This Research Infrastructure Improvement (RII) EPSCoR Research Fellows project provides a fellowship to an assistant professor and training for a graduate student at the University of Kansas (KU). This work is conducted in collaboration with the Department of Health Outcomes and Biomedical Informatics at the University of Florida (UF). Through the fellowship, the PI will advance the development of reliable machine learning (ML) methods designed to address pervasive data limitations in medical artificial intelligence (AI) systems. By integrating AI, ML, and clinical informatics, the research will investigate data challenges in clinical environments and build data-resilient learning methods validated on large-scale patient data repositories. The project results will improve the trustworthiness of AI-based clinical decision support and enable physicians to make more accurate, timely, and personalized treatment decisions. In addition, the project will provide hands-on training for graduate students in AI for healthcare and contribute to strengthening the future workforce in this critical domain. This project will address challenges of data quality in developing reliable AI models for medical applications such as disease prognosis, survival analysis, and treatment recommendation. It will investigate issues including data sparsity, domain shifts, and data noise in large-scale electronic health records, and will develop robust AI frameworks through algorithmic innovation, real-world evidence generation, and retrospective clinical validation. The project will strengthen research infrastructure at KU by supporting faculty professional advancement in AI for healthcare, establishing foundational research in data-resilient learning methods, and providing graduate students with hands-on training. The research activities will also strengthen KU’s partnerships with UF by fostering interdisciplinary collaboration that engages trainees from both computer science and medicine, whi