parkinson-epilepsy-behavior-recognition-analysis

Video recognition detects Parkinson tremors and epilepsy convulsions by analyzing monitoring video frames to identify abnormal limb movements, stiffness, and gait abnormalities for non-contact home risk monitoring.

To analyze abnormal behavior, upload a patient monitoring video and request recognition analysis using your username or phone as the open-id, which calls the cloud API to generate a structured risk assessment report with suggested actions.

The analysis identifies abnormal behaviors including limb tremors, convulsions, stiffness, and gait abnormalities, producing frequency statistics, duration statistics, and an overall risk assessment from the monitoring videos.

Yes, when history-query keywords are used, the workflow fetches the user's historical analysis report list from the cloud API and renders it as a Markdown table with clickable links to access past reports.

Yes, executing the provided scripts requires obtaining an open-id via the required configuration and interaction flow using your username or phone, which is necessary to call the cloud API for analysis and report retrieval.