MIDL Vision 2020

A fireside chat with the Program Chairs... 4 Tal Arbel from McGill University and Ismail Ben Ayed from ÉTS Montréal are two of the Program Chairs for MIDL 2020, alongside Marleen de Bruijne (Erasmus Medical Center) and Hervé Lombaert (ÉTS Montréal). Tal and Ismail have managed to find a slot in their busy schedule for a late Sunday night chat with our editor Ralph Anzarouth to talk about what is new this year, the joys of organizing a virtual conference at speed, and what attendees can expect when they log in. Can you tell us a bit about MIDL? Tal: MIDL is the intersection between two domains – deep learning and medical image analysis. It is trying to bridge the gap between these two fields. Important challenges in medical imaging are met with sophisticated and new deep learning architectures and models. I think that is what is unique about our conference. MIDL has gone virtual this year. How have you adapted to that? Ismail: Along with all the other conferences, we have had to juggle many decisions in a very short space of time. It has been hard work for the organizers! But something new and positive to come out of the fact that it is virtual is that it is free this year. Not all virtual conferences are free. We have had many thank you notes about this, particularly from countries where people cannot always afford to take part. Tal: We are the only conference that I know of that is fully free. Of course, our sponsors have helped. Our goal is to lower the barrier and be more inclusive. We already have over 3,500 people registered. That is a huge increase on last year. Ismail: Another idea borne out of the virtual conference, which is quite cool in my opinion, is to have a chat channel for every paper so that people can interact with the author throughout the event. We are considering keeping it for a period after the conference too to give the works more exposure and to accommodate different time zones. Tal : There will be Q&A in the all-day chat and the poster rooms as well. The virtual MIDL Vision Designer: Rotem Sm Publisher: RSIP Vision All rights reserved Unauthorized reproduction is strictly forbidden. Copyright: RSIP Vision Our editorial choices are fully independent from MIDL and the conference organizers. aspect can be constricting in some ways, but it does allow for greater access to the authors and a different kind of interaction between authors and participants. We strive for openness. What else is new this year? Tal: We have challenges for the first time this year. We are trying to bring the machine learning community closer to themedical imaging community and introduce them to some of the challenges that are specific to our community. We also have some very interesting