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 .
Notice to Applicant
This communication is in response to the amendment filed 12/1/2025. Claims 1, 4-6, 8, 9, 13, 14, and 18-20 have been amended. Claims 1-6, 8-14, and 16-20 remain pending and have been examined.
Response to Arguments
A. Applicant's arguments with respect to the rejection of claims 1-6, 8-14, and 16-20 under 35 USC 101 have been fully considered but they are not persuasive.
Applicant argues starting on page 8 of the response that claim 1 does not recite a method of organizing human activity in that it does not recite a personal behavior or a relationship or interaction between people.
Applicant argues starting on page 8 that claim 1 “provides specific elements, such as the use of a "distillation model", which could not be practically performed in the human mind,” and that the statement provided in Step 2A Prong 1 that the enumerated elements of claim 1 could be performed by a clinician analyzing information about a patient “indicates the Office is actually asserting that claim 1 falls within the grouping of mental processes.” Applicant further cites to paragraphs 35-68 of the specification and asserts that the distillation model cannot be practically performed in the human mind. Examiner respectfully disagrees.
Initially, the statement in Step 2A Prong 1 that the listed elements of claim 1 could be performed by a clinician analyzing information about a patient does not shift construction of the claim elements from a method of organizing human activity to a mental process. Methods of organizing human activity may include activities that include analyzing information, such as budgeting as provided in MPEP 2106.04(II)(C). However, even if the elements listed in Step 2A Prong 1 were construed as falling within the scope of a mental process they could still be practicably performed in the human mind and/or with a pen and paper.
Claim 1 recites the distillation model as “compris[ing] an uncertainty model configured to estimate the variance across the plurality of different risk predictions models and a prediction mean model configured to estimate the mean across predictions by the plurality of different risk prediction models.” Paragraph 34 of Applicant’s specification as originally filed states that a risk prediction model “can be any model trained or otherwise configured, programmed, or designed to generate a risk score based on one or more input features,” while paragraph 35 provides that a distillation model “may be any model or process capable of distilling the output of the plurality of different risk prediction models.” Claim 1 does not limit the “distillation model” to analyses which are incapable of being performed by an individual, and only recites the distillation model, uncertainty model, and prediction mean model as “configured to” provide the corresponding mean and variance values. With respect to Applicant’s citation to paragraphs 35-68 of the specification, the specific equations described therein are not recited in the rejected claim(s). Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993).
With respect to Applicant’s citation to the decision in Ex Parte Desjardins, Examiner finds that the present claims do not recite limitations analogous to those in Ex Parte Desjardins. The claims at issue in that decision were expressly directed to a method of training a machine learning model, and the decision stated that "the claimed subject matter provides technical improvements over conventional systems by addressing challenges in continual learning and model efficiency by reducing storage requirements and preserving task performance across sequential training." In contrast, the present claims only recite the models at a high level of generality as “configured to” determine the calculated values, and are not recited in a way that limits them to AI which cannot be performed in the human mind. Examiner also notes that the element of the risk prediction models being “trained or programmed” is not construed as falling within the scope of the abstract idea.
Applicant further argues starting on page 10 that elements “specific to the assessment of disease risk of the subject,” such as assessing risk of a condition or disease occurring in a subject, generating a single health risk score of the subject, and determining at least one missing or defective feature that would affect the generated risk score confidence interval, “are additional elements,” and further argue that “the claimed invention improves the technical field of medicine, particularly diagnosing of a future medical condition or disease.” Examiner respectfully disagrees.
Initially, Examiner maintains that the elements argued by Applicant fall within the scope of the abstract idea and are capable of being performed by an individual as part of caring for a patient. These elements would also fall within the scope of an abstract idea if the claims were construed as reciting elements within a mental process. Applicant cites to paragraphs 2, 3, 12, 13, and 27 as supporting an assertion that the claimed invention addresses issues with overconfident predictions for patients. However, the subject matter highlighted by Applicant, i.e. a plurality of different risk prediction models each of which analyses a received set of features about a subject, generating a plurality of health risk scores for the subject, a distillation model determining an estimated mean and variance among the generated health risk scores, utilizing that information to generate a single health risk score and a risk score confidence interval, determining an effect of one or more missing or defective features on the generated risk score confidence interval based on a predetermined feature impact score for each different type of feature, identifying a missing or defective feature for reporting if the missing or defective feature would narrow the generated risk score confidence interval if it were not missing or not defective, and generating a report with the information all fall within the scope of the abstract idea. Under Step 2A Prong 2 additional elements are analyzed individually and within the context of the claims as a whole to determine whether they integrate any recited idea into a practical application. The elements identified by Applicant however, are not additional elements, and subject matter falling entirely within the scope of the abstract idea is not sufficient to integrate the abstract idea into a practical application.
Applicant lastly argues on page 12 that the claims amount to significantly more than the abstract idea on the basis that “the additional limitations contribute the inventive concept of improving accuracy of predicting a health risk by determining, based on a predetermined feature impact score for each different type of feature of the plurality of features, at least one missing or defective feature of the plurality of features that would affect the generated risk score confidence interval of prediction of risk of the disease if the at least one missing or defective feature is not missing or defective, wherein the feature impact score is defined to measure the contribution of missingness or defectiveness of each feature to the risk score confidence interval" and "a recommendation to obtain data for the determined at least one missing or defective features so that the determined at least one missing or defective feature is not missing or defective" to achieve a more accurate health risk prediction (which enables improved assessment of the risk of disease or condition occurring in subject).” Examiner respectfully disagrees.
As noted above, determining missing or defective data and recommending to obtain data for the determined at least one or more of the missing or defective features to achieve a more accurate health risk prediction falls entirely within the scope of the abstract idea itself. MPEP 2106.05(I) stats that “[a]n inventive concept "cannot be furnished by the unpatentable law of nature (or natural phenomenon or abstract idea) itself” and that “[i]nstead, an "inventive concept" is furnished by an element or combination of elements that is recited in the claim in addition to (beyond) the judicial exception, and is sufficient to ensure that the claim as a whole amounts to significantly more than the judicial exception itself.”
The rejection under 35 USC 101 is maintained.
B. Applicant's arguments with respect to the rejection of claims 1-6, 8-14, and 16-20 under 35 USC 112(a) have been fully considered but they are not persuasive.
Applicant cites to MPEP 2161.01 on page 13 as specifying that "[t]o satisfy the enablement requirement of 35 U.S.C. 112(a) or pre-AIA 35 U.S.C. 112, first paragraph, the specification must teach those skilled in the art how to make and use the full scope of the claimed invention without 'undue experimentation',” and that those skilled in the art “would know how to generate a single health risk score and a risk score confidence interval from estimated mean and variance across a plurality of risk scores without undue experimentation. However, as acknowledged by Applicant on page 13, the claims are rejected based on failure to comply with the written description requirement of 35 USC 112(a), not the enablement requirement. As stated in MPEP 2161(II), “[t]he written description requirement is separate and distinct from the enablement requirement,” and an analysis of the Wands factors is not required to establish failure to comply with the written description requirement.
As provided below, paragraphs 36 and 106 of the specification reflect the language of the claims in describing generating a single health risk score and a risk score confidence interval for the subject from the determined estimated mean and variance. However, no disclosure is provided of how the system generates a single risk score of a disease risk separately from determining the estimated mean predicted risk value and estimated variance. Whether one skilled in the art could have devised some manner of generating a single health risk score and a risk score confidence interval from estimated mean and variance across a plurality of risk scores is not sufficient to establish written description support. See MPEP 2161.01(I)(“It is not enough that one skilled in the art could write a program to achieve the claimed function because the specification must explain how the inventor intends to achieve the claimed function to satisfy the written description requirement”).
The rejection under 35 USC 112(a) is maintained.
C. Applicant’s arguments with respect to the rejection of claims 1-6, 8-14, and 16-20 under 35 USC 103 have been considered but are not persuasive.
Applicant argues starting on page 15 that Krishnan does not teach or suggest that the recited feature impact score is defined to measure the contribution of missingness or defectiveness of each feature to the risk score confidence interval, and that Krishnan only discloses a score which “provides some measure or indication as to the potential usefulness of the particular imaging modality or feature(s) that would improve the confidence of an assessment or diagnosis.” Examiner respectfully disagrees.
Examiner initially notes that Kovacevie rather than Krishnan is relied upon to teach the calculated risk score confidence as specifically being an interval.
Claim 1 recites, in relevant part, determining at least one missing or defective feature of the plurality of features that would affect the generated risk score confidence interval of prediction of risk of the disease “based on” a feature impact score for each different type of feature of the plurality of features, and then recites the feature impact score as “defined to measure the contribution of missingness or defectiveness of each feature to the risk score confidence interval,” i.e. the positively recited function associated with the feature impact is determining at least one missing or defective feature of the plurality of features that would affect the generated risk score confidence.
Paragraphs 26, 30, 39, and 71 of Krishnan describe identifying additional tests or information which would improve the determined confidence, i.e. missing features that would affect the generated risk score confidence if not missing, using a score indicating the usefulness of each test or information in improving the confidence. Paragraphs 26, 30, and 39 describe determining the score “for each additional test, measurement or feature, which provides some measure or indication as to the potential usefulness of the particular imaging modality or feature(s) that would improve the confidence of an assessment or diagnosis determined by the CAD system,” as “determine which features would likely provide the greatest improvement in confidence in a diagnosis,” and as “showing the likely benefit of additional tests or features that would improving the confidence of diagnosis, etc.” Examiner maintains that the recited score is described as measuring in some manner the contribution of missingness or defectiveness of each feature to the risk score confidence.
The rejection under 35 USC 103 is maintained.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-6, 8-14, and 16-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more. Claims 1-6, 8, and 16 are drawn to a method, claims 9-14 and 17 are drawn to a system, and claims 18-20 are drawn to a non-transitory computer readable medium, each of which is within the four statutory categories.
Step 2A(1)
Claim 1 recites, in part, performing the steps of:
receiving a plurality of features obtained about the subject;
analyzing the plurality of features using a plurality of different risk prediction models, wherein each of the plurality of different risk prediction models generates a health risk score for the subject based on the plurality of features;
determining, using a distillation model configured to distill outputs of the plurality of different risk prediction models, an estimated mean and variance among the generated health risk scores for the subject, wherein the distillation model comprises an uncertainty model configured to estimate the variance across the plurality of different risk predictions models and a prediction mean model configured to estimate the mean across predictions by the plurality of different risk prediction models;
generating, from the determined estimated mean and variance, a single health risk score of a disease risk of the subject and a risk score confidence interval for the subject;
determining, based on a feature impact score for each different type of feature of the plurality of features, at least one missing or defective feature of the plurality of features that would affect the generated risk score confidence interval of prediction of risk of the disease if the at least one missing or defective feature is not missing or defective, wherein the feature impact score is defined to measure the contribution of missingness or defectiveness of each feature to the risk score confidence interval;
generating a report of the disease risk of the subject, the report comprising the single health risk score, the risk score confidence interval for the subject, and a recommendation to obtain data for the determined at least one missing or defective features so that the determined at least one missing or defective feature is not missing or defective; and
outputting the report for assessing the risk of the subject.
These steps fall within the scope of at least one form of abstract idea, principally that of a method of organizing human activity. Fundamentally the process is that of using information about a subject to determine a health risk score and confidence interval by determining a plurality of risk scores and estimating a mean and variance of those scores, identifying missing or defective information which would improve the confidence interval if acquired, and providing a report containing the risk score, confidence interval, and a recommendation to obtain the missing or defective information. These steps could be performed by a clinician analyzing information about a patient to assess their health risk, and determining what new information should be obtained about the patient in order to increase confidence or accuracy of the assessment.
Certain elements such as estimation of variance or mean may also be construed as mathematical calculations and would fall within the scope of an abstract idea on that basis as well.
Examiner additionally notes that the terms “risk prediction models,” “distillation model,” “uncertainty model,” and “prediction mean model” are not limited to models inherent to computer hardware. Examiner notes paragraphs 34 and 35 of Applicant’s specification.
Independent claims 9 and 18 recite similar limitations and also recite an abstract idea under the same analysis.
Step 2A(2)
This judicial exception is not integrated into a practical application because the additional elements within the claims only amount to:
A. Instructions to Implement the Judicial Exception. MPEP 2106.05(f)
Claim 1 recites the additional limitations of a) each of the risk prediction models trained or programmed, and b) a user interface used to output the report.
Claim 9 recites the additional elements of a) a memory used to store instructions, b) a processor recited as executing the instructions to perform the subsequently recited functions, c) each of the risk prediction models trained or programmed, and d) a user interface used to output the report.
Claim 18 recites the additional elements of a) a non-transitory computer readable medium used to store instructions, b) a processor recited as executing the instructions to perform the subsequently recited functions, c) each of the risk prediction models trained or programmed, and d) a user interface used to output the report.
Paragraphs 93-100 of the specification describe the system comprising a processor, and memory used to store instructions for execution by the processor, a user interface, and non-transitory computer readable media. Paragraph 95 states that the processor “may take any suitable form” including microprocessors, microcontrollers, FPGAs, and other processing devices. Paragraphs 96 and 100 state that memory 630 and computer readable storage 660 likewise “may take any suitable form” including RAM, flash memory, ROM, “or other similar memory devices.” Paragraph 97 states that the user interface “can be any device or system that allows information to be conveyed and/or received, and may include a display, a mouse, and/or a keyboard for receiving user commands” including “a command line interface or graphical user interface that may be presented to a remote terminal…”. Paragraph 100 lastly states that “memory 630 and storage 660 may both be considered to be non-transitory machine-readable media” and that “the term non-transitory will be understood to exclude transitory signals but to include all forms of storage, including both volatile and non-volatile memories.” Each of the processor, memory, user interface, and non-transitory computer readable medium are construed as encompassing corresponding generic computing elements.
Paragraph 34 states with respect to the risk prediction models and their training “[a] risk prediction model can be any model trained or otherwise configured, programmed, or designed to generate a risk score based on one or more input features” and that “one or more of the risk prediction models can be trained using training datasets that may be specific to the healthcare setting or a more generic training dataset.” The training or programming of the risk prediction models is therefore construed as encompassing generic forms and processes of training models.
The above elements only amount to mere instructions to implement functions within the abstract idea using computing elements as tools. Each of the processor, memory, non-transitory computer readable medium, and user interface are recited at a high level of generality as implementing a corresponding function, such as the memory being recited as used to store instructions and the user interface used to output the report, and are disclosed as encompassing generic forms of their respective computing element types. The training or programming of the risk prediction models is likewise only recited at a high level of generality and is disclosed broadly as including any type of model and training using various datasets. The above elements are therefore not sufficient to integrate the abstract idea into a practical application.
The above claims, as a whole, are therefore directed to an abstract idea.
Step 2B
The present claims do not include additional elements that are sufficient to amount to more than the abstract idea because the additional elements or combination of elements amount to no more than a recitation of:
A. Instructions to Implement the Judicial Exception. MPEP 2106.05(f)
As explained above, claims 1, 9, and 18 only recite the processor, memory, user interface, and non-transitory computer readable medium as tools for performing the steps of the abstract idea, and mere instructions to perform the abstract idea using a computer is not sufficient to amount to significantly more than the abstract idea. MPEP 2106.05(f)
Thus, taken alone, the additional elements do not amount to significantly more than the above-identified judicial exception. Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually.
Depending Claims
Claim 2 recites: receiving information regarding at least one or more of the missing or defective features identified for reporting to produce an updated plurality of features about the subject; analyzing the updated plurality of features using the plurality of different risk prediction models, wherein each of the plurality of different risk prediction models generates an updated health risk score for the subject; determining, using the distillation model, an estimated mean and variance among the updated health risk scores for the subject; generating, from the determined estimated mean and variance, an updated single health risk score and an updated risk score confidence interval for the subject; generating an updated report comprising the updated single health risk score and the risk updated score confidence interval for the subject; and outputting the updated report. These limitations fall within the scope of the abstract idea as set out above.
Claim 2 further recites the additional element of the user interface as used to output the updated report.
Paragraph 97 of the specification states that the user interface “can be any device or system that allows information to be conveyed and/or received, and may include a display, a mouse, and/or a keyboard for receiving user commands” including “a command line interface or graphical user interface that may be presented to a remote terminal…”. The user interface is construed accordingly as encompassing generic computer display elements.
The use of a user display to output the updated report only amounts to mere instructions to implement functions within the abstract idea using computing elements as tools. The user interface is only recited at a high level of generality as implementing the function of outputting the updated report, and is disclosed broadly as encompassing generic types interfaces. The recited user interface is therefore not sufficient to integrate the abstract idea into a practical application or to amount to significantly more than the abstract idea.
Claims 3 and 11 recite wherein at least some of the received plurality of features are vital signs and test results, and wherein at least some of the plurality of features are received from stored health data. These limitations fall within the scope of the abstract idea as set out above.
Claims 3 and 11 further recite the additional element(s) of an interface to an electronic health database.
Paragraph 33 of the disclosure states that “the disease risk analysis system can be in wired and/or wireless communication with an electronic health database.” While paragraph 98 describes a communication interface 650 including devices such as a network interface card communicating via Ethernet protocol or a TCP/IP stack, no further disclosure is provided of either the electronic health database or the interface. The interface and electronic health database are therefore construed as encompassing generic computer network elements and generic computer storage elements.
The above elements only amount to mere instructions to implement functions within the abstract idea using computing elements as tools. The interface and electronic health database are only recited at a high level of generality, where features are received “via an interface to an electronic health database,” and are disclosed broadly as encompassing generic forms of communication interfaces and electronic storage. These elements are therefore not sufficient to integrate the abstract idea into a practical application or to amount to significantly more than the abstract idea.
Claims 4 and 12 recite wherein the report comprises a ranking of the at least one missing or defective feature identified for reporting, the ranking based on the determined effect of each of the at least one missing or defective feature on the generated health risk score confidence interval. These limitations fall within the scope of the abstract idea as set out above.
Claims 5, 13, and 19 recite wherein the generated health risk score comprises a probability of a risk, and the confidence interval comprises a range for the probability. These limitations fall within the scope of the abstract idea as set out above.
Claims 6, 14, and 20 recite wherein the generated health risk score comprises a probability of a risk being within a confidence interval range. These limitations fall within the scope of the abstract idea as set out above.
Claim 8 recites receiving and carrying out instructions to pause or silence the recommendation. These limitations fall within the scope of the abstract idea as set out above.
Claim 10 recites receiving information regarding at least one or more of the missing or defective features identified for reporting to produce an updated plurality of features about the subject; and performing a new risk analysis with the updated plurality of features. These limitations fall within the scope of the abstract idea as set out above.
Claims 16 and 17 recite the additional element of the plurality of different risk prediction models being trained from a plurality of bootstrapping datasets.
As cited above, paragraph 34 states with respect to the risk prediction models and their training “[a] risk prediction model can be any model trained or otherwise configured, programmed, or designed to generate a risk score based on one or more input features” and that “one or more of the risk prediction models can be trained using training datasets that may be specific to the healthcare setting or a more generic training dataset.” Paragraph 34 further provides that “[p]ursuant to a bootstrapping approach, multiple models can be trained by creating multiple datasets through sampling from the training dataset with replacement.” Paragraph 61 additionally describes “training M models using M bootstrapping datasets, which are created by randomly sampling from the training set with replacement.”
However, the recited use of bootstrapping datasets to train the risk prediction models only amounts to using computing elements or techniques as tools to implement functions within the abstract idea. The bootstrapping datasets are only recited at a high level of generality as being used to train the risk prediction models, while the disclosure similarly only provides a high-level description of the bootstrapping technique according to its general meaning as the creation of multiple datasets by sampling a training dataset with replacement. The general use of bootstrapping as a data processing technique to create training datasets is not sufficient to integrate the abstract idea into a practical application or to amount to significantly more than the abstract idea.
Claims 1-6, 8-14, and 16-20 are therefore rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter.
Claim Rejections - 35 USC § 112(a)
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 1-6, 8-14, and 16-20 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.
With regard to claims 1, 9, and 18, the disclosure does not provide sufficient written description of the claimed subject matter to show that Applicant had possession of method or system which generates, from the determined estimated mean and variance, a single health risk score of a disease risk of the subject.
Paragraphs 57-65 describe determining a mean risk value and mean variance across a plurality of risk scores and variances corresponding to a plurality of models analyzing bootstrapped datasets. Paragraphs 36 and 106 of the specification reflect the language of the claims in describing generating a single health risk score and a risk score confidence interval for the subject from the determined estimated mean and variance. However, no disclosure is provided of how the system generates a single risk score of a disease risk separately from determining the estimated mean predicted risk value and estimated variance.
Claims 2-6, 8, 10-14, 16, 17, 19, and 20 inherit the deficiencies of claims 1, 9, and 18 through dependency and are likewise rejected.
With regard to claim 6, the disclosure does not provide sufficient written description of the claimed subject matter to show that Applicant had possession of method or system which generates a health risk score comprising a probability of a risk being within a confidence interval range.
Paragraph 12 provides the only description in the specification of a risk score comprising a probability of the risk being within a confidence interval range. However, no disclosure is provided in paragraph 12 or elsewhere of how such a probability is actually determined.
Claim Rejections - 35 USC § 112(b)
The previous rejection of claims 6 and 8 under 35 USC 112(b) is withdrawn based on the amendments filed 12/1/2025.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 4, 5, 9, 12, 13, and 16-19 are rejected under 35 U.S.C. 103 as being unpatentable over Krishnan et al (US Patent Application Publication 2010/0121178) in view of Kovacevic et al, Bootstrap Variance Estimation for Predicted Individual and Population-Average Risks (hereinafter Kovacevic).
With respect to claim 1, Krishnan discloses the claimed method for assessing risk of a condition or disease occurring in a subject, the method comprising:
receiving a plurality of features obtained about the subject (Figure 4, [16], [17], [22], [38], and [43] describe the system obtaining and extracting a plurality of patient features);
analyzing the plurality of features using a plurality of different risk prediction models, wherein one or more of the plurality of different risk prediction models are trained or programmed to generate a health risk score for the subject based on the plurality of features ([22], [23], [26], [29], [39], [43], [50], and [52] describe the system analyzing the patient features using “one or more classification models,” which have been trained using training datasets, and which output a probability of a particular diagnosis, i.e. a risk score for the subject);
generating a single health risk score of a disease risk of the subject and a risk score confidence for the subject ([17], [24], [26], [39], and [52] describe the classification models outputting a probability of diagnosis of a disease, i.e. a risk score, as well as a corresponding measure of confidence);
determining, based on a feature impact score for each different type of feature of the plurality of features, at least one missing or defective feature of the plurality of features that would affect the generated risk score confidence of prediction of risk of the disease if the at least one missing or defective feature is not missing or defective ([26], [30], [39], and [71] describe identifying additional tests or information which would improve the determined confidence as well as a score indicating the usefulness of each test or information in improving the confidence), wherein the feature impact score is defined to measure the contribution of missingness or defectiveness of each feature to the risk score confidence ([26], [30], and [39] describe determining the score “for each additional test, measurement or feature, which provides some measure or indication as to the potential usefulness of the particular imaging modality or feature(s) that would improve the confidence of an assessment or diagnosis determined by the CAD system,” as “determine which features would likely provide the greatest improvement in confidence in a diagnosis,” and as “showing the likely benefit of additional tests or features that would improving the confidence of diagnosis, etc,”, i.e. measuring the contribution to the confidence of the feature being missing or defective);
generating a report of the disease risk of the subject, the report comprising the single health risk score, the risk score confidence for the subject, and a recommendation to obtain data for the determined at least one missing or defective features so that the determined at least one missing or defective feature is not missing or defective ([23], [24], [26], [39], [54], and [68] describe the system outputting the determined probability of diagnosis, confidence, and recommended additional tests and/or data types); and
outputting the report via a user interface for assessing the risk of the subject ([23], [26], [40], and [54] describe a user interface which outputs the above information including the diagnosis, confidence, and recommended additional tests and/or data types);
but does not expressly disclose:
wherein each of the plurality of different risk prediction models generates a health risk score for the subject;
determining, using a distillation model configured to distill outputs of the plurality of different risk prediction models, an estimated mean and variance among the generated health risk scores for the subject, wherein the distillation model comprises an uncertainty model configured to estimate the variance across the plurality of different risk predictions models and a prediction mean model configured to estimate the mean across predictions by the plurality of different risk prediction models;
generating the single health risk score and risk score confidence interval from the estimated mean and variance.
However, Kovacevie teaches that it was old and well known in the art of patient risk estimation before the effective filing date of the claimed invention to analyze a plurality of patient features using a plurality of different risk prediction models each trained to generate a risk score for the patient (§§1-3, 3.2, and 5 describe calculating risk scores for a patient using models trained on and applied to a series of bootstrapped datasets, see e.g. “[u]sing the B sets of bootstrap weights it is also possible to calculate bootstrap replicates of the predicted probability for this individual.” Examiner notes paragraphs 38, 39, and 61 describing the recited plurality of models as models trained using bootstrapped datasets), determine, using a distillation model configured to distill outputs of the plurality of different risk prediction models, an estimated mean and variance among the generated health risk scores for the subject, wherein the distillation model comprises an uncertainty model configured to estimate the variance across the plurality of different risk predictions models and a prediction mean model configured to estimate the mean across predictions by the plurality of different risk prediction models (§§2, 3.1, 3.2, and 5 describe calculating average risk scores and variances across the bootstrapped models), and generate a single health risk score and risk score confidence interval from the estimated mean and variance (§§ 3.3 and 5 provide an example of determining a single risk score and confidence interval for a patient using the results).
Therefore it would have been obvious to one of ordinary skill in the art of patient risk estimation before the effective filing date of the claimed invention to modify the system of Krishnan to analyze a plurality of patient features using a plurality of different risk prediction models each trained to generate a risk score for the patient, determine, using a distillation model configured to distill outputs of the plurality of different risk prediction models, an estimated mean and variance among the generated health risk scores for the subject, wherein the distillation model comprises an uncertainty model configured to estimate the variance across the plurality of different risk predictions models and a prediction mean model configured to estimate the mean across predictions by the plurality of different risk prediction models, and generate a single health risk score and risk score confidence interval from the estimated mean and variance as taught by Kovacevie since the claimed invention is only a combination of these old and well known elements which would have performed the same function in combination as each did separately. In the present case Krishnan already discloses the use of a plurality of risk prediction models as part of the system as well as determining a single risk score and confidence for a patient, and doing so by calculating mean risk score and variance over a plurality of risk models and determining a single risk score and confidence interval therefrom as taught by Kovacevie would perform that same function in Krishnan, making the results predictable to one of ordinary skill in the art (MPEP 2143).
With respect to claim 4, Krishnan/Kovacevie teach the method of claim 1. Krishnan further discloses:
wherein the report comprises a ranking of the at least one missing or defective feature identified for reporting, the ranking based on the determined effect of each of the at least one missing or defective feature on the generated health risk score confidence ([26], [30], and [39] describe generating a score for each additional test, measurement, or feature indicating its relative usefulness in increasing confidence, i.e. ranking the effect of each feature; [23], [26], [40], and [54] describe a user interface which outputs the information including the recommended additional tests and/or data types);
but does not expressly disclose:
the risk score confidence being a confidence interval.
However, Kovacevie teaches that it was old and well known in the art of patient risk estimation before the effective filing date of the claimed invention to generate a risk score confidence interval (§§ 3.3 and 5 provide an example of determining a single risk score and confidence interval for a patient).
Therefore it would have been obvious to one of ordinary skill in the art of patient risk estimation before the effective filing date of the claimed invention to modify the combination of Krishnan and Kovacevie to generate a risk score confidence interval as taught by Kovacevie since the claimed invention is only a combination of these old and well known elements which would have performed the same function in combination as each did separately. In the present case Krishnan and Kovacevie already teach determining a confidence indicator for a patient risk score, and determining a confidence interval as the confidence indicator as taught by Kovacevie would perform that same function in Krishnan and Kovacevie, making the results predictable to one of ordinary skill in the art (MPEP 2143).
With respect to claim 5, Krishnan/Kovacevie teach the method of claim 1. Krishnan further discloses:
wherein the generated health risk score comprises a probability of a risk ([17], [23], and [31] describe the risk score being a probability of the diagnosis);
but does not expressly disclose:
the generated health risk score confidence being a confidence interval comprising a range for the probability.
However, Kovacevie teaches that it was old and well known in the art of patient risk estimation before the effective filing date of the claimed invention to generate a risk score confidence interval comprising a range for a risk probability (§§ 3.3 and 5 provide an example of determining a single risk score and confidence interval in the form of a 95% confidence range).
Therefore it would have been obvious to one of ordinary skill in the art of patient risk estimation before the effective filing date of the claimed invention to modify the combination of Krishnan and Kovacevie to generate a risk score confidence interval comprising a range for a risk probability as taught by Kovacevie since the claimed invention is only a combination of these old and well known elements which would have performed the same function in combination as each did separately. In the present case Krishnan and Kovacevie already teach determining a confidence indicator for a patient risk score, and determining a confidence interval comprising a range for the probability as the confidence indicator as taught by Kovacevie would perform that same function in Krishnan and Kovacevie, making the results predictable to one of ordinary skill in the art (MPEP 2143).
With respect to claim 9, Krishna discloses the claimed system for assessing risk of condition or disease occurring in a subject, the system comprising:
a processor ([19] describes the disclosed functions as implemented using processors); and
a non-transitory memory storing instructions that, when executed by the processor ([19] describes memory devices storing software executed by the processor), cause the processor to:
receive a plurality of features obtained about the subject (Figure 4, [16], [17], [22], [38], and [43] describe the system obtaining and extracting a plurality of patient features);
analyze the plurality of features using a plurality of different risk prediction models, wherein one or more of the plurality of different risk prediction models are trained or programmed to generate a health risk score for the subject based on the plurality of features ([22], [23], [26], [29], [39], [43], [50], and [52] describe the system analyzing the patient features using “one or more classification models,” which have been trained using training datasets, and which output a probability of a particular diagnosis, i.e. a risk score for the subject);
generate a single health risk score of a disease risk of the subject and a risk score confidence for the subject ([17], [24], [26], [39], and [52] describe the classification models outputting a probability of diagnosis of a disease, i.e. a risk score, as well as a corresponding measure of confidence);
determine, based on a feature impact score for each different type of feature of the plurality of features, at least one missing or defective feature of the plurality of features that would affect the generated risk score confidence of prediction of risk of the disease if the at least one missing or defective feature is not missing or defective ([26], [30], [39], and [71] describe identifying additional tests or information which would improve the determined confidence as well as a score indicating the usefulness of each test or information in improving the confidence), wherein the feature impact score is defined to measure the contribution of missingness or defectiveness of each feature to the risk score confidence ([26], [30], and [39] describe determining the score “for each additional test, measurement or feature, which provides some measure or indication as to the potential usefulness of the particular imaging modality or feature(s) that would improve the confidence of an assessment or diagnosis determined by the CAD system,” as “determine which features would likely provide the greatest improvement in confidence in a diagnosis,” and as “showing the likely benefit of additional tests or features that would improving the confidence of diagnosis, etc,”, i.e. measuring the contribution to the confidence of the feature being missing or defective);
generate a report of the disease risk of the subject, the report comprising the single health risk score, the risk score confidence for the subject, and a recommendation to obtain data for the determined at least one missing or defective features so that the determined at least one missing or defective feature is not missing or defective ([23], [24], [26], [39], [54], and [68] describe the system outputting the determined probability of diagnosis, confidence, and recommended additional tests and/or data types); and
a user interface configured to provide the output the report for assessing the risk of the subject ([23], [26], [40], and [54] describe a user interface which outputs the above information including the diagnosis, confidence, and recommended additional tests and/or data types);
but does not expressly disclose:
wherein each of the plurality of different risk prediction models generates a health risk score for the subject;
determining, using a distillation model configured to distill outputs of the plurality of different risk prediction models, an estimated mean and variance among the generated health risk scores for the subject, wherein the distillation model comprises an uncertainty model configured to estimate the variance across the plurality of different risk predictions models and a prediction mean model configured to estimate the mean across predictions by the plurality of different risk prediction models;
generating the single health risk score and risk score confidence interval from the estimated mean and variance.
However, Kovacevie teaches that it was old and well known in the art of patient risk estimation before the effective filing date of the claimed invention to analyze a plurality of patient features using a plurality of different risk prediction models each trained to generate a risk score for the patient (§§1-3, 3.2, and 5 describe calculating risk scores for a patient using models trained on and applied to a series of bootstrapped datasets, see e.g. “[u]sing the B sets of bootstrap weights it is also possible to calculate bootstrap replicates of the predicted probability for this individual.” Examiner notes paragraphs 38, 39, and 61 describing the recited plurality of models as models trained using bootstrapped datasets), determine, using a distillation model configured to distill outputs of the plurality of different risk prediction models, an estimated mean and variance among the generated health risk scores for the subject, wherein the distillation model comprises an uncertainty model configured to estimate the variance across the plurality of different risk predictions models and a prediction mean model configured to estimate the mean across predictions by the plurality of different risk prediction models (§§2, 3.1, 3.2, and 5 describe calculating average risk scores and variances across the bootstrapped models), and generate a single health risk score and risk score confidence interval from the estimated mean and variance (§§ 3.3 and 5 provide an example of determining a single risk score and confidence interval for a patient using the results).
Therefore it would have been obvious to one of ordinary skill in the art of patient risk estimation before the effective filing date of the claimed invention to modify the system of Krishnan to analyze a plurality of patient features using a plurality of different risk prediction models each trained to generate a risk score for the patient, determine, using a distillation model configured to distill outputs of the plurality of different risk prediction models, an estimated mean and variance among the generated health risk scores for the subject, wherein the distillation model comprises an uncertainty model configured to estimate the variance across the plurality of different risk predictions models and a prediction mean model configured to estimate the mean across predictions by the plurality of different risk prediction models, and generate a single health risk score and risk score confidence interval from the estimated mean and variance as taught by Kovacevie since the claimed invention is only a combination of these old and well known elements which would have performed the same function in combination as each did separately. In the present case Krishnan already discloses the use of a plurality of risk prediction models as part of the system as well as determining a single risk score and confidence for a patient, and doing so by calculating mean risk score and variance over a plurality of risk models and determining a single risk score and confidence interval therefrom as taught by Kovacevie would perform that same function in Krishnan, making the results predictable to one of ordinary skill in the art (MPEP 2143).
Claim 12 recites limitations similar to those recited in claim 4, and is rejected on the same grounds set out above with respect to claim 4.
Claim 13 recites limitations similar to those recited in claim 5, and is rejected on the same grounds set out above with respect to claim 5.
With respect to claim 16, Krishnan/Kovacevie teach the method of claim 1. Krishnan does not expressly disclose wherein the plurality of different risk prediction models are trained from a plurality of bootstrapping datasets.
However, Kovacevie teaches that it was old and well known in the art of patient risk estimation before the effective filing date of the claimed invention to train a plurality of different risk prediction models from a plurality of bootstrapping datasets (§§3 and 3.1 describe risk models trained on a series of bootstrapped datasets).
Therefore it would have been obvious to one of ordinary skill in the art of patient risk estimation before the effective filing date of the claimed invention to modify the combination of Krishnan and Kovacevie to train a plurality of different risk prediction models from a plurality of bootstrapping datasets as taught by Kovacevie since the claimed invention is only a combination of these old and well known elements which would have performed the same function in combination as each did separately. In the present case Krishnan and Kovacevie already teach training a plurality of risk prediction models, and training the models using a plurality of bootstrapping datasets as taught by Kovacevie would perform that same function in Krishnan and Kovacevie, making the results predictable to one of ordinary skill in the art (MPEP 2143).
Claim 17 recites limitations similar to those recited in claim 16, and is rejected on the same grounds set out above with respect to claim 16.
With respect to claim 18, Krishnan discloses the claimed non-transitory computer readable medium storing instructions for assessing risk of a condition or disease occurring in a subject, the instructions, when executed by one or more processors ([19] describes the disclosed functions performed via memory devices storing software executed by processors), cause the one or more processors to:
receive a plurality of features obtained about the subject (Figure 4, [16], [17], [22], [38], and [43] describe the system obtaining and extracting a plurality of patient features);
analyze the plurality of features using a plurality of different risk prediction models, wherein one or more of the plurality of different risk prediction models are trained or programmed to generate a health risk score for the subject based on the plurality of features ([22], [23], [26], [29], [39], [43], [50], and [52] describe the system analyzing the patient features using “one or more classification models,” which have been trained using training datasets, and which output a probability of a particular diagnosis, i.e. a risk score for the subject);
generate a single health risk score of a disease risk of the subject and a risk score confidence for the subject ([17], [24], [26], [39], and [52] describe the classification models outputting a probability of diagnosis of a disease, i.e. a risk score, as well as a corresponding measure of confidence);
determine, based on a feature impact score for each different type of feature of the plurality of features, at least one missing or defective feature of the plurality of features that would affect the generated risk score confidence of prediction of risk of the disease if the at least one missing or defective feature is not missing or defective ([26], [30], [39], and [71] describe identifying additional tests or information which would improve the determined confidence as well as a score indicating the usefulness of each test or information in improving the confidence), wherein the feature impact score is defined to measure the contribution of missingness or defectiveness of each feature to the risk score confidence ([26], [30], and [39] describe determining the score “for each additional test, measurement or feature, which provides some measure or indication as to the potential usefulness of the particular imaging modality or feature(s) that would improve the confidence of an assessment or diagnosis determined by the CAD system,” as “determine which features would likely provide the greatest improvement in confidence in a diagnosis,” and as “showing the likely benefit of additional tests or features that would improving the confidence of diagnosis, etc,”, i.e. measuring the contribution to the confidence of the feature being missing or defective);
generate a report of the disease risk of the subject, the report comprising the single health risk score, the risk score confidence for the subject, and a recommendation to obtain data for the determined at least one missing or defective features so that the determined at least one missing or defective feature is not missing or defective ([23], [24], [26], [39], [54], and [68] describe the system outputting the determined probability of diagnosis, confidence, and recommended additional tests and/or data types); and
output the report to a user via a user interface for assessing the risk of the subject ([23], [26], [40], and [54] describe a user interface which outputs the above information including the diagnosis, confidence, and recommended additional tests and/or data types);
but does not expressly disclose:
wherein each of the plurality of different risk prediction models generates a health risk score for the subject;
determining, using a distillation model configured to distill outputs of the plurality of different risk prediction models, an estimated mean and variance among the generated health risk scores for the subject, wherein the distillation model comprises an uncertainty model configured to estimate the variance across the plurality of different risk predictions models and a prediction mean model configured to estimate the mean across predictions by the plurality of different risk prediction models;
generating the single health risk score and risk score confidence interval from the estimated mean and variance.
However, Kovacevie teaches that it was old and well known in the art of patient risk estimation before the effective filing date of the claimed invention to analyze a plurality of patient features using a plurality of different risk prediction models each trained to generate a risk score for the patient (§§1-3, 3.2, and 5 describe calculating risk scores for a patient using models trained on and applied to a series of bootstrapped datasets, see e.g. “[u]sing the B sets of bootstrap weights it is also possible to calculate bootstrap replicates of the predicted probability for this individual.” Examiner notes paragraphs 38, 39, and 61 describing the recited plurality of models as models trained using bootstrapped datasets), determine, using a distillation model configured to distill outputs of the plurality of different risk prediction models, an estimated mean and variance among the generated health risk scores for the subject, wherein the distillation model comprises an uncertainty model configured to estimate the variance across the plurality of different risk predictions models and a prediction mean model configured to estimate the mean across predictions by the plurality of different risk prediction models (§§2, 3.1, 3.2, and 5 describe calculating average risk scores and variances across the bootstrapped models), and generate a single health risk score and risk score confidence interval from the estimated mean and variance (§§ 3.3 and 5 provide an example of determining a single risk score and confidence interval for a patient using the results).
Therefore it would have been obvious to one of ordinary skill in the art of patient risk estimation before the effective filing date of the claimed invention to modify the system of Krishnan to analyze a plurality of patient features using a plurality of different risk prediction models each trained to generate a risk score for the patient, determine, using a distillation model configured to distill outputs of the plurality of different risk prediction models, an estimated mean and variance among the generated health risk scores for the subject, wherein the distillation model comprises an uncertainty model configured to estimate the variance across the plurality of different risk predictions models and a prediction mean model configured to estimate the mean across predictions by the plurality of different risk prediction models, and generate a single health risk score and risk score confidence interval from the estimated mean and variance as taught by Kovacevie since the claimed invention is only a combination of these old and well known elements which would have performed the same function in combination as each did separately. In the present case Krishnan already discloses the use of a plurality of risk prediction models as part of the system as well as determining a single risk score and confidence for a patient, and doing so by calculating mean risk score and variance over a plurality of risk models and determining a single risk score and confidence interval therefrom as taught by Kovacevie would perform that same function in Krishnan, making the results predictable to one of ordinary skill in the art (MPEP 2143).
Claim 19 recites limitations similar to those recited in claim 5, and is rejected on the same grounds set out above with respect to claim 5.
Claims 2, 3, 10, and 11 are rejected under 35 U.S.C. 103 as being unpatentable over Krishnan et al (US Patent Application Publication 2010/0121178) in view of Kovacevic et al, Bootstrap Variance Estimation for Predicted Individual and Population-Average Risks (hereinafter Kovacevic) as applied to claims 1 and 9, and further in view of Gross et al (US Patent Application Publication 2016/0328525).
With respect to claim 2, Krishnan/Kovacevie teach the method of claim 1. Krishnan further discloses:
analyzing the plurality of features using the plurality of different risk prediction models, wherein one or more of the plurality of different risk prediction models generates a health risk score for the subject ([22], [23], [26], [29], [39], [43], [50], and [52] describe the system analyzing the patient features using “one or more classification models,” which have been trained using training datasets, and which output a probability of a particular diagnosis, i.e. a risk score for the subject);
generating a single health risk score and a risk score confidence interval for the subject ([24], [26], [39], and [52] describe the classification models outputting a probability of a particular diagnosis, i.e. a risk score, as well as a corresponding measure of confidence);
generating a report comprising the single health risk score and the risk score confidence interval for the subject ([23], [24], [26], [39], [54], and [68] describe the system outputting the determined probability of diagnosis, confidence, and recommended additional tests and/or data types); and
outputting the report via the user interface ([23], [26], [40], and [54] describe a user interface which outputs the above information including the diagnosis, confidence, and recommended additional tests and/or data types);
but does not expressly disclose:
receiving information regarding at least one or more of the missing or defective features identified for reporting to produce an updated plurality of features about the subject;
each of the plurality of different risk prediction models generating an updated health risk score for the subject;
determining, using the distillation model, an estimated mean and variance among the updated health risk scores for the subject;
generating an updated single health risk score and updated risk score confidence interval from the estimated mean and variance;
updating each of the health risk scores, single health risk score, risk score confidence interval for the subject, and report based on the updated plurality of features.
However, as cited above Kovacevie teaches that it was old and well known in the art of patient risk estimation before the effective filing date of the claimed invention to analyze a plurality of patient features using a plurality of different risk prediction models each trained to generate a risk score for the patient (§§1-3, 3.2, and 5 describe calculating risk scores for a patient using models trained on and applied to a series of bootstrapped datasets, see e.g. “[u]sing the B sets of bootstrap weights it is also possible to calculate bootstrap replicates of the predicted probability for this individual.” Examiner notes paragraphs 38, 39, and 61 describing the recited plurality of models as models trained using bootstrapped datasets), determine, using a distillation model, an estimated mean and variance among the generated health risk scores for the subject (§§2, 3.1, 3.2, and 5 describe calculating average risk scores and variances across the bootstrapped models), and generate a single health risk score and risk score confidence interval from the estimated mean and variance (§§ 3.3 and 5 provide an example of determining a single risk score and confidence interval for a patient using the results).
Therefore it would have been obvious to one of ordinary skill in the art of patient risk estimation before the effective filing date of the claimed invention to modify the combination of Krishnan and Kovacevie to analyze a plurality of patient features using a plurality of different risk prediction models each trained to generate a risk score for the patient, determine, using a distillation model, an estimated mean and variance among the generated health risk scores for the subject, and generate a single health risk score and risk score confidence interval from the estimated mean and variance as taught by Kovacevie since the claimed invention is only a combination of these old and well known elements which would have performed the same function in combination as each did separately. In the present case Krishnan and Kovacevie already teach the use of a plurality of risk prediction models as part of the system as well as determining a single risk score and confidence for a patient, and doing so by calculating mean risk score and variance over a plurality of risk models and determining a single risk score and confidence interval therefrom as taught by Kovacevie would perform that same function in Krishnan and Kovacevie, making the results predictable to one of ordinary skill in the art (MPEP 2143).
Gross then further teaches that it was old and well known in the art of patient risk estimation before the effective filing date of the claimed invention to receive information regarding at least one or more missing or defective features identified for reporting to produce an updated plurality of features about the subject and repeat risk calculations to produce updated risk scores and reports (Figure 4, [10], [32], [34]-[36], [49], [50], and [61]-[64] describe determining a risk score for a subject based on feature data, identifying missing features, receiving information for the identified missing features, and updating the values calculated using the features including updating risk scores and severity).
Therefore it would have been obvious to one of ordinary skill in the art of patient risk estimation before the effective filing date of the claimed invention to modify the combination of Krishnan and Kovacevie to receive information regarding at least one or more missing or defective features identified for reporting to produce an updated plurality of features about the subject and repeat risk calculations to produce updated risk scores and reports as taught by Gross since the claimed invention is only a combination of these old and well known elements which would have performed the same function in combination as each did separately. In the present case Krishnan and Kovacevie already teach identifying missing features for the subject as well as generating the recited risk scores and corresponding mean and variance, single risk score, risk score confidence interval, and report, and receiving the missing features and repeating the calculations using the updated features as taught by Gross would perform that same function in Krishnan and Kovacevie, making the results predictable to one of ordinary skill in the art (MPEP 2143).
With respect to claim 3, Krishnan/Kovacevie teach the method of claim 1. Krishnan further discloses:
wherein at least some of the received plurality of features are test results ([16] and [43] describe the features as including various forms of test results), and wherein at least some of the plurality of features are received via an interface to an electronic health database ([43] and [48] describe the system analyzing patient data retrieved from a database);
but does not expressly disclose:
the received plurality of features including vital signs.
However, Gross teaches that it was old and well known in the art of patient risk estimation before the effective filing date of the claimed invention to receive patient features including vital signs (Figure 3, [10], [27], and [41] describe receiving physiological sensor data including vital signs).
Therefore it would have been obvious to one of ordinary skill in the art of patient risk estimation before the effective filing date of the claimed invention to modify the combination of Krishnan and Kovacevie to receive patient features including vital signs as taught by Gross since the claimed invention is only a combination of these old and well known elements which would have performed the same function in combination as each did separately. In the present case Krishnan and Kovacevie already teach receiving feature data for a subject, and including vital signs in that feature data as taught by Gross would perform that same function in Krishnan and Kovacevie, making the results predictable to one of ordinary skill in the art (MPEP 2143).
With respect to claim 10, Krishnan/Kovacevie teach the system of claim 9. Krishnan does not expressly disclose wherein instructions further cause the processor to: receive information regarding at least one or more of the missing or defective features identified for reporting to produce an updated plurality of features about the subject; and perform a new risk analysis with the updated plurality of features.
However, Gross teaches that it was old and well known in the art of patient risk estimation before the effective filing date of the claimed invention to receive information regarding at least one or more missing or defective features identified for reporting to produce an updated plurality of features about the subject and perform a new risk analysis with the updated plurality of features (Figure 4, [10], [32], [34]-[36], [49], [50], and [61]-[64] describe determining a risk score for a subject based on feature data, identifying missing features, receiving information for the identified missing features, and updating the values calculated using the features including updating risk scores and severity).
Therefore it would have been obvious to one of ordinary skill in the art of patient risk estimation before the effective filing date of the claimed invention to modify the combination of Krishnan and Kovacevie to receive information regarding at least one or more missing or defective features identified for reporting to produce an updated plurality of features about the subject and perform a new risk analysis with the updated plurality of features as taught by Gross since the claimed invention is only a combination of these old and well known elements which would have performed the same function in combination as each did separately. In the present case Krishnan and Kovacevie already teach identifying missing features for the subject as well as performing a risk analysis with the features, and receiving the missing features and performing a new risk analysis with the updated plurality of features as taught by Gross would perform that same function in Krishnan and Kovacevie, making the results predictable to one of ordinary skill in the art (MPEP 2143).
Claim 11 recites limitations similar to those recited in claim 3, and is rejected on the same grounds set out above with respect to claim 3.
Claims 6, 14, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Krishnan et al (US Patent Application Publication 2010/0121178) in view of Kovacevic et al, Bootstrap Variance Estimation for Predicted Individual and Population-Average Risks (hereinafter Kovacevic) as applied to claims 1, 9, and 18, and further in view of Yang et al (US Patent Application Publication 2017/0360379).
With respect to claim 6, Krishnan/Kovacevie teach the method of claim 1. Krishnan does not expressly disclose wherein the generated risk score comprises a probability of a risk being within a confidence interval range.
However, Yang teaches that it was old and well known in the art of patient risk estimation before the effective filing date of the claimed invention to generate a risk score comprising a probability of a risk being within a confidence interval range (Figure 5, [4], [31]-[33], [42], [46], [47], and claim 18 describe generating a risk score for a patient based on whether some risk of the patient is withing a confidence interval).
Therefore it would have been obvious to one of ordinary skill in the art of patient risk estimation before the effective filing date of the claimed invention to modify the combination of Krishnan and Kovacevie to generate a risk score comprising a probability of the risk being within a confidence interval range as taught by Yang since the claimed invention is only a combination of these old and well known elements which would have performed the same function in combination as each did separately. In the present case Krishnan and Kovacevie already teach generating a risk score and a confidence interval, and generating a risk score comprising a probability of a risk being within a confidence interval range as taught by Yang would perform that same function in Krishnan and Kovacevie, making the results predictable to one of ordinary skill in the art (MPEP 2143).
Claim 14 recites limitations similar to those recited in claim 6, and is rejected on the same grounds set out above with respect to claim 6.
Claim 20 recites limitations similar to those recited in claim 6, and is rejected on the same grounds set out above with respect to claim 6.
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Krishnan et al (US Patent Application Publication 2010/0121178) in view of Kovacevic et al, Bootstrap Variance Estimation for Predicted Individual and Population-Average Risks (hereinafter Kovacevic) as applied to claim 1, and further in view of McNair et al (US 10,672,516).
With respect to claim 8, Krishnan/Kovacevie teach the method of claim 1. Krishnan does not expressly disclose receiving and carrying out instructions to pause or silence the recommendation.
However, McNair teaches that it was old and well known in the art of patient risk estimation before the effective filing date of the claimed invention to receive and carry out instructions to pause or silence a recommendation (Abstract, Figure 3B, Column 11 lines 19-31, Column 13 lines 5-12, and Column 15 lines 30-39 describe the system receiving an instruction rejecting a recommendation, i.e. pausing or silencing the recommendation. Examiner notes paragraph 73 of Applicant’s specification regarding the scope of instructions to pause or silence a recommendation).
Therefore it would have been obvious to one of ordinary skill in the art of patient risk estimation before the effective filing date of the claimed invention to modify the combination of Krishnan and Kovacevie to receive and carry out instructions to pause or silence a recommendation as taught by McNair since the claimed invention is only a combination of these old and well known elements which would have performed the same function in combination as each did separately. In the present case Krishnan and Kovacevie already teach providing recommendations, and receiving and carrying out instructions to pause or silence the recommendation as taught by McNair would perform that same function in Krishnan and Kovacevie, making the results predictable to one of ordinary skill in the art (MPEP 2143).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Allassonniere et al (US Patent Application Publication 2021/0158962);
Gluck et al (US Patent Application Publication 2020/0135336).
THIS ACTION IS MADE FINAL. 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 WILLIAM G LULTSCHIK whose telephone number is (571)272-3780. The examiner can normally be reached 9am - 5pm.
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, Fonya Long can be reached at (571) 270-5096. 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.
/Gregory Lultschik/Examiner, Art Unit 3682