What is the clinical evidence from this study?

Study design: Other. Population: Cardiac conditions (ECG analysis) (n=18869). Intervention: Explainable AI (XAI) methods (Saliency maps, GradCAM, LRP, IG).

May 26, 2023Open Access

Explaining Deep Learning for ECG Analysis: Building Blocks for Auditing and Knowledge Discovery

PWPatrick WagnerKU Leuven TMTemesgen MehariElectrophysiology WHWilhelm HaverkampElectrophysiology

Key Result

Saliency maps demonstrated the highest spatial and temporal specificity for ECG analysis, enabling glocal XAI techniques to align deep learning model behavior with cardiologists' decision rules.

Structured PICO

Can post-hoc explainable AI methods improve transparency and enable knowledge discovery in deep learning models for ECG analysis?

Population

ECG data analyzed by deep neural networks

Intervention

Post-hoc explainable AI (XAI) methods (local and global perspectives)

Outcome

Identification of sensible attribution methods and quantitative evidence in accordance with expert rules

Post-hoc explainable AI methods can enhance the transparency of deep learning models in ECG analysis, aiding in internal validity assessment and clinical knowledge discovery.

Limitations

Post-hoc attributions may not represent explanations from the perspective of human end-users
Inherent dependent nature of relationships in limb leads makes it intricate to isolate attributions

Abstract

Deep neural networks have become increasingly popular for analyzing ECG data because of their ability to accurately identify cardiac conditions and hidden clinical factors. However, the lack of transparency due to the black box nature of these models is a common concern. To address this issue, explainable AI (XAI) methods can be employed. In this study, we present a comprehensive analysis of post-hoc XAI methods, investigating the local (attributions per sample) and global (based on domain expert concepts) perspectives. We have established a set of sanity checks to identify sensible attribution methods, and we provide quantitative evidence in accordance with expert rules. This dataset-wide analysis goes beyond anecdotal evidence by aggregating data across patient subgroups. Furthermore, we demonstrate how these XAI techniques can be utilized for knowledge discovery, such as identifying subtypes of myocardial infarction. We believe that these proposed methods can serve as building blocks for a complementary assessment of the internal validity during a certification process, as well as for knowledge discovery in the field of ECG analysis.

Ask AI

Helpful

Bookmark

View Full Paper