Reliable PPG-based algorithm in atrial fibrillation detection

• Published in: 2016 IEEE Biomedical Circuits and Systems Conference (BioCAS) pp. 340 - 343
• Main Authors: Shih-Ming Shan, Sung-Chun Tang, Pei-Wen Huang, Yu-Min Lin, Wei-Han Huang, Dar-Ming Lai, Wu, An-Yeu Andy
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.10.2016
• Subjects: Atrial Fibrillation; Electrocardiography; Entropy; Feature extraction; Heart rate; Multiscale entropy; Photoplethysmogram; Sensitivity; Shannon entropy; Support vector machine; Support vector machines; Turning
• DOI: 10.1109/BioCAS.2016.7833801

Atrial Fibrillation (AF) is the most common type of arrhythmia. Since AF is a risk factor for stroke, automatic detection of AF is an important public health issue. Currently, the most useful and accurate tool for diagnosing AF is electrocardiography (EKG). On the other hand, PPG-based AF detection desires exploration. Photoplethysmogram (PPG) is an alternative technique to obtain the heart rate information by pulse oximetry. Convenience makes PPG promising in identifying arrhythmia like AF. The aim of this study is to investigate the potential of analyzing PPG waveforms to identify patients with AF. With the extracted features from multiple parameters, including interval and amplitude of PPG signals, patients were classified into AF and non-AF by support vector machine (SVM). The receiver operating characteristic curve (ROC) and statistical measures were applied to evaluate model performances. Among 468 patients' signals recorded in clinic environments, we achieve ROC area under curve, sensitivity and accuracy of 0.971, 0.942, and 0.957, respectively. The result suggests that the PPG-based AF detection algorithm is a promising pre-screening tool to help doctors monitoring patient with arrhythmia.