Clinical computer vision
Deep learning on clinical video and images: detecting, tracking and measuring what a specialist looks for, from an ordinary camera.
- Detection & tracking
- Pose & motion analysis
- Video-based assessment
AI research & engineering · Bengaluru
AI for medtech,
built for the clinic.
PoseLabs is a team of machine learning researchers and engineers building AI-powered solutions for healthcare: models that see, listen to signals and support clinical decisions, taken from research all the way to the bedside.
FIG. 01 — VISION
- —
- —
- —
- —
- —
- —
- —
- —
01 — WHAT WE DO
Hard problems in medicine, solved with AI.
We find where machine learning genuinely helps a clinician, and build it to the standard a hospital needs. Not demos that live in a slide deck.
01/ 04
NOW
AI-assisted clinical assessment from video, with hospital and academic research partners.
Biosignals & wearables
Models for sensor and physiological data, from small wearables to clinical recordings, turned into measures a clinician can read in a minute.
- Inertial & physiological sensors
- Signal quality & event detection
- Automated reports
Decision support
Screening and assessment tools that help clinicians prioritise and catch conditions earlier, with every output explainable and reviewable.
- Risk & screening models
- Calibrated, explainable outputs
- Clinician-in-the-loop review
Health data platforms
Secure systems for collecting, annotating and reviewing medical data remotely, which is the foundation every good model depends on.
- Remote data collection
- Expert annotation tools
- Consent & audit built in
02 — APPROACH
From research to the clinic.
- PROBLEM
We start with clinicians: what decision is hard, slow or unavailable today, and what would change if it weren't.
- DATA
Consented, well-annotated data, collected and stored securely in India, with the people who understand it.
- MODELS
Deep learning and classical methods, chosen for the problem, then validated against expert judgement.
- DEPLOYMENT
Software clinicians actually use. The model supports their call; it doesn't replace it.
How we work.
I
Privacy first
Health data is handled under India's Digital Personal Data Protection Act from the first build: clear notice, recorded consent, data kept in India and a way to erase it.
II
Clinicians decide
Every output traces back to the evidence it came from, so it can be checked rather than taken on trust.
III
Research-grade
Our work grows out of years of academic research with clinical partners, and is validated the same way.
Have a hard problem
in medtech?
Hospitals, researchers, device makers and partners: we'd like to hear from you.
[email protected]