Computer Vision News - July 2022

BEST OF CVPR 37 Medical Computer Vision The 11 keynote presentations covered cutting-edge research topics from leading researchers from academia and industry, including Q&A sessions with both live and virtual audiences. A total of 60+ live attendees, as well as 50+ remote attendees, joined the event, engaging in intensive discussion regarding the technological advances, the latest progress, challenges, and future promises. The detailed schedule, highlights, and recordings can be found here . Kensaku Mori from Nagoya University kicked off the morning session by introducing AI-based endoscopic procedure . Specifically, he introduced multiple advanced AI techniques for several critical tasks to improve performing endoscopic procedures including blood vessel recognition on laparoscopic video, left gastric artery segmentation, depth estimation and shape recovery, and classification of laparoscopic images, among other tooics. Prof. Mori also shared valuable and unique insights on the current challenges as well as future trends. Ender Konukoglu , from ETH Zurich , gave an overview of robust and trustworthy AI for medical imaging. He revealed a few key traits of trustworthy AI for medical imaging: 1) robust (adaptable systems robust to changes in input characteristics consistency), 2) interpretable , 3) Self- aware (identifying cases where confident predictions are not possible, 4) Sensitive (useful at realistic operating points that are applicable to clinical use cases). BenGlocker , from Imperial College London , introduced “safety nets in medical imaging AI ”. Safety nets provide a layer of protection against AI failures. In medical imaging AI, we need to make sure that the use of AI is safe and that any predictions made by an algorithm are trustworthy. Especially, Ben discussed various safeguards, including automatic quality control, failure detection, and stress testing, and also delved into robustness and reliability in the context of mismatched data between the method development and clinical deployment. Further, the use of causal reasoning and how it can help identify potential biases concluded his talk. In the first afternoon session, Xiaoxiao Li from the University of British Columbia (UBC) introduced different advanced federated learning (FL) methods to use more and diverse medical data under