Automated segmentation of deep brain nuclei using convolutional neural networks and susceptibility weighted imaging,Human Brain Mapping

当前位置： X-MOL 学术 › Hum. Brain Mapp. › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

Automated segmentation of deep brain nuclei using convolutional neural networks and susceptibility weighted imaging
Human Brain Mapping ( IF 3.5 ) Pub Date : 2021-07-29 , DOI: 10.1002/hbm.25604
Vincent Beliveau _{1,

2} , Martin Nørgaard _{3,

4} , Christoph Birkl ₅ , Klaus Seppi _{1,

2} , Christoph Scherfler _{1,

2}

Affiliation

The advent of susceptibility-sensitive MRI techniques, such as susceptibility weighted imaging (SWI), has enabled accurate in vivo visualization and quantification of iron deposition within the human brain. Although previous approaches have been introduced to segment iron-rich brain regions, such as the substantia nigra, subthalamic nucleus, red nucleus, and dentate nucleus, these methods are largely unavailable and manual annotation remains the most used approach to label these regions. Furthermore, given their recent success in outperforming other segmentation approaches, convolutional neural networks (CNN) promise better performances. The aim of this study was thus to evaluate state-of-the-art CNN architectures for the labeling of deep brain nuclei from SW images. We implemented five CNN architectures and considered ensembles of these models. Furthermore, a multi-atlas segmentation model was included to provide a comparison not based on CNN. We evaluated two prediction strategies: individual prediction, where a model is trained independently for each region, and combined prediction, which simultaneously predicts multiple closely located regions. In the training dataset, all models performed with high accuracy with Dice coefficients ranging from 0.80 to 0.95. The regional SWI intensities and volumes from the models' labels were strongly correlated with those obtained from manual labels. Performances were reduced on the external dataset, but were higher or comparable to the intrarater reliability and most models achieved significantly better results compared to multi-atlas segmentation. CNNs can accurately capture the individual variability of deep brain nuclei and represent a highly useful tool for their segmentation from SW images.

中文翻译：

使用卷积神经网络和磁化率加权成像自动分割深部脑核

磁化率敏感的 MRI 技术（例如磁化率加权成像 (SWI)）的出现，使得人脑内铁沉积的体内准确可视化和定量成为可能。尽管以前的方法已经被引入来分割富含铁的大脑区域，例如黑质、丘脑底核、红核和齿状核，但这些方法在很大程度上不可用，并且手动注释仍然是标记这些区域最常用的方法。此外，鉴于卷积神经网络 (CNN) 最近在优于其他分割方法方面取得的成功，它有望提供更好的性能。因此，本研究的目的是评估用于从 SW 图像中标记深部脑核的最先进的 CNN 架构。我们实现了五种 CNN 架构并考虑了这些模型的集成。此外，还包含多图集分割模型，以提供不基于 CNN 的比较。我们评估了两种预测策略：单独预测（针对每个区域独立训练模型）和组合预测（同时预测多个位置接近的区域）。在训练数据集中，所有模型均以高精度执行，Dice 系数范围为 0.80 至 0.95。模型标签中的区域 SWI 强度和体积与手动标签中获得的强度和体积密切相关。外部数据集的性能有所下降，但与评估者内部的可靠性相比更高或相当，并且与多图集分割相比，大多数模型取得了明显更好的结果。 CNN 可以准确捕获深部脑核的个体变异性，并且是从 SW 图像中进行分割的非常有用的工具。

更新日期：2021-09-19

点击分享查看原文

点击收藏

公开下载

阅读更多本刊最新论文本刊介绍/投稿指南11