Fast density-peaks clustering for registration-free pediatric white matter tract analysis
作者：
Highlights：
• We devise a fast density peak algorithm for data clustering that is able to automatically parcellate major WM tracts in a more efficient and accurate way than the existing clustering methods.
• The proposed clustering paves the possibility for tract-wise analysis on pediatric development in the native space with minimal human interactions. Compared with the popular atlas-based analysis derived from TBSS, our approach is free of additional skeletonization and registration steps that may introduce extra errors.
• We validate the effectiveness of both our clustering method and tract-wise analysis in two sections of experiments.
摘要
•We devise a fast density peak algorithm for data clustering that is able to automatically parcellate major WM tracts in a more efficient and accurate way than the existing clustering methods.•The proposed clustering paves the possibility for tract-wise analysis on pediatric development in the native space with minimal human interactions. Compared with the popular atlas-based analysis derived from TBSS, our approach is free of additional skeletonization and registration steps that may introduce extra errors.•We validate the effectiveness of both our clustering method and tract-wise analysis in two sections of experiments.
论文关键词：Fast clustering,White matter tracts,DTI,Pediatric development
论文评审过程：Received 30 September 2018, Revised 27 February 2019, Accepted 1 March 2019, Available online 2 March 2019, Version of Record 16 March 2019.
论文官网地址：https://doi.org/10.1016/j.artmed.2019.03.002
