Camera-tracked arm exercise games at home after stroke, reviewed by a physio
TL;DR Reachloop gives stroke survivors a tablet and camera that turn their prescribed arm exercises into short games, counting every repetition. The hospital physio sets the plan and reviews progress weekly from a dashboard. Families rent it for the first three months home, and hospitals use it to keep patients in their rehab programme.
A five-registry study found an age-standardised stroke incidence of 103.4 per 100,000 Indian adults in 2018–19[1], and an earlier review put it at 105–152 per 100,000 a year in city studies[2]. Many survivors go home with a weak arm and a sheet of exercises. In one Indian follow-up of 74 survivors, 36.5% did not stick to their home programme[3].
Why does that happen? The physio sessions that would keep them going are concentrated in cities and hard to reach for a patient who cannot travel alone, and nobody at the hospital sees what happens at home.
What it does, in regulatory terms: exercise guidance and remote supervision under a registered physiotherapist, not diagnosis or treatment decisions. Software that tracks rehab exercise may still count as medical device software under the Medical Devices Rules 2017, so get a CDSCO classification opinion before selling[4].
A tablet on a stand with its camera watches the patient's upper body. The physio picks exercises (reach, lift, wrist turns) and Reachloop wraps each in a simple game - popping targets, sorting fruit, steering a boat. Pose tracking counts repetitions, flags compensation such as shoulder hiking, and sends a weekly summary to the physio, who adjusts the plan in a short video call.
The seed pitched immersive VR, and I've reframed it. A Cochrane review found VR no better than standard therapy for arm function, though extra practice time on top of usual care helped[5]. Headsets also add balance and comfort problems for this group. So Reachloop sells extra supervised repetitions on a screen, not virtual worlds.
The patient uses it but usually an adult child pays. Families already pay for home physio visits out of pocket. I found no published price list, so test the rental price directly. Hospitals and neuro-rehab centres are the channel, since they prescribe it at discharge and earn from the remote review, which keeps the patient in their programme.
Rehab platforms are getting hospital traction in India - one Chennai rehab start-up said in September 2025 that it had 55+ partner centres[6]. Phone and tablet cameras now track joints well enough without markers. The NCAHP Act of 2021 brought physiotherapists under a national register[7], which gives hospitals a clear standard for who may supervise remotely.
No public figure exists for India's home stroke-rehab market. The bottom-up base is the incidence rate above[1] applied to India's adult population, which implies well over a million new strokes a year (my own arithmetic, so check it against current population estimates). Only a share survive with arm weakness and can afford a rental, so prove demand in one hospital before modelling more.
That start-up runs a VR rehab platform plus its own clinics in Tamil Nadu[6]. An Indian med-tech firm makes a game-based arm device, with a cloud platform for home monitoring[8]. Both lean on hardware or clinics. The gap is a low-cost, camera-only home kit that hospitals prescribe at discharge and supervise remotely.
| Difficulty | Moderate (49/100) |
| Starting capital | ₹1.2–2.2 Cr |
| Time to launch | 9–15 months |
Pose tracking on older, seated patients with weak limbs is harder than on gym users, and home lighting is often poor (one tube light in the corner, if you're lucky). The regulatory classification can add months. The longest task is a pilot that shows hospitals more repetitions and better completion than their exercise sheet.
Yes, this is regulated. Exercise-tracking software may count as a medical device, and the supervising physios must be registered.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't let AI change an exercise plan or judge recovery. The registered physio makes every treatment decision, and Reachloop gives exercise guidance and supervision, not diagnosis or treatment.
Want to work out the AI layer for Reachloop? Talk to Yash →
Here's what can go south -
I think this is a sensible idea for a founder with a physiotherapist co-founder and computer-vision skills. If you want a consumer game with no clinical partner then it's the wrong pick. Prove first that patients using it complete twice as many repetitions in the first month as those on a paper sheet.
Plan for roughly ₹1.2–2.2 Cr over 9–15 months, mostly on pose tracking, regulatory classification and a hospital pilot.
Possibly. Get a CDSCO classification view under the Medical Devices Rules 2017, use only NCAHP-registered physios and follow the DPDP Act for patient data.
Yes, but most countries regulate rehab software as a medical device. Check with your health regulator before your first pilot.
The free Reachloop validation kit has the line-by-line capital breakdown, the effort scoring, the device-classification and registration checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for families and physios, and 5 assumptions to test. Download the kit →
Thinking about building Reachloop or something close to it? The make-or-break work is settling the device classification and getting one hospital to prescribe it at discharge, and I can help you plan both. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.