Abstract Introduction/Rationale COPD exacerbations have a substantial impact on patients’ health and quality of life, with associated socio-economic burden.1 Home monitoring to facilitate earlier detection of exacerbations has demonstrated promising results in reducing the number of emergency visits and hospitalisations. However, the evidence supporting widespread adoption is inconclusive.2 Further development is required to explore the utility of signals measured within a person’s home in providing early, accurate, and objective diagnosis of COPD exacerbation. Method We evaluated the diagnostic accuracy of the DistaSense™ device, which provides overnight non-contact monitoring of heart rate (HR) and respiratory rate (RR) to facilitate early detection of COPD exacerbations. Patients were monitored every night, overnight and an exacerbation was defined as the need for oral antibiotics and/or corticosteroids. A predictive model using machine learning was developed to identify exacerbations, which includes analysis of motion, restlessness, and respiratory waveform. Result 49 people with COPD (55% male, mean (SD) age 70.1(6.8) years, mean (SD) FEV1 57.1(19) % predicted) were monitored. 32 exacerbations among 14 patients were detected, with a total of 2547 nights of monitoring. HR and RR signals were integrated into 326 features. The predictive model demonstrated that DistaSense™ was able to detect an exacerbation five days before the start of treatment, achieving a sensitivity of 72% for a specificity of 98%. The negative predictive value was 99% and the positive predictive value was 40%. Conclusion In the largest study of its kind reported to date, the DistaSense™ device offers exciting potential features for early diagnosis of COPD exacerbations five days prior to treatment. A high specificity was achieved, supporting the development of this remote home monitoring system to facilitate early diagnosis of COPD exacerbations. This abstract is funded by: LDT (Life Detection Technologies) and LPDP (Indonesian Endowment Fund for Education Agency)
Syam et al. (Fri,) studied this question.