This work introduces a novel unsupervised domain adaptation framework, DDSP, that addresses the instability issues in GAN-based medical image segmentation by making the model domain-agnostic and focusing on anatomical structure with semantic constraints, significantly outperforming existing methods on public datasets.

The miX-Lab's research findings in the field of pathological image generation have been accepted by the top conference on medical image analysis, MICCAI 2024.

The team proposed an unsupervised complex-valued network architecture based on physical information for efficient, high-quality color lens-free holographic reconstruction.

Congratulations Dr Tian for "Anatomy-Relation reasoning graph convolutional network for automatic fine-grained segmentation of organ's surgical anatomy" has been selected as cover article in IEEE Journal of Biomedical and Health Informatics.

Researchers from the Shenzhen Institute of Advanced Technology and their collaborators proposed a novel deep multi-magnification similarity learning (DSML) approach to improve performance in histopathological classification.

Congratulations to Xiaoqi Jia who has won a Scientific Prize as 29th (Top 3%) on the leaderboard in the “HuBMAP + HPA - Hacking the Human Body” Kaggle competition!

Congratulations to MixLab member Jiaxin Hou for winning the 2022 National Scholarship of UCAS!

The 1st Workshop on Computational Mathematics Modeling in Cancer Analysis (CMMCA2022) was held in conjunction with the 25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2022) on September 18th, 2022. The LNCS volume is available now.