Compressor-Based Classification for Atrial Fibrillation Detection

TL;DR

This study demonstrates that gzip compression-based methods can effectively classify atrial fibrillation from ECG data, achieving high sensitivity and specificity, and are adaptable to few-shot learning scenarios.

Contribution

The paper introduces the application of compressor-based text classification techniques to biomedical ECG data for AF detection, showing promising results.

Findings

Achieved sensitivity of 97.1% and specificity of 91.7% on MIT-BIH AF database.

Gzip-based classification performs well even with limited training data.

Method is adaptable to various data types and few-shot learning settings.

Abstract

Atrial fibrillation (AF) is one of the most common arrhythmias with challenging public health implications. Therefore, automatic detection of AF episodes on ECG is one of the essential tasks in biomedical engineering. In this paper, we applied the recently introduced method of compressor-based text classification with gzip algorithm for AF detection (binary classification between heart rhythms). We investigated the normalized compression distance applied to RR-interval and $\Delta$RR-interval sequences ($\Delta$RR-interval is the difference between subsequent RR-intervals). Here, the configuration of the k-nearest neighbour classifier, an optimal window length, and the choice of data types for compression were analyzed. We achieved good classification results while learning on the full MIT-BIH Atrial Fibrillation database, close to the best specialized AF detection algorithms (avg. sensitivity = 97.1\%, avg. specificity = 91.7\%, best sensitivity of 99.8\%, best specificity of 97.6\% with fivefold cross-validation). In addition, we evaluated the classification performance under the few-shot learning setting. Our results suggest that gzip compression-based classification, originally proposed for texts, is suitable for biomedical data and quantized continuous stochastic sequences in general.