Computer Vision News 40 Deep Learning for Medical Imaging The Summer School on Deep Learning for Medical Imaging, now in its fifth year, brings together medical imaging beginners and experts from diverse backgrounds keen to delve into the fundamentals of deep learning and its applications in medical imaging. Organiser Jose Dolz, an associate professor at ÉTS Montréal, is here to tell us more about July’s event. The collaborative effort behind the summer school involves a dedicated team from ÉTS Montréal, the University of Sherbrooke, and CREATIS – INSA Lyon in France, where it was established in 2019. Hosting duties alternate between Lyon and Montreal each year, enabling participation from different parts of the world and creating a genuinely global community. “One year, it’s in France, which attracts people from Europe,” Jose tells us. “The next year, it comes to Montreal so that we can attract more people from the US and North America.” Another distinctive feature is the school’s inclusive nature. Across five days, the curriculum spans a wide spectrum, accommodating students and professionals with varying levels of expertise. “It’s not limited to only AI or medical imaging experts,” Jose points out. “We cover many different topics, from the basics of machine learning and deep learning to more complex, advanced topics. This year, we have talks about foundation models, generative models, normalizing flows, diffusion tensor models, weakly supervised learning, and few-shot learning, so it actually covers many different aspects.” The event makes time for plenty of hands-on practice, reinforcing the theoretical knowledge shared during the talks. This interactive element ensures a deeper understanding for participants and facilitates engagement with tools for those on the clinical side of medical imaging, including doctors and physicians. Breaking down barriers between disciplines, the school enables a fruitful exchange of ideas and collaboration among professionals with different backgrounds.
RkJQdWJsaXNoZXIy NTc3NzU=