Community project · Non-profit
Edge AI for Health is a student-led project by Hoang Anh Nguyen at Hanoi University of Science and Technology. I design low-cost devices that run AI on the chip itself, so they work offline and patient data never leaves the device.
Clinically supported by a hospital in Hanoi. The partner is not named because this is an internal collaboration.
Why this project
A commune health station is often a patient’s first and only stop, yet heart and lung sounds take years of practice to judge well.
Many smart tools stop working when the connection drops — exactly in the places that need them most.
Each device uses common, affordable parts so a health station could own one, repair it, and build another.
Design principles
The AI runs on the device’s own chip. No signal, no problem.
Recordings are processed and kept locally, never uploaded by default.
Off-the-shelf parts that a local technician can source and replace.
Designs and code are shared as each device reaches a safe, documented stage.
The devices
Separates heart and lung sounds, filters noise and flags patterns worth a second listen — entirely on the device, with no internet.
For Commune health stations and village health workers who examine many patients without a cardiologist or pulmonologist nearby.
A small box for clinic waiting rooms that picks out coughs from background noise and suggests who may need TB or pneumonia screening.
A thin chest patch worn for several days that watches for irregular heart rhythms such as atrial fibrillation and saves only the moments that matter.
A headphone kit that lets a school nurse run a quick hearing check, with a built-in microphone that pauses the test when the room is too loud.
Illustrations are design drawings, not product photos.
How it works
Sound, heart rhythm or a hearing response, captured by low-cost sensors.
Separate overlapping sources and remove the noise of a real clinic.
A small model, sized for a microcontroller, highlights what needs attention.
A clear result on the device. The clinician makes the call.
Research foundation
Telling apart several sounds that overlap — three voices at once, a heartbeat under a breath, a cough in a noisy room — is the same problem. My separation research is the core of the acoustic devices.
Read the research notes →Who is behind it
I’m a third-year Electronics & Telecommunications student at Hanoi University of Science and Technology. I started with Linux and networking, moved into AI on small hardware, and found the problem I want to spend my time on: bringing careful listening and early screening to places that don’t have specialists.
More about me →Get involved
Tell me what your clinic actually needs, review a design, or help test a prototype safely.
Firmware, PCB layout, signal processing, small models — there is a piece for every skill.
Host a pilot at a health station or school, or sponsor parts for the next batch of devices.