Sector
Diagnostic devices
Device type
Single-lead wearable ECG patch, Class B
Services involved
Design and development, clinical research, quality and regulatory
Duration
22 months, concept to licence
The challenge
A prototype that worked in the lab and nowhere else.
A cardiology group came to us with a benchtop prototype that recorded clean single-lead ECG on healthy volunteers in a quiet room. On the ward it produced unusable traces within an hour. Electrode drying, motion artefact and a signal chain designed around a median chest were all contributing, and none of them were visible in the bench data. They needed a device that held signal quality across seventy-two hours of unsupervised wear, evidence that a cardiologist would accept, and a licence to sell it in India.
What we did
Rebuilt the requirement set before touching the design.
We rewrote the requirements from the use environment rather than the bench: skin impedance across hydration states, adhesive performance over seventy-two hours at 35 degrees and 70 percent humidity, motion artefact rejection during ordinary activity, and a defined behaviour when signal quality degrades rather than a silent failure. Electronics, firmware, mechanical and human factors then ran as one team under ISO 13485 design controls, with the risk file and the verification protocols built forward alongside the design instead of assembled at the end.
What it took
The adhesive, not the amplifier.
Nine weeks went to a problem nobody had scoped. The analogue front end met its noise budget on the first spin. The adhesive did not. Three candidate materials passed seventy-two hours on the bench and failed at around forty hours on real skin in Chennai summer conditions, and every one of those failures looked like an electronics fault in the recorded data. We instrumented the patch to log contact impedance continuously, which finally separated adhesive failure from signal-chain failure, and requalified against a fourth material. The finding went back into the requirement set as a measurable input rather than a material preference.
The outcome
What came out of it.
CDSCO licence granted twenty-two months after kick-off, with no major deficiency raised against the clinical evidence.
94.1 percent of recorded hours usable across the seventy-two hour wear period in the pivotal study, against an 85 percent target.
Agreement with the reference twelve-lead system inside the pre-specified limits at all three participating sites.
Production established on our own lines in Chennai, with unit-level traceability from component lot to shipped device.
Services involved
The disciplines behind this programme.
Medical device design and development
Requirement definition, electronics, firmware, mechanical and human factors through to verified design.
See the serviceClinical research (CRO)
Protocol, ethics approvals at three sites, data acquisition and the agreement analysis.
See the serviceQuality and regulatory affairs
Pathway, design controls, risk file and the CDSCO submission.
See the serviceAll three ran as one programme against a single plan, with one named programme lead accountable for the whole of it.
Next step
Want to talk about a similar programme?
Tell us what you are building and where it is stuck.



