Abstract
Detection of atrial fibrillation (AF) from a wrist watch photoplethysmogram (PPG) signal is important because the wrist watch form factor enables long term continuous monitoring of arrhythmia in an easy and non-invasive manner. We have developed a novel method not only to detect AF from a smart wrist watch PPG signal, but also to determine whether the recorded PPG signal is corrupted by motion artifacts or not. We detect motion and noise artifacts based on the accelerometer signal and variable frequency complex demodulation based time-frequency analysis of the PPG signal. After that, we use the root mean square of successive differences and sample entropy, calculated from the beat-to-beat intervals of the PPG signal, to distinguish AF from normal rhythm. We then use a premature atrial contraction detection algorithm to have more accurate AF identification and to reduce false alarms. Two separate datasets have been used in this study to test the efficacy of the proposed method, which shows a combined sensitivity, specificity and accuracy of 98.18%, 97.43% and 97.54% across the datasets.
Highlights
Atrial fibrillation (AF) is the most common sustained arrhythmia and is associated with significant morbidity and mortality[1]
To combat the challenges of atrial fibrillation (AF) detection with a smartwatch, in this study we present a novel method based on the wrist PPG signal for AF detection, which accounts for motion and noise artifacts (MNA)
For the ideal normal sinus rhythm (NSR) segments, almost all the patterns are inside the center quadrat while for the AF segments, there is no distinct parents as shown in Fig. 5(a,c), respectively
Summary
Atrial fibrillation (AF) is the most common sustained arrhythmia and is associated with significant morbidity and mortality[1]. In13, authors present an 8-layer deep neural network which can passively detect AF from a PPG signal obtained with a commercially available smartwatch Both irregularity of heart rhythm and absence of p-wave detection methods have been implemented in[14], in which the authors use a Kardia band to record both the ECG and PPG signals from a smartwatch and use it to classify AF from normal sinus rhythm. In[16], AF was discriminated from sinus rhythm using complex nonlinear combination analysis of pulse intervals preceded by a data quality check, the measurements were performed while the subjects were sitting in a comfortable position in a quiet hospital environment None of these smartwatch-based AF detection methods dealt with premature atrial contraction (PAC) and premature ventricular contraction (PVC) rhythms, which when present can cause false positive detection of AF. Details can be found in[21]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have