Assessment of U-Net in the segmentation of short tracts: Transferring to clinical MRI routine

IF 2.1 4区 医学 Q2 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING
Hohana Gabriela Konell , Luiz Otávio Murta Junior , Antônio Carlos dos Santos , Carlos Ernesto Garrido Salmon
{"title":"Assessment of U-Net in the segmentation of short tracts: Transferring to clinical MRI routine","authors":"Hohana Gabriela Konell ,&nbsp;Luiz Otávio Murta Junior ,&nbsp;Antônio Carlos dos Santos ,&nbsp;Carlos Ernesto Garrido Salmon","doi":"10.1016/j.mri.2024.05.009","DOIUrl":null,"url":null,"abstract":"<div><p>Accurately studying structural connectivity requires precise tract segmentation strategies. The U-Net network has been widely recognized for its exceptional capacity in image segmentation tasks and provides remarkable results in large tract segmentation when high-quality diffusion-weighted imaging (DWI) data are used. However, short tracts, which are associated with various neurological diseases, pose specific challenges, particularly when high-quality DWI data acquisition within clinical settings is concerned.</p><p>Here, we aimed to evaluate the U-Net network ability to segment short tracts by using DWI data acquired in different experimental conditions. To this end, we conducted three types of training experiments involving 350 healthy subjects and 11 white matter tracts, including the anterior, posterior, and hippocampal commissure, fornix, and uncinated fasciculus. In the first experiment, the model was exclusively trained with high-quality data of the Human Connectome Project (HCP) dataset. The second experiment focused on images of healthy subjects acquired from a local hospital dataset, representing a typical clinical routine acquisition. In the third experiment, a hybrid training approach was employed, combining data of the HCP and local hospital datasets. Then, the best model was also tested in unseen DWIs of 10 epilepsy patients of the local hospital and 10 healthy subjects acquired on a scanner from another company.</p><p>The outcomes of the third experiment demonstrated a notable enhancement in performance when contrasted with the preceding trials. Specifically, the short tracts within the local hospital dataset achieved Dice scores ranging between 0.60 and 0.65. Similar intervals were obtained with HCP data in the first experiment, and a substantial improvement compared to the scores between 0.37 and 0.50 obtained with the local hospital dataset at the same experiment. This improvement persisted when the method was applied to diverse scenarios, including different scanner acquisitions and epilepsy patients.</p><p>These results indicate that combining datasets from different sources, coupled with resolution standardization strengthens the neural network ability to generalize predictions across a spectrum of datasets. Nevertheless, short tract segmentation performance is intricately linked to the training composition, to validation, and to testing data. Moreover, curved tracts have intricate structural nature, which adds complexities to their segmenting. Although the network training approach tested herein has provided promising results, caution must be taken when extrapolating its application to datasets acquired under distinct experimental conditions, even in the case of higher-quality data or analysis of long or short tracts.</p></div>","PeriodicalId":18165,"journal":{"name":"Magnetic resonance imaging","volume":"111 ","pages":"Pages 217-228"},"PeriodicalIF":2.1000,"publicationDate":"2024-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Magnetic resonance imaging","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0730725X24001589","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}
引用次数: 0

Abstract

Accurately studying structural connectivity requires precise tract segmentation strategies. The U-Net network has been widely recognized for its exceptional capacity in image segmentation tasks and provides remarkable results in large tract segmentation when high-quality diffusion-weighted imaging (DWI) data are used. However, short tracts, which are associated with various neurological diseases, pose specific challenges, particularly when high-quality DWI data acquisition within clinical settings is concerned.

Here, we aimed to evaluate the U-Net network ability to segment short tracts by using DWI data acquired in different experimental conditions. To this end, we conducted three types of training experiments involving 350 healthy subjects and 11 white matter tracts, including the anterior, posterior, and hippocampal commissure, fornix, and uncinated fasciculus. In the first experiment, the model was exclusively trained with high-quality data of the Human Connectome Project (HCP) dataset. The second experiment focused on images of healthy subjects acquired from a local hospital dataset, representing a typical clinical routine acquisition. In the third experiment, a hybrid training approach was employed, combining data of the HCP and local hospital datasets. Then, the best model was also tested in unseen DWIs of 10 epilepsy patients of the local hospital and 10 healthy subjects acquired on a scanner from another company.

The outcomes of the third experiment demonstrated a notable enhancement in performance when contrasted with the preceding trials. Specifically, the short tracts within the local hospital dataset achieved Dice scores ranging between 0.60 and 0.65. Similar intervals were obtained with HCP data in the first experiment, and a substantial improvement compared to the scores between 0.37 and 0.50 obtained with the local hospital dataset at the same experiment. This improvement persisted when the method was applied to diverse scenarios, including different scanner acquisitions and epilepsy patients.

These results indicate that combining datasets from different sources, coupled with resolution standardization strengthens the neural network ability to generalize predictions across a spectrum of datasets. Nevertheless, short tract segmentation performance is intricately linked to the training composition, to validation, and to testing data. Moreover, curved tracts have intricate structural nature, which adds complexities to their segmenting. Although the network training approach tested herein has provided promising results, caution must be taken when extrapolating its application to datasets acquired under distinct experimental conditions, even in the case of higher-quality data or analysis of long or short tracts.

评估 U-Net 在短束分割中的作用:应用于临床常规磁共振成像
准确研究结构连接性需要精确的束分割策略。U-Net 网络在图像分割任务中的卓越能力已得到广泛认可,在使用高质量扩散加权成像(DWI)数据进行大束带分割时效果显著。然而,与各种神经系统疾病相关的短束带来了特殊的挑战,尤其是在临床环境中获取高质量的 DWI 数据时。在此,我们旨在利用在不同实验条件下获取的 DWI 数据,评估 U-Net 网络分割短束的能力。为此,我们进行了三种类型的训练实验,涉及 350 名健康受试者和 11 条白质束,包括前部、后部和海马、穹窿和无脊束。在第一项实验中,该模型完全使用人类连接组计划(HCP)数据集的高质量数据进行训练。第二个实验的重点是从当地医院数据集中获取的健康受试者图像,代表了典型的临床常规采集。在第三个实验中,采用了混合训练方法,将 HCP 数据集和本地医院数据集的数据结合起来。然后,最佳模型还在本地医院 10 名癫痫患者和另一家公司扫描仪采集的 10 名健康受试者的未见 DWI 中进行了测试。与之前的试验相比,第三次试验的结果表明性能明显提高。具体来说,当地医院数据集中的短路道获得了 0.60 到 0.65 之间的 Dice 分数。在第一次实验中,使用 HCP 数据也获得了类似的区间,与使用本地医院数据集获得的 0.37 到 0.50 之间的分数相比,有了大幅提高。在将该方法应用于不同场景(包括不同的扫描仪采集和癫痫患者)时,这种改进依然存在。这些结果表明,将来自不同来源的数据集与分辨率标准化相结合,可以增强神经网络对各种数据集进行泛化预测的能力。然而,短束分割性能与训练成分、验证和测试数据密切相关。此外,弯曲道具有复杂的结构性质,这也增加了其分割的复杂性。虽然本文测试的网络训练方法取得了令人鼓舞的结果,但在将其应用于在不同实验条件下获得的数据集时,必须谨慎行事,即使是在更高质量的数据或长、短束分析的情况下也是如此。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Magnetic resonance imaging
Magnetic resonance imaging 医学-核医学
CiteScore
4.70
自引率
4.00%
发文量
194
审稿时长
83 days
期刊介绍: Magnetic Resonance Imaging (MRI) is the first international multidisciplinary journal encompassing physical, life, and clinical science investigations as they relate to the development and use of magnetic resonance imaging. MRI is dedicated to both basic research, technological innovation and applications, providing a single forum for communication among radiologists, physicists, chemists, biochemists, biologists, engineers, internists, pathologists, physiologists, computer scientists, and mathematicians.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信