Back to top anchor

IMPRESS: Intelligent Multimodal imaging platform to PREdict Stroke motor outcomeS

36 months
Approved budget:
Associate Professor Alan Wang
Professor Cathy Stinear
Professor Peter Barber
Health issue:
Neurological (CNS)
Proposal type:
Lay summary
Movement is commonly impaired by stroke, and recovery of movement is critical for regaining independence. Biomarkers of the motor system can predict recovery of movement after stroke. In this study, we will use retrospective and prospective data to develop an intelligent prediction platform based on routine clinical neuroimaging. The platform will use machine learning methods to identify features of the sensorimotor network and whole brain that can be used to classify patients according to expected stroke outcome. There will be four outcome categories for hand and arm function, and three outcome categories for walking ability. Once validated, the platform will enable faster, smarter and more reliable predictions for stroke patients, with the potential to improve the quality and efficiency of rehabilitation. Being able to accurately predict recovery and outcomes soon after stroke enables clinicians, patients and families to set appropriate goals for rehabilitation and plan for life after stroke.