Reflections from MIDL 2024: Advancing Trustworthy AI in Healthcare

Posted on: 8 July 2024

Last Friday, 5 July, our lab had the honour of being represented by our director and ICREA professor Karim Lekadir, who delivered a keynote on trustworthy AI at the Medical Imaging with Deep Learning (MIDL) 2024 conference in Paris. This conference was an incredible opportunity to share our lab’s work and insights on AI projects across Europe, Africa, and Latin America.

 

Keynote Highlights

In the keynote, Karim discussed the critical importance of trustworthy AI in healthcare. The talk highlighted the challenges and successes of implementing AI solutions in diverse regions, emphasising the need for ethical and accessible AI. One of the standout points was the presentation of the FUTURE-AI guidelines, which provide a comprehensive framework for developing trustworthy and deployable artificial intelligence in healthcare. For those interested in diving deeper, you can access the guideline here: FUTURE-AI Guideline.

 

First Impressions of MIDL

This was our lab’s first experience at MIDL, and it certainly will not be the last. The conference was a remarkable gathering of brilliant minds dedicated to advancing the field of medical imaging and AI. Karim had the pleasure of meeting many friendly and passionate researchers, engaging in discussions about the future of ethical and accessible AI for global health.

 

A personal highlight for Karim was the gala dinner at the Musée des Arts Forains (Fairground Museum). The event was a delightful blend of history, culture, and networking, making it one of the most memorable gala dinners attended.

 

Spotlight on Research

MIDL 2024 featured an array of impressive papers covering topics from foundation models and federated learning to multi-modal models and trustworthy AI. You can explore the full list of papers presented at the conference here: MIDL 2024 Scientific Program.

 

We would also like to highlight an important paper from Veronika Cheplygina’s lab on “Data Usage and Citation Practices in Medical Imaging Conferences.” This paper addresses crucial aspects of data integrity and citation standards in our field. It’s a must-read for anyone involved in medical imaging research: Read the paper.

 

Looking Ahead

Our lab is excited to continue contributing to the advancement of trustworthy AI in healthcare. Maria Vakalopoulou, Caroline Petitjean, and the entire team organised a fantastic event. As Co-General Chair for MICCAI 2024, we aspire to host an equally outstanding conference in October in Marrakesh, Morocco.