Opening the black box: interpretability of machine learning algorithms in electrocardiography

Abstract

Recent studies have suggested that cardiac abnormalities can be detected from the electrocardiogram (ECG) using deep machine learning (DL) models. However, most DL algorithms lack interpretability, since they do not provide any justification for their decisions. In this study, we designed two new frameworks to interpret the classification results of DL algorithms trained for 12-lead ECG classification. The frameworks allow us to highlight not only the ECG samples that contributed most to the classification, but also which between the P-wave, QRS complex and T-wave, hereafter simply called ‘waves’, were the most relevant for the diagnosis. The frameworks were designed to be compatible with any DL model, including the ones already trained. The frameworks were tested on a selected Deep Neural Network, trained on a publicly available dataset, to automatically classify 24 cardiac abnormalities from 12-lead ECG signals. Experimental results showed that the frameworks were able to detect the most relevant ECG waves contributing to the classification. Often the network relied on portions of the ECG which are also considered by cardiologists to detect the same cardiac abnormalities, but this was not always the case. In conclusion, the proposed frameworks may unveil whether the network relies on features which are clinically significant for the detection of cardiac abnormalities from 12-lead ECG signals, thus increasing the trust in the DL models. This article is part of the theme issue ‘Advanced computation in cardiovascular physiology: new challenges and opportunities’.

Machine learning derived ECG risk score improves cardiovascular risk assessment in conjunction with coronary artery calcium scoring

“…Neural networks on ECGs have been shown to outperform manual QTc measurements for life-threatening ventricular arrhythmia prediction ( 33 , 34 ) and also as predictive tools for ventricular dysfunction ( 35 , 36 ), coronary artery disease ( 37 ), atrial fibrillation ( 38 , 39 ), myocardial hypertrophy ( 40 ) and ischemic heart disease ( 41 ). Although ML frameworks on ECGs lack direct interpretability, they have been used to detect the most relevant waves (P-wave, QRS complex or T-wave), contributing to diagnosis of CVDs ( 42 ). Additionally, ML frameworks have shown to detect both, clinically significant and other subtle features that are not traditionally used by cardiologists ( 42 ).…”

“…Although ML frameworks on ECGs lack direct interpretability, they have been used to detect the most relevant waves (P-wave, QRS complex or T-wave), contributing to diagnosis of CVDs ( 42 ). Additionally, ML frameworks have shown to detect both, clinically significant and other subtle features that are not traditionally used by cardiologists ( 42 ).…”

Advanced computation in cardiovascular physiology: new challenges and opportunities

“…In the context of AI methods for cardiology applications, Bodini et al [1] review the interpretability issues of state-of-the-art machine-learning algorithms using electrocardiogram (ECG) signals to detect cardiac abnormalities. They also propose two novel frameworks to interpret the classification results of deep learning algorithms trained for 12-lead ECG-based classification; here, interpretability refers to the signal samples and waves (e.g.…”

Clinical significance, challenges and limitations in using artificial intelligence for electrocardiography-based diagnosis

“…and non-cardiovascular diseases, for diagnosis, prognosis, and risk stratification [7][8][9][10][11][12]. Instead of being fed with handcrafted vectors, on which ML algorithms rely, the DL approach uses the end-toend learning strategy, which programmes the system to learn the necessary features from the raw data [13]. The Chung et al International Journal of Arrhythmia (2022) 23:24 benefit of deep neural networks is its ability to identify novel intervariable relationships independent of humanselected feature extraction, which can offer an enormous previously unrecognized insights in healthcare diagnosis and treatement [7].…”