Artificial intelligence is rapidly transforming the healthcare landscape, spanning from diagnosis to treatment. In recent years, deep learning has matured into a clinical tool capable of enhancing accuracy and efficiency across diverse medical imaging and diagnostic domains. Building on these foundations, generative AI has opened new frontiers in automated report drafting, clinical documentation, and the simulation of complex generative process of medical data. The emergence of multimodal large language models (LLMs) further enables the integration of heterogeneous clinical data—images, text, signals, and genomics—into unified reasoning systems, moving closer to holistic patient understanding. Beyond static decision support, agentic AI systems are beginning to demonstrate autonomy in orchestrating clinical workflows, optimizing resource allocation, and supporting personalized care pathways. Meanwhile, physical AI, embodied through robotics and sensor-rich systems, is expanding the role of AI from cognition to action with active and extensive applications in surgery. Together, these technologies illustrate a paradigm shift from narrow AI tools toward intelligent, adaptive, and integrated healthcare ecosystems. Nevertheless, challenges remain in areas of safety, interpretability, regulatory oversight, and ethical deployment. This lecture will survey the state of the art, highlight recent breakthroughs, and discuss the future prospects of medical AI as it evolves from diagnosis to treatment.

Kyu-Hwan Jung, Ph.D., is an Associate Professor at the Samsung Advanced Institute for Health Sciences and Technology (SAIHST), Sungkyunkwan University, and holds a joint appointment as an Assistant Professor at the Institute for Future Medicine, Samsung Medical Center. Prior to joining Sungkyunkwan University, he co-founded and served as CTO of VUNO Inc., a leading medical AI startup, where he directed the development, regulatory approval, and commercialization of multiple medical AI solutions. His research spans deep learning, software as a medical device, medical image analysis, and clinical decision support systems. He serves as Co-Chair of Industry Engagement for MICCAI 2025 and has held leadership roles in the Korean Society of Artificial Intelligence in Medicine and the Korean Society of Imaging Informatics in Medicine. With extensive publications in top-tier clinical journals such as Radiology and Ophthalmology, as well as leading AI conferences including MICCAI, NeurIPS, and AAAI, he bridges academia, industry, and healthcare practice to advance the adoption of AI in medicine.