End-to-end Deep Learning from Raw Sensor Data: Atrial Fibrillation Detection using Wearables

Abstract

We present a convolutional-recurrent neural network architecture with long short-term memory for real-time processing and classi cation of digital sensor data. The network implicitly performs typical signal processing tasks such as ltering and peak detection, and learns time-resolved embeddings of the input signal.We use a prototype multi-sensor wearable device to collect over 180 h of photoplethysmography (PPG) data sampled at 20 Hz, of which 36 h are during atrial brillation (AFib).We use end-to-end learning to achieve state-of-the-art results in detecting AFib from raw PPG data. For classi cation labels output every 0.8 s, we demonstrate an area under ROC curve of 0.9999, with false positive and false negative rates both below 2 × 10 −3 . This constitutes a signi cant improvement on previous results utilising domain-speci c feature engineering, such as heart rate extraction, and brings large-scale atrial brillation screenings within imminent reach.

Estimating Blood Pressure from the Photoplethysmogram Signal and Demographic Features Using Machine Learning Techniques

“…This method showed improvements in BP prediction compared to other existing methods. Gotlibovych et al investigated the potential of using raw PPG data to detect arrhythmia in 2018 [29] with reasonable success, which shows the possibility of using raw PPG signal as inputs to the deep learners. In [30], the authors have created a novel spectro-temporal deep neural network that took the PPG signal and its first and second derivative as inputs.…”

Deep Learning in Cardiology

“…An accuracy of 91.8% in AF detection was achieved by the method and in combination with its computational efficiency it is promising for real world deployment. Gotlibovych et al [117] trained an one layer CNN network followed by a LSTM using 180h of PPG wearable data to detect AF. Use of the LSTM layer allows the network to learn variable-length correlations in contrast with the fixed length of the convolutional layer.…”

“…b There is a wide variability in results reporting. [109] report specificity, [115] report results for SBP and DBP, [117] report sensitivity, specificity, [118] report positive predictive value, [119] report AUC for diabetest, results are also reported for high cholesterol sleep apnea and high BP.…”

Detection of Atrial Fibrillation Using a Machine Learning Approach

“…The irregular heartbeat indicates the symptoms of any accidental strokes that may result in further prolonged illness and, as a result, it leads to ultimate heart failure [4][5][6][7][8]. The AF detection methods are mainly focused on the RR intervals, short term study of heart rate variability, and sequential examination to check the existence of P-wave [9]. The current studies are mostly focused on the feature extraction techniques; therefore, the features have significant fictional effect on the final outcome of the models.…”