Speakers: Yugyung Lee, PhD

Yugyung (Yugi) Lee, PhD

Professor of Computer Science
University of Missouri Kansas City

Yugyung Lee, Professor of Computer Science at UMKC’s School of Science and Engineering, plays a pivotal role as coordinator for the PhD, MSDS, and MSCS programs. Her research interests lie in Deep Learning, Natural Language Processing, and AI, with a strong focus on leveraging computational models to enhance knowledge bases and Electronic Health Records (EHR), as well as to improve medical imaging segmentation. Dr. Lee’s scholarly work includes the development of perspective-based graph neural learning methods and advancements in autonomous learning strategies. She has published over 200 peer-reviewed articles and accumulated a significant number of citations, evidencing her scholarly impact. As a committed educator, she has successfully mentored 15 PhD graduates and is currently advising a substantial group of PhD students. Her projects have received backing from prestigious organizations such as the NIH, NSF, NEH, NSA, and Missouri Life Sciences, showcasing her recognized expertise in the field.

Unsupervised Segmentation with SAM and LLMs for Explainable Medical Image Analysis for Cancer and Alzheimer’s Detection

This talk will explore the latest advancements our lab has made in medical imaging analysis using deep learning. We have transitioned from traditional supervised methods to pioneering unsupervised approaches with our Segmentation Anything Model (SAM). This model enables effective segmentation of histopathological images for oral cancer and MRI/fMRI scans for Alzheimer’s detection without the need for pre-labeled datasets. Additionally, we are incorporating Large Language Models (LLMs) to enhance the interpretability of these images, making our research not only innovative but also more comprehensible and applicable. This presentation will demonstrate how these advanced tools significantly improve our diagnostic capabilities for these critical health conditions.