AI reads heart electrical signals to predict when patients will die
Researchers developed an artificial intelligence system that analyzes standard electrocardiograms to estimate individual patients' mortality risk and timing with significant accuracy. The breakthrough could reshape how hospitals identify high-risk patients and allocate resources, potentially preventing thousands of preventable deaths if widely adopted.
Originaltitel: Artificial intelligence-enabled electrocardiogram for mortality and cardiovascular risk estimation: a model development and validation study
<p>Background Artificial intelligence (AI)-enabled electrocardiography (ECG) can be used to predict risk of future disease and mortality but has not yet been adopted into clinical practice. Existing model predictions do not have actionability at an individual patient level, explainability, or biological plausibi. We sought to address these limitations of previous AI-ECG approaches by developing the AI-ECG risk estimator (AIRE) platform. Methods The AIRE platform was developed in a secondary care dataset (Beth Israel Deaconess Medical Center [BIDMC]) of 1 163 401 ECGs from 189 539 patients with deep learning and a discrete-time survival model to create a patient-specific survival curve with a single ECG. Therefore, AIRE predicts not only risk of mortality, but also time-to- mortality. AIRE was validated in five diverse, transnational cohorts from the USA, Brazil, and the UK (UK Biobank [UKB]), including volunteers, primary care patients, and secondary care patients. Findings AIRE accurately predicts risk of all-cause mortality (BIDMC C-index 0<middle dot>775, 95% CI 0<middle dot>773-0<middle dot>776; C-indices on external validation datasets 0<middle dot>638-0<middle dot>773), future ventricular arrhythmia (BIDMC C-index 0<middle dot>760, 95% CI 0<middle dot>756-0<middle dot>763; UKB C-index 0<middle dot>719, 95% CI 0<middle dot>635-0<middle dot>803), future atherosclerotic cardiovascular disease (0<middle dot>696, 0<middle dot>694-0<middle dot>698; 0<middle dot>643, 0<middle dot>624-0<middle dot>662), and future heart failure (0<middle dot>787, 0<middle dot>785-0<middle dot>789; 0<middle dot>768, 0<middle dot>733-0<middle dot>802). Through phenome-wide and genome-wide association studies, we identified candidate biological pathways for the prediction of increased risk, including changes in cardiac structure and function, and genes associated with cardiac structure, biological ageing, and metabolic syndrome. Interpretation AIRE is an actionable, explainable, and biologically plausible AI-ECG risk estimation platform that has the potential for use worldwide across a wide range of clinical contexts for short-term and long-term risk estimation. Funding British Heart Foundation, National Institute for Health and Care Research, and Medical Research Council. Copyright (c) 2024 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 license.</p>