Scholarlink

<p>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.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

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

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年8月14日

查看全文
  • 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

摘要

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.