TL;DR Recallwell gives stroke and head-injury patients short memory, attention and planning games on a tablet, set and adjusted by their therapist. Patients practise daily at home in their own language, and the clinic sees who is keeping up. Neuro-rehab centres buy it and pass the cost to families.
India sees more than 1.8 million new strokes a year, and therapists are scarce outside the big cities[1]. Physical recovery gets the attention, while thinking problems get a lot less of it. In a 2026 analysis of the national ageing survey, 15.3% of stroke survivors over 60 had cognitive impairment, against 11.6% of matched peers without a stroke[2].
Those deficits show up as missed pills, lost money and unsafe cooking. Rehab for them needs many hours of repetition, but a family can afford perhaps two or three clinic visits a week. And between visits? Most patients do nothing structured at all.
What it does, in regulatory terms: a rehabilitation exercise tool used under a clinician's prescription. It does not screen for, diagnose or score cognitive impairment. Because it aids treatment, it is likely to count as medical device software under CDSCO's October 2025 draft guidance, which classes software by how serious the condition is and how much its output matters[6]. Plan for low-risk registration and get regulatory advice early.
The therapist picks exercises from a library - remembering a shopping list, sorting coins, following a recipe's steps, spotting a changed object. Difficulty steps up or down by fixed rules the therapist sets. Sessions last 15–20 minutes, in Hindi, Tamil, Malayalam or English, with large buttons for one-handed use.
I've reframed the seed, which proposed VR and AR games to solve cognitive impairment. Headsets are costly, tiring for older patients and hard to use alone at home, and no game cures impairment. A 2026 meta-analysis found digital cognitive rehab gave a moderate gain in daily-living skills, but with uncertain long-term benefit[3]. Recallwell sells supervised practice volume, not recovery.
The patient is usually over 55 and recently discharged. The family pays, through the clinic. Neuro-rehab centres and hospital physiotherapy departments buy licences and add a home programme to their package. I found no published price for cognitive rehab software in India, so my working assumption, to be tested, is ₹1,500–3,000 a month per patient, charged by the clinic.
Rehab chains are scaling fast and already accept games - one Chennai rehab start-up reported 55+ partner centres and eight clinics of its own in September 2025[4]. CDSCO's draft software guidance of October 2025 finally gives a path for low-risk apps[6]. Tablets in Indian languages are now cheap enough to lend.
There is no public figure for the size of India's cognitive rehab software market. The bottom-up base is about 1.8 million strokes a year[1], with around one older survivor in seven showing cognitive impairment[2]. Many more patients with milder deficits could also benefit, but no Indian count exists. Treat clinic partnerships, not patients, as the first number to prove.
That start-up uses VR and AI for stroke physiotherapy across partner centres[4]. An Indian med-tech firm makes a game-based arm rehab device used in over 50 centres[1]. Both are mainly about movement. A US app is the closest model, a home app for speech and cognitive exercises that clinicians assign[5]. The gap is a cognitive home programme in Indian languages, tied to Indian clinics.
| Difficulty | Moderate (46/100) |
| Starting capital | ₹1.2–2.2 Cr |
| Time to launch | 12–18 months |
Building the games is the easy part. The hard parts are a quality system that satisfies device rules, exercise content that occupational therapists and neuropsychologists trust, and real outcome data from a pilot. Older users need hand-holding to keep going, and if nobody at home nudges them then the tablet ends up in a drawer.
Yes, this likely needs regulatory approval, because software that aids treatment can count as a medical device.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Recallwell is a supervised exercise tool. It must not use AI to screen for, diagnose or score cognitive impairment, and every clinical decision stays with the therapist.
Want to work out the AI layer for Recallwell? Talk to Yash →
Here's what can go south -
I think this fits a founder with a clinical rehab background, or one with a strong clinical co-founder. If you see it as a brain-games app for the mass market then please skip it. Prove first that patients complete at least four sessions a week for eight weeks when a therapist assigns them.
Plan for roughly ₹1.2–2.2 Cr, including a quality system for device rules, clinical content and a pilot.
Likely yes. Software that aids treatment may be a medical device under CDSCO's draft guidance, so plan for low-risk registration and get advice early.
Yes, a US app runs a similar model in the US. Each country has its own medical device and health data rules, so check with the local regulator first.
The free Recallwell validation kit has the line-by-line capital breakdown, the effort scoring, the device-software and data-protection checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for therapists and families, and 5 assumptions to test. Download the kit →
Thinking about building Recallwell or something close to it? The make-or-break work is getting the device classification right and showing clinics real adherence data, and I can help you sequence 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.