Abstract

In recent years, the integration of language models into medical AI has opened up a plethora of opportunities, revolutionizing the way healthcare professionals diagnose, treat, and manage patient care. This presentation explores the transformative potential of language models in medical AI, highlighting key applications such as the medical notes summary implemented at Changhua Christian Hospital and a research application of using large language models for ICD coding. Medical AI encompasses much more than just language models; we will also cover the application of AI in digital pathology and skin lesion diagnosis using deep learning models. However, alongside these advancements, there are significant limitations and challenges that must be addressed, including data privacy concerns, the need for robust validation, and, in the case of language models, hallucination. By examining both the opportunities and limitations, this presentation aims to provide a comprehensive overview of the current state and future directions of medical AI in the era of language models.

Bio.

Dr. Hackman is currently a Senior Engineer at Phison Electronics Corporation. He holds a Ph.D. in Computer Science from National Yang Ming Chiao Tung University (2015–2021) and an M.S. in Computer Science from Yuan Ze University (2012–2014). With extensive experience in artificial intelligence and data science, he previously served as a Data Scientist at the AI Development Center of Changhua Christian Hospital (2022–2024) and worked as a Postdoctoral Researcher at National Yang Ming Chiao Tung University (2021–2022). His international experience includes serving as a Lecturer (2014–2015) and Teaching Assistant (2010–2012) in the Department of Computer Science at the University of The Gambia. His expertise spans machine learning, deep learning, large language models, medical AI, and computer vision.   

All faculty and students are welcome to join.