Key result
CT-based AI enables automated coronary assessment and cancer therapy-related cardiotoxicity risk prediction on chest CT.
Why the study?
Cancer therapy-related cardiovascular toxicity often remains clinically silent until myocardial dysfunction or vascular disease develops, while CT is a valuable yet underutilized source of cardiovascular information during oncologic care.
Does CT-based artificial intelligence and radiomics improve the assessment and risk prediction of cardiovascular injury in cancer patients undergoing chemotherapy and radiotherapy?
Does CT-based artificial intelligence and radiomics improve the assessment and risk prediction of cardiovascular injury in cancer patients undergoing chemotherapy and radiotherapy?
This narrative review synthesizes how artificial intelligence and radiomics applied to dedicated cardiac CT and routine oncologic chest CT can facilitate early detection and risk stratification of chemotherapy- and radiotherapy-related cardiotoxicity, though prospective validation is needed.
Answers are grounded in this paper and cite their sources.
May aid opportunistic cardiotoxicity screening on routine CT; leaves open prospective validation before adoption.
Cancer therapy-related cardiovascular toxicity often remains clinically silent until myocardial dysfunction or vascular disease develops.Computed tomography (CT), integral to cancer screening, staging, radiotherapy planning, and follow-up, serves as a valuable yet underutilized source of cardiovascular information.This narrative minireview, derived from a nonsystematic PubMed search, examines the potential of artificial intelligence (AI) and radiomics to improve CT-based assessments of cardiovascular injury during oncologic therapy.An overview of the pathological basis of treatment-related cardiotoxicity is provided, along with applications of dedicated cardiac CT, including coronary artery calcium scoring, coronary CT angiography, and CT-derived fractional flow reserve.Opportunistic cardiovascular screening on routine non-electrocardiogram-gated chest CT is also emphasized.AI-enabled automated cardiac quantification and the capability of radiomics to extract quantitative features of myocardium and coronary plaques for risk prediction are highlighted.Integration of CT with other modalities, clinical decision-support systems, and multidisciplinary care is discussed.Current challenges, such as data standardization, model interpretability, and radiation exposure, are addressed, along with future directions involving therapyspecific models and multi-omics integration.Most existing CT-based AI and radiomics studies are retrospective, necessitating external validation and prospective evaluation to establish clinical utility.
No takes yet. Share an insight, caveat, or question.
Yao et al. (2026) conducted a review in Chemotherapy- and radiotherapy-related cardiotoxicity in cancer patients. Computed tomography-based artificial intelligence and radiomics vs. Conventional manual CT interpretation was evaluated. CT-based AI and radiomics enable automated coronary calcium scoring, plaque analysis, and risk prediction for cancer therapy-related cardiotoxicity on dedicated and routine chest CT.
Synapse has enriched 10 related papers. Consider them for comparative context: