US 11,749,405 B2
Systems and methods for dynamic biometric detection and response
Robin H. Stewart, Woodbridge, VA (US); and Qiliang Li, Fairfax, VA (US)
Filed by Robin H. Stewart, Woodbridge, VA (US); and Qiliang Li, Fairfax, VA (US)
Filed on Jul. 14, 2020, as Appl. No. 16/928,464.
Claims priority of provisional application 62/922,623, filed on Aug. 20, 2019.
Claims priority of provisional application 62/973,470, filed on Oct. 7, 2019.
Prior Publication US 2021/0052170 A1, Feb. 25, 2021
Int. Cl. G16H 50/20 (2018.01); A61B 5/0205 (2006.01); A61B 5/00 (2006.01); A61B 5/11 (2006.01); A61B 5/145 (2006.01); A61B 5/1455 (2006.01); A61J 7/04 (2006.01); G16H 10/60 (2018.01); G16H 50/30 (2018.01); G06F 18/2132 (2023.01); G06F 18/2135 (2023.01)
CPC G16H 50/20 (2018.01) [A61B 5/0022 (2013.01); A61B 5/02055 (2013.01); A61B 5/1112 (2013.01); A61B 5/1117 (2013.01); A61B 5/14532 (2013.01); A61B 5/14551 (2013.01); A61B 5/681 (2013.01); A61B 5/7264 (2013.01); A61B 5/746 (2013.01); A61J 7/0427 (2015.05); G06F 18/2132 (2023.01); G06F 18/2135 (2023.01); G16H 10/60 (2018.01); G16H 50/30 (2018.01); A61B 2562/0219 (2013.01); A61J 2205/60 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method of data processing for a dynamic biometric detection system, having by a processor-based medical detection agent, comprising:
retrieving a user profile associated with a user from a storage unit by the processor-based medical detection agent;
training the dynamic biometric detection system over a predetermined time period to form a behavior model associated with the user, wherein the training comprises,
detecting user location associated with the user, using an GPS device within a user behavior module of the processor-based medical detection agent;
sensing user motion associated with the user, using a plurality of sensors coupled to the processor-based medical detection agent;
detecting user activity based upon the sensed user motion, the detected location, and the user profile using an activity detection unit within the user behavior module;
sensing biomarkers of the user, using the processor-based medical detection agent;
detecting user behavior pattern to form the behavior model associated with the user over the training predetermined time period, using a user behavior module within the processor-based medical detection agent, wherein detecting user behavior pattern comprises,
generating a master matrix of sensed biomarkers, detected user location, sensed user motion, and detected user activity associated with the user,
extracting a set of one or more abridged matrices from the master matrix,
selecting a first matrix from the extracted set,
processing the first matrix using a Convolution Neural Network (CNN),
extracting a second matrix from the CNN processed matrix, and
generating the user behavior pattern from the extracted second matrix using a pattern recognition technique to identify repetitive data inputs associated with the user profile to form the behavior model;
detecting one or more parameters continuously from the group consisting of user location, user motion, user activity, and biomarkers;
determining an anomaly based upon the behavior model associated with the user and the detected one or more parameters, using an operations module and a policy manager within the processor-based medical detection agent;
generating a record of the detected anomaly, using an alert generator within the processor-based medical detection agent; and
updating the user profile with the generated record using the processor-based medical detection agent.