LightMedSeg-ISLES:参数比nnU-Net少81倍的卒中病灶分割
LightMedSeg-ISLES: Stroke Lesion Segmentation with 81x Fewer Parameters than nnU-Net
查看机构详情
- University of California, Berkeley(加州大学伯克利分校)
- University of California, San Francisco(加州大学旧金山分校)
机构由 AI 辅助整理,请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
提出LightMedSeg-ISLES,一个126万参数的轻量级流水线,用于ISLES'26卒中病灶分割,以81.4倍更少参数保留nnU-Net 97.5%的Dice,并提升病灶级F1,实现高效部署。
中文摘要 AI 辅助
大型网络和集成模型常能引领医学图像分割挑战,但其存储和推理需求使部署复杂化。我们提出LightMedSeg-ISLES,一个用于ISLES'26 T1加权卒中病灶分割的126万参数流水线。在146例保留测试队列上,翻转测试时增强获得0.618的平均Dice和0.599的病灶级F1。一个1.0235亿参数的nnU-Net ResEnc-L在尺寸过滤后产生0.634的Dice和0.544的病灶级F1。因此,LightMedSeg保留了nnU-Net Dice的97.5%,参数减少81.4倍,同时病灶级F1提高0.055。其四遍TTA操作点每个标准化patch的FLOPs比nnU-Net少4.7倍。它还略微超过过滤后的UNETR++和nnFormer。更长的训练和更强的增强在不增加容量的情况下额外增加0.0358的Dice,为更大的模型建立了一个强大的单检查点替代方案。
英文摘要
Large networks and ensembles often lead medical image segmentation challenges, but their storage and inference demands complicate deployment. We present LightMedSeg-ISLES, a 1.26-million-parameter pipeline for T1-weighted stroke lesion segmentation in ISLES'26. On a 146-case held-out cohort, flip test-time augmentation produces 0.618 mean Dice and 0.599 lesion-wise F1. A 102.35-million-parameter nnU-Net ResEnc-L produces 0.634 Dice and 0.544 lesion-wise F1 after size filtering. LightMedSeg therefore retains 97.5\% of nnU-Net's Dice with 81.4$\times$ fewer parameters while improving lesion-wise F1 by 0.055. Its four-pass TTA operating point requires 4.7$\times$ fewer FLOPs per standardized patch than nnU-Net. It also slightly exceeds filtered UNETR++ and nnFormer. Longer training and stronger augmentation add 0.0358 Dice without increasing capacity, establishing a strong single-checkpoint alternative to much larger models.