Deep Analysis of Biomedical Image and Omics Data

Abstract:
AI-driven by deep learning is transforming biomedical science and technology. The enormous success of deep learning stems from its unique capability of extracting essential features from big biomedical datasets for decision-making. In this talk, Dr. Islam will first give a broad overview of his works on AI-driven medical image analysis. He will then present his recent work on efficiently analyzing the deep-learning feature space for better interpretability of AI and using the feature space information for designing high-performance AI models for medical image analysis and radiation treatment planning. Dr. Islam is also going to discuss how to convert omics data into images and integrate the omics data and medical images in a single deep-learning workflow.

Bio:
Dr. Md Tauhidul Islam is a Research Scientist at the Laboratory of Artificial Intelligence in Medicine and Biomedical Physics at the Department of Radiation Oncology, Stanford University. He received his Ph.D. from the Department of Electrical and Computer Engineering, at Texas A&M University. In his PhD, Dr. Islam focused on developing novel ultrasound imaging techniques and medical image analysis methods for better diagnosis, prognosis, and treatment of various cancers. His research spans a broad range of topics, including artificial intelligence, medical image analysis, imaging and cancer physics, big omics data analytics, and image-guided interventions. He has published 41 journal articles and 11 conference papers. Dr. Islam has been awarded an NIH K99/R00 (Pathway to Independence) award.

Zoom:
https://umich.zoom.us/j/94801149707