Revisiting model scaling with a U-net benchmark for 3D medical image segmentation

作者：Ziyan Huang, Jin Ye, Haoyu Wang, Zhongying Deng, Zhikai Yang, Yanzhou Su, Jie Liu, Tianbin Li, Yun Gu, Shaoting Zhang, Yu Qiao, Lixu Gu, Junjun He · 发表于：Scientific Reports · 年份：2025 · DOI：10.1038/s41598-025-15617-1 · 被引用次数：7 · 研究领域：Radiomics and Machine Learning in Medical Imaging、Advanced Neural Network Applications、Medical Imaging and Analysis

Are larger models always better for 3D medical image segmentation? Despite the widespread adoption of 3D U-Net in various medical imaging tasks, this critical question remains underexplored. To challenge the common assumption, we systematically benchmark 18 U-Net variants-adjusting resolution stages, depth, and width-across 42 diverse public datasets. Our findings reveal that the answer is no: optimal architectures are highly task-specific, with smaller models often performing competitively. Specifically, we identify three key insights: (1) increasing resolution stages provides limited benefits for datasets with larger voxel spacing; (2) deeper networks offer limited advantages for anatomically complex shapes; and (3) wider networks provide minimal advantages for tasks with limited segmentation classes. Based on these insights, we provide practical guidelines for optimizing U-Net architectures according to dataset characteristics. Our findings highlight the limitations of the"bigger is better"paradigm while establishing a framework for balancing performance and computational efficiency in 3D medical image segmentation tasks.