Could learning to analyze artwork help novice medical residents better read radiology graphs? A new study by two Yale School of Medicine professors found just that.

In their paper, published on Aug. 24 in the Journal of the American College of Radiology, professors Rob Goodman and Michael Kelleher describe showing radiology residents 15 abnormal radiographs before and after undergoing a training session in analyzing artwork at the Yale Center for British Art. On average, the residents correctly identified four more abnormalities following the training session.

“Our study showed that our trainees’ perception rapidly improved following the session at YCBA,” Goodman, the first author, said. “This is incredibly important because lives depend on a radiologist’s ability to perceive a lesion. If a session at an art gallery improves this, it is time well spent.”

Linda Friedlaender, the senior curator of education at the YCBA and the founder of the Enhancing Observation program, said she adapted the existing program for Goodman’s radiology residents, since it was originally made for dermatology trainees. She added that both disciplines — radiology and dermatology — use visual inferences to make a diagnosis.

Analyzing radiographs depends on both interpretation and perception, Goodman said, adding that some believe only interpretation can be taught. “I wanted to prove them wrong,” he said.

The program consists of two and a half hours spent in the gallery. First, the trainees were given radiographs and asked to identify any abnormalities. Then, the residents were broken up into groups and assigned a painting, given 12 minutes to study it and write down notes in silence. Next, the residents were asked to describe their painting to trained gallery docents without using any conclusions or interpretations.

Friedlaender said she often has to stop participants during this portion and remind them to stay objective before they move on and draw conclusions based on their observations. She added that, afterward, she finds that participants often start describing the pictures on their posttest using the same vocabulary they began to use in the gallery.

“They’ll talk about the perimeter of a rash, they’ll talk about the texture — is it raised or flat — they’ll talk about the color,” Friedlaender said.

Goodman said training residents in an art gallery is especially useful since they do not yet have the words with which to describe radiological findings, but anyone can describe findings buried within paintings.

Additionally, he said a common pitfall in radiology is “satisfaction of search”: stopping when you see something and missing potentially more important findings elsewhere. Similarly, Friedlaender said students who undergo the program often reflect that once they believe they have found what the painting seems to be about, they stop looking and often overlook key details.

Since its inception in 1999, the program has grown to include undergraduate, medical and nursing students at both Yale and Quinnipiac University. Each year, 100 first-year Yale medical students also take part in the workshop.

Friedlaender said she was inspired to start the program after overhearing her husband, an orthopedic surgeon at Yale New Haven Hospital, talk to residents over the phone.

“I would hear him get a call late at night from a resident about someone that came to the ER, and Gary would ask them to describe what they were looking at and he would say, ‘Yes, but I need more information, can you be more detailed?’” she said.

The program was initially designed for dermatology residents at the medical school, and Friedlaender and two Yale dermatologists published a paper in the Journal of the American Medical Association in 2001 describing preliminary results, but this study marks the first time the program has been catered specifically to radiology residents.

The paper discussed two main limitations: a small sample size and a random selection of radiographs. Only 15 radiology trainees participated in the study, though all scored at least one point higher on the posttest than the pretest. Additionally, the radiographs used for both tests were not necessarily of equal difficulty.

Despite the limitations, Goodman said the program was a success, in terms of both the data and the effect on the residents, and that residents’ feedback was largely positive.

“One resident told me ‘I’ve never looked at one thing for so long,” he said. “The more I looked, the more I kept seeing.’ One other jokingly said, ‘Art made me smart.’”

He also added that it was much less intimidating for residents to be taken to an art gallery on day one rather than a technologically advanced radiology reading room.

Friedlaender said other universities and medical centers have introduced programs to expose their students to the humanities, but she has focused on the scientific benefits of the Enhancing Observation program.

“I’ve really stayed very focused on the program being about observation. If people learn something about British art on the way, well, so much the better,” she said.

Contact Madeline Bender at madeline.bender@yale.edu .