Faculty Candidate Seminar
AI-powered Computational Imaging and Processing for Advancing Visualization and Perception in Biomedical Applications
This event is free and open to the publicAdd to Google Calendar
In recent years, real-world challenges in biomedicine and healthcare emerge with the COVID-19 pandemic and increasing cancer cases. Deepening our understanding of human health is more important than ever before, where computational imaging and processing are critical to help understand human health at different levels. To address real-world challenges, my research develops efficient ML models for biomedical imaging and processing. Because of the unique data characteristics and requirements of biomedical tasks, many challenges exist in this emerging field of biomedical AI. In this talk, I will explore these challenges and present the two following lines of work:
First, I will introduce my work in designing AI-powered computational imaging models for advancing visualization. I will discuss how to integrate different kinds of prior knowledge from the physical world to develop reliable data-efficient ML models, by exploiting physics priors, longitudinal priors and data distribution priors. With the innovative ML models, the proposed approaches can be generally applied to various biomedical imaging applications including CT and MRI image reconstruction, X-ray projection synthesis, metal artifacts reduction, and molecular imaging such as Cryo-EM.
Second, I will introduce my work in developing AI-powered data processing models for advancing perception. I will discuss how to design ML models that are adaptive to unique characteristics of biomedical data including random noise and multi-modality. I will present a self-attention-guided ML model for quantitative image perception. Through international collaborations for cross-institute validation among four U.S. clinical centers and a Turkey institute, this work demonstrates the possibility for the developed ML model to characterize the in utero neurodevelopmental trajectory in real-world deployment.
Liyue Shen is a final-year Ph.D. candidate in Electrical Engineering at Stanford University, co-advised by John Pauly and Lei Xing. Her research focuses on Medical AI, which spans the interdisciplinary research areas of AI/ML, computer vision, biomedical imaging and data science. Her dissertation research develops efficient AI/ML-driven computational algorithms and techniques for biomedical imaging and processing to address real-world biomedical and healthcare problems through engineering and data science. Her works have been published in both ML/CV conferences (ICLR, ICCV, CVPR) and medical journals (Nature Biomedical Engineering, IEEE TMI, MedIA, Scientific Reports). She was the recipient of the Stanford Bio-X Bowes Graduate Student Fellowship, and was selected as Rising Star in EECS by MIT and Rising Star in Data Science by University of Chicago. She co-organized the Women in Machine Learning (WiML) Workshop at ICML 2021 and the Machine Learning for Healthcare (ML4H) Workshop at NeurIPS 2021. She received an M.Sc. from Stanford University. Before that, she conferred her bachelor’s degree in Electronic Engineering from Tsinghua University. In the past, she has interned at Nvidia and Waymo Research.