Predicting analyte levels based on intermittent sensor data
Inventors
Karinka, Shridhara Alva • Rao, Vivek S. • Wang, Chung-Che • ROBINSON, Peter Gary • OLSON, Jennifer Lynn • Hua, Xuandong • Kumar, Panganamala Ashwin • Jangam, Sujit R.
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12744127-B2
Publication Date
2026-09-22
Expiration Date
Abstract
Techniques for predicting analyte levels based on intermittent sensor data are disclosed. First sensor data is acquired from an analyte sensor. The first sensor data reflects analyte levels of a user who is wearing the analyte sensor. The first sensor data is collected over a first time period. Later, a determination is made as to whether data is still being acquired from the analyte sensor. As a result of determining that data is no longer being acquired from the analyte sensor, the first sensor data is classified as intermittent analyte data. The intermittent analyte data is then used to generate predicted analyte level data for the user during a second time period that is subsequent to the first time period. The predicted analyte level data is reflective of the intermittent analyte data.
Core Innovation
The disclosed system and methods use analyte sensor data collected during a first time period to generate predicted analyte levels or physiological response data for a later period when updated sensor data is no longer acquired. The predicted data updates a user profile or health model to maintain insights during sensor-data gaps. The approach also incorporates contextual inputs, population data, meal scoring, and coaching.
The system determines that predicted data is stale when elapsed time since the most recent sensor data exceeds a predetermined threshold, or when an associated confidence level or accuracy metric falls below a threshold value. In response, the system prompts the user to facilitate acquisition of new analyte sensor data from an existing or new sensor, and acquires the new sensor data.
Claims Coverage
The claims comprise three independent claims covering three inventive features directed to computer-based analyte prediction, analyte-profile generation, and health-model generation when updated sensor data is unavailable.
Computer system for predicting analyte levels
Independent claim 1 recites a processor system and storage system that use first intermittent sensor data to generate predicted analyte level data for a later time period when acquisition stops. The system determines that the prediction is stale when elapsed time exceeds a predetermined threshold or a confidence level or accuracy metric falls below a threshold value, prompts the user to facilitate acquisition of new analyte sensor data, and acquires the new sensor data after the prompt.
Method for generating an analyte level profile
Independent claim 10 recites generating a profile tracking user analyte levels throughout a first time period using first sensor data, and updating the profile during a subsequent period without acquired sensor data using predicted analyte level data generated from discontinuous analyte data. When the predicted data is stale, the method prompts the user to acquire new analyte sensor data from an existing or new sensor.
Method for generating a health model
Independent claim 15 recites generating a health model tracking user physiological response throughout a first time period using first sensor data, and updating the model during a subsequent period without updated sensor data using predicted physiological response data generated from periodic sensor data. When the predicted data is stale, the method prompts the user to acquire new analyte sensor data.
Together, the three independent claims cover generating predictions, analyte profiles, and health models from sensor data, maintaining them during periods without updated sensor data, and prompting for new sensor data when predictions are determined to be stale.
Stated Advantages
The described predictions allow insights to remain available during sensor-data gaps.
Documented Applications
Computer-based prediction of analyte levels during gaps in sensor data, including continuous glucose monitor (CGM) and mobile application implementations.
Generation and updating of a user analyte level profile or health model using sensor data and predicted physiological response data.
User-facing stale-data alerts and prompts to facilitate acquisition of new analyte sensor data from an existing or new sensor.
Profile-related coaching prompts, meal scoring using glucose and meal information, score trends with an alarm, and comparison of a user profile with population trends.
Interested in licensing this patent?