With the announcement of the winners of the MICCAI 2025 Challenges by the Medical Image Computing and Computer Assisted Intervention (MICCAI) Society, a team of PhD students from the University of Nottingham Ningbo China (UNNC), led by Professor Weng Ying, has once again claimed the global championships in the Robust 3D Brain MRI Inpainting with Random Masking Augmentation category. This marks the team’s third consecutive title in this prestigious competition. The team also achieved the global runner-up position in the Patch-Level Glioblastoma Subregion Classification with a Contrastive Learning-Based Encoder challenge.
As one of the most authoritative international competitions in the field of medical imaging, this year's MICCAI Challenge drew fierce competition, attracting more than 800 teams from leading institutions worldwide, including Harvard University, University College London, and the KTH Royal Institute of Technology, among others.
The team's breakthrough stemmed from an innovative approach that prioritised data augmentation over triditional algorithmic optimisation. "We trained our AI model much like teaching someone to recognise different cats - until it could identify even blurry or abstract versions with accuracy," explained Qifen Zhong, a PhD student at UNNC and a member of the team.
Rong Huang, a medical expert from the Ningbo Clinical Pathology Diagnosis Centre, anticipates that this AI-powered technology will soon revolutionise brain tumour diagnosis, shifting from manual interpretation to efficient AI-assisted analysis capable of detecting subtle details invisible to the human eye.
Remarkably, this hat-trick of global victories has been achieved by a university without a medical school.
"UNNC provides a truly resourceful and international academic platform," said Juexin Zhang, a PhD student at UNNC and another member of the team. "Although our backgrounds are in computer science, collaboration with experts from institutions such as the Queen's Medical Centre in the UK helped us understand clinical needs and avoid technical dead ends. This cross-disciplinary exchange empowered us to innovate and excel in global competitions."
The team is now establishing partnerships with medical institutions worldwide to translate their research achievements into industrial and clinical applications.
Published on 16 October 2025