David Anastasiu

Santa Clara University

Topic

Bio

David C. Anastasiu is an Assistant Professor in the Department of Computer Science and Engineering at Santa Clara University. His research interests fall broadly at the intersection of artificial intelligence/machine learning, data mining, computational genomics, and high performance computing. Much of his work has been focused on scalable and efficient methods for analyzing sparse data. He has developed serial and parallel methods for identifying near neighbors, characterizing how user behavior changes over time, analyzing traffic based on video sensors, and methods for personalized and collaborative presentation of Web search results, among others. In the biomedical domain, he has worked on methods for sensory-based prediction of Autism in children, searching related biochemical compounds, and designating the severity of kidney disease. Prof. Anastasiu serves on the program committees and senior program committees of the most prominent IEEE and ACM data science-related conferences and his work, which is funded by the National Science Foundation and several industrial partners, has been published in many top-tier conferences and journals.

Abstract

Biomedical AI has entered a new era with the rise of ChatGPT and large language models (LLMs), revolutionizing areas such as clinical decision support, literature analysis, and drug discovery. However, despite their impressive capabilities, LLMs face significant challenges in biomedical applications, including data scarcity, interpretability concerns, and regulatory hurdles. This talk will explore the impact of LLMs in healthcare, their limitations, and the enduring value of traditional machine learning methods in small-data settings. We will discuss how a hybrid AI approach can help bridge these gaps and drive meaningful progress in biomedical research.