用语言模型思路做脑白质纤维束分割，更准更快。
TrackletGPT: A Language-like GPT Framework for White Matter Tract Segmentation
- 将纤维束拆成小段作为'词'，用GPT框架建模序列关系
- 在多个数据集上平均分割精度提升，跨数据集也稳定有效
- 适合神经影像分析、脑连接研究和手术规划人员
白质纤维束分割对研究脑结构连接、神经疾病及神经外科至关重要。该任务复杂，因纤维束在个体间、条件间差异大，但跨半球和个体具有相似三维结构。为此，我们提出TrackletGPT，一种类语言的GPT框架，通过引入轨迹段（tracklets）作为令牌来重建序列信息。TrackletGPT可无缝跨数据集泛化，完全自动化，能编码细粒度的子纤维段，实现对追踪分割中GPT模型的扩展与优化。实验表明，其在TractoInferno和HCP数据集上的平均DICE、重叠率和过度延伸评分均优于当前最优方法，即使在跨数据集测试中表现依然优异。
原文摘要 · Abstract (English)
White Matter Tract Segmentation is imperative for studying brain structural connectivity, neurological disorders and neurosurgery. This task remains complex, as tracts differ among themselves, across subjects and conditions, yet have similar 3D structure across hemispheres and subjects. To address these challenges, we propose TrackletGPT, a language-like GPT framework which reintroduces sequential information in tokens using tracklets. TrackletGPT generalises seamlessly across datasets, is fully automatic, and encodes granular sub-streamline segments, Tracklets, scaling and refining GPT models in Tractography Segmentation. Based on our experiments, TrackletGPT outperforms state-of-the-art methods on average DICE, Overlap and Overreach scores on TractoInferno and HCP datasets, even on inter-dataset experiments.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。