Artificial intelligence-enhanced risk stratification of cancer therapeutics-related cardiac dysfunction using electrocardiographic images.
Evangelos K OikonomouVeer SanghaLovedeep Singh DhingraArya AminorroayaAndreas CoppiHarlan M KrumholzLauren A BaldassarreRohan KheraPublished in: medRxiv : the preprint server for health sciences (2024)
There is an unmet need for scalable and affordable biomarkers to stratify the risk of cancer therapeutics-related cardiac dysfunction (CTRCD). In this hospital system-based, decade-long cohort of patients without cardiomyopathy receiving anthracyclines or trastuzumab, a validated artificial intelligence algorithm applied to baseline electrocardiographic (AI-ECG) images identified individuals with a 2-fold and 4.8-fold risk of developing any cardiomyopathy or left ventricular ejection fraction <40%, respectively. This supports a role for AI-ECG interpretation of images as a scalable approach for the baseline risk stratification of patients initiating cardiotoxic chemotherapy.
Keyphrases
- artificial intelligence
- ejection fraction
- deep learning
- left ventricular
- machine learning
- aortic stenosis
- big data
- end stage renal disease
- heart failure
- newly diagnosed
- convolutional neural network
- chronic kidney disease
- peritoneal dialysis
- prognostic factors
- oxidative stress
- papillary thyroid
- emergency department
- epidermal growth factor receptor
- atrial fibrillation
- patient reported outcomes
- transcatheter aortic valve replacement
- blood pressure
- young adults
- aortic valve
- coronary artery disease
- acute coronary syndrome
- rectal cancer
- mitral valve
- tyrosine kinase
- childhood cancer
- locally advanced
- acute care