The abstract is read at the publisher; the summary is JClub's.
Multi-modal wearable sensors predict Rating of Perceived Exertion in end-stage kidney disease patients (N=13) with a Root Mean Square Error of 1.33, despite pulse-rate–based models failing.
- Why it matters: Heart-rate–driven exertion monitoring is unreliable in ESKD because beta-blockers and reduced capacity blunt typical pulse responses, creating a safety and efficacy gap during prehabilitation.
- What they did: During prehabilitation, continuous smartwatch data (pulse rate, electrodermal activity, skin temperature, accelerometer) were used to train individualized regression models, comparing pulse-only versus multi-feature pipelines in ESKD (N=13) and older controls (N=13).
- The result: Pulse-rate–only models performed substantially worse for the ESKD cohort but stayed valid for controls, while multi-feature models significantly improved ESKD prediction, indicating reliance on physiological signals beyond pulse rate.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.