Automatic identification of, and responding to, cognition impairment

US12431235B1 · kind B1 · utility

Assignee

• United Services Automobile Association (USAA) · US

Inventors

• Nolan Serrao · Plano, US
• Keegan Patrick Hayes · Whitestown, US
• Arthur Quentin Smith · Fredericksburg, US
• John Andrew Weems · San Antonio, US
• Salvador Adrian Bretado · San Antonio, US
• Justin Royell Nash · Little Elm, US
• David Joaquin Harris · San Antonio, US
• Carol Lyn Lawrence · Fair Oaks Ranch, US
• Theresa Marie Matowitz · San Antonio, US
• Courtney St. Martin · Forney, US

Key dates

• Filing date: Dec 21, 2021
• Grant date: Sep 30, 2025
• Expiry date: Apr 8, 2043

Classification

• Technology area (CPC G): Physics
• CPC primary: G16H50/20
• WIPO field: Medical technology
• WIPO sector: Instruments

Abstract

Aspects of the present disclosure are directed to a cognition analysis and response system that can A) identify and classify machine learning training data and use the training data to build a cognitive impairment machine learning model; B) identify cognitive impairments by applying the cognitive impairment machine learning model and performing a false positive/false negative analysis; and C) apply various automated adaptive measures for users with identified cognitive impairment. The cognition analysis and response system can generate training data by using an unsupervised training process and verifying results with validations of training items that have been pre-classified. The cognition analysis and response system can then iterate through each set of inputs paired with cognitive impairment classification, applying a machine learning model and updating the model based on a comparison of the model output to the cognitive impairment classification.