Join the Machine Learning in Medical Imaging Consortium (MaLMIC) for a networking opportunity
Monthly Virtual Forum Series on Zoom
Friday, April 16, 2021
3:30 to 5:30 p.m. EDT
Dr. David Berman, a pathologist, discussed unmet clinical needs in pathology machine learning followed by two researchers, Dr. Ali Bashasati and Dr. Chetan Srinidhi, describing their machine learning research work in the area. This was followed by discussion focused on opportunities to collaborate and share resources.
Monthly Virtual Forum Series on Zoom
Friday, April 16, 2021
3:30 to 5:30 p.m. EDT
Dr. David Berman, a pathologist, discussed unmet clinical needs in pathology machine learning followed by two researchers, Dr. Ali Bashasati and Dr. Chetan Srinidhi, describing their machine learning research work in the area. This was followed by discussion focused on opportunities to collaborate and share resources.
Cancer Grading in diagnostic surgical pathology: An unknown quantity
David Berman, MD, PhD
Professor, Queen’s Cancer Research Institute
Talk summary: I described current qualitative approaches to cancer grading. Focused on bladder cancer as an example, I highlighted the potential opportunity to improve patient care by implementing quantitative image analysis in the clinical realm
Revisiting Histopathology by AI: Progress and Outlook
Ali Bashashati, PhD
Director of AI Research, OVCARE, BC Cancer/VCHRI
Assistant Professor, University of British Columbia
Talk summary: AI applied to image analysis, fueled by the increased capacity of digitizing the pathology slides, offers the opportunity to objectively revisit the rich amount of information in tumor morphology. In this presentation, I highlighted the application of AI in digital pathology and how AI could be employed to accelerate translational research and biomarker discovery for cancer. I briefly talked about the BC Translational Digital Pathology AI Initiative where we intend to integrate imaging and omics for biomarker discovery. I then briefly highlighted a few use cases that we are currently working on.
Learning with self-supervision: An application to digital pathology
Chetan Srinidhi, PhD
Postdoctoral Fellow, Sunnybrook Research Institute
Talk summary: Training a neural network with copious amounts of labeled data is still a dominant paradigm in computational histopathology. However, collecting this annotated data is often expensive, tedious, and sometimes even infeasible in clinical practise. This talk explores how to exploit the hidden knowledge present in the unlabelled data to generate a powerful supervisory signal for unsupervised representation learning. We will investigate several histopathology use cases that can benefit from the application of self-supervised learning that leads to state-of-the-art supervised level performance with much reduced manual supervision.