Deep learning (DL)-based interpretation of medical images has reached a critical juncture of expanding outside research projects into translational ones, and is ready to make its way to the clinics. Advances over the last decade in data availability, DL techniques, as well as computing capabilities have accelerated this journey. Through this journey, today we have a better understanding of the challenges to and pitfalls of wider adoption of DL into clinical care, which, according to us, should and will drive the advances in this field for the next few years. The most important amongst these challenges are the lack of an appropriately digitized environment within healthcare institutions, the lack of adequate open and representative datasets on which DL algorithms can be trained and tested, and the lack of robustness of widely used DL training algorithms to certain pervasive pathological characteristics of medical images and repositories. In this talk, we provide an overview of the role of imaging in oncology, the different techniques that are shaping the way DL algorithms are being made ready for clinical use, and also the problems that DL techniques still need to address before DL can find a home in clinics. Finally, we also provide a summary of how DL can potentially drive the adoption of digital pathology.
Amit Sethi is an Associate Professor of Electrical Engineering at IIT Bombay, and a Visiting Instructor of Pathology at UIC. His research group works on computer vision, deep learning, and medical image analysis. His current research is focused on extracting valuable information, such as for prognosis, using deep learning on inexpensive medical modalities, such as whole slide pathology images of cancer tissues stained using hematoxylin and eosin. Technical challenges to solve these real-world problems include working with images that are very large in size but few in number, variable image quality and characteristics between hospitals, and the lack of reliable annotations. Search for better solutions involve concepts from deep neural networks, weak supervision, self-supervision, semi-supervision, domain adaptation, few-shot learning, and learning from noisy labels. Before joining IIT Bombay, he worked as a faculty member of Electronics and Electrical Engineering at IIT Guwahati, and as a management consultant at ZS Associates' Chicago where he worked producing business insights using data analysis. He obtained his PhD in Electrical and Computer Engineering from University of Illinois at Urbana-Champaign in 2005 with a focus on computer vision and machine learning, and a bachelors in Electrical Engineering from IIT Delhi in 1999.