Electrocardiography
Also known as ECG
Overview
Electrocardiography analyzes electrical waveforms recorded from the heart to identify rhythm disorders, measure cardiac intervals, and screen for conditions such as heart failure or myocardial infarction. Common methods include signal processing and waveform delineation, supervised CNN and Transformer classifiers, and self-supervised foundation models; multimodal approaches also align ECGs with clinical text or signals such as PPG. Researchers evaluate these systems on benchmarks including PTB-XL and MIT-BIH, as well as clinical datasets such as MIMIC-IV-ECG, for tasks ranging from beat classification to long-term monitoring and ECG reconstruction. A central challenge is making strong benchmark results hold across hospitals, patient groups, devices, and recording conditions while providing explanations and evidence that clinicians can trust.
Latest papers
108 papers11 papers in the last four weeks, down 31% on the four weeks before. 0.1% of all new papers.