基于深度学习的阿尔兹海默症影像学分类研究进展

doi:10.11938/cjmr20223013

摘要/Abstract

摘要：

随着全球老龄化的加剧与深度学习的发展，基于深度学习的阿尔兹海默症（AD）影像学分类成为当前的一个研究热点．本文首先阐述了AD影像学分类任务中常用的深度学习模型、评估标准及公开数据集；接着讨论了不同图像模态在AD影像学分类中的应用；然后着重探讨了应用于AD影像学分类的深度学习模型改进方法；进一步引入了对模型可解释性研究的探讨；最后总结并比较了文中提及的分类模型，归纳了与AD影像分类相关的大脑区域，并对该领域未来的研究方向进行了展望．

关键词: 阿尔茨海默症（AD）, 深度学习, 医学影像, 分类, 可解释性

Abstract:

As global aging worsens and deep learning advances, the imaging classification of Alzheimer’s disease (AD) based on deep learning has become a hot topic of research. This paper reviewed the common deep learning models, evaluation criteria and public datasets in AD imaging classification tasks, discussed the application of different image modalities in AD imaging classification. The content was focused on the improvement of deep learning models applied to AD imaging classification. The studies of model interpretability were also introduced. Finally, the paper summarized and compared the classification models mentioned, identified the brain regions related to AD image classification, and outlined the future research directions in this field.

Key words: Alzheimer’s disease (AD), deep learning, medical imaging, classification, interpretability

中图分类号:

TP391

钱程一,王远军. 基于深度学习的阿尔兹海默症影像学分类研究进展[J]. 波谱学杂志, 2023, 40(2): 220-238.

QIAN Chengyi,WANG Yuanjun. Research Progress on Imaging Classification of Alzheimer’s Disease Based on Deep Learning[J]. Chinese Journal of Magnetic Resonance, 2023, 40(2): 220-238.

图/表 8

图1

图2

图3

图4

表1

表2

各分类模型综合比较

第一作者	分类任务	数据模态		数据集		测试方法	分类模型	准确率%
郁松^[21]	AD/HC	sMRI		1015 HC 575 AD 1709 MCI		训练集60% 验证集20% 测试集20%	3D ResNet-101	97.425
Parmar^[24]	AD/HC	fMRI		30 AD 30 HC		训练集60% 验证集20% 测试集20%	3D CNN	94.58
Bi^[19]	AD/HC MCI/HC AD/MCI AD/HC/MCI	fMRI		118 AD 295 HC 335 MCI		五折交叉验证	RNN，ELM	91.3 (AUC) 80.5 (AUC) 82.4 (AUC) 84.7 (AUC)
Yi??i?T^[20]	AD/HC MCI/HC	sMRI		训练集 30 AD 316 HC 70 MCI	测试集 46 AD+MCI 23 HC	/	2D CNN	83 82
Punjabi^[33]	AD/HC	PET sMRI+PET		共1299		/	3D CNN	85.15 92.34
Zhang^[49]	AD/HC pMCI/sMCI	sMRI		200 AD 231 HC 172 pMCI 232 sMCI		五折交叉验证	3D ResAttNet	91.3 82.1
Qiu^[52]	AD/HC	sMRI+性别+年龄+MMSE		/		训练集60% 验证集20% 测试集20%	FCN	96.8
Deng^[28]	AD/HC	DTI+sMRI		98 AD 100 HC		训练集60% 验证集20% 测试集20%	CNN	90.00
Marzban^[29]	AD/HC MCI/HC	DTI+sMRI		115 AD 185 HC 106 MCI		十折交叉验证	CNN	93.5 79.6
Kang^[30]	EMCI/HC		DTI+sMRI	50 HC 70 EMCI		训练集80% 测试集20%	VGG-16，LASSO		94.2
Kwak^[54]	AD/HC sMCI/pMCI		sMRI	110 AD 109 HC 34 pMCI 81 sMCI		五折交叉验证	DenseNet		93.75 73.90
Ju^[16]	MCI/HC		fMRI	91 MCI 79 HC		十折交叉验证	AE		86.47
Baydargil^[17]	AD/MCI/HC		PET	141 AD 105 MCI 70 HC		训练集80% 验证集10% 测试集10%	CAE		98.67
Guan^[51]	AD/HC pMCI/sMCI		sMRI	384 AD 392 HC 401 sMCI 197 pMCI		训练集 90% 测试集 10%	ResNet、pABN		90.7 79.3
Li^[34]	AD/MCI/HC		MRI	237 AD 288 MCI 262 HC		训练集65% 测试集 35%	ResNet-200 （迁移学习）		83
Basaia^[40]	AD/HC sMCI/pMCI		sMRI	294 AD 352 HC 253 pMCI 510 sMCI		训练集90% 测试集10%	3D FCN（迁移学习）		99.2 75.1
孔伶旭^[36]	EMCI/HC		fMRI	32 HC 32 EMCI		五折交叉验证	Mobilenet（迁移学习）		73.67
Bin^[37]	AD/HC		sMRI	100 AD 100 HC		五折交叉验证	Inception-v3 （迁移学习）		99.45
Massalimova^[27]	AD/MCI/HC		DTI	训练集 59 AD 308 HC 7 MCI	测试集 16 AD 74 HC 1 MCI	/	ResNet-18（迁移学习）		97
Mehmood^[38]	AD/HC EMCI/LMCI		sMRI	75 AD 85 HC 70 EMCI 70 LMCI		训练集64% 验证集16% 测试集20%	VGG-19（迁移学习）		95.33 83.72
Raju^[50]	非常轻度痴呆/轻度痴呆/中度痴呆/HC（四分类）		sMRI	1013非常轻度痴呆 896轻度痴呆 64中度痴呆 3200 HC（训练集）		334非常轻度痴呆 139轻度痴呆 10中度痴呆 530 HC（测试集）	VGG-16（迁移学习）		99
Lian^[39]	AD/HC sMCI/pMCI	sMRI		数据集1 199 AD 229 HC 167 pMCI 226 sMCI	数据集2 159 AD 200 HC 38 pMCI 239 sMCI	两个数据集间交叉验证	H-FCN（迁移学习）	90.3 80.9
Oh^[18]	AD/HC sMCI/pMCI	sMRI		198 AD 230 HC 166 pMCI 101 sMCI		五折交叉验证	3D CNN，ICAE （迁移学习）	88.6 73.95
金祝新^[53]	AD/MCI MCI/HC AD/HC		sMRI	267 AD 574 HC 446 MCI		训练集90% 测试集10%	3D CNN（迁移学习）		94.6 92.5 90.9
曾安^[42]	AD/HC pMCI/HC pMCI/sMCI		sMRI	137 AD 162 HC 76 pMCI 134 sMCI		五折交叉验证	2D CNN（集成学习）		81 79 62
Bi^[22]	AD/HC MCI/HC AD/MCI AD/MCI/HC		sMRI	243 AD 307 HC 525 MCI		/	PCANet，k-means （集成学习）		89.15 92.6 97.01 91.25
Choi^[41]	AD/HC		sMRI	715 AD 335 HC 305 MCI		训练集60% 验证集20% 测试集20%	VGG-16，GoogLeNet，AlexNet（集成学习）		93.84
Venugopalan^[55]	AD/HC		sMRI+电子病历+基因数据	共220		十折交叉验证	AE，CNN，随机森林（集成学习）		88
Zeng^[46]	sMCI/pMCI AD/HC sMCI/HC pMCI/HC AD/sMCI AD/pMCI		sMRI	92 AD 92 HC 92 sMCI 95 pMCI		训练集70% 测试集30%	PCA，DBN（多任务）		87.78 98.62 92.31 96.67 99.62 91.89
Spasov^[43]	sMCI/pMCI		sMRI	192 AD 184 HC 409 MCI		十折交叉验证	CNN（多任务）		86
Liu^[48]	AD/HC MCI/HC		sMRI	97 AD 11 HC 233 MCI		五折交叉验证	V-Net，DenseNet （多任务）		88.9 76.2
Abuhmed^[15]	AD/MCI/HC		PET+sMRI+神经心理学数据+神经病理学数据+认知评分	共1371		十折交叉验证	BiLSTM，随机森林（多任务）		84.95

表2

表3

表A1

中英文全称及对应缩写表

缩写	中文全称	英文全称
AD	阿尔兹海默症	Alzheimer disease
NFT	神经原纤维缠结	neurofibrillary tangles
CDR	临床痴呆分级	clinical dementia rating
MMSE	简易智力状态检查量表	mini-mental state examination
MCI	轻度认知障碍	mild cognitive impairment
EMCI	早期轻度认知障碍	early mild cognitive impairment
LMCI	晚期轻度认知障碍	later mild cognitive impairment
sMCI	稳定型轻度认知障碍	stable mild cognitive impairment
pMCI	渐近型轻度认知障碍	progressive mild cognitive impairment
HC	健康对照组	healthy control
ADNI	阿尔兹海默症神经影像学计划	Alzheimer’s disease neuroimaging initiative
OASIS	开放存取影像研究	open access series of imaging studies
IXI	从图像中提取信息数据集	information eXtraction from images
sMRI	结构磁共振成像	structural magnetic resonance imaging
PET	正电子发射型计算机断层显像	positron emission computed tomography
DTI	扩散张量成像	diffusion tensor imaging
fMRI	功能磁共振成像	functional magnetic resonance imaging
FA	各向异性分数	fractional anisotropy
MD	平均扩散率	mean diffusivity
RD	径向扩散系数	radial diffusivity
DC	中心度特征	degree centrality
ROI	感兴趣区域	region of interest
AAL	自动解剖标记	automated anatomical labeling
TP	真阳性	true positive
FP	假阳性	false positive
TN	真阴性	true negative
FN	假阴性	false negative
ROC	接受者操作特征曲线	receiver operating characteristic curve
AUC	曲线下面积	area under curve
CNN	卷积神经网络	convolutional neural networks
ILSVRC	ImageNet大规模视觉识别挑战赛	ImageNet large scale visual recognition challenge
RNN	循环神经网络	recurrent neural network
BRNN	双向循环神经网络	bidirectional recurrent neural network
AE	自动编码机	autoencoder
CAE	卷积自动编码机	convolutional autoencoder
ICAE	带Inception模块的卷积自动编码机	inception modal based convolutional autoencoder
ResNet	残差网络	residual network
DenseNet	密集连接网络	densely connected convolutional network
pABN	并行注意力增强双线性网络	parallel attention-augmented bilinear network
k-means	k均值聚类	k-means clustering algorithm
PCA	主成分分析法	principal components analysis
ELM	极限学习机	extreme learning machine
H-FCN	分层全卷积神经网络	hierarchical fully convolutional network
FCN	全卷积神经网络	fully convolutional network
ADAS-cog	阿尔兹海默症评定量表-认知分量表	Alzheimer’s disease assessment scale
BiLSTM	双向长短期记忆网络	bidirectional long short-term memory
MAE	平均绝对误差	mean absolute error
CAM	类激活映射	class activation mapping
Grad-CAM	梯度加权类激活映射	gradient-weighted class activation mapping
GAP	全局平均池化	global average pooling
CSV	类显著映射	class saliency visualization

表A1

参考文献 58

[1]	JACK Jr C R, BENNETT D A, BLENNOW K, et al. NIA-AA research framework: Toward a biological definition of Alzheimer’s disease[J]. Alzheimers Dement, 2018, 14(4): 535-562. doi: 10.1016/j.jalz.2018.02.018
[2]	SCHELTENS P, DE STROOPER B, KIVIPELTO M, et al. Alzheimer’s disease[J]. The Lancet, 2021, 397(10284): 1577-1590. doi: 10.1016/S0140-6736(20)32205-4
[3]	ANDERSON N D. State of the science on mild cognitive impairment[J]. J Gerontol B Psycholo Sci Soc Sci, 2020, 75(7): 1359-1360.
[4]	MAGGIPINTO T, BELLOTTI R, AMOROSO N, et al. DTI measurements for Alzheimer’s classification[J]. Phys Med Biol, 2017, 62(6): 2361-2375. doi: 10.1088/1361-6560/aa5dbe
[5]	LAHMIRI S, SHMUEL A. Performance of machine learning methods applied to structural MRI and ADAS cognitive scores in diagnosing Alzheimer’s disease[J]. Biomed Signal Proces, 2019, 52: 414-419. doi: 10.1016/j.bspc.2018.08.009
[6]	ELSHATOURY H, AVOTS E, ANBARJAFARI G, et al. Volumetric histogram-based Alzheimer’s disease detection using support vector machine[J]. J Alzheimers Dis, 2019, 72(2): 515-524. doi: 10.3233/JAD-190704
[7]	UYSAL G, OZTURK M. Hippocampal atrophy based Alzheimer’s disease diagnosis via machine learning methods[J]. J Neurosci Meth, 2020, 337: 108669. doi: 10.1016/j.jneumeth.2020.108669
[8]	CHU Y, XU W L. Review of early classification of Alzheimer’s disease based on computer-aided diagnosistechnology[J]. Computer Engineering & Science, 2022, 44(5): 879-893.
	楚阳, 徐文龙. 基于计算机辅助诊断技术的阿尔兹海默症早期分类研究综述[J]. 计算机工程与科学, 2022, 44(5): 879-893.
[9]	YAO X F, YUAN Z B, BU X X. The early predication of Alzheimer's disease based on intelligent radiomics technology[J]. Sci Technol Rev, 2021, 39(20): 101-109. doi: 10.20506/rst.issue.39.1.3056
	姚旭峰, 袁增贝, 卜溪溪. 基于智能影像基因组学技术的阿尔茨海默病预测进展[J]. 科技导报, 2021, 39(20): 101-109. doi: 10.20506/rst.issue.39.1.3056
[10]	ZHU Y X, FENG W, GUO X H. Application of deep learning method in brain image of Alzheimer’s disease[J]. Med Recapitulate, 2019, 25(18): 3562-3566.
	朱映璇, 冯巍, 郭秀花. 深度学习方法在阿尔茨海默病脑图像方面的应用进展[J]. 医学综述, 2019, 25(18): 3562-3566.
[11]	TANVEER M, RICHHARIYA B, KHAN R U, et al. Machine learning techniques for the diagnosis of Alzheimer’s disease: a review[J]. ACM T Multim Comput, 2020, 16(1s): 30.
[12]	FAN Z, XU F, QI X, et al. Classification of Alzheimer’s disease based on brain MRI and machine learning[J]. Neural Comput Appl, 2020, 32(7): 1927-1936. doi: 10.1007/s00521-019-04495-0
[13]	KARAMI V, NITTARI G, TRAINI E, et al. An optimized decision tree with genetic algorithm rule-based approach to reveal the brain’s changes during Alzheimer’s disease dementia[J]. J Alzheimers Dis, 2021, 84(4): 1577-1584. doi: 10.3233/JAD-210626
[14]	ALICKOVIC E, SUBASI A, ALZHEIMERS DIS N. Automatic detection of Alzheimer disease based on histogram and random forest[C]// Banja Luka, BOSNIA & HERCEG: International Conference on Medical and Biological Engineering in Bosnia and Herzegovina (CMBEBIH), 2019, 73: 91-96.
[15]	ABUHMED T, EI-SAPPAGH S, ALONSO J M. Robust hybrid deep learning models for Alzheimer’s progression detection[J]. Knowl-Based Syst, 2021, 213: 106688. doi: 10.1016/j.knosys.2020.106688
[16]	JU R H, HU C H, ZHOU P, et al. Early diagnosis of Alzheimer’s disease based on resting-state brain networks and deep learning[J]. IEEE ACM T Comput Bi, 2019, 16(1): 244-257.
[17]	BAYDARGIL H B, PARK J S, KANG D Y, et al. Classification of Alzheimer’s disease using stacked sparse convolutional autoencoder[C]// Jeju, SOUTH KOREA:19th International Conference on Control, Automation and Systems (ICCAS), 2019, 891-895.
[18]	OH K, CHUNG Y C, KIM K W, et al. Classification and visualization of Alzheimer’s disease using volumetric convolutional neural network and transfer learning[J]. Sci Rep, 2019, 9(1): 18150. doi: 10.1038/s41598-019-54548-6
[19]	BI X, ZHAO X G, HUANG H, et al. Functional brain network classification for Alzheimer’s disease detection with deep features and extreme learning machine[J]. Cogn Comput, 2020, 12(3): 513-527. doi: 10.1007/s12559-019-09688-2
[20]	YIĞIT A, ISIK Z. Applying deep learning models to structural MRI for stage prediction of Alzheimer’s disease[J]. Turk J Elec Eng & Comp Sci, 2020, 28(1): 196-210.
[21]	YU S, LIAO W H. An alzheimer’s disease classificationalgorithm based on 3D-ResNet[J]. Computer Engineering & Science, 2020, 42(6): 1068-1075.
	郁松, 廖文浩. 基于3D-ResNet的阿尔兹海默症分类算法研究[J]. 计算机工程与科学, 2020, 42(6): 1068-1075.
[22]	BI X L, LI S T, XIAO B, et al. Computer aided Alzheimer's disease diagnosis by an unsupervised deep learning technology[J]. Neurocomputing, 2020, 392: 296-304. doi: 10.1016/j.neucom.2018.11.111
[23]	GUO H B, ZHANG Y J. Resting state fMRI and improved deep learning algorithm for earlier detection of Alzheimer’s disease[J]. IEEE Access, 2020, 8: 115383-115392. doi: 10.1109/Access.6287639
[24]	PARMAR H S, NUTTER B, LONG R, et al. Deep learning of volumetric 3D CNN for fMRI in Alzheimer’s disease classification[C]// Houston, TX: SPIE Medical Imaging Conference - Biomedical Applications in Molecular, Structural, and Functional Imaging, 2020, 11317.
[25]	REGINOLD W, LUEDKE A C, ITORRALBA J, et al. Altered superficial white matter on tractography MRI in Alzheimer’s disease[J]. Dementia and Geriatric Cognitive Disorders Extra, 2016, 6(2): 233-241. doi: 10.1159/000446770
[26]	BIGHAM B, ZAMANPOUR S A, ZEMORSHIDI F, et al. Identification of superficial white matter abnormalities in Alzheimer’s disease and mild cognitive impairment using diffusion tensor imaging[J]. J Alzheimers Dis Rep, 2020, 4(1): 49-59.
[27]	MASSALIMOVA A, VAROL H A, IEEE. Input agnostic deep learning for Alzheimer’s disease classification using multimodal MRI images[C]// Electr Network:43rd Annual International Conference of the IEEE-Engineering-in-Medicine-and-Biology-Society (IEEE EMBC), 2021, 10.1109/embc46164.2021.96298072875-2878. doi: 10.1109/embc46164.2021.96298072875-2878
[28]	DENG L, WANG Y J. Hybrid diffusion tensor imaging feature-based AD classification[J]. J X-Ray Sci Technol, 2021, 29(1): 151-169. doi: 10.3233/XST-200771 pmid: 33325450
[29]	MARZBAN E N, ELDEIB A M, YASSINE I A, et al. Alzheimer's disease diagnosis from diffusion tensor images using convolutional neural networks[J]. PLoS ONE, 2020, 15(3): e0230409. doi: 10.1371/journal.pone.0230409
[30]	KANG L, JIANG J W, HUANG J J, et al. Identifying early mild cognitive impairment by multi-modality MRI-based deep learning[J]. Front Aging Neurosci, 2020, 12: 206. doi: 10.3389/fnagi.2020.00206 pmid: 33101003
[31]	GAO F. Integrated positron emission tomography/magnetic resonance imaging in clinical diagnosis of Alzheimer’s disease[J]. Eur J Radiol, 2021, 145: 1100170.
[32]	RABINOVICI G D, GATSONIS C, APGAR C, et al. Association of amyloid positron emission tomography with subsequent change in clinical management among medicare beneficiaries with mild cognitive impairment or dementia[J]. JAMA-J Am Med Assoc, 2019, 321(13): 1286-1294. doi: 10.1001/jama.2019.2000
[33]	PUNJABI A, MARTERSTECK A, WANG Y R, et al. Neuroimaging modality fusion in Alzheimer’s classification using convolutional neural networks[J]. PLoS ONE, 2019, 14(12): e0225759. doi: 10.1371/journal.pone.0225759
[34]	LI Y, DING W, WANG X, et al. Alzheimer’s disease classification model based on MED-3D transfer learning[C]// Proceedings of the 2nd International Symposium on Artificial Intelligence for Medicine Sciences, 2021, 394-398.
[35]	CHEN S, MA K, ZHENG Y. Med3d: Transfer learning for 3d medical image analysis[J]. arXiv preprint, arXiv:.00625, 2019.
[36]	KONG L X, WU H F, ZENG Y, et al. Transfer-learning feature extraction from rs-fMRI for classification of early mild cognitive impairment[J]. CAAI Transactions on Intelligent Systems, 2021, 16(4): 662-672.
	孔伶旭, 吴海锋, 曾玉, 等. 迁移学习特征提取的rs-fMRI早期轻度认知障碍分类[J]. 智能系统学报, 2021, 16(4): 662-672.
[37]	BIN TUFAIL A, MA Y K, ZHANG Q N. Binary classification of Alzheimer’s disease using sMRI imaging modality and deep learning[J]. J Digit Imaging, 2020, 33(5): 1073-1090. doi: 10.1007/s10278-019-00265-5
[38]	MEHMOOD A, YANG S Y, FENG Z X, et al. A transfer learning approach for early diagnosis of Alzheimer’s disease on MRI images[J]. Neuroscience, 2021, 460: 43-52. doi: 10.1016/j.neuroscience.2021.01.002
[39]	LIAN C F, LIU M X, ZHANG J, et al. Hierarchical fully convolutional network for joint atrophy localization and Alzheimer’s disease diagnosis using structural MRI[J]. IEEE Trans Pattern Anal Mach Intell, 2020, 42(4): 880-893. doi: 10.1109/TPAMI.34
[40]	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-Clin, 2019, 21: 101645. doi: 10.1016/j.nicl.2018.101645
[41]	CHOI J Y, LEE B H. Combining of multiple deep networks via ensemble generalization loss, based on mri images, for Alzheimer’s disease classification[J]. IEEE Signal Process Lett, 2020, 27: 206-210. doi: 10.1109/LSP.97
[42]	ZENG A, JIA L F, PAN D, et al. Early prognosis of Alzheimer’s disease based on convolutional neural networks and ensemble learning[J]. J Biomed Eng, 2019, 36(5): 711-719.
	曾安, 贾龙飞, 潘丹, 等. 基于卷积神经网络和集成学习的阿尔茨海默症早期诊断[J]. 生物医学工程学杂志, 2019, 36(5): 711-719.
[43]	SPASOV S, PASSAMONTI L, DUGGENTO A, et al. A parameter-efficient deep learning approach to predict conversion from mild cognitive impairment to Alzheimer’s disease[J]. Neuroimage, 2019, 189: 276-287. doi: 10.1016/j.neuroimage.2019.01.031
[44]	AMIEVA H, OUVRARD C, GIULIOLI C, et al. Self-reported hearing loss, hearing aids, and cognitive decline in elderly adults: A 25-year study[J]. J Am Geriatr Soc, 2015, 63(10): 2099-2104. doi: 10.1111/jgs.13649 pmid: 26480972
[45]	SARATXAGA C L, MOYA I, PICON A, et al. MRI deep learning-based solution for Alzheimer’s disease prediction[J]. J Pers Med, 2021, 11(9): 902. doi: 10.3390/jpm11090902
[46]	ZENG N Y, LI H, PENG Y H. A new deep belief network-based multi-task learning for diagnosis of Alzheimer’s disease[J]. Neural Comput Appl, 2021, 10.1007/s00521-021-06149-6. doi: 10.1007/s00521-021-06149-6
[47]	KATABATHULA S, WANG Q Y, XU R. Predict Alzheimer’s disease using hippocampus MRI data: a lightweight 3D deep convolutional network model with visual and global shape representations[J]. Alzheimers Res Ther, 2021, 13: 104. doi: 10.1186/s13195-021-00837-0
[48]	LIU M H, LI F, YAN H, et al. A multi-model deep convolutional neural network for automatic hippocampus segmentation and classification in Alzheimer's disease[J]. Neuroimage, 2020, 208: 116459. doi: 10.1016/j.neuroimage.2019.116459
[49]	ZHANG X, HAN L, ZHU W, et al. An explainable 3D residual self-attention deep neural network for joint atrophy localization and Alzheimer’s disease diagnosis using structural MRI[J]. IEEE J Biomed Health Inform, 2022, 26(11): 5289-5297. doi: 10.1109/JBHI.2021.3066832
[50]	RAJU M, THIRUPALANI M, VIDHYABHARATHI S, et al. Deep learning based multilevel classification of Alzheimer’s disease using MRI scans[J]. IOP Conf Ser: Mater Sci Eng, 2021, 1084: 012017. doi: 10.1088/1757-899X/1084/1/012017
[51]	GUAN H, WANG C Y, CHENG J, et al. A parallel attention-augmented bilinear network for early magnetic resonance imaging-based diagnosis of Alzheimer’s disease[J]. Hum Brain Mapp, 2022, 43(2): 760-772. doi: 10.1002/hbm.v43.2
[52]	QIU S R, JOSHI P S, MILLER M I, et al. Development and validation of an interpretable deep learning framework for Alzheimer’s disease classification[J]. Brain, 2020, 143(6): 1920-1933. doi: 10.1093/brain/awaa137
[53]	金祝新. 迁移学习与可视化辅助的阿尔茨海默症早期诊断[D]. 杭州: 杭州电子科技大学, 2019.
[54]	KWAK K, STANFORD W, DAYAN E, et al. Identifying the regional substrates predictive of Alzheimer’s disease progression through a convolutional neural network model and occlusion[J]. Hum Brain Mapp, 2022, 43(18): 5509-5519. doi: 10.1002/hbm.v43.18
[55]	VENUGOPALAN J, TONG L, HASSANZADEH H R, et al. Multimodal deep learning models for early detection of Alzheimer’s disease stage[J]. Sci Rep, 2021, 11(1): 3254. doi: 10.1038/s41598-020-74399-w
[56]	SHAHAMAT H, ABADEH M S. Brain MRI analysis using a deep learning based evolutionary approach[J]. Neural Netw, 2020, 126: 218-234. doi: 10.1016/j.neunet.2020.03.017
[57]	BEHESHTI I, GANAIE M A, PALIWAL V, et al. Predicting brain age using machine learning algorithms: A comprehensive evaluation[J]. IEEE J Biomed Health Inform, 2022, 26(4): 1432-1440. doi: 10.1109/JBHI.2021.3083187
[58]	BASHYAM V M, ERUS G, DOSHI J, et al. MRI signatures of brain age and disease over the lifespan based on a deep brain network and 14 468 individuals worldwide[J]. Brain, 2020, 143(7): 2312-2324. doi: 10.1093/brain/awaa160

数据集	类别	数据集大小	图像模态
ADNI	AD MCI HC	192 398 229	sMRI、DTI、fMRI、PET
OASIS	AD HC	100 316	sMRI、DTI、fMRI、PET
IXI	HC	约600	sMRI、DTI、磁共振血管造影