A training pipeline of an arrhythmia classifier for atrial fibrillation detection using Photoplethysmography signal

Photoplethysmography (PPG) signal is potentially suitable in atrial fibrillation (AF) detection for its convenience in use and similarity in physiological origin to electrocardiogram (ECG). There are a few preceding studies that have shown the possibility of using the peak-to-peak interval of the PPG signal (PPIp) in AF detection. However, as a generalized model, the accuracy of an AF detector should be pursued on the one hand; on the other hand, its generalizability should be paid attention to in view of the individual differences in PPG manifestation of even the same arrhythmia and the existence of sub-types. Moreover, a binary classifier for atrial fibrillation and normal sinus rhythm is not convincing enough for the similarity between AF and ectopic beats. In this study, we project the atrial fibrillation detection as a multiple-class classification and try to propose a training pipeline that is advantageous both to the accuracy and generalizability of the classifier by designing and determining the configurable options of the pipeline, in terms of input format, deep learning model (with hyperparameter optimization), and scheme of transfer learning. With a rigorous comparison of the possible combinations of the configurable components in the pipeline, we confirmed that first-order difference of heartbeat sequence as the input format, a 2-layer CNN-1-layer Transformer hybrid(R) model as the learning model and the whole model fine-tuning as the implementing scheme of transfer learning is the best combination for the pipeline (F1 value: 0.80, overall accuracy: 0.87)(R).

Automated summary

Photoplethysmography (PPG) signal shows potential in atrial fibrillation (AF) detection due to its convenience and physiological similarity to electrocardiogram (ECG). This study proposes a multiple-class classification model for AF detection, taking into consideration individual differences and sub-types in PPG manifestation. The best combination of configurable components in the pipeline includes first-order difference of heartbeat sequence as input format, a 2-layer CNN-1-layer Transformer hybrid model as the learning model, and the whole model fine-tuning as the transfer learning scheme (F1 value: 0.80, overall accuracy: 0.87).