Clinical pre/post-surgery video motion data trains one model across neurology, spine, senior care
zawthet · x · 2026-11-12
zawthet describes a healthcare AI data strategy that differs from public motion datasets (healthy volunteers, studio, one session): his team uses real clinical visits ordered by surgeons, pairing the same patient pre-op and at 3 months with clinical reasons attached. Coverage spans knees, neurology, PM&R, spine and senior care — tremor amplitude, sit-to-stand wattage, sway, hand and wrist signals — as video-derived pose, joint angles and continuous signals labeled by encounter. Goal: one model for every department.
More from Apps
- Hermes launches open-source workspace where Claude Code, Codex agents collaborate like teammates — ycombinator · 2026-11-12
- LukeW turns 20 years of design expertise into an MCP server that reviews and fixes UI code — LukeW · 2026-11-12
- Unmetered launches a marketplace for people to sell the AI tools they built for their own jobs — jacob_posel · 2026-11-12
- fal plugin brings image & video generation into ChatGPT, producing a 60-second ad — OpenAIDevs · 2026-11-12
- Treat Claude as a design collaborator, not a genie: how to unlock real creative alpha — kyleturman · 2026-11-12
- AI Pitch Deck Teardown Scores Your Fundraising Deck for $10 — MartinGTobias · 2026-11-12