GoogleGranted patentHealth & bodyInterest 8/10

Google patents privacy-preserving maps for tracking health conditions

Google describes a system that uses machine learning on personal devices to guess if users have a specific health condition based on their daily routines and search habits. Instead of sending personal data to a central server, the devices share only anonymous, aggregated statistics to help build detailed maps showing where the condition is spreading.

Why it mattersThis approach could allow public health officials to track disease outbreaks in real-time at a neighborhood level without compromising individual privacy.

In the filing’s own words

Predictive modeling using statistical evaluation combined with federated learning is described to enable incident map creation while preserving anonymity and achieving fine granularity. Machine-learned models ( 116) are trained to infer when persons associated with individual user devices ( 102) do or do not have a condition. The models ( 116) are deployed to the user devices ( 102) and return generalized location data and aggregated statistics about inferences made by the models. A remote system collects the aggregated statistics ( 120) and builds an incidence map ( 122 - 1, 122 - 2) identifying hotspots and coldspots for the condition. The incidence map identifies affected regions, with fine granularity down to a neighborhood or street level. The subregion-level information is regularly updated as new inferences, and new aggregated statistics, are made. Hence, the incidence map ( 122 - 1, 122 - 2) is current and highly detailed.

What exactly they claim (claim 1)

1. A computer-implemented method for maintaining privacy during attribution of a condition, the method comprising: training, by a remote system, a machine-learned model to infer, based on signals received by a user device in a geographic region, whether a person associated with the user device has the condition, wherein the signals comprise daily routine signals and text-based signals associated with user devices, wherein the daily routine signals are based on location and mobility of user devices, wherein the signals further comprise indexed search information comprising search queries associated with the user device; deploying, by the remote system, copies of the machine-learned model for local execution at each of a group of user devices; for each individual machine-learned model of the copies of the machine-learned model, instructing a respective user device of the group of user devices to perform federated learning on a respective copy of the machine-learned model on the respective user device of the group of user devices; collecting, by the remote system and from the group of user devices, aggregated statistics of inferences made by the copies of the machine-learned model; determining, by the remote system, based on the aggregated statistics, an incidence rate of the condition in a particular subregion of the geographic region; and outputting, by the remote system, based on the incidence rate of the condition in the particular subregion, an incidence map of the geographic region indicating a different incidence rate between multiple subregions of the geographic region.

Facts

Company
Google (Google LLC)
Official title
Maintaining privacy during attribution of a condition
Patent number
US12792926B2
Application number
17772883
Filed
Oct 28, 2019
Granted
Nov 10, 2026
Inventors
1
Claims
15

Classified under

G06N 3/098 Physics & computingG06F 16/2477 Physics & computingG06F 21/6218 Physics & computingG06N 20/00 Physics & computingG16H 50/80 Physics & computingG06F 2221/2111 Physics & computing

More from Google in Health & body

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.