深度学习在老年人轻度认知障碍诊断中应用的Scoping综述

doi:10.3969/j.issn.1006-9771.2025.06.008

Abstract

Abstract:

Objective To systematically review the application and effectiveness of deep learning (DL) in diagnosis of mild cognitive impairment (MCI) among older adults.

Methods PubMed, Web of Science, CNKI and Wanfang databases were searched for literatures related to the application of DL in MCI among older adults, from database inception to December, 2024. A scoping review was conducted. The literature screening process followed the Scoping Review Report Specification list, and the quality assessment was conducted using the cross-sectional study quality evaluation tool developed by the Evidence-based Health Care Center.

Results A total of eleven papers were included, from Italy, USA, South Korea, China, India and Switzerland, involving 11 829 elderly participants, publicated mainly between 2014 and 2024, reflecting the rapid development trend of the field in the last decade, which was in line with the timing of the development of DL technology. The quality scores of the included literatures were all six to seven. The types of studies were all cross-sectional studies with significant cross-disciplinary characteristics, mainly originating from the fields of clinical medicine, biology and neuroimaging. The literature data were mainly based on the Alzheimer's disease Neuroimaging Program database and integrated other data resources. In terms of data type, in addition to brain imaging data, one study based on text data was also included. In terms of models used, five of the studies were mainly based on convolutional neural networks, and the rest used different DL modeling frameworks. The task types contained binary and triple classification. In terms of prediction results, the DL models constructed on multimodal data, such as brain images, could be used to construct high-precision prediction models for MCI classification, and the models were all good, with accuracy more than 70% and AUC values more than 0.7. The diagnostic accuracy of some of the models was more than 90%, and the model with the highest prediction accuracy was the one that used the Biceph-Net lightweight framework, with accuracy close to 100%, and the text analysis model based on Transformer made the AUC value of 0.846, which provided new ideas for the diagnosis of non-imaging data.

Conclusion DL can not only provide strong support for the accurate identification of MCI in the elderly, but also provide auxiliary prediction tools for clinicians, which can help delay the progression of the disease and improve the prognosis of patients.

Key words: older adults, mild cognitive impairment, deep learning, scoping review

CLC Number:

R749.1

WU Xiaohui, JIANG Lei, ZHU Jingru, XIE Lihui. Deep learning for diagnosis of mild cognitive impairment in older adults: a scoping review[J]. Chinese Journal of Rehabilitation Theory and Practice, 2025, 31(6): 674-681.

Figures/Tables 3

Figure 1

Table 1

Table 2

Basic characteristics of included literatures"

纳入文献	国家	年龄/岁	被试数量	研究方法	数据来源与类型	预测指标	评价指标	深度学习模型
Ciarmiello等^[20](2023)	意大利	72±7.4 MCI: 73.6±8.0 HC:70.8±6.2	n = 328 (MCI = 179；HC = 149) 测试集：15% 训练集：70% 验证集：15%	横断面研究	ADNI数据库；¹⁸F florbetaben PET成像数据	MCI与HC诊断	准确率80% AUC：测试集：0.9 整体样本：0.94	模型：FNN
Pourramezan Fard等^[21] (2024)	美国	75~92	n = 68 (MCI = 34；NC = 34) 5折交叉验证测试集 = 13~14 训练集 = 55~56	横断面研究	I-CONECT项目，对68例75岁及以上老人(分为MCI组和NC组)进行每次30 min、每周4次、共6个月的半结构化视频访谈，采集访谈音频并转录为文本，构建研究数据集	MCI与NC诊断	准确率85.16% AUC：0.85	模型：NLP+MLP 优化器：Adam
Khatri等^[22] (2024)	韩国	56~91	n = 1075 (AD = 390；MCI = 370；HC=315) 测试集 = 107 训练集 = 968	横断面研究	ADNI数据库；sMRI数据	AD、MCI与HC诊断	准确率：多分类：94.31% MCI/HC：92.15% AUC = 0.96	模型：CNN+自注意力机制优化器：AdamW+AdamWGC
Suk等^[23] (2014)	美国	55~89	n = 398 (AD = 93；MCI = 204；NC = 101) 10折交叉验证训练集 = 约358 测试集 = 约40	横断面研究	ADNI数据库；基线的T₁加权sMRI数据、¹⁸F-FDG-PET数据	AD、MCI与NC诊断 MCI-C与MCI-NC区分	MCI/NC：准确率85.67% AUC = 0.8808 MCI-C/MCI-NC：准确率75.92% AUC = 0.7466	模型：DBM
Yoon等^[24] (2024)	韩国	ADNI1(超分辨率) AD：57~91 MCI：60~87 NC：56~88 ADNI1(分类) AD：55~88 MCI：55~89 NC：62~86 ADNI3(分类) AD：56~90 MCI：55~91 NC：52~91	ADNI1(超分辨率) n₁ = 170 (AD = 38；MCI = 54；NC = 78) ADNI1(分类) n₂ = 403 (AD = 95；MCI = 206；NC = 102) ADNI3(分类) n₃ = 1014(AD = 122；MCI = 387；NC = 605) AIBL(外部验证)：n = 116	横断面研究	ADNI数据库和AIBL数据库；1.5 T和3 T的MRI数据	AD与MCI诊断	MCI/NC：准确率85.2% AUC = 0.847 AD/MCI：准确率81.7% AUC=0.813	模型：CNN、diffusion model-based generative AI
Kim等^[25] (2018)	韩国	55~90	n = 202 (HC = 52；MCI = 99；AD = 51) 采用10折交叉验证训练集 = 181 测试集 = 21	横断面研究	ADNI数据库；基线T₁加权sMRI数据、¹⁸F-FDG-PET数据、CSF数据	AD、MCI与HC诊断	MCI/HC：准确率87.09%	模型：MSH-ELM
Suk等^[26] (2016)	韩国、新加坡、美国	ADNI2 MCI：73.9 NC：73.8 In-house MCI：75.0 NC：72.9	ADNI2 n₁ = 62(MCI = 31；NC = 31) In-house n₂ = 37(MCI = 12；NC = 25) 留一法交叉验证	横断面研究	ADNI2数据集和In-house数据集；rs-fMRI数据	MCI与NC诊断	ADNI2：准确率72.58% In-house：准确率81.08%	模型：DAE+HMM
Liu等^[27] (2020)	中国	AD：75.9±6.8 MCI：75.2±7.3 NC：75.9±5.0	n = 449 (AD = 97；MCI = 233；NC = 119) 5折交叉验证训练集 = 89 测试集 = 360	横断面研究	ADNI数据库；基线的T₁加权sMRI扫描数据	AD、MCI与NC诊断	MCI/NC：准确率76.2%，AUC = 0.775	模型：CNN 优化器：Adam
Rashid等^[28] (2023)	印度	未说明	n = 6765 (AD = 1365；MCI = 3100；CN = 2500) 训练集：80% 测试集：20%	横断面研究	ADNI、AIBL、OASIS数据库；2D MRI数据	AD、MCI与NC诊断	MCI/AD：准确率98.16% CN/MCI/AD：准确率97.80%	模型：Biceph-net轻量级框架优化器：Adam
Lee等^[29](2021)	韩国	CN：62~89 平均71.1 eMCI：56~92 平均75.2	n = 101 (eMCI = 53；NC = 48) 10折交叉验证	横断面研究	ADNI数据库；rs-fMRI数据	eMCI与NC诊断	准确率74.42% AUC = 0.7438	模型：CNN+强化学习+GCN 优化器：Adam
Etminani等^[30](2022)	多国合作	ADNI 男性56~92，平均76.7；女性56~96，平均76.2 EDLB 男性48~91，平均72.7；女性50~86，平均72.9	n = 757 (AD = 200；MCI-AD = 200；DLB = 157；NC = 200) 训练集 = 684 测试集 = 73	横断面研究	ADNI数据库和EDLB数据库；¹⁸F-FDG PET扫描数据	AD、DLB、MCI-AD与NC诊断	DLB：AUC = 0.962 MCI-AD：AUC = 0.714 NC：AUC = 0.947	模型：3D CNN 优化器：Adadelta

Table 2

References 39

[1]	SPRUNG J, ROBERTS R O, WEINGARTEN T N, et al. Postoperative delirium in elderly patients is associated with subsequent cognitive impairment[J]. Br J Anaesth, 2017, 119(2): 316-323. doi: 10.1093/bja/aex130 pmid: 28854531
[2]	STEPHAN B C M, COCHRANE L, KAFADAR A H, et al. Population attributable fractions of modifiable risk factors for dementia: a systematic review and meta-analysis[J]. Lancet Healthy Longev, 2024, 5(6): e406-e421.
[3]	LEHMANN S, SCHRAEN-MASCHKE S, VIDAL J S, et al. Plasma phosphorylated tau 181 predicts amyloid status and conversion to dementia stage dependent on renal function[J]. J Neurol Neurosurg Psychiatry, 2023, 94(6): 411-419. doi: 10.1136/jnnp-2022-330540 pmid: 37012068
[4]	LEDUC V, DE BEAUMONT L, THÉROUX L, et al. HMGCR is a genetic modifier for risk, age of onset and MCI conversion to Alzheimer's disease in a three cohorts study[J]. Mol Psychiatry, 2015, 20(7): 867-873.
[5]	YIN C, IMMS P, CHENG M, et al. Anatomically interpretable deep learning of brain age captures domain-specific cognitive impairment[J]. Proc Natl Acad Sci U S A, 2023, 120(2): e2214634120.
[6]	HUCKANS M, HUTSON L, TWAMLEY E, et al. Efficacy of cognitive rehabilitation therapies for mild cognitive impairment (MCI) in older adults: working toward a theoretical model and evidence-based interventions[J]. Neuropsychol Rev, 2013, 23(1): 63-80. doi: 10.1007/s11065-013-9230-9 pmid: 23471631
[7]	METZLER-BADDELEY C, HUNT S, JONES D K, et al. Temporal association tracts and the breakdown of episodic memory in mild cognitive impairment[J]. Neurology, 2012, 79(23): 2233-2240.
[8]	LIU Z, ZHANG L, BAI L, et al. Repetitive transcranial magnetic stimulation and Tai Chi Chuan for older adults with sleep disorders and mild cognitive impairment: a randomized clinical trial[J]. JAMA Netw Open, 2025, 8(1): e2454307.
[9]	MASSOUD F, BELLEVILLE S, BERGMAN H, et al. Mild cognitive impairment and cognitive impairment, no dementia: Part B, therapy[J]. Alzheimers Dement, 2007, 3(4): 283-291. doi: 10.1016/j.jalz.2007.07.002 pmid: 19595949
[10]	COGNÉ É, POSTUMA R B, CHASLES M J, et al. Montreal Cognitive Assessment and the Clock Drawing Test to identify MCI and predict dementia in isolated REM sleep behavior disorder[J]. Neurology, 2024, 102(4): e208020.
[11]	ZHANG L, WANG L, GAO J, et al. Deep fusion of brain structure-function in mild cognitive impairment[J]. Med Image Anal, 2021, 72: 102082.
[12]	GROOT C, SMITH R, COLLIJ L E, et al. Tau positron emission tomography for predicting dementia in individuals with mild cognitive impairment[J]. JAMA Neurol, 2024, 81(8): 845-856. doi: 10.1001/jamaneurol.2024.1612 pmid: 38857029
[13]	FRIZZELL T O, GLASHUTTER M, LIU C C, et al. Artificial intelligence in brain MRI analysis of Alzheimer's disease over the past 12 years: a systematic review[J]. Ageing Res Rev, 2022, 77: 101614.
[14]	ZHU W, SUN L, HUANG J, et al. Dual attention multi-instance deep learning for Alzheimer's disease diagnosis with structural MRI[J]. IEEE Trans Med Imaging, 2021, 40(9): 2354-2366.
[15]	JO T, KIM J, BICE P, et al. Circular-SWAT for deep learning based diagnostic classification of Alzheimer's disease: application to metabolome data[J]. EBioMedicine, 2023, 97: 104820.
[16]	HWANG D, KIM K Y, KANG S K, et al. Improving the accuracy of simultaneously reconstructed activity and attenuation maps using deep learning[J]. J Nucl Med, 2018, 59(10): 1624-1629.
[17]	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 Trans Pattern Anal Mach Intell, 2020, 42(4): 880-893.
[18]	TRICCO A C, LILLIE E, ZARIN W, et al. PRISMA extension for scoping reviews (PRISMA-ScR): checklist and explanation[J]. Ann Intern Med, 2018, 169(7): 467-473. doi: 10.7326/M18-0850 pmid: 30178033
[19]	周英风, 顾莺, 胡雁, 等. JBI循证卫生保健中心对关于不同类型研究的质量评价工具:患病率及分析性横断面研究的质量评价[J]. 护士进修杂志, 2018, 33(3): 219-221.
	ZHOU Y F, GU Y, HU Y, et al. The Joanna Briggs Institute critical appraisal tools for use in systematic review: prevalence study and analytical cross sectional study[J]. J Nurs Train, 2018, 33(3): 219-221.
[20]	CIARMIELLO A, GIOVANNINI E, PASTORINO S, et al. Machine learning model to predict diagnosis of mild cognitive impairment by using radiomic and amyloid brain PET[J]. Clin Nucl Med, 2023, 48(1): 1-7.
[21]	POURRAMEZAN FARD A, MAHOOR M H, ALSUHAIBANI M, et al. Linguistic-based mild cognitive impairment detection using informative loss[J]. Comput Biol Med, 2024, 176: 108606.
[22]	KHATRI U, KWON G R. Diagnosis of Alzheimer's disease via optimized lightweight convolution-attention and structural MRI[J]. Comput Biol Med, 2024, 171: 108116.
[23]	SUK H I, LEE S W, SHEN D. Hierarchical feature representation and multimodal fusion with deep learning for AD/MCI diagnosis[J]. NeuroImage, 2014, 101: 569-582.
[24]	YOON D, MYONG Y, KIM Y G, et al. Latent diffusion model-based MRI superresolution enhances mild cognitive impairment prognostication and Alzheimer's disease classification[J]. NeuroImage, 2024, 296: 120663.
[25]	KIM J, LEE B. Identification of Alzheimer's disease and mild cognitive impairment using multimodal sparse hierarchical extreme learning machine[J]. Hum Brain Mapp, 2018, 39(9): 3728-3741. doi: 10.1002/hbm.24207 pmid: 29736986
[26]	SUK H I, WEE C Y, LEE S W, et al. State-space model with deep learning for functional dynamics estimation in resting-state fMRI[J]. NeuroImage, 2016, 129: 292-307.
[27]	LIU M, 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.
[28]	RASHID A H, GUPTA A, GUPTA J, et al. Biceph-Net: a robust and lightweight framework for the diagnosis of Alzheimer's disease using 2D-MRI scans and deep similarity learning[J]. IEEE J Biomed Health Inform, 2023, 27(3): 1205-1213.
[29]	LEE J, KO W, KANG E, et al. A unified framework for personalized regions selection and functional relation modeling for early MCI identification[J]. NeuroImage, 2021, 236: 118048.
[30]	ETMINANI K, SOLIMAN A, DAVIDSSON A, et al. A 3D deep learning model to predict the diagnosis of dementia with Lewy bodies, Alzheimer's disease, and mild cognitive impairment using brain ¹⁸F-FDG PET[J]. Eur J Nucl Med Mol Imaging, 2022, 49(2): 563-584.
[31]	KAM T E, ZHANG H, JIAO Z, et al. Deep learning of static and dynamic brain functional networks for early MCI detection[J]. IEEE Trans Med Imaging, 2020, 39(2): 478-487.
[32]	ZHOU T, THUNG K, ZHU X, et al. Effective feature learning and fusion of multimodality data using stage‐wise deep neural network for dementia diagnosis[J]. Hum Brain Mapp, 2018, 40(3): 1001-1016.
[33]	GU D, LV X, SHI C, et al. A stable and scalable digital composite neurocognitive test for early dementia screening based on machine learning: model development and validation study[J]. J Med Internet Res, 2023, 25: e49147.
[34]	GHORAANI B, BOETTCHER L N, HSSAYENI M D, et al. Detection of mild cognitive impairment and Alzheimer's disease using Dual-task Gait Assessments and machine learning[J]. Biomed Signal Process Control, 2021, 64: 102249.
[35]	YUAN C, LINN K A, HUBBARD R A. Algorithmic fairness of machine learning models for Alzheimer disease progression[J]. JAMA Netw Open, 2023, 6(11): e2342203.
[36]	OSMAN Y B M, LI C, HUANG W, et al. Sparse annotation learning for dense volumetric MR image segmentation with uncertainty estimation[J]. Phys Med Biol, 2023, 69(1).
[37]	WEN J, THIBEAU-SUTRE E, DIAZ-MELO M, et al. Convolutional neural networks for classification of Alzheimer's disease: overview and reproducible evaluation[J]. Med Image Anal, 2020, 63: 101694.
[38]	SUK H I, LEE S W, SHEN D, et al. Deep ensemble learning of sparse regression models for brain disease diagnosis[J]. Med Image Anal, 2017, 37: 101-113.
[39]	LIANG G, XING X, LIU L, et al. Alzheimer's disease classification using 2D convolutional neural networks[J]. Annu Int Conf IEEE Eng Med Biol Soc, 2021, 2021: 3008-3012. doi: 10.1109/EMBC46164.2021.9629587 pmid: 34891877

纳入文献	1	2	3	4	5	6	7	8	总分
Ciarmiello等^[20] (2023)	√	√	√	√		√	√	√	7
Pourramezan Fard等^[21](2024)	√	√	√	√			√	√	6
Khatri等^[22](2024)	√	√	√	√			√	√	6
Suk等^[23](2016)	√	√	√	√			√	√	6
Yoon等^[24](2024)	√	√	√	√		√	√	√	7
Kim等^[25](2018)	√	√	√	√		√	√	√	7
Suk等^[26](2014)	√	√	√	√			√	√	6
Liu等^[27](2020)	√	√	√	√	√		√	√	7
Rashid等^[28](2023)	√	√	√	√			√	√	6
Lee等^[29](2021)	√	√	√	√			√	√	6
Etminani等^[30](2022)	√	√	√	√			√	√	6

Deep learning for diagnosis of mild cognitive impairment in older adults: a scoping review

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 3

References 39

Related Articles 15

Metrics

Comments

Recommended 0

[1]	CHEN Junli, ZHANG Han, SHI Zhixue, LIU Ya, ZHAO Yingzhe, DONG Zhiwei, JI Lihong, LI Haiyan, CHEN Fangfang, WANG Chunping, MA Anning, JING Qi. Development of a community toolkit for identifying and managing mild cognitive impairment among older adults [J]. Chinese Journal of Rehabilitation Theory and Practice, 2025, 31(6): 692-702.
[2]	PAN Yi, HOU Shuanglong, WEN Xiaoni. Transcutaneous vagus nerve stimulation for executive function in adults: a scoping review [J]. Chinese Journal of Rehabilitation Theory and Practice, 2025, 31(5): 520-528.
[3]	XU Xiaohong, ZHONG Xiaoke, SUN Siyi, ZHANG Qi, CHENG Huaichun. Psychological benefits of art therapy for older adults with mild cognitive impairment: a systematic review [J]. Chinese Journal of Rehabilitation Theory and Practice, 2025, 31(2): 165-171.
[4]	WANG Xiaoxiao, QU Lu, WANG Jiaxin, ZHAO Meiping, LI Muzi, LIU Qiaoyun. Structure, content and psychometric properties of assessment tools for interoceptive function: a scoping review based on COSMIN and ICF [J]. Chinese Journal of Rehabilitation Theory and Practice, 2025, 31(2): 172-183.
[5]	JIANG Changhao, JIANG Xianxin, HUANG Chen, ZHONG Xiaoke. Application of artificial intelligence in diagnosis and intervention in sleep disorder for older adults: a scoping review using ICF [J]. Chinese Journal of Rehabilitation Theory and Practice, 2024, 30(8): 922-929.
[6]	CAO Yaping, LI Ju, HUANG Minghao, LI Zhongcheng, LANG Jian. Risk factors, health and rehabilitation intervention strategies for low back pain in adult golfers: a scoping review [J]. Chinese Journal of Rehabilitation Theory and Practice, 2024, 30(6): 657-664.
[7]	WEN Nana, ZHANG Xinhui, LONG Qing, WANG Yuhao, YU Qunping, ZHANG Hanchun, ZHENG Guohua. Predictive value of gait and balance on frailty in community-dwelling older adults in Shanghai, China [J]. Chinese Journal of Rehabilitation Theory and Practice, 2024, 30(6): 731-736.
[8]	SU Rufeng, ZHONG Xiaoke, GAO Xiaoyan, JIANG Changhao. Application of artificial intelligence in anxiety and depression among children and adolescents: a scoping review [J]. Chinese Journal of Rehabilitation Theory and Practice, 2024, 30(5): 513-519.
[9]	SU Zhaoyin, KANG Weihan, LIU Yatao, LÜ Yuan, Michael NERLICH. Relationship between physical activity levels and stroke risk among middle-aged and older adults in China based on CHARLS data [J]. Chinese Journal of Rehabilitation Theory and Practice, 2024, 30(4): 449-453.
[10]	LIU Hui, YIN Hang, HU Chenghong, JIA Shaohui, FENG Yewenzhe, HU Qingkui. Content and psychometric properties of functional measurements in patients with sarcopenia based on ICF [J]. Chinese Journal of Rehabilitation Theory and Practice, 2024, 30(3): 273-280.
[11]	WU Jinlong, REN Zhanbing, YI Zizhen, PENG Li. Characteristics of interbrain synchrony during interactions among individuals with autism spectrum disorder: a scoping review [J]. Chinese Journal of Rehabilitation Theory and Practice, 2024, 30(2): 168-175.
[12]	MA Xiaochen, LI Shufan, JIA Shuqi, LIU Cong, ZHANG Zhenyu, HAN Dongyang. Relationship of physical activity and cognitive function to sleep quality in older adults with cognitive impairment: a mediating effect of resting EEG [J]. Chinese Journal of Rehabilitation Theory and Practice, 2024, 30(12): 1442-1451.
[13]	HUANG Zhaoxin, TANG Letian, XIAO Xiaofei, LOU Yantao. Effect of orthopedic insoles on lower extremities function in children and adolescents with flexible flatfoot: a scoping review [J]. Chinese Journal of Rehabilitation Theory and Practice, 2024, 30(11): 1343-1350.
[14]	SHAO Weiting, LEI Jianghua. Effect of response interruption and redirection as a behavioral intervention on vocal stereotypy in children with autism spectrum disorder: a scoping review [J]. 《Chinese Journal of Rehabilitation Theory and Practice》, 2024, 30(1): 10-20.
[15]	WANG Hangyu, GE Keke, FAN Yonghong, DU Lilu, ZOU Min, FENG Lei. Effect of active music therapy on cognitive function for older adults with cognitive impairment: a systematic review based on ICD-11 and ICF [J]. 《Chinese Journal of Rehabilitation Theory and Practice》, 2024, 30(1): 36-43.