Automated Neuromuscular Assessment with Robotics + CV

Marker-free 3D hand pose estimation and closed-loop feedback in unsupervised rehabilitation sessions, mediated by Socially Assistive Robots.

A non-invasive system for automating standardized neuromuscular tests using Socially Assistive Robots (SAR) as the patient-facing interface and computer vision as the measurement layer.

What I’m building

  • Computer vision pipelines (Python / OpenCV) for 3D hand pose estimation that quantify fine motor deficits without markers.
  • Closed-loop algorithms that generate real-time feedback to patients during unsupervised rehabilitation sessions.
  • Integration of SAR with the perception stack so that the assessment runs end-to-end without a clinician in the loop.