INDACO.ONE / Knowledge Centre / Computer Vision in Care

Technology

Computer Vision in Care

Computer vision in care is the use of ordinary cameras and machine learning to observe movement, posture, technique and physiological signals — without contact, and without additional hardware.

The camera as a general-purpose sensor

A camera is unusual among sensors: it captures many things at once. From one video stream it is possible to derive gait and posture, the quality with which a technique is performed, and — through remote photoplethysmography (rPPG) — contactless estimates of heart rate and its variability.

This is what makes it economically distinctive. One sensor, already present in the phones and tablets people own, can serve several care domains without anything being purchased, shipped or installed.

Observing technique, not replacing measurement

A significant and underappreciated problem in respiratory care is that many manoeuvres are performed incorrectly — inhalers used with poor technique, spirometry efforts that fail acceptability criteria. The consequence is not merely inefficiency: it is data that cannot be trusted, and treatment that does not reach the lungs.

Computer vision can guide the person performing the manoeuvre and flag likely quality problems. It does not produce the clinical measurement — a certified device does that. Coaching the technique and measuring the result are two different responsibilities, and keeping them separate is both scientifically and regulatorily correct.

Limits, stated plainly

Vision-based methods degrade with poor lighting, occlusion and movement. Contactless physiological estimation is known to be sensitive to skin tone, illumination and motion, and should be treated as a trend indicator rather than a clinical vital sign. Reporting these limitations is part of using the technology responsibly.

Privacy by design

Cameras in care settings raise legitimate concerns. The defensible approach is to process on-device or to extract features rather than retain raw video wherever possible, to make consent granular and revocable, and to be explicit about what is stored, for how long, and who can see it. Video captured in a person's home is among the most sensitive data that exists.

How INDACO.ONE approaches it

The same vision engine serves multiple modules — motor and gait signals in Cognitive Care, inhaler and spirometry technique in Respiratory Care, contactless physiological trends in both. One engine, several domains: more understanding without more software, and no hardware to deploy.

Related concepts

INDACO.ONE is a software-only Connected Care Intelligence Platform. It is not a diagnostic device and does not diagnose, treat or replace clinical judgment. It supports understanding, coordination and longitudinal insight under human oversight.