Artificial intelligence offers significant potential for early detection and personalized management of cardiotoxicity in cancer patients, though widespread adoption requires rigorous validation and workflow integration.
Artificial intelligence (AI) offers new opportunities in cardio-oncology for early detection, risk stratification, and personalized management of cardiovascular complications in cancer patients. By leveraging data from electronic health records, blood biomarkers, imaging tests such as echocardiography, electrocardiograms, and wearables, AI models can facilitate prediction, detection and response to treatment of cardiovascular disease entities, pre-existing and developing as a consequence of cancer therapy. Specific to the latter, referred to as cardiotoxicity, widespread adoption has been hindered by the limited availability of large datasets for model training, insufficient external validation, and challenges in integrating AI tools into routine clinical workflows. Future progress will depend on advancements in AI technologies, rigorous multi-center validation, development of explainable models, and seamless integration into clinical practice. Barriers, not only from a systems perspective, but also from a provider and most importantly from a patient perspective will need to be addressed for successful implementation. With a broad multidisciplinary perspective and patient focus, AI can advance cardio-oncology care and improve outcomes for patients with cancer.
Ma et al. (Sun,) studied this question.