Patent Application 16916004 - SYSTEM AND METHODS UTILIZING ARTIFICIAL

Title: SYSTEM AND METHODS UTILIZING ARTIFICIAL INTELLIGENCE ALGORITHMS TO ANALYZE WEARABLE ACTIVITY TRACKER DATA

Application Information

Invention Title: SYSTEM AND METHODS UTILIZING ARTIFICIAL INTELLIGENCE ALGORITHMS TO ANALYZE WEARABLE ACTIVITY TRACKER DATA
Application Number: 16916004
Submission Date: 2025-05-14T00:00:00.000Z
Effective Filing Date: 2020-06-29T00:00:00.000Z
Filing Date: 2020-06-29T00:00:00.000Z
National Class: 600
National Sub-Class: 483000
Examiner Employee Number: 95210
Art Unit: 3791
Tech Center: 3700

Rejection Summary

102 Rejections: 0
103 Rejections: 1

Cited Patents

The following patents were cited in the rejection:

Office Action Text

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 .

Response to Amendment
The Amendment filed January 31, 2025 has been entered. Claims 1-3, 5-9, 11-13, 15-19 and 21-24 remain pending in the application.

Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.

The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.

Claims 1-3, 5-9, 11-13, 15-19 and 21-24 are rejected under 35 U.S.C. 103 as being unpatentable over Leventhal et al. (US 2020/0185102 A1) (hereinafter – Leventhal) in view of Shariff et al. (US 2016/0302671 A1) (hereinafter – Shariff) in further view of Malhotra et al. (US 2018/0211010 A1) (“Malhotra”).

Regarding claims 1 and 12, Leventhal discloses A system for performing health monitoring, the system comprising (Abstract and entire document):
a wearable device that includes a wireless interface (Para. [0086], “Using different sources of electronic medical records (or population medical database, or integrated health records)” and Para. [0167], “wearable sensors data” see also para. [0386 – 0387]. It is further noted that wireless interfaces on wearable devices are considered well known within the art.); and
a server device including a memory and one or more processors configured to (Para. [0383], “As shown in FIG. 33, the server system 200 can include one or more additional hardware components as desired. Exemplary additional hardware components include, but are not limited to, a memory 202 (alternatively referred to herein as a non-transitory computer readable medium). Exemplary memory 202 can include, for example, random access memory (RAM), static RAM, dynamic RAM, read-only memory (ROM), programmable ROM, erasable programmable ROM, electrically erasable programmable ROM, flash memory, secure digital (SD) card, and/or the like. Instructions for implementing the server system 200 can be stored on the memory 202 to be executed by the processor 201.”):
receive monitored parameter data from the wearable device via the wireless interface (Para. [0086], “Using different sources of electronic medical records (or population medical database, or integrated health records)” and Para. [0167], “wearable sensors data”);
determine an artificial intelligence algorithm to use for processing the monitored parameter data (Para. [0230], “In one embodiment, probabilistic models can be selected based on certain criteria. For example, probabilistic models can be selected in order to adapt for different data types for the underlying feature vector. For example, the feature vectors can contain profile features, symptoms, attributes, and values that are binary, multinomial, continuous, or a combination thereof. Models such as neural networks can be selected for a great variety of data types. Hidden Markov models can be selected for some specific combination of multinomial and continuous data types.” A model is selected based on the data and the data type.);
wherein the codes included in the health records comprise at least one of ICD (International Statistical Classification of Diseases) 9/10 diagnostic codes, CPT (Current Procedural Terminology) procedure codes, GPI (Generic Product Identifier) drug codes, or LOINC (Logical Observation Identifiers Names and Codes) lab codes (Para. [0133], “Doctor visit summary diagnosis (e.g., represented as coded International Classification of Diseases 10 (ICD10), or other coding)”);
process the filtered monitored parameter data and the embedding data to generate an input vector, wherein the input vector comprises a first portion corresponding to the filtered monitored parameter data and a second portion corresponding to the embedding data (Para. [0159] – [0171], “Performing model building on the “clean medical records” data. Model building can use various “Machine Learning” or “Deep Learning” algorithms used in the field of algorithm development or machine learning. Model building may use some parts of the clean medical records data as the feature vector for model training, such as:” an input vector is generated based on the embedding data and the parameter data.);
process the input vector by the artificial intelligence algorithm to generate an output vector (Para. [0098] – [0099], “Performing model building using a model building module (or model training module) 212 (shown in FIG. 3B) on the “clean structured medical history” data. Model building can use various “Machine Learning” or “Deep Learning” methods, for example. Model building may use: [0099] a. the SAV feature structure” and para. [0071], “The process continues with model training. The server system 200 can create a mathematical and learning model that, given an input of SAV's, can provide an output of the most probable conditions (or prescribed medication, or lab test referral). The output of such model can be in the form of a probability vector 214 (shown in FIG. 12A) including a plurality of probabilities, each associated with a medical condition.” And para. [0357], “For that combined vectors state, using a machine learned model (for example, the health predictive model 210), calculate the distribution of probabilities (that is, a vector for each condition the system knows, where for that condition the probability of having that given the above input vectors is presented). The output vector can be denoted as a current probability vector, or f (x.sub.profile,x.sub.symptoms).” see also [0230]); and
schedule an appointment with a healthcare provider based on the output vector ( Para. [0071], “At least part of the probability vector can be presented via the user interface 340 (shown in FIGS. 15 and 16).” And para. [0048], “Turning to FIG. 1, the environment 100 is shown as including a server system 200 communicating with one or more client devices 300 via at least one communication network 400. The server system 200 can include one or more computer systems that are individually and/or collectively configured to transfer information and/or data among the client devices 300.”Para. [0003], “Digital healthcare solutions provide an inexpensive and quick way for people to receive health information pertaining to certain symptoms and diseases (or medical conditions) prior to, or in additional to, deciding to visit a doctor's office in person.” And para. [0119], “Allowing the users to connect to care providers to receive further help” and para. [0188], “Allowing care provider to respond back to user with simple options (for example: stay home, sending you a prescription, ordered a lab test, go to ER)” depending on results, the best course of action is suggested which includes going to ER or scheduling the appointment.).
Leventhal fails to disclose discard a subset of the monitored parameter data from the wearable device based on the determined artificial intelligence algorithm to generate filtered monitored parameter data;
obtain embedding data by processing health records corresponding to a user registered to the wearable device using a natural language processing algorithm, wherein the embedding data comprises an n-dimensionless vector of dimensionless features obtained based on codes included in the health records corresponding to the user, and
However, in the same field of endeavor, Shariff teaches discard one or more types of the monitored parameter data from the wearable device based on the determined artificial intelligence algorithm to generate filtered monitored parameter data ([0025], “The wearable electronic device 102 may also include additional physiological sensors (not shown) such as a skin temperature sensor, a galvanic skin response (GSR) sensor, a blood pressure sensor, and the like. The wearable electronic device 102 may additionally include sensors that are used to detect non-physiological data; such sensors may include a gyroscope, a light sensor, and the like, although these sensors may be used to determine physiological data either in addition or instead.” And [0028] and [0042] and [0053], the input to the machine learning algorithm includes the data from the wearable sensors, but not all of the data from all of the sensors. Data is chosen for the algorithm depending on the desired output, such as heart rate and respiration rate data is input, but the temperature, GSR, gyroscope data, etc. may not be included and is then considered a discarded data type resulting in filtered monitored parameter data.);
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, to modify the system/method/device as taught by Leventhal to include discard one or more types of the monitored parameter data from the wearable device based on the determined artificial intelligence algorithm to generate filtered monitored parameter data as taught by Shariff to monitor outside of clinical settings ([0015], “Because these devices are wearable they may provide continuous, or approximately continuous, monitoring of physiological data for people outside of clinical settings.”).
Leventhal as modified fails to disclose obtain embedding data by processing health records corresponding to a user registered to the wearable device using a natural language processing algorithm, wherein the embedding data comprises an n-dimensionless vector of dimensionless features obtained based on codes included in the health records corresponding to the user, and
However, in the same field of endeavor, Malhotra teaches obtain embedding data by processing health records corresponding to a user registered to the wearable device using a natural language processing algorithm, wherein the embedding data comprises an n-dimensionless vector of dimensionless features obtained based on codes included in the health records corresponding to the user (Para. [0084 – [0086], “An embedding layer is a type of layer that usable in deep neural networks and used in Natural Language Processing (NLP) applications. An embedding layer is a kind of matrix and an input vector of the deep neural network, which is a one-hot or multi-hot vector in NLP in preferred embodiments, is multiplied by this matrix. One preferred embodiment uses either a matrix initialized with some random numbers or a matrix of which values are trained by other deep neural network.” Thus, the specific embedding vectors can be used in the model of Leventhal as modified), and
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, to modify the system/method/device as taught by Leventhal as modified to include obtain embedding data by processing health records corresponding to a user registered to the wearable device using a natural language processing algorithm, wherein the embedding data comprises an n-dimensionless vector of dimensionless features obtained based on codes included in the health records corresponding to the user as taught by Malhotra in order to process the data that is similar to natural language (Para. [0085], “The Med2Vec model is an advanced variation of the Word2Vec model that is based on the fact that the nature of medical data is similar with that of natural languages.”).
Regarding claims 2, 13, 21 and 22, Leventhal as modified discloses The system of claim 1, Leventhal further discloses wherein the wearable device is an activity tracker, and wherein the monitored parameter data includes data points related to one or more of: a heart rate; an oxygen level; an activity level including at least one of a number of steps, a number of flights climbed, or a duration of exercise; or a number of calories burned (Para. [0104], ‘measurements (e.g., blood pressure, weight, height, heart rate, etc.)”).
Regarding claim 3, Leventhal as modified discloses The system of claim 2, Leventhal further discloses wherein the monitored parameter data further includes information logged by the user manually (Para. [0078], “The user interface 340 (for example, shown in FIGS. 14-18 and 21-32) can emulate a typical physician-patient dialogue. Additional and/or alternative user interfaces can be provided where a feature vector is constructed from free text input in both offline or during an active conversation (on line), or from forms filled by physicians, etc.”).
Regarding claims 5 and 15, Leventhal as modified discloses The system of claim 1, Leventhal further discloses wherein the health records are stored in a database and comprise at least one of: claims records received from a health care provider; prescription records received from a pharmacy; or laboratory results received from a laboratory or other health care provider (Para. [0054], “The operator 320 can open the app on the client device 300, log in to a user account that is registered in a user database (not shown) in the server system 200, and communicate with the server system 200. In some embodiments, the server system 200 can represent a backend functionality for the app. Each user account can be associated with an entity, such as a patient. The user database can include profile information and/or medical history for each user account. The profile information can include name (optional), age, gender, sex, weight, height, smoking habit, and/or the like.” And para. [0086], “Using different sources of electronic medical records (or population medical database, or integrated health records). And Para. [0081], “Such modules can include automatic case summary generation, communicating case details between different care givers, communication between care giver and patient, digital prescription of medications, lab test referrals, communication between patient and doctor that interacting into two different languages, etc.” and para. [0103] – [0112]).
Regarding claims 6 and 16, Leventhal as modified discloses The system of claim 1, Leventhal further discloses wherein the input vector data further includes social determinants data including one or more of: economic information; neighborhood information; education information; nutritional information; or other environmental information (Para. [0101], “demographic data (e.g., gender, age, district, etc.)” and [0103], “lifestyle information (e.g., smoking, drinking, diet, etc.)”).
Regarding claims 7 and 17, Leventhal as modified discloses The system of claim 1, Leventhal further discloses wherein the input vector data further includes demographic data including one or more of: age information; gender information; neighborhood type information; family size information; or employment indicator information (Para. [0054], “The operator 320 can open the app on the client device 300, log in to a user account that is registered in a user database (not shown) in the server system 200, and communicate with the server system 200. In some embodiments, the server system 200 can represent a backend functionality for the app. Each user account can be associated with an entity, such as a patient. The user database can include profile information and/or medical history for each user account. The profile information can include name (optional), age, gender, sex, weight, height, smoking habit, and/or the like.”).
Regarding claim 8, Leventhal as modified discloses The system of claim 1, Leventhal further discloses wherein the artificial intelligence algorithm comprises one or more of: a multi-layer perceptron (MLP) algorithm; a convolution neural network (CNN); or a recurrent neural network (RNN) (Para. [0071], “Model building is done by taking feature vectors (a vector per one patient case) where the vectors are known to have a certain diagnosis (encoded, for example, by the Doctor as ICD10 in the visit record), using the diagnosis (or any medical insight) as a category label, and applying one or more of common machine learning algorithms to train a model. Such algorithms can be of statistical, connected neural networks, and/or any other suitable methods. Non-limiting examples of such machine learning algorithms are: Naïve Bayes, Random Forest, Fully-connected neural network, convolutional neural network, Bayesian model, Bayesian Network, logistic regression, and others as well as combinations of them.”).
Regarding claims 9 and 18, Leventhal as modified discloses The system of claim 1, Leventhal further discloses wherein the server device is further configured to generate a notification message based on the output vector, wherein the notification message is transmitted to one of the wearable device or a mobile device associated with the wearable device to provide the user of the wearable device with a suggested action (See FIG. 17, and para. [0323], “FIG. 17 shows the user interface 340 as presenting questions via a follow-up conversation to obtain a diagnosis, and optionally a treatment, determined by a doctor. Advantageously, the follow-up conversation provides a simple way to mark the accuracy of the health predictive model 210, hence creating “closed loop” data for improving the health predictive model 210 (shown in FIG. 3A).” The output vector of the machine learning model is a probability vector that shows a percentage it is a certain diagnosis for example. And para. [0188], “Allowing care provider to respond back to user with simple options (for example: stay home, sending you a prescription, ordered a lab test, go to ER)” see further para. [0049], “An exemplary client device 300 can include a personal computer (PC) and/or mobile computer device, such as a tablet computer and/or a smart phone.).
Regarding claims 11 and 19 Leventhal as modified discloses The system of claim 1, Leventhal further discloses wherein the server device is further configured to generate a notification message based on the output vector, wherein the notification message is transmitted to a healthcare provider to suggest treatment options that may be applicable to a patient (See FIG. 17, and para. [0323], “FIG. 17 shows the user interface 340 as presenting questions via a follow-up conversation to obtain a diagnosis, and optionally a treatment, determined by a doctor. Advantageously, the follow-up conversation provides a simple way to mark the accuracy of the health predictive model 210, hence creating “closed loop” data for improving the health predictive model 210 (shown in FIG. 3A).” The output vector of the machine learning model is a probability vector that shows a percentage it is a certain diagnosis for example. And para. [0188], “Allowing care provider to respond back to user with simple options (for example: stay home, sending you a prescription, ordered a lab test, go to ER)”).
Regarding claim 23, the rejection is substantially the same as claims 1 and 12 and the same rejections are applied.
Regarding claim 24, the rejection is substantially the same as claims 9 and 18 and the same rejections are applied.

Response to Arguments
Applicant’s arguments, see Remarks pages 8-11, filed January 31, 2025, with respect to the 101 rejections have been fully considered and are persuasive. Since the amendment now includes “schedule an appointment with a healthcare provider based on the output vector.” The 101 rejection has been withdrawn.
Applicant’s arguments with respect to claims 1-3, 5-9, 11-13, 15-19 and 21-24 have been considered but are moot because the new ground of rejection does not solely rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.

Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOSEPH A TOMBERS whose telephone number is (571)272-6851. The examiner can normally be reached on M-TH 7:00-16:00, F 7:00-11:00(Eastern).
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, Robert Chen can be reached on 571-272-3672. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see https://ppair-my.uspto.gov/pair/PrivatePair. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.

/J.A.T./Examiner, Art Unit 3791

/TSE W CHEN/Supervisory Patent Examiner, Art Unit 3791