Measuring Activities of Daily Living in Stroke Patients with Motion Artificial Intelligence

The goal this project is to create a motion artificial intelligence (AI) analysis software to monitor activities of daily living (ADL) for stroke survivors with disabilities. The project utilizes wearable smart devices and a supervised machine learning (ML) algorithm to monitor activities that are meaningful and useful for clinicians and healthcare providers. Objectives are: (1) create a working ML algorithm to recognize 23 ADLs with 10 post-inpatient rehabilitation stroke survivors, (2) create a basic user-interface and clinical practice protocol for use in clinic to record movements, (3) create the pipeline and framework to process recorded data, and (4) create the framework to remote-access activity logs. The resulting technology will inform healthcare providers on the duration, metabolic equivalent, and frequency of engagement in particular activities, which can be used to monitor recovery post-inpatient care, reduce risks such as re-hospitalization, and provide objective indicators of independence and rehabilitation efficacy.