Detecting and interpreting myocardial infarction using fully convolutional neural networks

Abstract

Objective: We aim to provide an algorithm for the detection of myocardial infarction that operates directly on ECG data without any preprocessing and to investigate its decision criteria. Approach: We train an ensemble of fully convolutional neural networks on the PTB ECG dataset and apply state-of-the-art attribution methods. Main results: Our classifier reaches 93.3% sensitivity and 89.7% specificity evaluated using 10-fold cross-validation with sampling based on patients. The presented method outperforms state-of-the-art approaches and reaches the performance level of human cardiologists for detection of myocardial infarction. We are able to discriminate channel-specific regions that contribute most significantly to the neural network's decision. Interestingly, the network's decision is influenced by signs also recognized by human cardiologists as indicative of myocardial infarction. Significance: Our results demonstrate the high prospects of algorithmic ECG analysis for future clinical applications considering both its quantitative performance as well as the possibility of assessing decision criteria on a per-example basis, which enhances the comprehensibility of the approach.

arXiv comment11 pages, 4 figures

References in corpus (7)

• A systematic study of the class imbalance problem in convolutional neural networks
• Methods for Interpreting and Understanding Deep Neural Networks
• LSTM Fully Convolutional Networks for Time Series Classification
• ECG Heartbeat Classification: A Deep Transferable Representation
• Systematic evaluation of CNN advances on the ImageNet
• Detection of Inferior Myocardial Infarction using Shallow Convolutional Neural Networks
• Abductive reasoning as the basis to reproduce expert criteria in ECG Atrial Fibrillation identification

Cited by in corpus (6)

• Deep learning for time series classification: a review
• The Virtual Doctor: An Interactive Artificial Intelligence based on Deep Learning for Non-Invasive Prediction of Diabetes
• Application of artificial intelligence techniques for automated detection of myocardial infarction: A review
• Explaining Deep Learning for ECG Analysis: Building Blocks for Auditing and Knowledge Discovery
• Improving ECG Classification Interpretability using Saliency Maps
• A Conceptual Framework for Establishing Trust in Real World Intelligent Systems