基于监督图正则化和信息融合的轻度认知障碍分类方法

doi:10.6040/j.issn.1672-3961.0.2023.025

摘要/Abstract

摘要：

为提高转化型和稳定型轻度认知障碍(mild cognitive impairment, MCI)的分类精度, 利用阿尔茨海默病样本和正常对照组样本学习投影矩阵, 使用监督图正则化项优化样本的局部近邻关系, 基于投影矩阵对MCI样本进行空间变换, 提取对转化型和稳定型MCI具有判别性的特征。将提取的特征与mini精神状态检查评分, 以及与载脂蛋白E4等位基因信息融合, 通过信息互补增强特征的判别性。使用融合特征训练支持向量机分类器对转化型和稳定型MCI分类。在ADNI数据库上进行试验, 分类精度达到73.33%。与已有方法相比, 本研究提出方法的分类精度、敏感度和特异度显著提高。

关键词: 轻度认知障碍, 监督图正则化, 信息融合, 特征提取, 分类

Abstract:

To precisely distinguish progressive and stable mild cognitive impairment (MCI). The projection matrix was learned from Alzheimer′s disease samples and normal control samples. The supervised graph regularization was used to optimize the local nearest neighbor relationship of the samples. Based on the projection matrix, the spatial transformation of the MCI samples was carried out to extract the discriminative features of progressive and stable MCI. The proposed features were fused with the scores of Mini-Mental State Examination and apolipoprotein E4. The SVM classifier was trained using the fused features for the MCI classification. The experiments were conduct on the Alzheimer′s Disease Neuroimaging Initiative (ADNI) database. The classification accuracy reached to 73.33%. Compared with the existing approaches, the proposed method significantly improved the classification accuracy, sensitivity and specificity.

Key words: mild cognitive impairment, supervised graph regularization, information fusion, feature extraction, classification

中图分类号:

TP391

李颖,王建坤. 基于监督图正则化和信息融合的轻度认知障碍分类方法[J]. 山东大学学报 (工学版), 2023, 53(4): 65-73.

Ying LI,Jiankun WANG. The classification of mild cognitive impairment based on supervised graph regularization and information fusion[J]. Journal of Shandong University(Engineering Science), 2023, 53(4): 65-73.

图/表 6

图1

表1

表2

表3

表4

图2

参考文献 42

1	2021 Alzheimer's disease facts and figures[J]. Alzheimers & Dementia, 2021, 17(3): 327-406.
2	MORRIS J C , CUMMINGS J . Mild cognitive impairment (MCI) represents early-stage Alzheimer's disease[J]. Journal of Alzheimer's Disease, 2005, 7 (3): 235- 239. doi: 10.3233/JAD-2005-7306
3	PETERSEN R C , DOODY R , KURZ A , et al. Current concepts in mild cognitive impairment[J]. Archives of Neurology, 2001, 58 (12): 1985- 1992. doi: 10.1001/archneur.58.12.1985
4	WARD A , TARDIFF S , DYE C , et al. Rate of conversion from prodromal Alzheimer's disease to Alzheimer's dementia: a systematic review of the literature[J]. Dementia & Geriatric Cognitive Disorders Extra, 2013, 3 (1): 320- 332.
5	RATHORE S , HABES M , IFTIKHAR M A , et al. A review on neuroimaging-based classification studies and associated feature extraction methods for Alzheimer's disease and its prodromal stages[J]. NeuroImage, 2017, 155, 530- 548. doi: 10.1016/j.neuroimage.2017.03.057
6	ZHENG W , YAO Z , XIE Y , et al. Identification of Alzheimer's disease and mild cognitive impairment using networks constructed based on multiple morphological brain features[J]. Biol Psychiatry Cogn Neurosci Neuroimaging, 2018, 3 (10): 887- 897.
7	YANG D , MASURKAR A . Early-stage MRI volumetric differences in white matter hyperintensity and temporal lobe volumes between autopsy-confirmed Alzheimer's disease, cerebral small vessel disease, and mixed pathologies[J]. Dementia and Geriatric Cognitive Disorders Extra, 2022, (12): 69- 75.
8	陈全, 彭永, 甘棋心, 等. 阿尔茨海默病的影像学研究进展[J]. 国际神经病学神经外科学杂志, 2022, 49 (5): 60- 66. doi: 10.16636/j.cnki.jinn.1673-2642.2022.05.013
	CHEN Quan , PENG Yong , GAN Qixin , et al. Research advances in the imaging findings of Alzheimer's disease[J]. Journal of International Neurology and Neurosurgery, 2022, 49 (5): 60- 66. doi: 10.16636/j.cnki.jinn.1673-2642.2022.05.013
9	张月芳, 邓红霞, 呼春香, 等. 融合残差块注意力机制和生成对抗网络的海马体分割[J]. 山东大学学报(工学版), 2020, 50 (6): 76- 81.
	ZHANG Yuefang , DENG Hongxia , HU Chunxiang , et al. Hippocampal segmentation combining residual attention mechanism and generative adversarial networks[J]. Journal of Shandong University(Engineering Science), 2020, 50 (6): 76- 81.
10	MORADI E , PEPE A , GASER C , et al. Machine learning framework for early MRI-based Alzheimer's conversion prediction in MCI subjects[J]. NeuroImage, 2015, 104, 398- 412. doi: 10.1016/j.neuroimage.2014.10.002
11	接标, 张道强. 面向脑网络的新型图核及其在MCI分类上的应用[J]. 计算机学报, 2016, 39 (8): 1667- 1680.
	JIE Biao , ZHANG Daoqiang . The novel graph kernel for brain networks with application to MCI classification[J]. Chinese Journal of Computers, 2016, 39 (8): 1667- 1680.
12	姜世香, 杨艳杰. 轻度认知障碍的发展演化及识别诊断[J]. 中国临床心理学杂志, 2017, 25 (1): 88- 91. doi: 10.16128/j.cnki.1005-3611.2017.01.020
	JIANG Shixiang , YANG Yanjie . The development and the diagnosis of mild cognitive impairment[J]. Chinese Journal of Clinical Psychology, 2017, 25 (1): 88- 91. doi: 10.16128/j.cnki.1005-3611.2017.01.020
13	TONG T , GAO Q , GUERRERO R , et al. A novel grading biomarker for the prediction of conversion from mild cognitive impairment to Alzheimer's disease[J]. IEEE Transactions on Biomedical Engineering, 2017, 64 (1): 155- 165. doi: 10.1109/TBME.2016.2549363
14	FAN Y , BATMANGHELICH N , CLARK C M , et al. Spatial patterns of brain atrophy in MCI patients, identified via high-dimensional pattern classification, predict subsequent cognitive decline[J]. NeuroImage, 2008, 39 (4): 1731- 1743. doi: 10.1016/j.neuroimage.2007.10.031
15	YOUNG J , MODAT M , CARDOSO M J , et al. Accurate multimodal probabilistic prediction of conversion to Alzheimer's disease in patients with mild cognitive impairment[J]. NeuroImage: Clinical, 2013, (2): 735- 745.
16	BRON E E , KLEIN S , PAPMA J M , et al. Cross-cohort generalizability of deep and conventional machine learning for mri-based diagnosis and prediction of Alzheimer's disease[J]. NeuroImage: Clinical, 2021, 102712.
17	LIN W , GAO Q , YUAN J , et al. Predicting Alzheimer's disease conversion from mild cognitive impairment using an extreme learning machine-based grading method with multimodal data[J]. Frontiers in Aging Neuroscience, 2020, (12): 1- 9.
18	SHEN H T , ZHU X , ZHANG Z , et al. Heterogeneous data fusion for predicting mild cognitive impairment conversion[J]. Information Fusion, 2021, 66 (1): 54- 63.
19	YE D H, POHL K M, DAVATZIKOS C. Semi-supervised pattern classification: application to structural MRI of Alzheimer's disease[C]//IEEE International Workshop on Pattern Recognition in Neuroimaging. Seoul, Korea: IEEE, 2011: 1-4.
20	FILIPOVYCH R , DAVATZIKOS C . Semi-supervised pattern classification of medical images: application to mild cognitive impairment (MCI)[J]. NeuroImage, 2011, 55 (3): 1109- 1119. doi: 10.1016/j.neuroimage.2010.12.066
21	BATMANGHELICH K N, YE D H, POHL K M, et al. Disease classification and prediction via semi-supervised dimensionality reduction[C]//IEEE International Symposium on Biomedical Imaging. Chicago, USA: IEEE, 2011: 1086-1090.
22	CHENG B , LIU M , ZHANG D , et al. Domain transfer learning for MCI conversion prediction[J]. IEEE Transactions on Biomedical Engineering, 2015, 62 (7): 1805- 1817. doi: 10.1109/TBME.2015.2404809
23	COUPÉ P , ESKILDSEN S F , MANJÓN J V , et al. Scoring by nonlocal image patch estimator for early detection of Alzheimer's disease[J]. NeuroImage: Clinical, 2012, 1 (1): 141- 152. doi: 10.1016/j.nicl.2012.10.002
24	LI Y , FANG Y , ZHANG H , et al. Self-weighting grading biomarker based on graph-guided information propagation for the prediction of mild cognitive impairment conversion[J]. IEEE Access, 2019, (7): 116632- 116642.
25	LIU X , TOSUN D , WEINER M W , et al. Locally linear embedding (LLE) for MRI based Alzheimer's disease classification[J]. NeuroImage, 2013, (83): 148- 157.
26	ZHU X , SUK H I , LEE S W , et al. Subspace regularized sparse multitask learning for multiclass neurodegenerative disease identification[J]. IEEE Transactions on Biomedical Engineering, 2016, 63 (3): 607- 618. doi: 10.1109/TBME.2015.2466616
27	LI Y , FANG Y , WANG J , et al. Biomarker extraction based on subspace learning for the prediction of mild cognitive impairment conversion[J]. BioMed Research International, 2021, 5531940.
28	MWANGI B , TIAN T S , SOARES J C . A review of feature reduction techniques in neuroimaging[J]. Neuroinformatics, 2014, 12 (2): 229- 244. doi: 10.1007/s12021-013-9204-3
29	FOLSTEIN M F , FOLSTEIN S E , MCHUGH P R . "Mini-mental state": a practical method for grading the cognitive state of patients for the clinician[J]. Journal of Psychiatric Research, 1975, 12 (3): 189- 198. doi: 10.1016/0022-3956(75)90026-6
30	LEHTOVIRTA M , SOININEN H , LAAKSO M P , et al. SPECT and MRI analysis in Alzheimer's disease: relation to apolipoprotein E ε4 allele[J]. Journal of Neurology Neurosurgery & Psychiatry, 1996, 60 (6): 644- 649.
31	VERGHESE P B , CASTELLANO J M , HOLTZMAN D M . Apolipoprotein E in Alzheimer's disease and other neurological disorders[J]. Lancet Neurology, 2011, 10 (3): 241- 252. doi: 10.1016/S1474-4422(10)70325-2
32	HALL A , MUÑOZ-RUIZ M , MATTILA J , et al. Generalizability of the disease state index prediction model for identifying patients progressing from mild cognitive impairment to Alzheimer's disease[J]. Journal of Alzheimer's Disease, 2015, 44 (1): 79- 92. doi: 10.3233/JAD-140942
33	杨小惠, 吴芹, 石京山. ApoE4在阿尔茨海默病中的研究进展[J]. 临床与病理杂志, 2017, 37 (3): 627- 631.
	YANG Xiaohui , WU Qin , SHI Jingshan . Research progress of apolipoprotein E4 in Alzheimer's disease[J]. Journal of Clinical and Pathological Research, 2017, 37 (3): 627- 631.
34	MORRIS J C . The clinical dementia rating (CDR): current version and scoring rules[J]. Neurology, 1993, 43 (11): 2412- 2414.
35	FISCHL B , SERENO M I , DALE A M . Cortical surface-based analysis. Ⅱ: Inflation, flattening, and a surface-based coordinate system[J]. NeuroImage, 1999, 9 (2): 195- 207. doi: 10.1006/nimg.1998.0396
36	DALE A M , FISCHL B , SERENO M I . Cortical surface-based analysis. I. Segmentation and surface reconstruction[J]. NeuroImage, 1999, 9 (2): 179- 194. doi: 10.1006/nimg.1998.0395
37	FISCHL B , LIU A , DALE A M . Automated manifold surgery: constructing geometrically accurate and topologically correct models of the human cerebral cortex[J]. IEEE Transactions on Medical Imaging, 2001, 20 (1): 70- 80. doi: 10.1109/42.906426
38	FISCHL B , DALE A M . Measuring the thickness of the human cerebral cortex from magnetic resonance images[J]. Proceedings of the National Academy of Sciences of the United States of America, 2000, 97 (20): 11050- 11055.
39	TZOURIO-MAZOYER N , LANDEAU B , PAPATHANASSIOU D , et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain[J]. NeuroImage, 2002, 15 (1): 273- 289.
40	FAN R E , CHANG K W , HSIEH C J , et al. LIBLINEAR: a library for large linear classification[J]. Journal of Machine Learning Research, 2008, 9 (9): 1871- 1874.
41	CHANG C C , LIN C J . LIBSVM: a library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology, 2011, 2 (3): 27.
42	ZHOU Q , GORYAWALA M , CABRERIZO M , et al. An optimal decisional space for the classification of Alzheimer's disease and mild cognitive impairment[J]. IEEE Transactions on Biomedical Engineering, 2014, 61 (8): 2245- 2253.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

样本	人数		年龄/岁		临床痴呆评分^[34]	mini精神状态检查评分
样本	男	女	平均	标准差	临床痴呆评分^[34]	均值	标准差
NC	78	87	76.40	5.37	0	29.19	0.96
sMCI	63	32	74.94	7.32	0.5	27.69	1.73
pMCI	73	53	73.40	9.25	0.5	26.49	1.70
AD	72	70	76.10	7.51	0.5/1	23.20	2.01

特征提取方式	分类精度/%	敏感度/%	特异度/%	A_uc
CT特征+CV特征+无监督图^[27]	69.37	75.39	61.23	0.695 1
CV特征+无监督图	67.21	73.01	59.46	0.702 8
CV特征+有监督图+自适应近邻结构优化	69.15	77.81	57.57	0.709 2
CV特征+有监督图+自适应近邻结构优化+特征选择(SGRIP特征)	70.24	78.76	58.64	0.700 9

特征	分类精度/%	敏感度/%	特异度/%	A_uc
MMSE	62.44	53.27	74.65	0.658 3
APOE4	60.67	63.50	56.84	0.417 2
MMSE-APOE4	59.17	57.78	61.15	0.685 0
SGRIP特征	70.24	78.76	58.64	0.700 9
SGRIP特征-MMSE	71.99	78.23	63.63	0.733 3
SGRIP特征-APOE4	69.83	77.95	59.03	0.721 1
SGRIP特征-MMSE-APOE4	73.33	78.84	65.93	0.761 2

特征	分类器	分类精度/%	敏感度/%	特异度/%	A_uc
MFN特征^[6]	SVM(高斯核)	65.61	70.63	58.95	0.667 0
分级生物特征^[13]	SVM(线性核)	67.43	70.04	63.96	0.698 1
基于子空间学习的生物特征^[27]	SVM(线性核)	69.37	75.39	61.23	0.695 1
SGRIP特征-MMSE-APOE4	SVM(高斯核)	72.79	81.71	60.88	0.754 2
SGRIP特征-MMSE-APOE4	SVM(线性核)	73.33	78.84	65.93	0.761 2

[1]	白琳,俱通,王浩,雷明珠,潘晓英. 面向不平衡数据的提升均衡集成学习算法[J]. 山东大学学报 (工学版), 2024, 54(4): 59-66.
[2]	陈晓江,杨晓奇,陈广豪,刘伍颖. 混合BERT和宽度学习的低时间复杂度短文本分类[J]. 山东大学学报 (工学版), 2024, 54(4): 51-58.
[3]	宋辉,张轶哲,张功萱,孟元. 基于类权重和最小化预测熵的测试时集成方法[J]. 山东大学学报 (工学版), 2024, 54(3): 36-43.
[4]	聂秀山,巩蕊,董飞,郭杰,马玉玲. 短视频场景分类方法综述[J]. 山东大学学报 (工学版), 2024, 54(3): 1-11.
[5]	徐金华,罗义凯,李昱燃,李岩. 基于时频分解与深度学习的轨道客流预测[J]. 山东大学学报 (工学版), 2024, 54(2): 60-68.
[6]	马坤,刘筱云,李乐平,纪科,陈贞翔,杨波. 用于意图识别的自适应多标签信息学习模型[J]. 山东大学学报 (工学版), 2024, 54(1): 45-51.
[7]	那绪博,张莹,李沐阳,陈元畅,华云鹏. 基于ODCG的网约车需求预测模型[J]. 山东大学学报 (工学版), 2023, 53(5): 48-56.
[8]	于泓,杜娟,魏琳,张利. 计及行为特征的市场化用户电量数据拟合方法[J]. 山东大学学报 (工学版), 2023, 53(4): 113-119.
[9]	张喜龙,韩萌,陈志强,武红鑫,李慕航. 动态集成选择的不平衡漂移数据流Boosting分类算法[J]. 山东大学学报 (工学版), 2023, 53(4): 83-92.
[10]	刘财辉,周琪,叶晓文. 一种基于改进ReliefF算法的入侵检测模型[J]. 山东大学学报 (工学版), 2023, 53(2): 1-10.
[11]	孟令灿,聂秀山,张雪. 基于遮挡目标去除的公交车拥挤度分类算法[J]. 山东大学学报 (工学版), 2022, 52(4): 83-88.
[12]	孙志巍,宋明阳,潘泽华,景丽萍. 上下文感知的判别式主题模型[J]. 山东大学学报 (工学版), 2022, 52(4): 131-138.
[13]	王丽,于明仟,刘文鹏,周瑜,郑蕊蕊,贺建军. 面向类不平衡数据的K近邻偏标记学习算法[J]. 山东大学学报 (工学版), 2022, 52(3): 18-24.
[14]	张学思,张婷,刘兆英,江天鹏. 基于轻量型卷积神经网络的海面红外显著性目标检测方法[J]. 山东大学学报 (工学版), 2022, 52(2): 41-49.
[15]	龚楷伦,翟婷婷,唐鸿成. 一种面向多标签分类的在线主动学习算法[J]. 山东大学学报 (工学版), 2022, 52(2): 80-88.