koko体育app

欢迎来到《koko体育app 学报(医学版)》
田恒屹, 王瑜, 计亚荣, 等. 3DGE-UNet磁共振成像的脑胶质瘤全自动分割算法:对不充分全局特征提取的改进[J]. koko体育app 学报(医学版), 2024, 55(2): 447-454. DOI:
引用本文: 田恒屹, 王瑜, 计亚荣, 等. 3DGE-UNet磁共振成像的脑胶质瘤全自动分割算法:对不充分全局特征提取的改进[J]. koko体育app 学报(医学版), 2024, 55(2): 447-454. DOI:
TIAN Hengyi, WANG Yu, JI Yarong, et al. Fully Automatic Glioma Segmentation Algorithm of Magnetic Resonance Imaging Based on 3D-UNet With More Global Contextual Feature Extraction: An Improvement on Insufficient Extraction of Global Features[J]. Journal of Sichuan University (Medical Sciences), 2024, 55(2): 447-454. DOI:
Citation: ᩚᩚᩚᩚᩚᩚ⁤⁤⁤⁤ᩚ⁤⁤⁤⁤ᩚ⁤⁤⁤⁤ᩚ𒀱ᩚᩚᩚ TIAN Hengyi, WANG Yu, JI Yarong, et al. Fully Automatic Glioma Segmentation Algorithm of Magnetic Resonance Imaging Based on 3D-UNet With More Global Contextual Feature Extraction: An Improvement on Insufficient Extraction of Global Features[J]. Journal of Sichuan University (Medical Sciences), 2024, 55(2): 447-454. DOI:

3DGE-UNet磁共振成像的脑胶质瘤全自动分割算法:对不充分全局特征提取的改进

Fully Automatic Glioma Segmentation Algorithm of Magnetic Resonance Imaging Based on 3D-UNet With More Global Contextual Feature Extraction: An Improvement on Insufficient Extraction of Global Features

  • 摘要:
    目的 脑胶质瘤及其子区域的全自动分割是计算机辅助肿瘤临床诊断的基础,卷积神经网络在脑部磁共振成像(magnetic resonance imaging, MRI)分割过程中,小卷积核只能提取局部特征而无法有效融合全局特征,缩小了对影像信息的感知范围,从而导致分割精度不足的问题。本研究旨在利用膨胀卷积,针对三维(three-dimensional, 3D)-UNet不能充分提取全局特征的问题进行改进。
    方法  ①算法构建:本文提出一种带有三条全局上下文特征提取通路的3D-UNet模型,即3DGE-UNet模型。使用脑部肿瘤分割挑战赛(Brain Tumor Segmentation Challenge, BraTS)2019提供的公开数据集(335例患者),设计一种全局上下文特征提取(more global contextual feature extraction, GE)模块,并将该模块引入到3D-UNet网络的第一、二、三条跳跃连接处,利用该模块充分提取图像的不同尺度全局特征,并与上采样得到的特征图叠加,扩大模型感受野,实现不同尺度特征的深层融合,从而完成端到端的脑肿瘤全自动分割。②算法验证:图像数据来源于BraTs2019数据集(335例患者术前MRI图像,包括T1、T1ce、T2和FLAIR四种模态图像和一张医生标注的肿瘤标签图像),将数据集分别按照8∶1∶1划分为训练集、验证集和测试集。以医生标注的肿瘤标签图像为金标准,使用Dice系数(综合评价有效性)、敏感度(病灶区域的查全率)和95%Hausdorff距离(肿瘤边界的分割精度)在测试集中评估算法对完整肿瘤区域(whole tumor, WT)、核心肿瘤区域(tumor core, TC)和增强肿瘤区域(enhancing tumor, ET)三种区域的分割性能,分别使用无GE模块的3D-UNet模型和有GE模块的3DGE-UNet模型进行检测,内部验证GE模块设置的有效性。使用3DGE-UNet模型、ResUNet、UNet++、nnUNet、UNETR检测上述性能指标,并比较5种算法模型的收敛性,外部验证3DGE-UNet模型的有效性。
    结果 ①在内部验证中,改进的3DGE-UNet模型在测试集中分割WT、TC和ET区域的Dice均值分别为91.47%、87.14%和83.35%,与传统的3D-UNet模型相比(89.79%、85.13%和80.90%),综合评价达到了最优值,对于三个区域的分割精度都有提升(P<0.05)。3DGE-UNet模型与3D-UNet模型相比,ET区(86.46% vs. 80.77%)具有更高的敏感度(P<0.05),在查全率方面表现更为优越,在面对病灶区域时,3DGE-UNet模型更倾向于正确识别并捕获更全面的阳性区域,更为有效地避免漏诊发生。3DGE-UNet模型在肿瘤WT区域边缘分割的出色性能,其95%Hausdorff距离均值优于3D-UNet模型(8.17 mm vs. 13.61 mm,P<0.05),但在TC区(8.73 mm vs. 7.47 mm)和ET区(6.21 mm vs. 5.45 mm)的性能表现与3DUNet模型相似。②在外部验证中,其余4种算法仅TC的Dice均值(87.25%)、WT的敏感度均值(94.59%)、TC的敏感度均值(86.98%)和ET的平均95%Hausdorff距离(5.37 mm)优于3DGE-UNet模型,但差异无统计学意义。3DGE-UNet模型可以在训练阶段快速收敛,速度优于其他外部模型。
    结论 3DGE-UNet模型可以有效地提取和融合不同尺度的特征信息,提高脑肿瘤分割的准确性。
     
    Abstract:
    Objective The fully automatic segmentation of glioma and its subregions is fundamental for computer-aided clinical diagnosis of tumors. In the segmentation process of brain magnetic resonance imaging (MRI), convolutional neural networks with small convolutional kernels can only capture local features and are ineffective at integrating global features, which narrows the receptive field and leads to insufficient segmentation accuracy. This study aims to use dilated convolution to address the problem of inadequate global feature extraction in 3D-UNet.
    Methods 1) Algorithm construction: A 3D-UNet model with three pathways for more global contextual feature extraction, or 3DGE-UNet, was proposed in the paper. By using publicly available datasets from the Brain Tumor Segmentation Challenge (BraTS) of 2019 (335 patient cases), a global contextual feature extraction (GE) module was designed. This module was integrated at the first, second, and third skip connections of the 3D UNet network. The module was utilized to fully extract global features at different scales from the images. The global features thus extracted were then overlaid with the upsampled feature maps to expand the model's receptive field and achieve deep fusion of features at different scales, thereby facilitating end-to-end automatic segmentation of brain tumors. 2) Algorithm validation: The image data were sourced from the BraTs 2019 dataset, which included the preoperative MRI images of 335 patients across four modalities (T1, T1ce, T2, and FLAIR) and a tumor image with annotations made by physicians. The dataset was divided into the training, the validation, and the testing sets at an 8∶1∶1 ratio. Physician-labelled tumor images were used as the gold standard. Then, the algorithm's segmentation performance on the whole tumor (WT), tumor core (TC), and enhancing tumor (ET) was evaluated in the test set using the Dice coefficient (for overall effectiveness evaluation), sensitivity (detection rate of lesion areas), and 95% Hausdorff distance (segmentation accuracy of tumor boundaries). The performance was tested using both the 3D-UNet model without the GE module and the 3DGE-UNet model with the GE module to internally validate the effectiveness of the GE module setup. Additionally, the performance indicators were evaluated using the 3DGE-UNet model, ResUNet, UNet++, nnUNet, and UNETR, and the convergence of these five algorithm models was compared to externally validate the effectiveness of the 3DGE-UNet model.
    Results 1) In internal validation, the enhanced 3DGE-UNet model achieved Dice mean values of 91.47%, 87.14%, and 83.35% for segmenting the WT, TC, and ET regions in the test set, respectively, producing the optimal values for comprehensive evaluation. These scores were superior to the corresponding scores of the traditional 3D-UNet model, which were 89.79%, 85.13%, and 80.90%, indicating a significant improvement in segmentation accuracy across all three regions (P<0.05). Compared with the 3D-UNet model, the 3DGE-UNet model demonstrated higher sensitivity for ET (86.46% vs. 80.77%) (P<0.05) , demonstrating better performance in the detection of all the lesion areas. When dealing with lesion areas, the 3DGE-UNet model tended to correctly identify and capture the positive areas in a more comprehensive way, thereby effectively reducing the likelihood of missed diagnoses. The 3DGE-UNet model also exhibited exceptional performance in segmenting the edges of WT, producing a mean 95% Hausdorff distance superior to that of the 3D-UNet model (8.17 mm vs. 13.61 mm, P<0.05). However, its performance for TC (8.73 mm vs. 7.47 mm) and ET (6.21 mm vs. 5.45 mm) was similar to that of the 3D-UNet model. 2) In the external validation, the other four algorithms outperformed the 3DGE-UNet model only in the mean Dice for TC (87.25%), the mean sensitivity for WT (94.59%), the mean sensitivity for TC (86.98%), and the mean 95% Hausdorff distance for ET (5.37 mm). Nonetheless, these differences were not statistically significant (P>0.05). The 3DGE-UNet model demonstrated rapid convergence during the training phase, outpacing the other external models.
    Conclusion The 3DGE-UNet model can effectively extract and fuse feature information on different scales, improving the accuracy of brain tumor segmentation.
     
© 2024 《koko体育app 学报(医学版)》编辑部 版权所有 cc

开放获取🤡 本文遵循知识共享署名—非商业性使用4.0国际许可协议(CC BY-NC 4.0),允许第三方对本刊发表的论文自由共享(即在任何媒介以任何形式复制、发行原文)、演绎(即修改、转换或以原文为基础进行创作),必须给出适当的署名,提供指向本文许可协议的链接,同时标明是否对原文作了修改;不得将本文用于商业目的。CC BY-NC 4.0许可协议详情请访问

/

返回文章
返回
var _hmt = _hmt || []; (function() { var hm = document.createElement("script"); hm.src = "https://hm.baidu.com/hm.js?90c4d9819bca8c9bf01e7898dd269864"; var s = document.getElementsByTagName("script")[0]; s.parentNode.insertBefore(hm, s); })(); !function(p){"use strict";!function(t){var s=window,e=document,i=p,c="".concat("https:"===e.location.protocol?"https://":"http://","sdk.51.la/js-sdk-pro.min.js"),n=e.createElement("script"),r=e.getElementsByTagName("script")[0];n.type="text/javascript",n.setAttribute("charset","UTF-8"),n.async=!0,n.src=c,n.id="LA_COLLECT",i.d=n;var o=function(){s.LA.ids.push(i)};s.LA?s.LA.ids&&o():(s.LA=p,s.LA.ids=[],o()),r.parentNode.insertBefore(n,r)}()}({id:"K9y7iMpaU8NS42Fm",ck:"K9y7iMpaU8NS42Fm"}); koko体育-koko体育app koko体育-koko体育网页版koko体育app koko体育-全站app下载(官网) m6米乐app|下载 m6米乐app|主頁欢迎您!!