IBIIS-AIMI Postdoc Seminar: Indrani Bhattacharya, PhD and Rogier van der Sluijs, PhD

Talk 1: Multimodal Data Fusion for Selective Identification of Aggressive and Indolent Prostate Cancer on Magnetic Resonance Imaging

Abstract: Automated methods for detecting prostate cancer and distinguishing indolent from aggressive disease on Magnetic Resonance Imaging (MRI) could assist in early diagnosis and treatment planning. Existing automated methods of prostate cancer detection mostly rely on ground truth labels with limited accuracy, ignore disease pathology characteristics observed on resected tissue, and cannot selectively identify aggressive (Gleason Pattern≥4) and indolent (Gleason Pattern=3) cancers when they co-exist in mixed lesions. This talk will cover multimodal and multi-scale fusion approaches to integrate radiology images, pathology images, and clinical domain knowledge about prostate cancer distribution to selectively identify and localize aggressive and indolent cancers on prostate MRI.

About:  Dr. Indrani Bhattacharya is a Postdoctoral Scholar at Stanford University School of Medicine. Prior to joining Stanford, Dr. Bhattacharya received her Ph.D. and M.S. in Electrical Engineering from Rensselaer Polytechnic Institute (RPI), NY. Her research interests are in machine learning, computer vision, and multimodal data analytics applied to several interdisciplinary real-world problems in precision medicine, biomedical image processing, and human-centered computing. Her current postdoctoral research focuses on multimodal and multi-scale data fusion for developing automated methods for cancer detection. Her doctoral research focused on the development of multi-sensor fusion algorithms for estimating and analyzing human behavior in group interactions. She has been the recipient of multiple awards and recognitions, including Rising Stars in EECS 2020, MICCAI 2020 NIH award, the Founder’s Award of Excellence from RPI, best poster, perfect pitch and travel awards from DoE, NSF, NIH, C3E Initiative, Women in Computer Vision, RPI and Stanford.

Talk 2: Pretraining Neural Networks for Medical AI

 Abstract: Transfer learning has quickly become standard practice for deep learning on medical images. Typically, practitioners repurpose existing neural networks and their corresponding weights to bootstrap model development. This talk will cover several methods to pretrain neural networks for medical tasks. The current challenges for pretraining neural networks in Radiology will be discussed and recent advancements that address these bottlenecks will be highlighted.

About: Dr. Rogier van der Sluijs is a postdoctoral researcher at the Center for Artificial Intelligence in Medicine & Imaging of the Department of Radiology at Stanford University. Dr. van der Sluijs was trained as a medical doctor at the University of Utrecht, holds a MSc in Epidemiology, and received his Ph.D. in Trauma Surgery. His research is at the intersection of machine learning, medical imaging, and representation learning.

Contact Email: 
aimicenter@stanford.edu