Automated Dentate Nucleus Segmentation from QSM Images Using Deep Learning

Author:

Shiraishi Diogo HORCID,Saha Susmita,Adanyeguh Isaac M,Cocozza SirioORCID,Corben Louise A,Deistung Andreas,Delatycki Martin B,Dogan Imis,Gaetz William,Georgiou-Karistianis Nellie,Graf Simon,Grisoli Marina,Henry Pierre-Gilles,Jarola Gustavo M,Joers James M,Langkammer ChristianORCID,Lenglet ChristopheORCID,Li Jiakun,Lobo Camila C,Lock Eric F,Lynch David R,Mareci Thomas H,Martinez Alberto R M,Monti Serena,Nigri Anna,Pandolfo Massimo,Reetz KathrinORCID,Roberts Timothy P,Romanzetti Sandro,Rudko David A,Scaravilli Alessandra,Schulz Jörg BORCID,Subramony S H,Timmann Dagmar,França Marcondes C,Harding Ian H,Rezende Thiago J R,

Abstract

AbstractPurposeTo develop a dentate nucleus (DN) segmentation tool using deep learning (DL) applied to brain quantitative susceptibility mapping (QSM) images.Materials and MethodsBrain QSM images from 132 healthy controls and 170 individuals with cerebellar ataxia or multiple sclerosis were collected from nine different datasets worldwide for this retrospective study. Manual delineation of the DN (gray matter and white matter hilus) was first undertaken by experienced raters with a robust quality control process. Performance of automated segmentation was compared following training using several DL architectures. A two-step approach was implemented, composed of a localization model followed by DN segmentation.ResultsThe manual tracing protocol produced ground-truth data with high intra-rater (average ICC 0.906) and inter-rater reliability (average ICC 0.776). Initial DL architecture exploration indicated that the nnU-Net framework performed best. The two-step localization plus segmentation pipeline achieved a Dice score of 0.898±0.031 and 0.894±0.036 for left and right DN, respectively. In external validation, our algorithm outperformed the leading existing automated tool (left/right DN Dice 0.863±0.038/0.843±0.066 vs. 0.568±0.222/0.582±0.239). The model demonstrated generalizability across unseen datasets during the training step. The measures showed a superior correlation index with manual annotations and performed well in both isotropic and anisotropic QSM datasets.ConclusionWe provide a model that accurately and efficiently segments the DN from brain QSM images. The model can be readily deployed for use in observational, natural history, and treatment trials for biomarker discovery.

Publisher

Cold Spring Harbor Laboratory

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3