Using data to improve physical therapy for people with neurological diseases
summary
Collects synchronized sensor streams — 50 Hz motor and joint-angle data, pressure insoles, finger-position encoders, SpO2 and blood pressure — and attaches Raspberry Pi loggers to devices, funnels files into a Docker pipeline and Databricks for analysis. Uses the WHO ICF to codify therapist assessments via an app and builds dashboards that estimate patient versus motor contribution to each step (goal: patient does more, motor less). Creates digital twins for benchmarking across hundreds of patients; work is at data‑engineering stage with a new therapy center opening and research underway.
AI-written summary · about
chapters (7)
- 0:00Robotic therapy introduction
- 3:04Data-driven robotic devices
- 8:07High-frequency sensor data
- 12:23Digital twins for outcomes
- 14:10Data engineering challenges
- 17:43Therapist feedback dashboards
- 20:36Future plans and collaboration
related talks
33mAI Healthcare – A Cybernetic Perspective
03mDBTB INT Sai Moturu r
31mData as an instrument for more efficient collaboration with robots
17mLearning to assess the quality of stroke rehabilitation exercises
01h 14mChallenges for Leveraging the Potential of Digital Technology in Medi
44mQuantified Self: Analyzing the Big Data of our Daily Life
50mInnovation and data analytics: Solving real-world challenges
01h 04mTake Control of the Data of You