Work

Remote illness detection faces a trust barrier

Year: 2024

Type: article

Abstract: On average, wearable devices can detect when a person is becoming unwell with an acute respiratory illness.1 There is real opportunity for health systems to become dynamic and responsive moment to moment for very large and dispersed populations through integration of diverse emergent data sources and artificial intelligence (AI). But what differentiates clear signs of illness for one person, or one group of people, might differ substantially in another. This is important because health services are rightly wary of adopting algorithm-based tools when there is a real danger that the performance of the algorithms will bias results against minoritised populations,2–4 as happens when an algorithm fails to generalise outside of the context in which it was trained.

Source: The Lancet Digital Health

DOI: 10.1016/s2589-7500(24)00145-6

Author: Benjamin L Smarr

Institution: University of California San Diego

Language: English

Topic: Non-Invasive Vital Sign Monitoring

Subfield: Biomedical Engineering

Field: Engineering

Domain: Physical Sciences

SDGs: Good health and well-being, Reduced inequalities

Open Access status: gold

Estimated APC paid: $5,780