Detecting and interpreting myocardial infarction using fully convolutional neural networks
This work addresses the need for automated, interpretable ECG analysis in clinical settings, though it is incremental as it applies existing neural network and attribution methods to a specific medical domain.
The researchers tackled the problem of detecting myocardial infarction directly from ECG data without preprocessing, achieving 93.3% sensitivity and 89.7% specificity, matching human cardiologist performance.
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.