Researchers at Johns Hopkins presented an explainable artificial intelligence tool at the International Conference on Medical Image Computing and Computer Assisted Intervention. The system was trained on videos of expert surgeons and on tracked hand movements recorded as surgeons closed incisions. Using this data, the AI can both rate a student’s suturing technique and explain, in actionable terms, how the student should change their movements.
The tool gives immediate, personalized feedback by sending students a text message after practice. Each message compares the student’s movements to those of experts and highlights specific adjustments to refine technique. The team designed the approach to reduce the need for attending surgeons to watch, score and then give detailed feedback, a process that is time consuming.
To test the idea, the researchers ran a randomized study with 12 medical students who already had some suturing experience. All participants practiced closing an incision with stitches. One group received immediate AI feedback; the other learned from recorded surgeon videos before practicing again. Results varied by prior experience: students with a solid foundation learned much faster with AI coaching, while beginners struggled and showed less benefit at this stage. Senior author Mathias Unberath said, "The next best thing might be our explainable AI that shows students how their work deviates from expert surgeons." First author Catalina Gomez noted that the team can calculate performance before and after the intervention to see whether students move closer to expert practice.
The researchers plan to refine the model to make it easier to use and more widely available, and they hope to build a version students could use at home with a suturing kit and a smartphone. They say such a system could scale training by providing more practice opportunities outside the clinic. Additional coauthors are from Johns Hopkins and the University of Arkansas. The work was supported by the Johns Hopkins DELTA Grant IO 80061108 and the Link Foundation Fellowship in Modeling, Simulation, and Training.
Difficult words
- explainable — designed to show how decisions are made
- track — follow and record movements or datatracked
- actionable — clear practical steps people can follow
- personalize — make something fit a specific personpersonalized
- suture — to close a wound with stitchessuturing
- refine — make small improvements to a skill
- intervention — an action taken to change an outcome
- randomize — assign things or people by chancerandomized
Tip: hover, focus or tap highlighted words in the article to see quick definitions while you read or listen.
Discussion questions
- Do you think immediate text feedback from AI can replace feedback from an attending surgeon? Why or why not?
- How might having a home suturing kit plus AI feedback change medical students' practice opportunities?
- What changes could help beginners get more benefit from AI coaching, based on the article?
Related articles
Uganda report urges reform of science and innovation
A national report launched on 21 June says Uganda must reform its science, technology and innovation systems to move faster toward middle-income status. It highlights gender gaps, weak funding and calls for stronger links between research, government and business.
Young men in South Korea move to the political right
Surveys after the June 2025 snap presidential election show many young men in South Korea have shifted to the political right, creating a large gender gap on feminism, redistribution and immigration, even as most still support democratic rules.
Brain differences in WTC responders with PTSD
New imaging research of World Trade Center responders finds measurable brain structure differences linked to long-term PTSD. Researchers used gray-white contrast (GWC) MRI and other markers to distinguish responders with and without PTSD.
