Summer school on deep learning for medical imaging — Recap

Last summer I had the opportunity to attend a summer school on deep learning for medical imaging at the École de Technologie Supérieure in Montreal, Canada. In this post, I will highlight some of my experiences at this event.

First, I’ll start by noting that this is the first time in two years after the pandemic that I and some of the other attendees were in a physical classroom. I was very excited to be meeting colleagues in real life again since in-person meetings provide a sense of connection that is difficult to replicate via video.

This summer school was five days full of learning activities, hands-on workshops, cultural experiences, and networking spaces.

Since this event was attended by professionals in different industries and career stages the first day started by giving us a gentle introduction to the world of machine learning and deep learning to make sure we all were speaking the same language.

One thing that caught my attention was that one of the professors said that traditional machine learning is now obsolete — most problems are now solved using deep learning. (This was the precise reason why I took this course to begin with! The last time I interviewed for data science roles I felt that I was lacking deep learning skills.)

Although I have been studying data science for many years, I finally had a powerful insight into what makes neural networks so powerful:

Once we had a refresher on the basic concepts of deep learning, we went into more depth about the specific challenges and the different neural network architectures used to tackle image-related problems.

One of the things I find fantastic about being an AI professional is that the skills that you have are relevant and transferable to various industries. Although I’m not in the medical field, the strategies I learned during this summer school are applicable in other contexts. However, that doesn’t mean I didn’t take the opportunity to learn a little more about medicine.

At the end of this summer school, we also had the chance to learn about Geometric Deep Learning, one of the current cutting-edge research topics in the industry nowadays.

Cultural outings in Montreal during the summer school:

New colleagues from all walks of life:

For anyone wanting to dive deeper into what we learned during that week, you can explore the hashtag #DLMI2002 on Twitter. And if you want to attend, next year this summer school will be held in Lyon, France.