Artificial Intelligence for Risk Assessment of Cancer Therapy-Related Cardiotoxicity and Precision Cardio-Oncology

Through the success of basic and translational research, cancer survivors are one of the largest growing subsets of the patients in the healthcare systems. However, the growing awareness of cardiotoxicity resulting from cancer treatment has led to the emerging field of cardio-oncology, which centers on screening, monitoring, and treating patients with cardiac dysfunction during or after cancer treatment or even in early of anticancer drug discovery. Applications of artificial intelligence (AI) and machine learning (ML) application in clinical decision-making and drug development are rapidly evolving. Additionally, the breadth and scale of biological and phenotypic data (biological phenotyping, wearable devices, imaging, and access to large-scale electronic health record) are unprecedented and will continue to expand and become too complicated to interpret by eye. AI and ML models, with clinical oversight, have the potential to revolutionize cardio-oncology patient management and improve cardiac risk stratification during drug discovery and post-marketing surveillance. This review will provide overall review of data resources and AI/ML models used for cardio-oncology and case studies for patient risk stratification and post-marketing drug surveillance and discuss limitations and future directions for the field.