Automated Stroke Detection via Audio-Video Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current manual evaluation methods for stroke severity and early detection, such as the RACE scale, are prone to inter-observer variability, require extensive training and resources, and can lead to delays in treatment due to biases and inefficiencies.
Innovation Solution
An automated system using a user-friendly AI agent that detects and tracks facial features and body parts through an audio-video interface, analyzing facial palsy, motor impairments, and speech to generate a standardized stroke score, thereby reducing human error and increasing efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual evaluation methods (RACE scale) are used for stroke detection, then human expertise and judgment can be applied, but inter-observer variability and treatment delays occur
Solution Approach 1:
The patent replaces the manual mechanical assessment process with an automated audio-video-based system using machine learning models. The system captures video of patient movements and audio of patient responses, then automatically analyzes these inputs to generate stroke scale scores, eliminating human observer variability while maintaining assessment accuracy.
2Measurement precision
If manual stroke scale evaluation is performed, then clinical judgment can be applied, but extensive training and personnel resources are required
Solution Approach 1:
The system enables prehospital settings to perform their own stroke assessments without requiring specialized trained personnel. The automated audio-video analysis system allows any responder to obtain reliable stroke scale scores through simple video recording and audio capture, making the technology self-sufficient and easily deployable in resource-limited environments.
3Reliability
If automated audio-video analysis is used for stroke detection, then inter-observer variability is eliminated, but system complexity increases
Solution Approach 1:
The system uses a single multi-functional audio-video platform that can perform multiple stroke assessment functions simultaneously. The same system captures both video for motor function analysis and audio for speech/language assessment, then processes both inputs through integrated machine learning models to generate comprehensive stroke scale scores, reducing the need for multiple separate complex devices.
Data Source
AI summary
An example system for detecting stroke includes an imaging device; a microphone; and a computing device. The computing device is configured to receive a sequence of images capturing a state of a patient; analyze the sequence of images to detect one or more of limb impairment, gaze impairment, or facial palsy; and assign a respective numeric score to each of the detected one or more limb impairment, gaze impairment, or facial palsy. The computing device is also configured to receive an audio signal capturing a voice of the patient; analyze the audio signal to detect aphasia or agnosia; assign a respective numeric score to the detected aphasia or agnosia. The computing device is further configured to generate a stroke score, which is a sum of the respective numeric scores for the detected one or more of limb impairment, gaze impairment, or facial palsy and the detected aphasia or agnosia.