Motion and Noise Artifact-Resilient Atrial Fibrillation Detection Using a Smartphone

Abstract

We have recently found that our previously-developed atrial fibrillation (AF) detection algorithm for smartphones can give false positives when subjects’ fingers or hands move, as we rely on proper finger placement over the smartphone camera to collect the signal of interest. Specifically, smartphone camera pulsatile signals that are obtained from normal sinus rhythm (NSR) subjects but are corrupted by motion and noise artifacts (MNAs) are frequently detected as AF. AF and motion-corrupted episodes have the similar characteristic that pulse-to-pulse intervals (PPIs) are irregular. We have developed an MNA-resilient smartphone-based AF detection algorithm that first discriminates and eliminates MNA-corrupted episodes in smartphone camera recordings, and then detects AF in MNA-free recordings. We found that MNA-corrupted episodes have highly-varying pulse slope, large turning point ratio, or large kurtosis values in smartphone signals compared to MNA-free AF and NSR episodes. We first use these three metrics for MNA discrimination and exclusion. Then, AF is detected in MNA-free signals using our previous algorithm. The capability to discriminate MNAs and AFs separately in smartphone signals increases the specificity of AF detection. To evaluate the performance of the proposed MNA-resilient AF algorithm, 99 subjects, including 88 study participants with AF at baseline and in NSR after electrical cardioversion as well as 11 participants with MNA-corrupted NSR, were recruited. Using iPhone 4S, 5S, and 6S models, we collected 2-minute pulsatile time series from each subject. The clinical results show that the accuracy, sensitivity and specificity of the proposed AF algorithm are 0.97, 0.98, 0.97, respectively, which are higher than those of the previous AF algorithm.

Photoplethysmography based atrial fibrillation detection: a review

“…For example, the thresholds of some features were first used to exclude poor pulses, then an ML model was built for the detection of AF in the clean pulses. 64,65 Table 2 is a chronological summary of the selected ML studies and the reported performance results. All the studies reported in the Table 2 were based in short length of PPG segment, with maxima of 2 min.…”

“…This situation can lead sinus rhythm signals corrupted by motion artifacts to be incorrectly detected as AF and vice versa. 65,73 The most common approach to deal with this issue is to simply discard the corrupted segments and use only the clean parts of PPG signals. 49,50,58,60,64,74 Some of the works followed a two-step approach: first, to identify motion artifacts by using accelerometer data, or by performing PPG signal quality assessment; and second, to perform AF detections with only good quality signals.…”

Enhancing the Robustness of Smartphone Photoplethysmography: A Signal Quality Index Approach

“…Third, there are many methods for improving smartphone PPG accuracy [27,[29][30][31][32][33]. For example, adding a suitable bandpass filter for signal processing [28] or excluding data with RR intervals that differ more than a certain threshold [70] are simple and effective approaches to reduce noise.…”

“…Given that smartphones with in-built cameras have become a part of modern life, using them to access health information is an ideal alternative when ECGs or similar medical devices are not available [26]. In addition, there have been several reported techniques for increasing the accuracy of smartphone PPG, such as point-of-interest selection [27], bandpass filtering [28], adaptive signal thresholding [29], motion detection techniques [30][31][32], interpolation techniques [33], and signal decomposition methods [34][35][36]. Bioengineering studies indicate that the average HR [37] and HRV measured using smartphone PPG are comparable with those measured using gold standard ECGs [21,28,[38][39][40].Although it is a promising solution for practical data collection and has an accuracy that has been well proved in several experiments, using smartphone PPG to measure HRV has received limited research attention in applied disciplines such as medicine or psychology [41]; a possible explanation is the lack of robustness in practical scenarios.…”

Cuff-Less Blood Pressure Monitoring System Using Smartphones

“…These smartphone methods can be used for heart rate estimation, atrial fibrillation detection, oxygen saturation measurement, and BP estimation [39]- [45]. Two smartphones were used in [42] to record the audio signal and the PPG signal together to estimate BP.…”