The curriculum for the AI Literacy Course curriculum was developed with a focus on promoting practical and clinically relevant AI experience for radiology trainees and establishing a foundational knowledge base. The course begins with Two 30-minute lectures were held each day from October 4-8 and broadcast via Zoom; lecturers were not recorded. To align with the “noon conference” time frame for the host institution, lectures were held from 12-1pm CST, with additional extra time allotted on the first day of the course to allow for the course introduction. The course was available free of charge to all residents, fellows, attendings, and medical students on radiology rotations at participating institutions at the time of the course.

The course will begin with introductory didactic lectures on fundamental terminology and methods of AI, followed by a series of subspecialty-based lectures (Pediatric Radiology, Neuroradiology, Abdominal Imaging, etc.), and then special topics based lectures (Economics and Ethics of AI, Algorithm Bias, The Future of AI in Radiology, etc.). The course concludes with a “Hands-on” session with an FDA approved AI algorithm. In this session, participants will be given a guided introduction to the application and will have the opportunity to complete curated cases/experiences on their own computer.

The course is designed to incorporate feedback from participants and partner programs. Lecturers are encouraged to highlight their areas of expertise, emerging technologies, and clinical AI tools they find relevant to learners. Participants can suggest topics or recommend areas of interest for future courses during the midweek feedback session or in the post-course survey. Most subspecialty and special topics sessions will rotate on a two year cycle, however “Ethics of AI” is planned to recur with each session. A session on the ethics of AI allows course directors to discuss the role physicians and researchers must play to ensure AI advances health equity instead of deepening health disparities and is fundamental to the mission of the AI Literacy Course.

In a survey of 62 radiology training programs in the U.S., a recent study found that less than half had any formal artificial intelligence (AI) educational initiatives and only 3% of programs advertised their training pathway to residents or fellows [1]. AI education and hands-on experience with AI have been shown to mitigate fear of AI. Perhaps more importantly, radiology trainees who have experience with AI are able to develop more realistic expectations of what AI can and cannot do. Radiology trainees often report that AI is important to their medical training, however few radiology trainees feel that they have had enough AI exposure through their training program.

Opportunities for AI education are becoming more available for radiology trainees, with more radiology residencies instituting AI training curricula, radiological societies hosting online courses, in addition to the variety of free courses on the basics of coding and machine learning available online [2,3]. The field of radiology has recognized the importance of AI education in radiology and is slowly taking steps to address this need, however substantial barriers to accessible and practical AI education remain. The disparity in the number of programs with AI education opportunities and the number of radiology trainees is stark, and this disparity becomes even more pronounced in under-resourced healthcare settings and outside the United States.

The AI Literacy Course, hosted by Artificial Intelligence in Radiology Education (AIRE), is the largest free Radiology AI education resource in the world, reaching 500 participants at 25 US programs and in 10 countries last year. This course serves to provide accessible, fundamental, and practical AI education and to provide the tools for radiologists to thrive in the future of radiology.

References

1. Li D, Morkos J, Gage D, Yi PH. Artificial intelligence education and research initiatives and leadership positions in academic radiology departments. Current Problems in Diagnostic Radiology. 2022; 51(4): 552-555.
2. Lindqwister AL, Hassanpour S, Lewis PJ, Sin JM. AI-RADS: An artificial intelligence curriculum for residents. Academic Radiology. 2021. 28(12): 1810-1816. https://doi.org/10.1016/j.acra.2020.09.017
3. RSNA AI Certificate Program. Radiological Society of North America. https://www.rsna.org/ai-certificate. Accessed 11 April 2022.