New Pathway Could Advance Artificial Intelligence in Radiology

Combining formal instruction with practical problem-solving for fourth-year residents could accelerate the use of artificial intelligence in radiology, according to a report published in Radiology: Artificial Intelligence.

Artificial intelligence and machine learning have demonstrated their potential to improve medical imaging, from enhancing the quality of CT scans to detecting breast cancer.

However, formal training in AI and machine learning is still rare in the medical imaging field: The report noted that organized AI and machine learning curricula are limited to a few institutions and training opportunities are lacking.

To better prepare radiology residents to use AI and machine learning, and to advance the use of these technologies in medical imaging, three senior radiology residents at Brigham and Women’s Hospital (BWH) recently developed a data science pathway to provide a well-rounded introductory experience to AI and machine learning for fourth-year residents.

The pathway incorporates formal instruction and practical problem-solving into residents’ curriculum, allowing them to collaborate with data scientists.

"Across the nation there are a number of radiology residency programs that are trying to figure out how to integrate AI into their training," said the paper's co-lead author Walter F. Wiggins, MD, PhD.

"We thought that perhaps our experience would help other programs figure out ways to integrate this type of training into either their elective pathways or their more general residency curriculum."

The pathway provides residents with an immersion into AI and machine learning through a flexible schedule of educational, experimental, and research activities at Massachusetts General Hospital (MGH) and BWH Center for Clinical Data Science (CCDS).

Through the pathway, residents were exposed to all aspects of AI and machine learning application development, such as data curation, model design, quality control, and clinical testing. Residents contributed to model and tool development at different stages, and their work during the pilot led to 12 accepted abstracts for presentation at national meetings.

Feedback from the pilot project led to the establishment of a formal AI and machine learning curriculum for future residents.

"Radiologists have always had to manage, analyze and process data in order to be able to do their work," Wiggins said. "We already have the underlying skill sets and infrastructure that we can tap into to allow residents with an interest in AI and ML to really develop and become leaders in applying these skills clinically."

The pathway provides radiology residents with the opportunity to work directly with data scientists to better understand how they approach image analysis issues with machine learning tools.

This partnership enabled the data scientists to better understand how radiologists approach a radiology problem in a clinical setting. The data scientists could be easily integrated into clinical practice.

"An important component of a curriculum like this is to learn the language the data scientists speak and teach them a little bit about the language that we as radiologists speak so that we can have better, more effective collaborations. Going through that process over several different projects was where I think I gained the best experience throughout all of this," said Wiggins.

"I also hope that people from other institutions might read this manuscript and find something useful for integrating into their residency curricula or for developing specialized pathways for informatics and/or data science.”

The team stated that the framework for the pathway could be easily implemented in other organizations, which could help contribute to the widespread use of AI and machine learning in radiology.

“The key factors that contributed to the success of the program that could be extrapolated to any other training institution were motivated learners, dedicated mentors, and integration into a multidisciplinary AI-ML research environment,” researchers concluded.

“It is our belief that each of these facets of the data science pathway experience will prepare residents to serve as leaders in multidisciplinary clinical data science an in radiology as a whole.”