The RFS AI Journal Club: A Roadmap for Foundational Research
on AI in Medical Imaging
AI tools are progressively making an impact on many industries, including healthcare and radiology. As we recognize their potential to improve patient care, we are exploring ways to implement these technologies. We are still in the early phase, but there is no doubt that AI will play a pivotal role in the practice of radiology over the next decade and beyond.
As a current resident, one obvious question is how do I learn more about disruptive technologies like AI and machine learning (ML)? Most of the residency curriculum includes clinical topics and even non-interpretive skills, such as economics and quality. However, with all the media coverage on the potential impact of AI/ML, there are few resources available during residency to learn about AI/ML. It was not until later that I discovered the National Imaging Informatics Curriculum and Course.
With the need for more information, I was fortunate to discover the AI journal club, hosted by the ACR-RFS. The club is organized by residents from multiple institutions and a variety of topics are discussed online (learn more at acr.org/Member-Resources/rfs/Journal-Club).
Until now, journal clubs have covered a variety of both radiology and non-radiology topics. The panel includes leaders in the field of radiology, as well as computer science and research professionals. The goal of the journal club is to provide wide exposure to the field of AI and increase collaboration with different disciplines.
As my curiosity peaked, I found myself supplementing my knowledge with online courses on ML and starting to work on research projects. Judy Gichoya, MD, one of the founders of the club, invited me to participate in one of the sessions. It was a welcoming challenge to give a lecture online. We wanted to find an article that focused on what was new in AI and on the horizon. The article “A Roadmap for Foundational Research on AI in Medical Imaging” was a perfect fit. We reached out to Curtis P. Langlotz, MD, PhD, and Bibb Allen Jr., MD, FACR, for guidance, and another journal club was in motion.
While the article describes the pathway for AI/ML algorithm development, it also gives insight into the current challenges researchers face as they develop the technology. This article is perfect for trainees as it delineates different research opportunities for anyone with an interest in playing an active role in informatics.
Thanks to the initiatives of residents from multiple institutions and the ACR, a source of knowledge was made available for everyone in the world.
The RFS journal club plays a pioneering role in the dissemination of knowledge in regards to AI and radiology and might offer an early foundational step toward a future curriculum in informatics for residents worldwide.
This has been an amazing experience and I have been fortunate to learn a lot about AI and ML in radiology. I am very grateful and appreciative to the RFS-AI Advisory Council for this incredible opportunity.
Dan Cohen-Addad, MD is a radiology chief resident at SUNY Downstate Medical Center.