The 6th “2026 SNU FastMRI Challenge,” held at Seoul National University for 50 days from July 1 to August 20, concluded successfully, with the winners announced.

The FastMRI Challenge is a competition in deep-learning-based image reconstruction technology that seeks to reduce MRI scan times while producing high-quality images. Led by Professor Jong-Ho Lee of the Department of Electrical and Computer Engineering, the challenge marked its sixth year in 2026. A total of 271 undergraduate students from 186 teams participated in this year’s competition. Students from diverse majors including electrical and computer engineering, computer science and engineering, biological sciences, and liberal studies took part, demonstrating notable collaboration across disciplinary boundaries. As participation has grown, more students have returned to take on the challenge again in subsequent years or looked forward to it each summer break, establishing the FastMRI Challenge as a representative program at Seoul National University. The event was hosted by the Department of Electrical and Computer Engineering at Seoul National University and sponsored by AIRS Medical, KLIM Ventures, SNU Information Technology Services, and the Brain Imaging Center.

One notable tradition of the challenge is that award recipients from the previous year return as teaching assistants for the following year’s competition, passing on their experience to new participants. This year, Woojin Lee, a fourth-year student in the Department of Materials Science and Engineering, and Seongsu Lee, a fourth-year student in the Department of Physics Education, who both achieved strong results in last year’s competition, participated as teaching assistants and supported the operation of the event. This tradition has created a virtuous cycle in which the award recipients’ experience contributes to the growth of the next generation of participants.

Following intense competition for the total prize of KRW 30 million, the team “Suwon Gorani Hunters,” consisting of Seungmin Baek from the Department of Electrical and Computer Engineering and Seoyeon Lee from the School of Biological Sciences, won first place and received a prize of KRW 15 million. Both participants drew attention for taking part in the competition while serving in the military.

Second place went to the team of Woojin Ahn and Chaewon Im from the Department of Computer Science and Engineering. Third place was awarded to the team of Jehyeok Choi from the Department of Electrical and Computer Engineering and Juwon Seo from the Department of Computer Science and Engineering. The teams of Jeongmin Shin and Daehyeop Kim from the College of Liberal Studies and Sehwa Im and Yunsuk Ra from the Department of Electrical and Computer Engineering placed fourth and fifth, respectively.

In a post-competition survey, many participants expressed satisfaction with the program, commenting that the challenge allowed them to make productive use of their summer break and left them with positive memories. Participants also reported high satisfaction with the opportunity to train relatively large models and gain hands-on experience with medical data. One participant noted, “I had previously worked only on one-train, one-inference research, such as 3DGS and PINNs, but it was valuable to have the opportunity to train a relatively large model through this competition.” Another stated, “It was especially valuable to collaborate with colleagues using GPUs, work directly with medical data that is usually difficult to access, and apply concepts learned in major courses to an actual project.”

In their acceptance remarks, the first-place team cited the extensive discussions with their teammates and the opportunity to experiment with diverse approaches, even in an environment with limited internet access during military service, as their most meaningful experiences. They also noted that clearly dividing their roles between idea implementation and data analysis greatly helped them conduct experiments efficiently within a limited timeframe.

Professor Lee stated, “Whenever I see students working day and night with real-world data during their summer break, I feel a great sense of fulfillment. I hope that the talented individuals who grow through this challenge will lead the future development of AI technology.”

Since its first launch in 2021, the SNU FastMRI Challenge has continued for six years and has become a representative summer program for SNU engineering students. With its tradition of award recipients returning as teaching assistants, as well as its provision of hands-on experience with medical imaging data and opportunities to train large models, participation and interest have continued to grow each year.

Professor Lee stated that he plans to continue holding competitions in which undergraduate students and researchers participate together, centered on challenging research topics across diverse fields, thereby sustaining a virtuous cycle of talent and technology.

Source: https://ece.snu.ac.kr/ece/news?md=v&bbsidx=58100