All testing data were collected from “Pulsewatch” smartwatches worn by the participants in the intervention group in Part II of our clinical trial study (NIH study ID: NCT03761394), which was a 14-day at-home, free-living protocol [23 (link)]. The participants wore a continuous AF monitoring ECG patch, approved by the FDA, which did not restrain their daily activities [23 (link)] and allowed us to collect a reference ECG signal. The “Pulsewatch” system was developed in Part I of our clinical trial study, and consisted of a smartwatch (Samsung Gear S3 or Galaxy Watch 3) and a smartphone (Samsung Galaxy S6, S10e, A50, or XCover Pro) pre-installed with our own applications for data collection and analysis. Only the smartwatch application software performed all signal processing tasks in near real-time including motion artifact and AF detection and our proposed PPG peak detection and PAC/PVC detection [18 (link)]; the smartphone application functioned as a data transfer hub to upload PPG data to our secure cloud server. Written informed consent forms were collected from all patient participants, and formal ethical approval for this study was obtained from the UMMS Institutional Review Board (IRB ID: H00016067) [23 (link)].
Free full text: Click here