DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Claim priority PCT/JP2023/023269 06/23/2023 is acknowledge.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 11/04/2025 was filed. The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
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 because they recite an abstract idea and do not demonstrate practical application or an inventive concept.
Step 1
Claims 1-20 recite statutory machines, systems, processes, or a manufacture, as follows:
Claims 1-8 recite machines or systems, including an information generation device, functional units, and, in claim 7, a measurement device.
Claims 9-14 recite computer-implemented processes.
Claims 15-20 recite a manufacture, namely a computer-readable non-transitory recording medium having a program recorded therein.
The claims are statutory under Step 1, and the analysis proceeds to Step 2A.
Step 2A, Prong One
Claims 1-20 recite the mental process of evaluating sensor-derived information to estimate disease risk and generate a health-measure recommendation; the additional claim details do not change that exception.
Independent claims
Non-bold language identifies the asserted judicial-exception language. Bold language identifies the additional elements preserved for Prong Two and Step 2B.
Claim 1.
[1] An information generation device comprising: a memory storing instructions; and a processor connected to the memory and configured to execute the instructions to:
[2] acquire sensor data including acceleration and angular velocity measured by a measurement device mounted on footwear of at least one subject to be managed;
[3] estimate a disease risk for each disease related to at least one of the subjects to be managed using the acquired sensor data;
[4] generate proposal information including a health measure according to a disease risk related to at least one of the subjects to be managed; and
[5] output the generated proposal information.
Claims 1, 9, and 15 are mirror claims. Claim 1 recites the subject matter as an information generation device, claim 9 recites the same acquisition, risk-estimation, proposal-generation, and output operations as an information generation method, and claim 15 recites those operations as a program on a computer-readable non-transitory recording medium. The different statutory forms do not change the subject matter analysis because each requires the same sequence of information evaluation and health-measure recommendations. Claims 2-8, 10-14, and 16-20 depend from those mirror forms and are analyzed by their added limitations below.
Under the broadest reasonable interpretation, the non-bold limitation [3] means determining a disease risk for each disease from acquired sensor data. Limitation [4] means selecting or constructing information that includes a health measure based on that risk, and limitation [5] means outputting the resulting recommendation. Non-bold language limitations [3]-[5] recite a mental process because a person can evaluate information to determine disease risk, select or construct a health-measure recommendation according to that evaluation, and output the recommendation in a paper.
Dependent claims
Claims 2-8, 10-14, and 16-20 retain the inherited mental-process exception from claims 1, 9, and 15. The groups below identify whether each dependent addition remains within that exception or supplies an additional element for the downstream stages.
Claims 2, 10, and 16. The added gait-index evaluation, model input, and disease-risk information estimate are directed to the same mental process because a person can calculate or evaluate the index, consider it with other information, and determine a risk score.
Claims 3, 11, and 17. The operation of the added health measure estimation model is directed toward the same mental process, as a person can use a disease-risk score to select or formulate a health measure.
Claims 4, 12, 18, 5, 13, and 19. These claims direct the recommendation according to a way of working or an industry type of an organization. The claimed activity remains mental because a person can evaluate disease risk together with work or industry context and tailor a health-measure recommendation accordingly. The context changes the subject matter considered in the recommendation but does not change the mental-process exception.
Claims 6, 14, and 20. These claims add machine-learning models and specify an incomplete heterogeneous variational autoencoder. The named model is an additional element for Prong Two and Step 2B, but the inherited abstract idea remains the mental process of evaluating risk and generating a health-measure recommendation. Whether the named model provides a practical application or significantly more is evaluated below; it is not treated as a separate Prong One exception.
Claims 7 and 8. Claim 7 adds a measurement device installed on footwear that measures acceleration and angular velocity, generates sensor data, and transmits it to the information generation device. Claim 8 adds displaying proposal information including a health measure optimized for an organization on a terminal screen. Neither claim recites a new exception. By dependency, both retain the mental process of evaluating disease risk and generating a health-measure recommendation from claims 1-6, while their physical measurement and display limitations are evaluated as additional elements below.
All claims 1-20 therefore recite the mental-process exception through their independent limitations or dependency. Claims 6, 7, 8, 14, and 20 also recite the additional model, measurement, or display limitations identified above. Prong One is satisfied for every claim, and the additional elements proceed to Prong Two.
Step 2A, Prong Two
Claims 1-20 do not integrate the recited mental process into a practical application. The additional elements provide computer implementation, obtain the information evaluated, identify a model, or present the resulting recommendation without changing a technical operation or performing the proposed health measure.
Independent Claims Analysis
Additional elements. Claims 1, 9, and 15 respectively recite an information generation device, execution by a computer, and a computer-readable non-transitory recording medium memory storing instructions and a processor connected to the memory, together with acquiring sensor data including acceleration and angular velocity measured by a measurement device mounted on footwear.
Individual Additional Elements Evaluation
MPEP §§2106.04(d)(1), 2106.05(a), and 2106.05(f) require the claim to reflect the components or operations that produce a technological improvement, rather than merely use a computer to perform the exception. The information generation device, computer, and a computer-readable non-transitory recording medium memory storing instructions and a processor connected to the memory in claims 1, 9, and 15 identify where the risk evaluation and recommendation are executed or stored. The claims recite no changed processor, memory organization, communication protocol, or computer operation. Because the claims use those implementation forms to execute the mental process without claiming a technical change to them, these elements do not integrate the exception into a practical application.
MPEP §§2106.05(b), 2106.05(e), and 2106.05(g) require the machine relationship or data-gathering operation to meaningfully limit how the exception is applied. The acquiring sensor data including acceleration and angular velocity measured by a measurement device mounted on footwear in claims 1, 9, and 15 identifies the source and content of the information evaluated. It does not control the measurement device, require a sensor arrangement, prescribe calibration or waveform processing, or return a control signal to the device. The acquisition therefore gathers the selected input before the risk evaluation rather than applying that evaluation through a particular machine. This limitation is insignificant pre-solution data gathering under MPEP §2106.05(g).
Claims 1, 9, and 15. As an ordered combination, the device, computer, or a computer-readable non-transitory recording medium memory storing instructions and a processor connected to the memory, first acquires footwear-sensor data and then provides the environment for the recited risk evaluation and recommendation. The combination therefore supplies input and computer implementation for the exception without imposing a meaningful practical limit under MPEP §§2106.04(d), 2106.05(e), 2106.05(f), and 2106.05(g).
Dependent Claims Analysis
The dependent claims do not alter the Prong Two result. Claims adding only further exception language inherit the parent analysis; claims adding a model, measurement device, or display do not claim the technical operation needed to turn the risk evaluation and recommendation into a practical application.
Claims 2, 10, and 16, and claims 3, 11, and 17, add gait-index evaluation, model input, disease-risk information, and use of a health-measure estimation model. Prong One identified those limitations as parts of the mental evaluation, not additional elements. These claims therefore add no new element for Prong Two and inherit the implementation and acquisition analysis of claims 1, 9, and 15.
Claims 4, 5, 12, 13, 18, and 19 tailor the recommendation according to a way of working or industry type. Prong One identified that contextual tailoring as part of the mental process. These claims add no new additional element and inherit the Prong Two result of their respective parent claims.
Claims 6, 14, and 20 add models learned using a machine-learning method and an incomplete heterogeneous variational autoencoder. Under MPEP §§2106.04(d)(1), 2106.05(a), and 2106.05(f), a named model integrates the exception only when the claim reflects a technical operation that produces the asserted improvement. The model is invoked to produce the disease-risk estimate used by the mental recommendation. It therefore does not integrate the exception.
Claim 7 adds a measurement device installed on footwear, measurement of spatial acceleration and spatial angular velocity, generation of sensor data using the measurements, and transmission of the generated sensor data. MPEP §§2106.05(a) and 2106.05(b) recognize that a claimed measurement device or technique generating new data may provide a technological improvement when the claim recites the particular configuration or technique producing that improvement. Claim 7 specifies measured quantities and location but no sensor configuration, calibration, data conversion, or processing technique. It connects the device to the mental process only by transmitting the generated input. The measurement and transmission are therefore pre-solution data gathering under MPEP §2106.05(g), not a technological improvement or meaningful particular-machine application, and are evaluated independently under Step 2B.
Claim 8 adds displaying, on a screen of a terminal device used in an organization, proposal information including a health measure optimized for the organization. Under MPEP §§2106.05(e), 2106.05(g), and 2106.05(h), presenting the result of an exception or limiting its audience or setting does not impose a meaningful practical limit. Claim 8 recites no display architecture, visualization technique, or changed terminal operation. The screen presents the completed recommendation in its intended organizational setting.
The dependent ordered combinations retain the parent sequence and place the added model, measurement device, or display at the risk-evaluation, input-gathering, or result-presentation stage. No claim recites an interaction that changes the operation of the computer, model, sensor, or display. Claims 2-8, 10-14, and 16-20 therefore do not integrate the exception under MPEP §2106.04(d).
Claims 1-20 remain directed to the recited mental process after Prong Two. The identified additional elements proceed unchanged to Step 2B.
Step 2B
Claims 1-20 do not recite significantly more than the mental process. The additional elements individually provide generic computer implementation, input gathering, a result-oriented model, or result presentation, and their ordered combinations add no non-generic technical arrangement.
Independent Claims Analysis
Under MPEP §2106.05(d), an additional element recited at a high level of generality does not supply an inventive concept when function-matched evidence establishes that the activity is well-understood, routine, and conventional. The information generation device, computer, and a computer-readable non-transitory recording medium memory storing instructions and a processor connected to the memory of claims 1, 9, and 15 perform execution, storage, input/output, and communication at that level. Spec. 0150-0159 identifies a processor, ordinary storage, input/output and communication interfaces, a network, and recording media, and states that the hardware configuration does not limit the scope. This description establishes the functional match to the generic computer, system, and media implementation evaluated in Alice. The Court held that the system and media claims add nothing of substance to the underlying abstract idea. Alice Corp. v. CLS Bank International, 573 U.S. 208, 226 (2014). These generic implementation elements are WURC for the claimed functions and do not add significantly more.
Under MPEP §§2106.05(d) and 2106.05(g), obtaining information at a high level so that the exception can be performed is both WURC computer activity and insignificant pre-solution activity. The acquiring of acceleration and angular-velocity data from a footwear measurement device in claims 1, 9, and 15 requires receipt of the selected input but no acquisition protocol, sensor control, calibration, or raw-signal processing. Spec. 0032 permits wired or wireless receipt, including receipt of information calculated elsewhere. The “use of a computer to obtain data” among “well-understood, routine, conventional activities.” Alice, 573 U.S. at 225. The claimed acquisition is functionally the same high-level receipt of data before analysis and therefore does not add significantly more. Also refer to US 20230306455 A1, par. 0122; US 20230273048 A1 par. 0030; US 20220000430 A1 par. 0007, 0028; US 20110054358 A1 par. 0046; US 20120024061 A1 abstract; US 12667280 B2, claim 1;
Under MPEP §§2106.05(a), 2106.05(c), 2106.05(e), and 2106.05(f), claims 1, 9, and 15 claim no improved computer or other technology, no transformation of a particular article, no performance of the proposed health measure, and no other technical operation beyond using the implementation elements to perform the exception. The implementation form and selected sensor-data source therefore supply neither an inventive technical function nor another meaningful limit.
Claims 1, 9, and 15. As an ordered combination, the device, computer, or recording medium performs generic execution and storage, while the acquisition operation supplies the data used by the mental process. The claim places no specialized processing, feedback, control, or changed component function in that arrangement. Each additional element performs the same input or implementation role it performs separately. Under MPEP §§2106.05(d), 2106.05(e), 2106.05(f), and 2106.05(g), the combination does not provide an inventive concept.
Dependent Claims Analysis
The dependent claims also lack significantly more. Claims adding only exception content inherit the parent Step 2B analysis; the model, footwear-measurement, and display additions remain result-oriented tools or extra-solution activity and create no inventive technical arrangement with the inherited elements.
Claims 2, 10, and 16, and claims 3, 11, and 17, add only the gait-index, model-input, disease-risk, and health-measure evaluations identified in Prong One as the mental process. They add no new additional element and therefore inherit the Step 2B result of claims 1, 9, and 15.
Claims 4, 5, 12, 13, 18, and 19 add only work-context or industry-context tailoring identified in Prong One as part of the mental process. They add no new additional element and inherit the Step 2B result of their respective parent claims.
Claims 6, 14, and 20 add models learned using a machine-learning method and an incomplete heterogeneous variational autoencoder. MPEP §2106.05(f) treats a result-oriented invocation of a tool as an instruction to apply the exception when the claim recites the desired result without the mechanism that achieves it. These claims name the model but require no architecture, training operation, missing-data condition, imputation process, or changed computer function. The model is required only to perform the claimed disease-risk evaluation. That functional invocation does not add significantly more, individually or with the inherited generic implementation and data acquisition.
Claim 7 adds the footwear measurement device, measurement of spatial acceleration and spatial angular velocity, generation of sensor data, and transmission of that data. Under MPEP §§2106.05(b), 2106.05(d), and 2106.05(g), a physical device does not supply an inventive concept when its claimed relation to the exception is only gathering and forwarding the input. The claim recites no particular sensor configuration, calibration, signal processing, or feedback relationship. The WURC finding to the high-level functions supported by the specification. Spec. 0002 states that a gait-analysis technique using data from “a sensor mounted on footwear such as shoes has been developed”; Spec. 0016-0018 identifies interchangeable acceleration, angular-velocity, and inertial sensors; and Spec. 0029 identifies Bluetooth, WiFi, or wired transmission. These disclosures match the claimed measurement and transmission functions and describe them through known sensor types and standard communication alternatives without making any particular one part of the claim. Appending those pre-solution functions to the inherited mental evaluation does not add significantly more.
Claim 8 adds displaying the proposal information and optimized health measure on a terminal screen. Under MPEP §§2106.05(d) and 2106.05(g), displaying the result after the exception is insignificant post-solution activity when no particular presentation tool or technique changes the display operation. Claim 8 requires only a terminal screen and the content to be shown. Spec. 0140 describes that same result display, while Spec. 0157 identifies the display device only through its ordinary function of displaying information. This establishes that the terminal performs the high-level display function recited by the claim. Electric Power explains that “merely presenting the results of abstract processes of collecting and analyzing information, without more” is insufficient. Electric Power Group, LLC v. Alstom S.A., 830 F.3d 1350, 1354 (Fed. Cir. 2016). The claimed WURC result display does not add significantly more.
For the dependent ordered combinations, the machine-learning model and incomplete heterogeneous variational autoencoder perform the risk evaluation, the footwear measurement device supplies and transmits its input, and the terminal screen presents its result. The claim language recites no technical feedback, specialized placement, or cooperative operation that changes any component’s function. Under MPEP §§2106.05(d) through 2106.05(g), the combinations do not provide an inventive concept.
Claims 1-20, considered individually and as ordered combinations, do not add significantly more than the recited mental process. Claims 1-20 are therefore ineligible under 35 U.S.C. 101.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1-3, 6, 7-11 and 14-17 are rejected under 35 U.S.C. 103 as being unpatentable over US20180279915A1- Huang, and Oono, K., Charoenphakdee, et, al. refer to pto-892-U.
[Claim 1]
Huang teaches, An information generation device comprising: a memory storing instructions; and a processor connected to the memory and configured to execute the instructions to: acquire sensor data including acceleration and angular velocity measured by a measurement device mounted on footwear of at least one subject to be managed;(Huang, at 52 , quantized data from sensors embedded within a wearable device ( e . g . , wearable device 12 ) can be received ( e . g . , by mobile computing device 14. par. 0041, fig. 5; sensors can include…a three-axis accelerometer and gyroscope 26 par. 0021; wearable device 12 is implemented as an insert 22 for a shoe. par. 0018; )
estimate a disease risk for each disease related to at least one of the subjects to be managed using the acquired sensor data;(Huang, the classifier 46 can employ a regression model configured to provide calculate a its parameter representing a likelihood that the subject exhibits the clinical parameter par. 0035; …common in the construction industry or retail trades ) .
The wireless portable gait system can be used to provide real - time and continuous sensory surveillance , risk identification and just - in - time intervention methods related to a STF. par. 0016; The clinical parameter assigned by the classifier 46 can relate to a certain medical condition)
generate proposal information; (Huang, The alert generator 47…the alert can be , for example , a tactile signal…an audio signal or a visual signal, redefined time period , if the subject does not respond to the tactile signal ( e . g . , does not take corrective steps to eliminate the detected risk based on the clinical parameter ) , the alert generator 47 can generate a visual signal or an audio signal… par. 0037; …. In other examples , the clinical parameter can be used for just - in - time intervention before the occurrence of an event ( like syncope , falling , loss of balance , etc . )… par. 0036; …clinical parameter can be used to detect , quantify or characterize a risk of a slip , trip and fall ( STF ) event par. 0016; ease of display and analysis” 0049.)
output the generated proposal; (Huang, provide personalized healthcare 0016; just-in-time intervention [0036]; corrective steps to eliminate the detected risk 0037; alert generated upon detection of the risk to prevent the event 0038.)
Obvious Rational:
Huang's describe obtaining footwear sensor data by means of a shoe insert which includes a three-axis accelerometer and a gyroscope and then applying machine learning to the resulting gait data. Moreover, Huang calculates the probability that the subject has the clinical parameter, associates that parameter with medical conditions, and makes use of the parameter in the area of personal healthcare, in interventions, and for issuing risk-responsive alerts. Huang [0016], [0018], [0021], [0035]-[0038], [0041].
Huang does not explicitly teach how to estimate the risk of each disease and then produce proposal information involving a health measure based on that disease risk, Oono teaches the use of health data to estimate the probabilities of various diseases and then recommends that the user take supplements, specifically by linking the measurements to intervention services. Oono Abstract, §6.1, p.1.
A POSITA would have modified Huang with Oono because Huang already seeks personalized healthcare, risk identification, and just-in-time intervention from sensed gait data, expressly to significantly enhance the subject’s safety and health. Par. 0016 Huang wit Oono because is vital for understanding and improving physical and mental well-being and that ML is promising for modeling their relationships and offering actionable suggestions. Abstract, p.1 Oono
[Claim 2]
Huang, in combination with Oono, teaches,
The information generation device according to claim 1, wherein the processor is configured to execute the instructions to:
calculate a gait index using the sensor data; Huang, features related to gait ... include walking speed, stride height, stride length ... cadence or swing time [0029]; classifiers ... utilize the extracted features [0031].
Under BRI, gait index encompasses a derived gait characteristic; the applicant itself identifies distance, height, angle, speed, and time as gait indices. Huang therefore calculates the claimed gait index because its sensor data are processed into corresponding gait features such as walking speed, stride length, cadence, and swing time.
and input data including the gait index calculated using the sensor data to a disease risk estimation model that outputs a disease risk score indicating a degree of a disease risk , (Huang, classifiers ... utilize the extracted features and from the provided feature vector ... a confidence ... can be calculated [0031]; final layer ... provides the confidence values for the output classes [0034]; parameter representing a likelihood that the subject exhibits the clinical parameter [0035]; clinical parameter ... can relate to a certain medical condition [0036].)
and estimates disease risk information according to the disease risk score output from the disease risk estimation model (Huang states that the resulting clinical parameter is used for diagnosis, stratification or monitoring of a medical condition [0036].).
Huang derives gait features such as walking speed, stride height, stride length ... cadence or swing time from sensor data [0029], and its classifier utilizes the extracted features to select an outcome class and calculate a confidence [0031]. Huang further provides confidence values for the output classes, each confidence corresponding to an output class [0034], alternatively calculates a likelihood that the subject exhibits the clinical parameter [0035], and states that the clinical parameter can relate to a certain medical condition [0036].
Thus, a POSITA using Huang diagnostically would have had an express reason to treat its candidate conditions [0030] as the classifier classes and use the already-disclosed confidence for each class as the degree of risk for the corresponding condition.
[Claim 3]
Huang, in combination with Oono, teaches,
The information generation device according to claim 2, the processor execute is configured to execute the instructions to:
estimate the health measure according to the disease risk score of at least one of the subjects to be managed using a health measure estimation model that outputs the health measure according to an input of the disease risk score. (Huang, provide an alert based on the clinical parameter [0037]; generate the alert when the clinical parameter satisfies the alert condition based on comparison with a threshold [0038]; generate a just-in-time alert to prevent the occurrence of the detected risk [0038].)
[Claim 6]
Huang, in combination with Oono, teaches,
The information generation device according to claim 3, wherein the health measure estimation model and the disease risk estimation model are models learned using a machine learning method, and the disease risk estimation model includes an incomplete heterogeneous variational autoencoder.
Huang teaches machine-learning classification and regression for estimating a subject’s clinical parameter and the likelihood that the subject exhibits that parameter. Huang states that a statistical or machine learning-based classification can be applied to the pattern related to gait, balance or posture to assign a clinical parameter to the pattern and that a regression model configured to calculate a parameter representing a likelihood that the subject exhibits the clinical parameter may be used. Huang therefore teaches machine-learning models for personalized health and risk estimation, but does not disclose an incomplete heterogeneous variational autoencoder. See Huang, paragraphs 0006, 0016, 0035, 0038, and 0042.
Oono et al. teaches a machine learning model for estimating attributes about healthcare, lifestyles, and personalities and a deep generative model trained with masked modeling to learn the joint distribution of attributes conditioned on known ones. The reference expressly states, As a deep generative model, we employed Heterogeneous-Incomplete Variational Autoencoder. It further explains that the model handles heterogeneous attributes and missing values by estimating missing healthcare attributes from available health information. VHGM also describes estimating the probabilities of various diseases and healthcare measures such as blood pressure. See VHGM v1, abstract and sections 1, 2, 3, 6.1, and 7.1.
A person of ordinary skill in the art would have combined Oono’s HIVAE architecture with Huang’s machine-learning health-risk system. Both references address using learned models to estimate health-related outputs from subject data. Oono identifies the specific technical benefit of using HIVAE to model heterogeneous healthcare attributes and estimate missing values, while Huang provides the personalized clinical-parameter and risk-estimation application. A POSITA would therefore have used Oono’s HIVAE as the learned architecture for Huang’s health-measure or disease-risk estimation model to permit estimation from heterogeneous or incomplete subject information and to provide uncertainty associated with the estimate.
Notes: Claims 7-11, 14-17, and 20 are rejected with claims 1-3 and 6 for being very similar.
Claim(s) 4-5, 12-13 and 18-19 are rejected under 35 U.S.C. 103 as being unpatentable over US20180279915A1- Huang, and Oono, K., Charoenphakdee, et, al. refer to pto-892-U and Xue -CN112002429A.
[Claim 4]
Huang, in combination with Oono, teaches,
The information generation device according to claim 3, wherein the processor execute is configured to execute the instructions to:
estimate the health measure . (Huang, …A statistical or machine learning-based classification can be applied to the pattern related to gait, balance or posture to assign a clinical parameter to the pattern risk of a slip, trip and fall (STF) event can be determined based on the clinical parameter satisfying a STF condition. par. 0006; A regression model configured to provide calculate a parameter representing a likelihood that the subject exhibits the clinical parameter. par. 0036; "personalized healthcare related to gait, balance or posture to a subject" to enhance the subject's safety (par. [0016]) by applying a "regression model configured to calculate a parameter representing a likelihood that the subject exhibits the clinical parameter" (par. [0035]), wherein the system evaluates movement risk against a baseline where "the threshold can be set for the individual subject" or "established for the subject" rather than a generic population (pars. [0038], [0041-0042]). )
Huang teaches applying machine-learning classification to a subject’s movement pattern to assign a clinical parameter and using a regression model configured to calculate a parameter representing a likelihood that the subject exhibits the clinical parameter. Huang further uses the clinical parameter for personalized healthcare related to gait, balance or posture to a subject and permits a threshold to be set for the individual subject.
Huang does not explicitly teach how to use a disease-risk score to estimate a separate health measure or to customize the estimation model according to the subject’s way of working.
Oono et al. teaches a model for estimating attributes about healthcare, lifestyles, and personalities and learning attributes conditioned on known ones. VHGM further provides that users may query any combination of attributes as input and that the model estimates other attributes. Its applications include estimating health measures such as systolic blood pressure and estimating the probabilities of various diseases from health information. See VHGM v1, abstract and sections 2, 3, 6.1, and 7.1.
A person of ordinary skill in the art would have combined VHGM’s conditional health-attribute estimation with Huang’s disease-risk and personalized clinical-parameter analysis. The disease probability or disease-risk score would be provided as a known input attribute, and VHGM’s model would estimate another healthcare attribute, such as blood pressure, glucose, or another health measure. This combination would use VHGM’s disclosed input-output model for its established purpose of estimating an unknown healthcare attribute from known health information.
Xue further teaches customizing health-risk analysis using positions, work types, and working years, as well as work/rest behavior, business trips, and working behavior. Xue also establishes a model using physical-examination information and professional characteristics to determine whether a subject has a health risk. In addition, Xue sets an abnormal-value threshold according to age, work, and race. See p.4, p.7
A person of ordinary skill in the art would have further combined Xue work-related characteristics with the Huang and VHGM model inputs or training data. The purpose would be to customize the health-measure estimation model according to the managed subject’s working conditions, because Xue expressly associates work characteristics and work behavior with health-risk determination. The resulting system would predictably estimate a health measure using a disease-risk score while accounting for the subject’s way of working. The combination applies known machine-learning inputs and personalization techniques to their established functions would personalize healthcare related to gait, balance, or posture to a subject, and permits a threshold to be set for the individual subject.
[Claim 5]
Huang, in combination with Oono, teaches,
The information generation device according to claim 3, wherein the processor execute is configured to execute the instructions to:
estimate the health measure according to the disease risk score of at least one of the subjects to be managed using the health measure estimation model customized according to an industry type of an organization to which the subject to be managed belongs.
Huang teaches applying machine-learning classification and regression to estimate a subject-related clinical parameter and a likelihood that the subject exhibits that parameter. Huang further teaches personalized healthcare related to gait, balance or posture to a subject and setting a threshold for the individual subject. See Huang, paragraphs 0006, 0016, 0035, 0038, and 0042.
Huang does not expressly disclose estimating a separate health measure from a disease-risk score using a model customized according to the industry type of the subject’s organization.
Oono et al. teaches a model for estimating attributes about healthcare, lifestyles, and personalities and learning attributes conditioned on known ones. VHGM further provides that users may query any combination of attributes as input and that the model estimates other attributes. Its applications include estimating health measures such as systolic blood pressure and estimating the probabilities of various diseases from health information. See VHGM v1, abstract and sections 2, 3, 6.1, and 7.1.
A person of ordinary skill in the art would have combined VHGM’s conditional health-attribute estimation with Huang’s disease-risk and personalized clinical-parameter analysis. The disease probability or disease-risk score would be provided as a known input attribute, and VHGM’s model would estimate another healthcare attribute, such as blood pressure, glucose, or another health measure. This combination would use VHGM’s disclosed input-output model for its established purpose of estimating an unknown healthcare attribute from known health information.
Xue further teaches using organization and professional-category information in health-risk modeling. Xue identifies personal information including affiliated enterprise, positions, work types, and working years. Xue also establishes a professional category comparison model using personal information, physical-examination information, and professional characteristics to determine whether the subject has a health risk. The system further creates enterprise-level portraits from employee information and enterprise attributes and environment. See p.4, p.7
A person of ordinary skill in the art would have further combined Xue’s enterprise and professional-category information with the Huang and VHGM model inputs or training data. The reason would be to customize the health-measure estimation model according to the organizational context in which the subject belongs, because Xue expressly uses enterprise and professional-category characteristics to determine health risk. The resulting system would predictably estimate a health measure using a disease-risk score while accounting for the organization’s industry or professional classification.
Note: Claims 12-13 and 18-19 are rejected with claims 4-5 for being very similar.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOSHUA DAMIAN RUIZ whose telephone number is (571)272-0409. The examiner can normally be reached 0800-1800.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Shahid Merchant can be reached at (571) 270-1360. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/J.D.R./Examiner, Art Unit 3684
/Shahid Merchant/Supervisory Patent Examiner, Art Unit 3684
Prosecution Insights