Integrating Computed Tomography and Magnetic Resonance Imaging education into a clinical clerkship through specialist-led, case-based learning: a mixed-methods evaluation
This mixed-methods study demonstrates that a single-session, specialist-led, case-based learning module effectively integrates CT and MRI education into neurology clerkships, significantly improving medical students' diagnostic accuracy, self-reported confidence, and interest in radiology without requiring extensive curricular redesign.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In modern hospitals, doctors rely heavily on pictures of the inside of the body to figure out what is wrong with a patient. Two of the most common tools for taking these pictures are Computed Tomography, or CT, and Magnetic Resonance Imaging, or MRI. A CT scan uses X-rays to create detailed cross-sections of the body, while an MRI uses powerful magnets to produce incredibly clear images of soft tissues like the brain. For a patient with a sudden headache or weakness, these images are often the first clue a doctor has about a stroke, a tumor, or an injury. However, learning how to read these complex pictures is a specialized skill usually reserved for radiologists, the doctors who specialize in imaging. For medical students training to become neurologists or other specialists, the time to learn how to interpret these scans is often limited. They might see the images in a textbook or a lecture, but they rarely get to practice looking at the actual, moving slices of a patient's brain in a real clinical setting before they start working with patients. This gap between knowing the theory and being able to apply it in a busy hospital can leave future doctors feeling unprepared when they need to make critical decisions based on a scan.
A team of educators at Weill Cornell Medicine decided to test a new way to bridge this gap. They wanted to see if they could teach medical students how to read brain scans effectively without overhauling their entire curriculum. Instead of adding a long, separate course, they designed a single, focused session that fit directly into the students' existing neurology rotation. The approach was built on a simple idea: let the students try to solve the puzzle on their own first, then bring in an expert to guide them. Before the session, the students were given seven real, anonymized cases involving brain conditions like strokes and tumors. They were asked to look at the actual CT and MRI scans on a computer, scrolling through the images themselves to find abnormalities, just as a doctor would in a real clinic. This was not a passive lecture where students watched a screen; it was an active exercise where they had to navigate the technology and form their own initial thoughts.
After the students had spent time reviewing these cases independently, they gathered for a one-hour session led by a neuroradiologist, a doctor who specializes in brain imaging. In this meeting, the expert walked through the same seven cases, but this time, the teaching was adaptive. The instructor listened to the students' interpretations, identified where they were confused, and tailored the discussion to address those specific uncertainties. The goal was to connect the dots between what the students saw on the screen and the patient's symptoms, showing how a specific dark spot on an image translates to a diagnosis like a stroke or a tumor. The session was designed to be flexible, with most students attending in person and some joining via video call, proving that this kind of specialized training could work in different formats.
The results of this experiment were clear and encouraging. When the students answered questions about their own abilities before and after the session, their confidence grew significantly. They felt much more capable of recognizing what they were looking at on a brain scan. Their understanding of how imaging fits into the broader picture of patient care also improved, and they reported a stronger interest in the field of radiology. More importantly, their actual performance on test cases went up. Before the session, about 85 percent of the students correctly identified a specific type of stroke in a test case; after the session, that number rose to over 94 percent. Similarly, their ability to spot a brain tumor in another case improved from roughly 89 percent to nearly 98 percent. The students rated the session very highly, giving it an average score of 8.6 out of 10, and many said they wished for even more opportunities to learn this way.
This study suggests that a short, well-designed intervention can make a real difference in how medical students learn to read medical images. By pairing independent practice with expert feedback, the educators were able to help students connect the visual data on a screen with the clinical reality of a sick patient. The findings indicate that this model is not only feasible but also highly valued by learners. It offers a practical path forward for medical schools that want to integrate imaging education into their existing programs without needing to build entirely new courses from scratch. While the study was conducted at a single institution and involved a specific group of students, the success of this approach points to a promising way to ensure that future doctors are comfortable and skilled in using the powerful imaging tools that define modern medicine.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.