Two-week ECG patches with signed reports for physicians in smaller towns
TL;DR Rhythmbay rents licensed ECG patches that a patient wears for 7–14 days, then returns a report that software has sorted and a cardiologist has signed. Physicians in tier-2 and tier-3 towns order it for palpitations, fainting or unexplained stroke. Patients pay per episode. It works if it catches rhythms that a 24-hour recording misses.
Atrial fibrillation often comes and goes, so one ECG can miss it. In a village screening of 354 people over 50, 5.1% showed AF, yet only one of the 12 positives had it continuously and nine tested positive just once[2]. A larger rural survey in Maharashtra found AF in 1.6% of adults, rising to 5.6% in the oldest group[1].
Missing it is costly. Of 3,421 AF patients in Kerala's registry, 12.9% died within a year and 1.8% died of stroke[3].
So what does a physician in a district town do when they suspect an intermittent rhythm? They have two options - a 24-hour Holter if one is nearby, or a referral to a city cardiologist.
What it does, in regulatory terms - an ambulatory ECG recording service that supports a doctor's diagnosis. The patch must hold a CDSCO licence under the Medical Devices Rules 2017, and the analysis software is likely to count as a device too[5]. The software only sorts and flags strips for a technician, and a cardiologist signs every report.
The physician orders a recording on a phone. A field technician fits the patch at the clinic, the patient wears it through showers and sleep, and couriers it back. Within 72 hours the physician gets a two-page report covering AF burden, pauses and the longest runs, with strips matched to the times the patient pressed a symptom button.
The seed imagined a consumer wearable watching hearts for life. I narrowed it. Nobody in India pays for lifelong monitoring, and a stream of alerts from healthy people would bury cardiologists alive. A doctor-ordered episode has a clear question, a clear payer and a clear end.
The patient pays per recording, on the physician's advice. A 24-hour Holter is the price anchor, but I couldn't verify current chain prices, so collect three local quotes before you set yours. My working assumption is ₹2,500–5,000 for 7–14 days. Hospitals without in-house cardiology are a second buyer, paying per report. Insurers and PM-JAY don't cover outpatient rhythm recording, as far as I could find.
CDSCO brought Class A and B devices under licensing on 1 October 2022, and Class C and D on 1 October 2023[5]. A licensed patch now has a clear route to market, and an unlicensed import is easier to challenge. CDSCO listed a guidance document on medical device software in July 2026[5]. India's Telemedicine Practice Guidelines (2020) also allow remote reporting to sit inside normal practice[6].
There's no public figure for India's ambulatory ECG market. The narrow signals are volume - one state registry enrolled 3,421 AF patients from 53 hospitals in a year[3], and one Indian maker says its spot-ECG devices have run more than 2.5 million ECGs across 5,000-plus clinics[4]. So Indian clinics will pay for ECGs outside hospitals. Whether they'll pay for multi-day recording is the open question.
Indian portable-ECG makers sell 12-lead devices with doctor review in under 15 minutes, but they record a spot reading, not days of rhythm[4]. US patch-monitor services are the closest model, with a patch worn for days, with an algorithm sorting the data and technicians checking every report[7]. Diagnostic chains offer 24-hour Holters in cities. The gap is multi-day recording that a district-town physician can order without sending the patient away.
| Difficulty | Moderate (50/100) |
| Starting capital | ₹1.2–2.2 Cr |
| Time to launch | 9–15 months |
You'll probably license or import a patch rather than design one, but a device licence still takes months. The harder work is building a reading desk that turns 14 days of signal into a report a busy physician trusts, and a field network that fits patches and recovers them in towns where couriers are slow.
Yes. This is a regulated medical device business, and you can't take a single order until the device side is settled.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't let AI sign off a rhythm call. A missed pause is a patient safety problem, so a cardiologist reads and signs every report, and the software only sorts.
Want to work out the AI layer for Rhythmbay? Talk to Yash →
Here's what can go south -
I think this suits a founder with cardiology connections and some device-regulatory experience. It does not suit anyone hoping to sell a consumer gadget, full stop.
Before raising money, show that 14-day recordings in 100 patients find a clinically relevant rhythm that a 24-hour recording would have missed, and that physicians reorder.
Plan for roughly ₹1.2–2.2 Cr. Device licensing, patch stock and the reading desk take most of it.
Yes. The patch needs a CDSCO licence under the Medical Devices Rules 2017, the analysis software probably counts as a device too, and reporting cardiologists follow the telemedicine guidelines.
Yes, and the model already exists in the US with patch-monitor services. You'd need your own country's device approval and medical practice rules before you start.
The free Rhythmbay validation kit has the line-by-line capital breakdown, the effort scoring, the device and data-protection checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for physicians and patients, and 5 assumptions to test. Download the kit →
Thinking about building Rhythmbay or something close to it? The make-or-break work is securing a licensed patch and a reading desk physicians trust, and I'm happy to work through both with you. 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.