Multi-label classification of reduced-lead ECGs using an interpretable deep convolutional neural network

Abstract

Objective. We propose a model that can perform multi-label classification on 26 cardiac abnormalities from reduced lead Electrocardiograms (ECGs) and interpret the model. Approach. PhysioNet/Computing in Cardiology (CinC) challenge 2021 datasets are used to train the model. All recordings shorter than 20 seconds are preprocessed by normalizing, resampling, and zero-padding. The frequency domains of the recordings are obtained by applying Fast Fourier Transform. The time domain and frequency domain of the signals are fed into two separate deep convolutional neural networks. The outputs of these networks are then concatenated and passed through a fully connected layer that outputs the probabilities of 26 classes. Data imbalance is addressed by using a threshold of 0.13 to the sigmoid output. The 2-lead model is tested under noise contamination based on the quality of the signal and interpreted using SHapley Additive exPlanations (SHAP). Main results. The proposed method obtained a challenge score of 0.55, 0.51, 0.56, 0.55, and 0.56, ranking 2nd, 5th, 3rd, 3rd, and 3rd out of 39 officially ranked teams on 12-lead, 6-lead, 4-lead, 3-lead, and 2-lead hidden test datasets, respectively, in the PhysioNet/CinC challenge 2021. The model performs well under noise contamination with mean F1 scores of 0.53, 0.56 and 0.56 for the excellent, barely acceptable and unacceptable signals respectively. Analysis of the SHAP values of the 2-lead model verifies the performance of the model while providing insight into labeling inconsistencies and reasons for the poor performance of the model in some classes. Significance. We have proposed a model that can accurately identify 26 cardiac abnormalities using reduced lead ECGs that performs comparably with 12-lead ECGs and interpreted the model behavior. We demonstrate that the proposed model using only the limb leads performs with accuracy comparable to that using all 12 leads.

A comprehensive review on efficient artificial intelligence models for classification of abnormal cardiac rhythms using electrocardiograms

“…A classification model among two binary supervised classifiers and one hybrid unsupervised-supervised classification system was then selected for each class. Wickramasinghe et al [84] preprocessed recordings shorter than 20s by normalizing, resampling, and zero-padding. The frequency domains of the recordings were obtained by applying a fast fourier transform.…”

xECGArch: A trustworthy deep learning architecture for interpretable ECG analysis considering short-term and long-term features

“…In contrast to that, Singh & Sharma [22] conducted a more systematic comparison of four xAI methods based on Shapley additive explanations (SHAP), local interpretable model-agnostic explanations (LIME), and GradCAM. SHAP was also used in [23][24][25][26] and exemplary compared to LIME and permutation feature relevance in [12]. Besides these post-hoc methods, applied to a model after classification, multiple works visualized ante-hoc generated attention layer values to explain ECG classifications, showing the samples' relevance for classification [27][28][29][30][31].…”

An Automated Cardiac Arrhythmia Classification Network for 45 Arrhythmia Classes Using 12-Lead Electrocardiogram

“…Deep learning techniques have shown great promise in ECG signal analysis, enabling automatic and accurate detection of various cardiac arrhythmia [19]- [29] and [38]- [41]. Deep neural network (DNN) realizes the effective combination of feature extraction and cardiac arrhythmia classification through end-to-end learning.…”

“…However, deep learning-based cardiac arrhythmia classification still has some limitations. First, the existing studies [19]- [29] and [38]- [41] have either utilized many publicly available datasets or collected their own datasets using individual approaches. However, there is still a lack of studies that classify various arrhythmia classes.…”

“…However, there is still a lack of studies that classify various arrhythmia classes. Specifically, previous studies [19]- [29] and [38]- [41] focused on classifying a single class or only around ten arrhythmia classes. According to [18], more than 60 types of arrhythmia classes exist in various manifestations.…”