claims 3 and 4 are rejected under 35 U.S.C. 103 as being unpatentable over Goldner in view of Lefkofsky in further view of Wang et al. “Method For Identifying High-Risk AML Patients” (US20210371937A1).
Regarding claim 3, the combination of Goldner and Lefkofsky teaches the limitations of claim 1. Both Goldner and Lefkofsky do not explicitly recite wherein the activity dissipation model is a multivariate logit model.
Wang teaches wherein the activity dissipation model is a multivariate logit model. ([0128], “In analyses assessing prediction of treatment response, uni- and multi-variate logistic regression models were used with the bootstrap-adjusted AUROC metric to determine the ability of various clinical parameters and GE scores to predict initial induction response.”)
Regarding claim 4, the combination of Goldner and Lefkofsky teaches the limitations of claim 1. Both Goldner and Lefkofsky do not explicitly recite wherein the activity magnitude model is a multivariate logit model.
Wang teaches wherein the activity magnitude model is a multivariate logit model. ([0128], “In analyses assessing prediction of treatment response, uni- and multi-variate logistic regression models were used with the bootstrap-adjusted AUROC metric to determine the ability of various clinical parameters and GE scores to predict initial induction response.”
Examiner’s Note: multivariate logit model is another way of phrasing multivariate logistic regression model.)
It would have been obvious to one of ordinary skill in the art, before the effective filing date, to combine the teachings of Goldner and Lefkofsky with Wang, as the references deal with building healthcare related models to help patients find appropriate treatment. Wang would modify Goldner and Lefkofsky by including the use of multivariate logistic regression models as possible model choices, as Goldner and Lefkofsky only specify logistic regression models as being possible choices. The benefit of doing so is that multivariate logit models can handle multiple independent and dependent variables at the same time, meaning that the outcome can be more complex and comprehensive than with a standard logistic regression model. (Hidalgo et al. “Multivariate or Multivariable Regression”).
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Goldner in view of Lefkofsky in further view of Sengupta et al. “Video Based Matching And Tracking By Analyzing One Or More Image Abstractions” (US8600172B2).
Regarding claim 10, the combination of Goldner and Lefkofsky teaches the limitations of claim 1. Both Goldner and Lefkofsky do not explicitly recite wherein the covariance matrix is associated with the at least one of a bimodal distribution or a multi-modal distribution.
Sungupta teaches wherein the covariance matrix is associated with the at least one of a bimodal distribution or a multi-modal distribution. (Section 11, “instead of fusing the abstraction of the image patch to the abstraction of the object to maintain one single statistical description, a “multimodal” distribution may be used, such as a multimodal Gaussian distribution for the underlying BoF model of the object. In this way, an abstraction of an object may have several underlying Gaussian distributions, with each such distribution having an underlying mean and covariance matrix associated with it.”)
It would have been obvious to one of ordinary skill in the art, before the effective filing date, to combine the teachings of Goldner and Lefkofsky with Sengupta, as the references deal with the properties of covariance matrices in relation to the data. Sengupta would modify Goldner and Lefkofsky by clarifying the use of multimodal distributions within covariance matrices. The benefit of doing so is that it allows multimodal data to be better represented within the covariance matrices, allowing for a more thorough analyzation of the data.
Claim 13 is rejected under 35 U.S.C. 103 as being unpatentable over Goldner in view of Lefkofsky in further view of Solomon and Wang.
Regarding claim 13, the combination of Goldner and Lefkofsky teaches the limitations of claim 11. Goldner and Lefkofsky does not explicitly recite wherein the at least one response model includes a lognormal distribution for state dynamics, a first multivariate logit model for dissipation of program effect, and a second multivariate logit model for program magnitude.
Solomon teaches wherein the at least one response model includes a lognormal distribution for state dynamics, (Figure 8 and [0193], “The statistical model may be a lognormal model (e.g., a lognormal distribution graph 800).”
Examiner’s Note: The Specification does not give a definition for the use of “state dynamics” within the context of the case, as such the examiner is using the definition of “how the internal variables of a system change, evolve, and transition from one condition to another over time.” As presented in the distribution graph, the lognormal distribution changes its fit as the pitting factor increases. This qualifies as an example of a state dynamic. In addition, a POSITA would surmise that a lognormal distribution would change depending on the length of time or program status.)
It would have been obvious to one of ordinary skill in the art, before the effective filing date, to combine the teachings of Goldner and Lefkofsky with Solomon as the references deal with the use of models and the properties associated with them. Solomon would modify Goldner and Lefkofsky by explicitly defining the use of lognormal distributions within model usage and as an example of usage within state dynamics. The benefit of doing so is that this establishes lognormal distribution within the parameters used within the state model, allowing researchers and analysts to assume that quantitative data should follow a lognormal distribution, as well as further defining the use of lognormal distributions when describing state dynamics.
Wang teaches a first multivariate logit model for dissipation of program effect, and a second multivariate logit model for program magnitude ([0128], “In analyses assessing prediction of treatment response, uni- and multi-variate logistic regression models were used with the bootstrap-adjusted AUROC metric to determine the ability of various clinical parameters and GE scores to predict initial induction response.”
Examiner’s Note: The Specification does not explain what a dissipation of program effect or what a program magnitude refers to in context to the case, so the examiner has interpreted them to be related to the activity dissipation and activity magnitude models explained earlier in claims 3 and 4 respectively.)
It would have been obvious to one of ordinary skill in the art, before the effective filing date, to combine the teachings of Goldner and Lefkofsky with Wang, as the references deal with building healthcare related models to help patients find appropriate treatment. Wang would modify Goldner and Lefkofsky by including the use of multivariate logistic regression models as possible model choices, as Goldner and Lefkofsky only specify logistic regression models as being possible choices. The benefit of doing so is that multivariate logit models can handle multiple independent and dependent variables at the same time, meaning that the outcome can be more complex and comprehensive than with a standard logistic regression model. (Hidalgo et al. “Multivariate or Multivariable Regression”).
Claim 14 is rejected under 35 U.S.C. 103 as being unpatentable over Goldner, Lefkofsky, Solomon, and Wang in view of IEEE TechnologyNavigator “Covariance Matrices” and Sarabandi et al. “Automotive Radar Scene Simulator” (US20200111382A1).
Regarding claim 14, the combination of Goldner, Lefkofsky, Solomon, and Wang teaches the limitations of claim 13. Goldner teaches parameterize the at least one response model based on the first covariance matrix portion, the second covariance matrix portion, and the third covariance matrix portion. ([0056], “The Monitoring System use an implied conditional independence and normality assumption within each cluster to extract a change in a feature Y given a change of rX in another from feature X: rY=rX·CYXCXX −1, where C is the covariance matrix for a cluster and CYX is the corresponding entry for features X and Y. Modelling the changes of related features with less conditional independencies implied is also possible, and can require a more computationally involved inference. The Monitoring System can select a single feature from each cluster, preferably one feature that contributes the most to the predictions of the outcome model. This can be established via sensitivity analysis and measuring contribution via Shapley values or similar common estimation methods.”)
Goldner, Lefkofsky, Solomon, and Wang does not explicitly recite generate a first covariance matrix portion associated with the first multivariate logit model, a second covariance matrix portion associated with the second multivariate logit model, and a third covariance matrix portion associated with the lognormal distribution, the first covariance matrix portion, the second covariance matrix portion, and the third covariance matrix portion being diagonally arranged; and
IEEE teaches generate a first covariance matrix portion associated with the first multivariate logit model, a second covariance matrix portion associated with the second multivariate logit model (“Covariance matrices are square symmetric matrices that encode the second-order statistical relationships among a set of random variables.”, “The matrix summarizes the multivariate spread of a random vector in a compact algebraic form that is directly amenable to decomposition and manipulation.”, “Knowledge of a full covariance matrix, rather than individual pairwise covariances, is needed whenever the joint behavior of multiple variables matters.”)
IEEE also teaches the first covariance matrix portion, the second covariance matrix portion, and the third covariance matrix portion being diagonally arranged; and (“Each diagonal entry gives the variance of a single variable, while each off-diagonal entry records the covariance between a pair of variables, measuring how much they tend to move together.”
Examiner’s Note: Based on the examiner’s interpretation, the majority of this claim recites mathematical properties relating to covariance matrices. The covariance matrices have a direct association with a multivariate spread of data as long as said data is longitudinal in nature, which has been established as a necessity for building a health monitoring system as described within the case. In addition, due to covariance matrices always being square to represent the number of variables assigned to it, the covariance matrices would be diagonally arranged due to the inherit structure of them. The examiner is not sure how these claim limitations are meant to represent anything other than an understanding of the properties and relations of covariance matrices.)
Sarabandi teaches a third covariance matrix portion associated with the lognormal distribution, ([0043], “In another example, the parametric statistical model may be defined by a mean, the standard deviation, and a covariance matrix of an n-dimensional distribution function, where the distribution function is one of an exponential distribution function, a Lognormal distribution function or a Weibull distribution function.”)
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Neumann “Systems And Methods For Selecting A Treatment Schema Based On User Willingness” (US20210098099A1): Also teaches building a system designed to help treat or correct a patient’s problem(s).
Aimone et al. “System And Method For Brain Modelling” (US20200337625A1): Also teaches using user data to create a biological model.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALEXANDER JOSEPH MEKOVSKY whose telephone number is (571)272-0428. The examiner can normally be reached 9am-6pm M-F.
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, Ryan Pitaro can be reached at (571) 272-4071. 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.
/ALEXANDER JOSEPH MEKOVSKY/ Examiner, Art Unit 2188
/RYAN F PITARO/Supervisory Patent Examiner, Art Unit 2188