64
Research Trials
29
Peer-reviewed publications
17
Clinical Conditions

Updated interim data from 54 COPD patients and 135,000 monitored hours, showing cough rates rise in the days before exacerbations and correlate only weakly with CAT scores.

A month of continuous monitoring in stable chronic cough shows that multi-day averaging makes cough frequency a more reliable trial endpoint, allowing smaller studies, and reveals a short-lived Hawthorne effect.

Mixed-methods study in Kampala assessing 48-hour cough frequency monitoring with the Hyfe Research app for TB screening and triage, alongside staff-reported implementation barriers.
24 June 2026

This decentralized study (NCT06775301) validated Hyfe’s smartwatch, which detects and timestamps coughs entirely on-device with no audio recorded or retained. Thirty-one adults wore it for 20 to 24 hours during normal daily activities, while a separate device recorded continuous audio that expert annotators reviewed to mark every cough.
Median sensitivity was 87%, with a median false positive rate of 1.6 per hour. Agreement with ground truth was high: Lin’s concordance correlation coefficient was 0.958 for each patient’s 24-hour total and 0.975 for hourly counts. Performance did not differ significantly by cough rate, sex, age, cough etiology or urban versus rural residence. The device placed 84% of participants in the correct cough-rate category, and the few misclassifications were all minor.
