1.首都医科大学附属北京积水潭医院医学工程部(北京 100035)
2.山东中医药大学青岛中医药科学院(山东青岛 266112)
3.首都医科大学生物医学工程学院(北京 100069)
陈振鹏,E-mail:m18366108084@163.com
李海云,E-mail:haiyunli@ccmu.edu.cn
收稿:2024-11-16,
修回:2025-02-03,
纸质出版:2026-04-20
移动端阅览
周伟斌,陈振鹏,李海云.应用多模态磁共振影像基于多层感知机的轻度认知障碍分类识别方法[J].北京生物医学工程,2026,45(2):119-126.
ZHOU Weibin,CHEN Zhenpeng,LI Haiyun.Classification and recognition of mild cognitive impairment using multimodal magnetic resonance imaging based on multi-layer perceptron[J].Beijing Biomedical Engineering,2026,45(2):119-126.
周伟斌,陈振鹏,李海云.应用多模态磁共振影像基于多层感知机的轻度认知障碍分类识别方法[J].北京生物医学工程,2026,45(2):119-126. DOI: 10.3969/j.issn.1002-3208.2026.02.002.
ZHOU Weibin,CHEN Zhenpeng,LI Haiyun.Classification and recognition of mild cognitive impairment using multimodal magnetic resonance imaging based on multi-layer perceptron[J].Beijing Biomedical Engineering,2026,45(2):119-126. DOI: 10.3969/j.issn.1002-3208.2026.02.002.
目的
2
轻度认知障碍(mild cognitive impairment
MCI)是阿尔茨海默病(Alzheimer’s disease
AD)的前驱阶段,MCI包括稳定型轻度认知障碍(stable mild cognitive impairment
sMCI)与进展型轻度认知障碍(progressive mild cognitive impairment
pMCI)。准确区分sMCI与pMCI对于AD临床早期诊断干预至关重要。
方法
2
本文提出了一种应用多模态多图谱基于多层感知机模型的MCI分类识别方法,用于sMCI与pMCI分类识别。应用FreeSurfer,从结构磁共振成像(structural magnetic resonance imaging
sMRI)影像数据中计算得到的形态学特征参数包括大脑皮层分割区域厚度均值、表面积、灰质体积、沟壑深度、褶皱指数以及指定解剖区域的体素强度均值。采用DPABI,从静息态功能磁共振成像(resting state functional MRI
rs-fMRI)影像数据中计算得到的功能特征参数包括低频振幅和局部一致性。采用岭回归进行特征参数筛选,并根据岭回归输出的系数矩阵确定最具区分度的脑区。
结果
2
建立的方法在ADNI数据集上进行了10折交叉验证。集成来自sMRI和rs-fMRI两种模态影像的特征参数,sMCI与pMCI的分类识别准确率为82.7%,AUC为93.0%。实验结果发现内嗅皮层、颞叶厚度和体积、海马、海马旁回的体素强度均值等形态学特征具有明显组间差异。
结论
2
构建的分类识别方法具有潜在的区分sMCI与pMCI的能力。融合结构磁共振和静息态功能磁共振影像模态特征集成多图谱能够显著地提高模型对MCI分类识别性能。
Objective
2
Mild cognitive impairment (MCI) is the prodromal stage of Alzheimer's disease (AD). MCI includes stable mild cognitive impairment (sMCI) and progressive mild cognitive impairment (pMCI). It is very important to distinguish sMCI and pMCI accurately for the early diagnosis and intervention of AD.
Methods
2
In this paper, a multi-modal and multi-atlas MCI classification recognition method based on multi-layer perceptron was proposed to differentiate pMCI from sMCI. Applying FreeSurfer, the morphological characteristic parameters derived from structural magnetic resonance imaging (sMRI) data were achieved including average cortical thickness, surface area, gray matter volume, sulcal depth, and folding index, as well as the average voxel intensity of specific anatomical regions. Adopting DPABI, the functional characteristic parameters derived from resting state functional MRI (rs-fMRI) data were obtained including amplitude of low frequency fluctuations (ALFF) and regional homogeneity (ReHo). In addition, ridge regression was applied for characteristic parameters selection, and the brain regions with the most distinguishing features were determined based on the coefficients matrix of ridge regression output.
Results
2
The proposed method was cross-validated by 10 fold on the ADNI dataset, integrating the characteristic parameters from two modalities of sMRI and rs-fMRI, the accuracy of distinguishing between sMCI and pMCI was 82.7%, with an AUC of 93.0%. In addition, it was found that the thickness and volume of entorhinal cortex and temporal lobe, and the mean value of voxel intensity of hippocampus and parahippocampal gyrus had significant differences between groups.
Conclusions
2
The proposed method has the potential to distinguish pMCI from sMCI. The performance of MCI classification recognition can be significantly improved by incorporating modality features of structural MRI and rs-fMRI and integrating multi-atlas.
周震,景斌. 基于体素的和基于形变的形态学测量在轻度认知障碍识别上的比较研究[J]. 北京生物医学工程, 2021, 40(5): 494-498, 511.
Zhou Z, Jing B. Comparative study of voxel-based and deformation-based morphometry in the identification of mild cognitive impairment[J]. Beijing Biomedical Engineering, 2021, 40(5): 494-498, 511.
Grundman M, Petersen RC, Ferris SH, et al. Mild cognitive impairment can be distinguished from Alzheimer disease and normal aging for clinical trials[J]. Archives of Neurology, 2004, 61(1): 59-66.
Misra C, Fan Y, Davatzikos C. Baseline and longitudinal patterns of brain atrophy in MCI patients, and their use in prediction of short-term conversion to AD: results from ADNI[J]. Neuroimage, 2009, 44(4): 1415-1422.
Lian C, Liu M, Zhang J, et al. Hierarchical fully convolutional network for joint atrophy localization and Alzheimer’s disease diagnosis using structural MRI[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 42(4): 880-893.
Samper-González J, Burgos N, Bottani S, et al. Reproducible evaluation of classification methods in Alzheimer’s disease: Framework and application to MRI and PET data[J]. NeuroImage, 2018, 183: 504-521.
Huang Y, Xu J, Zhou Y, et al. Diagnosis of Alzheimer’s disease via multi-modality 3D convolutional neural network[J]. Frontiers in Neuroscience, 2019, 13: 509.
Cho Y, Seong JK, Jeong Y, et al. Individual subject classification for Alzheimer’s disease based on incremental learning using a spatial frequency representation of cortical thickness data[J]. NeuroImage, 2012, 59(3): 2217-2230.
Li S, Yuan X, Pu F, et al. Abnormal changes of multidimensional surface features using multivariate pattern classification in amnestic mild cognitive impairment patients[J]. Journal of Neuroscience, 2014, 34(32): 10541-10553.
Ma Z, Jing B, Li Y, et al. Identifying mild cognitive impairment with random forest by integrating multiple MRI morphological metrics[J]. Journal of Alzheimer’s Disease, 2020, 73(3): 991-1002.
Buvaneswari PR, Gayathri R. Detection and classification of Alzheimer’s disease from cognitive impairment with resting-state fMRI[J]. Neural Computing and Applications, 2023, 35: 22797-22812.
董国昭, 曾晓天, 杨柳, 等. 基于动态图论特征的阿尔茨海默病早期预测[J]. 北京生物医学工程, 2019, 38(6): 560-567.
Dong GZ, Zeng XT, Yang L, et al. Early prediction of Alzheimer disease based on dynamic graph theory [J]. Beijing Biomedical Engineering, 2019, 38(6): 560-567.
Hojjati SH, Ebrahimzadeh A, Khazaee A, et al. Predicting conversion from MCI to AD by integrating rs-fMRI and structural MRI[J]. Computers in Biology and Medicine, 2018, 102: 30-39.
Jack Jr CR, Bernstein MA, Fox NC, et al. The Alzheimer’s disease neuroimaging initiative (ADNI): MRI methods[J]. Journal of Magnetic Resonance Imaging, 2008, 27(4): 685-691.
Fischl B. FreeSurfer[J]. NeuroImage, 2012, 62(2): 774-781.
Yan CG, Wang XD, Zuo XN, et al. DPABI: data processing analysis for (resting-state) brain imaging[J]. Neuroinformatics, 2016, 14(3): 339-351.
Zang Y, Jiang T, Lu Y, et al. Regional homogeneity approach to fMRI data analysis[J]. NeuroImage, 2004, 22(1): 394-400.
Zou QH, Zhu CZ, Yang Y, et al. An improved approach to detection of amplitude of low-frequency fluctuation (ALFF) for resting-state fMRI: fractional ALFF[J]. Journal of Neuroscience Methods, 2008, 172(1): 137-141.
Markello RD, Hansen JY, Liu ZQ, et al. Neuromaps: structural and functional interpretation of brain maps[J]. Nature Methods, 2022, 19(11): 1472-1479.
Teipel SJ, Grothe M, Lista S, et al. Relevance of magnetic resonance imaging for early detection and diagnosis of Alzheimer disease[J]. Medical Clinics of North America, 2013, 97(3): 399-424.
Moon SW, Lee B, Choi YC. Changes in the hippocampal volume and shape in early-onset mild cognitive impairment[J]. Psychiatry Investigation, 2018, 15(5): 531-537.
0
浏览量
2
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621