现代仪器与医疗

基于3D体素残差网络的阿尔兹海默症患者脑部MRI图像多模态分类研究

PDF下载

陆安 ¹ , 钟慧芳 ²

现代仪器与医疗 | 临床研究 2025,31(2): 55-60

收起

现代仪器与医疗 |临床研究 2025 , 31 (2) : 55 -60

基于3D体素残差网络的阿尔兹海默症患者脑部MRI图像多模态分类研究

全屏

[Author(id=1320388912867037867, tenantId=1146029695717560320, journalId=1319373289681944729, articleId=1320388912028177057, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1320388912959312557, tenantId=1146029695717560320, journalId=1319373289681944729, articleId=1320388912028177057, authorId=1320388912867037867, language=EN, stringName=An Lu, firstName=An, middleName=null, lastName=Lu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=¹, address=^1.Zhongshan Hospital of Traditional Chinese Medicine, Zhongshan 528400, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1320388913043198638, tenantId=1146029695717560320, journalId=1319373289681944729, articleId=1320388912028177057, authorId=1320388912867037867, language=CN, stringName=陆安, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=¹, address=^1.中山市中医院，中山　528400, bio={"content":"

陆安，男，本科，工程师，研究方向：医疗设备管理和医学图像处理

"}, bioImg=null, bioContent=

陆安，男，本科，工程师，研究方向：医疗设备管理和医学图像处理

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1320388912669905572, tenantId=1146029695717560320, journalId=1319373289681944729, articleId=1320388912028177057, xref=1., ext=[AuthorCompanyExt(id=1320388912682488485, tenantId=1146029695717560320, journalId=1319373289681944729, articleId=1320388912028177057, companyId=1320388912669905572, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=^1.Zhongshan Hospital of Traditional Chinese Medicine, Zhongshan 528400, China), AuthorCompanyExt(id=1320388912695071398, tenantId=1146029695717560320, journalId=1319373289681944729, articleId=1320388912028177057, companyId=1320388912669905572, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=^1.中山市中医院，中山　528400)])]), Author(id=1320388914712531632, tenantId=1146029695717560320, journalId=1319373289681944729, articleId=1320388912028177057, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1320388914817389234, tenantId=1146029695717560320, journalId=1319373289681944729, articleId=1320388912028177057, authorId=1320388914712531632, language=EN, stringName=Huifang Zhong, firstName=Huifang, middleName=null, lastName=Zhong, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=², address=^2.College of Biomedical Engineering, Guangdong Medical University, Dongguan 523000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1320388914909663923, tenantId=1146029695717560320, journalId=1319373289681944729, articleId=1320388912028177057, authorId=1320388914712531632, language=CN, stringName=钟慧芳, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=², address=^2.广东医科大学生物医学工程学院，东莞　523000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1320388912757985959, tenantId=1146029695717560320, journalId=1319373289681944729, articleId=1320388912028177057, xref=2., ext=[AuthorCompanyExt(id=1320388912766374568, tenantId=1146029695717560320, journalId=1319373289681944729, articleId=1320388912028177057, companyId=1320388912757985959, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=^2.College of Biomedical Engineering, Guangdong Medical University, Dongguan 523000, China), AuthorCompanyExt(id=1320388912778957481, tenantId=1146029695717560320, journalId=1319373289681944729, articleId=1320388912028177057, companyId=1320388912757985959, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=^2.广东医科大学生物医学工程学院，东莞　523000)])])]

陆安¹, 钟慧芳²

作者信息

^1.中山市中医院，中山　528400

^2.广东医科大学生物医学工程学院，东莞　523000

作者简介:

陆安，男，本科，工程师，研究方向：医疗设备管理和医学图像处理

Multimodal Classification of Brain MRI Images in Alzheimer's Disease Patients Using 3D VoxResNet

An Lu¹, Huifang Zhong²

Affiliations

^1.Zhongshan Hospital of Traditional Chinese Medicine, Zhongshan 528400, China

^2.College of Biomedical Engineering, Guangdong Medical University, Dongguan 523000, China

出版时间: 2025-04-10 doi: 10.11876/mimt202502012

文章导航

摘要

收起

目的

探讨利用深度学习的识别分类方法来解决传统阿尔兹海默症检测与评估手段所存在的繁琐、耗时等问题，以提高诊断和预测的速度与准确度。

方法

从ADNI获取核磁共振成像图像数据。预处理阶段，利用MIPAV医学图像处理软件和3D Slicer对MRI图像进行AC-PC校准、空间标准化、颅骨剥离、偏置校正。预处理后的数据转化为5D张量，并输入至Pytorch环境中的体素残差网络（VoxResNet）模型进行训练。训练过程中，采用L2正则化、带动量的SGD优化器和交叉熵损失函数对模型进行优化。模型性能通过分析训练集损失、验证集损失和验证集准确率的变化曲线进行评估。

结果

二分类任务中，模型表现出色，验证集准确率90.18%，ROC-AUC值为0.882，表明其对阿尔茨海默症患者和正常认知者的区分能力强。三分类任务中，模型验证集准确率较低，出现欠拟合现象。

结论

基于3D VoxResNet的脑部MRI识别方法，可为AD早期诊断和治疗提供一定的参考价值。

关键词

阿尔兹海默症 / 深度学习 / 核磁共振成像 / VoxResNet模型

Abstract

收起

Objective

To explore deep learning-based recognition and classification methods to address the cumbersome and time-consuming issues of traditional Alzheimer's disease (AD) detection and assessment, thereby improving diagnostic and predictive speed and accuracy.

Methods

MRI data were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI). During preprocessing, MIPAV medical image processing software and 3D Slicer were used for AC-PC alignment, spatial normalization, skull stripping, and bias correction. The preprocessed data were converted into 5D tensors and input into a voxel residual network (VoxResNet) model in PyTorch for training. The training process employed L2 regularization, stochastic gradient descent (SGD) with momentum, and cross-entropy loss for optimization. Model performance was evaluated by analyzing the training loss, validation loss, and validation accuracy curves.

Results

In binary classification tasks, the model performed exceptionally well, achieving a validation accuracy of 90.18% and an ROC-AUC value of 0.882, demonstrating strong discriminative ability between AD patients and cognitively normal individuals. However, in ternary classification tasks, the model exhibited lower validation accuracy, indicating underfitting.

Conclusion

The 3D VoxResNet-based MRI recognition method can provide valuable insights for early AD diagnosis and treatment.

Key words

Alzheimer's disease / Deep learning / Magnetic resonance imaging / VoxResNet model

引用本文

陆安, 钟慧芳. 基于3D体素残差网络的阿尔兹海默症患者脑部MRI图像多模态分类研究. 现代仪器与医疗, 2025 , 31 (2) : 55 -60 . DOI: 10.11876/mimt202502012

An Lu, Huifang Zhong. Multimodal Classification of Brain MRI Images in Alzheimer's Disease Patients Using 3D VoxResNet[J]. Modern Instruments & Medical Treatment, 2025 , 31 (2) : 55 -60 . DOI: 10.11876/mimt202502012

基金

收起

中山市社会公益与基础研究项目(2024B1062)
广东省医学会医学工程学分会青年委员会面上项目课题(2025-GDMAMS-02)

参考文献

收起

文献

收起

参考文献引证文献

排序方式：

[1]

Fargo K. Alzheimer's Association Report: 2023 Alzheimer's disease facts and figures[J]. Alzheimer's & Dementia, 2023, 19: 1598-1695.

[2]

王宗敏, 福林, 高云玥. 基于深度学习的多模态医学影像研究进展[J]. 分子影像学杂志, 2022, 45(3): 459-464.

[3]

俞益洲, 马杰超, 石德君, 等. 深度学习在医学影像分析中的应用综述[J]. 数据与计算发展前沿, 2019, 1(6): 37-52.

[4]

Pan D, Zeng A, Jia L, et al. Early detection of Alzheimer's disease using magnetic resonance imaging: a novel approach combining convolutional neural networks and ensemble learning[J]. Front Neurosci, 2020, 14: 259.

[5]

Basheera S, Ram MSS. Convolution neural network–based Alzheimer's disease classification using hybrid enhanced independent component analysis based segmented gray matter of T2 weighted magnetic resonance imaging with clinical valuation[J]. Alzheimers Dement, 2019, 5: 974-986.

[6]

Abrol A, Bhattarai M, Fedorov A, et al. Deep residual learning for neuroimaging: an application to predict progression to Alzheimer's disease[J]. J Neurosci Meth, 2020, 339: 108701.

[7]

Basaia S, Agosta F, Wagner L, et al. Automated classification of Alzheimer's disease and mild cognitive impairment using a single MRI and deep neural networks[J]. NeuroImage: Clinical, 2019, 21: 101645.

[8]

Chen H, Dou Q, Yu L, et al. VoxResNet: Deep voxelwise residual networks for brain segmentation from 3D MR images[J]. NeuroImage, 2018, 170: 446-455.

[9]

闫志强, 董秋峰, 陈丹, 等. 个体化AC-PC线体表投影在脑立体定向手术中的应用[J]. 神经解剖学杂志, 2022, 38(3): 334-338.

[10]

李基臣, 亓玉龙, 胡海瑞, 等. 数字图像处理技术在医学影像中的研究与应用[J]. 电子技术与软件工程, 2022, 29(9): 194-197.

[11]

苏奎, 董默. Matlab在医学图像分割中的应用[J]. 中华消化病与影像杂志(电子版), 2022, 12(2): 108-112.

[12]

王大溪, 陈鑫鑫. 基于形态学的脑部MRI图像颅骨剥离算法[J]. 计算机技术与发展, 2015, 25(12): 206-209+215.

[13]

朱毓华, 张慧玮, 吴平, 等. 空间标准化对帕金森病脑代谢网络应用的影响[J]. 中国医学计算机成像杂志, 2015, 21(5): 487-491.

[14]

姜虎成, 林科. 基于MICO+FCM的MRI脑组织分割[J]. 计算机仿真, 2020, 37(3): 238-242.

[15]

王磊. 基于自组织神经网络的多模态MRI图像分割[J]. 太赫兹科学与电子信息学报, 2018, 16(2): 307-311+322.

[16]

黄臣程, 李兆飞. 基于局部偏差的图像分割与偏置场校正模型[J]. 四川理工学院学报(自然科学版), 2017, 30(5): 44-50.

[17]

陈伟, 孙康康, 李奇轩, 等. 多模态医学图像融合方法的研究进展[J]. 中国辐射卫生, 2023, 32(5): 580-585.

[18]

曹玉红, 徐海, 刘荪傲, 等. 基于深度学习的医学影像分割研究综述[J]. 计算机应用, 2021, 41(8): 2273-2287.

[19]

史豪斌, 王少熙, 黄睿茜. 融合注意力机制与神经网络的三维点云分类算法[J]. 微电子学与计算机, 2023, 40(5): 12-19.

[20]

金燕, 薛智中, 姜智伟. 基于循环残差卷积神经网络的医学图像分割算法[J]. 计算机辅助设计与图形学学报, 2022, 34(8): 1205-1215.

[21]

李炯逸, 李彬, 邱前辉, 等. 基于MRI与优化3D-ResNet18的鼻咽癌复发预测模型[J]. 中国生物医学工程学报, 2023, 42(5): 583-593.

[22]

Zhang Y, Wang S, Ji G, et al. An MR brain images classifier via principal component analysis and kernel support vector machine[J]. Sci World J, 2012, 2(3): 369-388.

[23]

Olfa BA, Jenny BP, Michelle A, et al. Recognition of Alzheimer's disease and Mild Cognitive Impairment with multimodal image-derived biomarkers and Multiple Kernel Learning[J]. Neurocomputing, 2017, 220: 98-110.

[24]

Litjens G, Kooi T, Bejnordi BE, et al. A survey on deep learning in medical image analysis[J]. Med Image Anal, 2017, 42: 60-88.

[25]

董广伟, 张亦荣, 徐嘉弈, 等. 基于基因信息挖掘的阿尔茨海默症早期诊断[J]. 枣庄学院学报, 2023, 40(2): 43-54.

[26]

韩坤, 潘海为, 张伟, 等. 基于多模态医学图像的Alzheimer病分类方法[J]. 清华大学学报(自然科学版), 2020, 60(8): 664-671+682.

[27]

林冰洁, 王梅云. 深度学习在医学影像学中的研究现状及发展前景[J]. 山东大学学报(医学版), 2023, 61(12): 21-29.

[28]

许晶晶, 张敏鸣. 人工智能机器学习方法在阿尔茨海默病中的应用现状[J]. 诊断学理论与实践, 2018, 17(4): 466-470.

2025年第31卷第2期

PDF下载

引用本文

BibTeX

文章信息

doi: 10.11876/mimt202502012

首发时间：2026-10-23
出版时间：2025-04-10

补充材料

相关文章

文章信息

作者

出版历史

基金

中山市社会公益与基础研究项目(2024B1062)

广东省医学会医学工程学分会青年委员会面上项目课题(2025-GDMAMS-02)

作者信息

^1.中山市中医院，中山　528400

^2.广东医科大学生物医学工程学院，东莞　523000

参考文献

分享链接

https://castjournals.cast.org.cn/joweb/xdyqyyl/CN/10.11876/mimt202502012

分享至

全文二维码

扫描看全文

引用本文

BibTeX

本文的引用情况

2种不同金属材料的力学参数

科 Family	属数 Number of genus	种数 Number of species	占总种数比例 Percentage of total species (%)	属 Genus	种数 Number of species	占总种数比例 Percentage of total species (%)
鹅膏菌科Amanitaceae	2	11	5.26	鹅膏菌属 Amanita	10	4.78
小菇科 Mycenaceae	2	12	5.74	丝盖伞属 Inocybe	5	2.39
多孔菌科 Polyporaceae	8	14	6.70	蜡蘑属 Laccaria	5	2.39
红菇科 Russulaceae	3	23	11.00	小皮伞属 Marasmius	6	2.87
				小菇属 Mycena	11	5.26
				光柄菇属 Pluteus	5	2.39
				红菇属 Russula	17	8.13
				栓菌属 Trametes	5	2.39

关闭全屏

BibTeX
EndNote
RefWorks
TxT