DETAILED ACTION
Claims 1, 3-11, & 13-22 are currently pending and have been examined.
This action is in response to the amendment filed on 4/29/2026.
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 .
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.
Claims 1, 3-11, & 13-21 are rejected under 35 U.S.C. 103 as being unpatentable over Wexler (US20230255556) in view of Mrowka (US20170018199A1), Cox (US20170286622), and Estes (US20200147305).
As per Claim 1, Wexler teaches a method for collecting user information for determining parameters for adjusting operation of a user’s insulin delivery pump, the method comprising:
determining a large language model trained to process spoken data and associate categories to the spoken data (para. 46: models described including generative pre-trained transformer model);
determining a set of rules adapted to cause the large language model to associate first portions of the spoken data with a first category corresponding to food (para. 53, 59, 172, 258: unstructured text input into system converted into logical pairings describing categories of text including food entity, food-related word, quantifiers);
determining user spoken data corresponding to the user of the insulin delivery pump (para. 52, 64, 278: input received from user including text or speech and parsing input including meal description);
receiving an adjustment regarding the user record from the user (para. 45: user can provide updates to data into the system);
Wexler does not expressly teach determining a set of rules defining categories for parsing the spoken data, the set of rules adapted to determining a set of rules adapted to cause the large language model to associate second portions of the spoken data corresponding to activity; providing the set of rules and the user spoken data to the large language model, wherein the large language model is configured to parse the user spoken data based on the set of rules to generate food data associated with the first category and activity data associated with the second category; receiving, from the large language model, the food data associated with the first category and the activity data associated with the second category; automatically populating a user record corresponding to the user based on the food data and the activity data.
Mrowka, however, teaches to using a voice-based system for diet and exercise tracking where a machine learning model is used to analyze a user’s voice data (para. 32). The input audio voice data is parsed into a text segment and is analyzed by the system using rules to determine if a match exists with previously input data (para. 41). Mrowka teaches to matching the parsed segment with previous data and assigning the segment to a diet tracking component or exercise tracking component (para. 41). The Examiner interprets the diet tracking and exercise tracking components as categories for the parsed segment data. Mrowka also teaches to parsing the data and generating output based on the detected diet or exercise activities identified (Fig. 5b; para. 49). Mrowka further teaches to storing the data into a database (para. 40). The Examiner asserts that Mrowka teaches to using natural language processing and machine learning models and a large language model is a type of natural language processing model and therefore Mrowka teaches to this limitation. Furthermore, Wexler teaches to using one or more machine learning models configured to analyze the input data and generate the output data and where the machine learning model can include a generative pre-trained transformer.
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the aforementioned features in Mrowka with Wexler based on the motivation of diet and exercise tracking device that has one-submission tracking capability (Mrowka – para. 12).
Wexler and Mrowka do not expressly teach parsing the user spoken data using the model based on the set of rules to generate activity data associated with the second category; automatically populating a user record corresponding to the user based the activity data.
Cox, however, teaches to using speech-to-text algorithms with natural language processing to process and parse natural language input and extracting key features from the input and associating the features with corresponding concepts and values (para. 151). Cox further teaches to receiving various lifestyle information and classifying the data in accordance with a number of defined lifestyle categories (para. 52). These categories can include level of physical activity, level of availability to healthy food sources, quality of home and work environment (lighting, air quality, quietness, safety, etc.), level of access to exercise facilities, various qualitative aspects of the patient's home and work life, and the like (para. 126). Cox further teaches to storing the data in a database (para. 102).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the aforementioned features in Cox with Wexler and Mrowka based on the motivation of personalize a patient's on-going treatment and care based on both their medical condition and the patient's own personal lifestyle, taking into account multiple lifestyle conditions and the facilities and resources available to that particular patient based on their lifestyle (Cox – para. 36).
Wexler, Mrowka, and Cox do not expressly teach
automatically generating adjusted pump parameters based on the adjustment, the food data and the activity data for adjusting operation of the insulin delivery pump; and
automatically sending a control signal to the insulin delivery pump once the adjusted pump parameters are generated, the control signal instructing the insulin delivery pump to deliver an adjusted amount of insulin based on the adjusted pump parameters.
Estes, however, teaches to a user-configurable insulin pump system where various input patient data can be used by an infusion pump system to calculate an updated dosage for medication treatment (para. 4). Estes also teaches to using various parameter data to provide adjusted medication dosage parameters such as physical activity data, food data (para. 25, 30-32).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the aforementioned features in Estes with Wexler, Mrowka, and Cox based on the motivation of safety and efficacy of an infusion pump system may be enhanced because the convenient manner of inputting data to the infusion pump using speech or image recognition may facilitate more timely and complete data entry by the user (Estes – para. 12).
As per claim 3, Wexler, Mrowka, Cox, and Estes teaches the method of claim 1. Wexler teaches further comprising determining analyzed data based on the food data, wherein the analyzed data comprises food values (para. 61: nutritional data acquired and is correlated to user meal data).
As per claim 4, Wexler, Mrowka, Cox, and Estes teaches the method of claim 3. Wexler teaches wherein the food values correspond to a calorie value and/or a carbohydrate value (para. 33: caloric and carbohydrate data acquired).
As per claim 5, Wexler, Mrowka, Cox, and Estes teaches the method of claim 4. Wexler teaches wherein the analyzed data comprises a food type and wherein the food values are determined based on a database associating food types with food values (para. 40, 61-62: food logging determining by connecting to external database containing nutritional information and querying with scaled nutrition data; individual result contains multiple nutrition information mappings, each with its own serving description which must subsequently be evaluated).
As per claim 6, Wexler, Mrowka, Cox, and Estes teaches the method of claim 3. Wexler teaches further comprising: sending the food data to a remote device (para. 60-62: query sent to external system); and determining the food values from the remote device (para. 60-62: external resources provide nutritional data to food logger system).
As per claim 7, Wexler, Mrowka, Cox, and Estes teaches the method of claim 3. Wexler, Mrowka, & Cox do not expressly teach wherein the adjusted pump parameters are further based on the analyzed data.
Estes, however, teaches to a user-configurable insulin pump system where various input patient data can be used by an infusion pump system to calculate an updated dosage for medication treatment (para. 4).
The motivations to combine the above mentioned references are discussed in the rejection of claim 1, and incorporated herein.
As per claim 8, Wexler, Mrowka, Cox, and Estes teaches the method of claim 1. Wexler teaches further comprising determining activity values based on the activity data, the activity values corresponding to an amount of energy expended by the user (para. 33: health related data obtained from user including physical activity or exercise data).
As per claim 9, Wexler, Mrowka, Cox, and Estes teaches the method of claim 1. Wexler, Mrowka, & Cox do not expressly teach further comprising causing a user device to present the adjusted pump parameters and/or a request to change operation of the insulin delivery pump based on the adjusted pump parameters.
Estes, however, teaches to a user-configurable insulin pump system where various input patient data can be used by an infusion pump system to calculate an updated dosage for medication treatment (para. 4). Estes also teaches to displaying the dosage to a user (para. 38).
The motivations to combine the above mentioned references are discussed in the rejection of claim 1, and incorporated herein.
As per claim 10, Wexler, Mrowka, Cox, and Estes teaches the method of claim 1. Wexler teaches further comprising:
causing a user device to present at least a portion of the user record (para. 62: output of user data via food logger system using display).
Claims 11 & 13-20 recite substantially similar limitations as those already addressed in claims 1 & 3-10, and, as such, are rejected for similar reasons as given above.
As per claim 21 Wexler, Mrowka, Cox, and Estes teaches the method of claim 1. Wexler teaches wherein: the set of rules defines the first category corresponding to food; the set of rules further requires that (i) the food data associated with the first category comprises one or more food values including fiber, fat, protein, sugar, calories, or carbohydrates (para. 33: caloric and carbohydrate data acquired); and generating the food data comprises, respectively, (i) generating the one or more food values based on the first portions of the spoken data (para. 33, 273: caloric and carbohydrate data acquired; user can enter information via speech to text functionality).
Wexler and Mrowka do not expressly teach teaches wherein: the set of rules defines the second category corresponding to activity (ii) the activity data associated with the second category comprises one or more exercise values including energy expended or exercise intensity; generating the activity data comprises (ii) generating the one or more exercise values based on the second portions of the spoken data.
Cox, however, teaches to using speech-to-text algorithms with natural language processing to process and parse natural language input and extracting key features from the input and associating the features with corresponding concepts and values (para. 151). Cox, however, teaches to extracting user input data and associating the data with exercise category such as determining patterns of activity including levels of physical activity (para. 41).
The motivations to combine the above mentioned references are discussed in the rejection of claim 1, and incorporated herein.
Allowable Subject Matter
Claim 22 is objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Prior Art Rejection
All of the cited references fail to expressly teach or suggest, either alone or in combination, the features found within claim 22. In particular, the cited prior art of record fails to expressly teach or suggest the combination of: wherein: the set of rules further defines a third category corresponding to sleep and a fourth category corresponding to operation of the insulin delivery pump; the set of rules further requires that (i) sleep data associated with the third category comprises one or more sleep values including REM cycles or duration and (ii) operation data associated with the fourth category comprises one or more operation values including supply information, installation information, or wear information; parsing the user spoken data further generates (i) the sleep data based on third portions of the spoken data associated with the third category and (ii) the operation data based on fourth portions of the spoken data associated with the fourth category; and automatically generating adjusted pump parameters is further based on the sleep data and the operation data for adjusting operation of the insulin delivery pump.
The most relevant prior art of record includes:
Wexler (US20230255556) teaches to systems and methods for converting text inputs, that describe an instance of dietary intake in the form of unstructured text, to a detailed description of the nutritional content of the instance in the form of structured data. The food logging system can incorporate a natural language processing (NLP) system tailored for the food logging. The food logging system can extract information relevant to estimating nutrient values from unconstrained user text, namely, food items and their associated portions. The extracted information is used to make database queries and automate the steps that a user would perform when logging food. The system automates all portion conversions and calculations required to reach a final estimate of the nutrient content of a user's meal.
Mrowka (US20170018199) teaches to a wearable diet and exercise tracking device that has one-submission tracking capability. One-submission tracking capability of a wearable diet and exercise tracking device in accordance with the invention enables a user of such device to track (log) data related to a discrete incidence of food consumption or physical activity such as food eaten and/or exercises performed, including the associated food(s) and/or exercise(s) input quantity(ies) (e.g., 1 cup, 2 sets of 25 repetitions, etc.), as well as to automatically associate with this data and track information on the related calories and nutrients and/or calories burned, upon a single submission (input) into such device of voice information by the user speaking describing the particulars of such food consumption or physical activity without further action by the user.
Cox (US20170286622) teaches to an improved data processing apparatus and method and more specifically to mechanisms for performing patient risk assessment based on machine learning of health risks of a patient population.
Response to Arguments
Applicant’s arguments on pages 8-9 regarding claims 1, 3-11, & 13-21 being rejected under 35 USC § 103 have been fully considered but are moot in view of the amended grounds of rejection. Applicant also argues that:
Mrowka does not expressly teach using a large language model to perform the claimed steps described above.
The Examiner, however, asserts that that Mrowka teaches to using natural language processing and machine learning models and a large language model is a type of natural language processing model and therefore Mrowka teaches to this limitation. Furthermore, the claims broadly describe the use of a large language model to perform various functions including associating data with categories and generating additional data but there is no description as to how the large language model performs these claimed functions. Therefore, the Examiner considers the claimed large language model as analogous to a generic machine learning model which is described in Mrowka with the use of natural language processing with comprehensive text parsing and text cleaning methods; v) machine learning with large, real-time user data sets (see Mrowka – para. 14). Furthermore, Wexler teaches to using one or more machine learning models configured to analyze the input data and generate the output data and where the machine learning model can include a generative pre-trained transformer.
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 Jonathan K Ng whose telephone number is (571)270-7941. The examiner can normally be reached M-F 8 AM - 5 PM.
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, Anita Coupe can be reached on 571-270-7949. 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.
/Jonathan Ng/Primary Examiner, Art Unit 3619
Prosecution Insights