慢性病防治专栏

多模态人工智能在阿尔茨海默病早期诊断中的应用:现状、挑战与临床转化路径

  • 仇晓蕊 ,
  • 张译丹 ,
  • 杨国锋
展开
  • 河北医科大学第二医院老年病科 (河北 石家庄 050000 )

收稿日期: 2025-11-24

  网络出版日期: 2026-04-29

基金资助

国家自然科学基金项目(82471453)

Applications of multimodal artificial intelligence in early Alzheimer's diagnosis: A review of current trends, obstacles, and clinical translational strategies

  • Xiaorui QIU ,
  • Yidan ZHANG ,
  • Guofeng YANG
Expand
  • Department of Geriatrics,the Second Hospital of Hebei Medical University,Shijiazhuang 050000,Hebei,China

Received date: 2025-11-24

  Online published: 2026-04-29

摘要

阿尔茨海默病(Alzheimer's disease, AD)是一种中枢神经系统退行性疾病,目前治疗手段有限,其早期诊断对于延缓病程发展至关重要。然而,传统诊断方法在早期或不典型病例中仍存在主观性强、准确性不足等局限性。在此背景下,本文旨在探讨多模态人工智能(artificial intelligence, AI)在AD早期诊断中的应用、挑战与临床转化路径。通过整合神经影像、生物标志物、临床评估等多维度数据,多模态AI能够挖掘深层的疾病模式,为AD的早期识别与鉴别提供了更敏感、更客观的新工具。然而,AD病理机制的复杂性、数据的异质性以及模型可解释性不足等问题,仍制约着AI技术的临床转化。未来需进一步推动多学科合作,深化可解释性研究,并探索其与临床工作流程的有效融合,从而最终实现AD的精准诊疗目标。

本文引用格式

仇晓蕊 , 张译丹 , 杨国锋 . 多模态人工智能在阿尔茨海默病早期诊断中的应用:现状、挑战与临床转化路径[J]. 实用医学杂志, 2026 , 42(9) : 1579 -1585 . DOI: 10.3969/j.issn.1006-5725.2026.09.012

Abstract

Alzheimer's disease (AD) is a progressive neurodegenerative disorder with currently limited therapeutic options. Early diagnosis is pivotal for delaying disease progression; however, traditional diagnostic methods remain constrained by significant subjectivity and insufficient accuracy, particularly in early-stage or atypical cases. Against this backdrop, this review explores the applications, challenges, and clinical translation pathways of multimodal artificial intelligence (AI) in the early diagnosis of AD. By integrating multidimensional data-including neuroimaging, biomarkers, and clinical assessments-multimodal AI can uncover latent disease patterns, offering a novel, highly sensitive, and objective tool for the early identification and differential diagnosis of AD. Nevertheless, several barriers hinder its clinical translation, including the complexity of AD pathophysiology, data heterogeneity, and the lack of model interpretability. Future efforts must prioritize multidisciplinary collaboration, advance research into explainable AI, and facilitate seamless integration into clinical workflows. Ultimately, these strides will pave the way for precision diagnosis and personalized treatment of AD.

参考文献

[1] JACK C R J, ANDREWS J S, BEACH T G, et al. Revised criteria for diagnosis and staging of Alzheimer's disease: Alzheimer's Association Workgroup[J].Alzheimers Dement, 2024, 20(8): 5143-5169. doi:10.1002/alz.13859 .
[2] 申杰, 徐桂华. 阿尔茨海默病与血脑屏障的相关性研究进展[J]. 实用医学杂志, 2024, 40 (11): 1602-1606. doi: 10.3969/j.issn.1006-5725.2024.11.023 .
[3] CASTELLANI R J. Biomarker-only diagnosis of Alzheimer's disease: Not ready for prime time[J].J Alzheimers Dis, 2025,103(2): 356-357. doi:10.1177/13872877241301805 .
[4] QIU S, MILLER M I, JOSHI P S, et al. Multimodal deep learning for Alzheimer's disease dementia assessment[J]. Nat Commun, 2022, 13 (1): 3404. doi: 10.1038/s41467-022-31037-5 .
[5] MONSOUR R, DUTTA M, MOHAMED A Z, et al. Neuroimaging in the Era of Artificial Intelligence: Current Applications[J]. Fed Pract, 2022, 39 (): S14-S20. doi: 10.12788/fp.0231 .
[6] MORA-RUBIO A, BRAVO-ORTIZ M A, QUINONES ARREDONDO S, et al. Classification of Alzheimer's disease stages from magnetic resonance images using deep learning[J].PeerJ Comput Sci, 2023, 9: e1490. doi:10.7717/peerj-cs.1490 .
[7] XU M, OUYANG Y, YUAN Z. Deep Learning Aided Neuroimaging and Brain Regulation[J]. Sensors (Basel), 2023, 23 (11) :4993. doi: 10.3390/s23114993 .
[8] HNILICOVA P, KANTOROVA E, SUTOVSKY S, et al. Imaging Methods Applicable in the Diagnostics of Alzheimer's Disease, Considering the Involvement of Insulin Resistance[J]. Int J Mol Sci, 2023, 24 (4) :3325. doi: 10.3390/ijms24043325 .
[9] YEN C, LIN C L, CHIANG M C. Exploring the Frontiers of Neuroimaging: A Review of Recent Advances in Understanding Brain Functioning and Disorders[J]. Life (Basel), 2023, 13 (7) :1472. doi: 10.3390/life13071472 .
[10] 韩英妹, 李一杰, 张衡, 等. 基于MRI分析阿尔茨海默病大尺度脑网络研究进展[J]. 实用医学杂志, 2024, 40(4): 575-579. doi: 10.3969/j.issn.1006-5725.2024.04.024 .
[11] TAE W S, HAM B J, PYUN S B, et al. Current Clinical Applications of Structural MRI in Neurological Disorders[J]. J Clin Neurol, 2025, 21(4): 277-293. doi:10.3988/jcn.2025.0185 .
[12] ZHAO G, ZHANG H, XU Y, et al. Research on magnetic resonance imaging in diagnosis of Alzheimer's disease[J]. Eur J Med Res, 2024, 29(1): 632. doi:10.1186/s40001-024-02172-0 .
[13] LEW C O, ZHOU L, MAZUROWSKI M A, et al. MRI-based Deep Learning Assessment of Amyloid, Tau, and Neurodegeneration Biomarker Status across the Alzheimer Disease Spectrum[J]. Radiology, 2023, 309 (1): e222441. doi: 10.1148/radiol.222441 .
[14] HWANG U, KIM S W, JUNG D, et al. Real-world prediction of preclinical Alzheimer's disease with a deep generative model[J]. Artif Intell Med, 2023, 144: 102654. doi: 10.1016/j.artmed.2023.102654 .
[15] QIU A, XU L, LIU C. Predicting diagnosis 4 years prior to Alzheimer's disease incident[J]. Neuroimage Clin, 2022, 34: 102993. doi: 10.1016/j.nicl.2022.102993 .
[16] PAN D, ZENG A, YANG B, et al. Deep Learning for Brain MRI Confirms Patterned Pathological Progression in Alzheimer's Disease[J]. Adv Sci (Weinh), 2023, 10 (6): e2204717. doi: 10.1002/advs.202204717 .
[17] LI Y, YI P, JIN M, et al. A radiomics model predicts progression from mild cognitive impairment to alzheimer's disease using structural MRI[J]. Sci Rep, 2025, 15 (1): 35679. doi: 10.1038/s41598-025-19586-3 .
[18] SHETTY J, SHENOY M K, KOLEKAR S V, et al. Early detection of Alzheimer's disease progression: Comparative evaluation of deep learning models[J]. Sci Rep, 2025, 15 (1): 43242. doi: 10.1038/s41598-025-27360-8 .
[19] LIU J, XU Y, LIU Y, et al. Attention-Guided 3D CNN With Lesion Feature Selection for Early Alzheimer's Disease Prediction Using Longitudinal sMRI[J]. IEEE J Biomed Health Inform, 2025, 29 (1): 324-332. doi: 10.1109/jbhi.2024.3482001 .
[20] BAI J, ZHANG Z, YIN Y, et al. LGG-NeXt: A Next Generation CNN and Transformer Hybrid Model for the Diagnosis of Alzheimer's Disease Using 2D Structural MRI[J]. IEEE J Biomed Health Inform, 2025, 29 (4): 2808-2818. doi: 10.1109/jbhi.2024.3495835 .
[21] NOH J H, KIM J H, YANG H D. Classification of Alzheimer's Progression Using fMRI Data[J]. Sensors (Basel), 2023, 23 (14) :6330. doi: 10.3390/s23146330 .
[22] ALAYBA A M, SENAN E M, ALSHUDUKHI J S. Enhancing early detection of Alzheimer's disease through hybrid models based on feature fusion of multi-CNN and handcrafted features[J]. Sci Rep, 2024, 14 (1): 31203. doi: 10.1038/s41598-024-82544-y .
[23] ZHANG S, WANG Q, WEI M, et al. FROG: A Fine-Grained Spatiotemporal Graph Neural Network With Self-Supervised Guidance for Early Diagnosis of Alzheimer's Disease[J]. IEEE J Biomed Health Inform, 2025, 29(8):5900-5911. doi:10.1109/JBHI.2025.3552638 .
[24] HIGUCHI M, TAGAI K, TAKAHATA K, et al. Advances in PET imaging of protein aggregates associated with neurodegenerative disease[J]. Nat Rev Neurol, 2025, 21(9): 506-522. doi:10.1038/s41582-025-01126-2 .
[25] BORCHERT R J, AZEVEDO T, BADHWAR A, et al. Artificial intelligence for diagnostic and prognostic neuroimaging in dementia: A systematic review[J]. Alzheimers Dement, 2023, 19 (12): 5885-5904. doi: 10.1002/alz.13412 .
[26] OSSENKOPPELE R, KANT R, HANSSON O. Tau biomarkers in Alzheimer's disease: Towards implementation in clinical practice and trials[J]. Lancet Neurol, 2022, 21 (8): 726-734. doi: 10.1016/s1474-4422(22)00168-5 .
[27] SANAAT A, BOCCALINI C, MATHOUX G, et al. A deep learning model for generating [(18)F]FDG PET Images from early-phase [(18)F]Florbetapir and [(18)F]Flutemetamol PET images[J]. Eur J Nucl Med Mol Imaging, 2024, 51 (12): 3518-3531. doi: 10.1007/s00259-024-06755-1 .
[28] GAO X, SHI F, SHEN D, et al. Multimodal transformer network for incomplete image generation and diagnosis of Alzheimer's disease[J]. Comput Med Imaging Graph, 2023, 110: 102303. doi: 10.1016/j.compmedimag.2023.102303 .
[29] YOO H B, KANG S K, SHIN S A, et al. Artificial Intelligence-Powered Quantification of Flortaucipir PET for Detecting Tau Pathology[J]. J Nucl Med, 2025, 66(11): 1827-1833. doi:10.2967/jnumed.125.269636 .
[30] RAO Y L, GANARAJA B, MURLIMANJU B V, et al. Hippocampus and its involvement in Alzheimer's disease: A review[J]. 3 Biotech, 2022, 12 (2): 55. doi: 10.1007/s13205-022-03123-4 .
[31] HU W, DU Q, Wei L, et al. Novel multi-task learning for Alzheimer's stage classification using hippocampal MRI segmentation, feature fusion, and nomogram modeling[J]. Eur J Med Res, 2025, 30(1): 910. doi:10.1186/s40001-025-03204-z .
[32] IBRAHIM N S N, SUPPIAH S, IBRAHIM B, et al. Comparing deep learning CNN method with traditional MRI-based hippocampal segmentation and volumetry for early Alzheimer's disease diagnosis across diverse populations[J]. Sci Rep, 2025, 16(1): 58. doi:10.1038/s41598-025-29366-8 .
[33] YIN T T, CAO M H, YU J C, et al. T1-Weighted Imaging-Based Hippocampal Radiomics in the Diagnosis of Alzheimer's Disease[J]. Acad Radiol, 2024, 31 (12): 5183-5192. doi: 10.1016/j.acra.2024.06.012 .
[34] DU Y, ZHANG S, FANG Y, et al. Radiomic Features of the Hippocampus for Diagnosing Early-Onset and Late-Onset Alzheimer's Disease[J]. Front Aging Neurosci, 2022, 13: 789099. doi:10.3389/fnagi.2021.789099 .
[35] ZUBAIR M, JAFFAR A, HUSSAIN S, et al. Alzheimer's disease classification using a hybrid deep learning approach with multi-layer U-net segmentation and XAI driven analysis[J]. PLoS One, 2025, 20(9): e0332572. doi:10.1371/journal.pone.0332572 .
[36] CONTADOR J, VARGAS-MARTINEZ A M, SANCHEZ-VALLE R, et al. Cost-effectiveness of Alzheimer's disease CSF biomarkers and amyloid-PET in early-onset cognitive impairment diagnosis[J]. Eur Arch Psychiatry Clin Neurosci, 2023, 273(1): 243-252. doi:10.1007/s00406-022-01439-z .
[37] SABBAGH M N, DECOURT B. Biomarker-based diagnosis of preclinical Alzheimer disease: time for the clinic?[J] Nat Rev Neurol, 2023, 19 (2): 71-72. doi: 10.1038/s41582-022-00767-x .
[38] ALEX J S R, ROSHINI R, MANEESHA G, et al. Enhanced detection of mild cognitive impairment in Alzheimer's disease: A hybrid model integrating dual biomarkers and advanced machine learning[J]. BMC Geriatr, 2025, 25 (1): 54. doi: 10.1186/s12877-025-05683-5 .
[39] BAYAT S, ROE C M, SCHINDLER S, et al. Everyday Driving and Plasma Biomarkers in Alzheimer's Disease: Leveraging Artificial Intelligence to Expand Our Diagnostic Toolkit[J]. J Alzheimers Dis, 2023, 92 (4): 1487-1497. doi: 10.3233/jad-221268 .
[40] AQEEL A, HASSAN A, KHAN M A, et al. A Long Short-Term Memory Biomarker-Based Prediction Framework for Alzheimer's Disease[J]. Sensors (Basel), 2022, 22 (4) :1475. doi: 10.3390/s22041475 .
[41] CHOULIARAS L, THOMAS A, MALPETTI M, et al. Differential levels of plasma biomarkers of neurodegeneration in Lewy body dementia, Alzheimer's disease, frontotemporal dementia and progressive supranuclear palsy[J]. J Neurol Neurosurg Psychiatry, 2022, 93 (6): 651-658. doi: 10.1136/jnnp-2021-327788 .
[42] CHIU S I, FAN L Y, LIN C H, et al. Machine Learning-Based Classification of Subjective Cognitive Decline, Mild Cognitive Impairment, and Alzheimer's Dementia Using Neuroimage and Plasma Biomarkers[J]. ACS Chem Neurosci, 2022, 13(23): 3263-3270. doi:10.1021/acschemneuro.2c00255 .
[43] DU L, LANGHOUGH R E, WILSON R E, et al. Longitudinal plasma phosphorylated-tau217 and other related biomarkers in a non-demented Alzheimer's risk-enhanced sample[J]. Alzheimers Dement, 2024, 20(9): 6183-6204. doi:10.1002/alz.14100 .
[44] ZUO Q, SHEN Y, ZHONG N, et al. Alzheimer's Disease Prediction via Brain Structural-Functional Deep Fusing Network[J]. IEEE Trans Neural Syst Rehabil Eng, 2023, 31: 4601-4612. doi: 10.1109/tnsre.2023.3333952 .
[45] CHEN Z, BI S, SHAN Y, et al. Multiparametric hippocampal signatures for early diagnosis of Alzheimer's disease using 18F-FDG PET/MRI Radiomics[J]. CNS Neurosci Ther, 2024, 30(4): e14539. doi:10.1111/cns.14539 .
[46] LEI B, ZHU Y, LIANG E, et al. Federated Domain Adaptation via Transformer for Multi-Site Alzheimer's Disease Diagnosis[J]. IEEE Trans Med Imaging, 2023, 42 (12): 3651-3664. doi: 10.1109/tmi.2023.3300725 .
[47] HU W, DU Q, WEI L, et al. Novel multi-task learning for Alzheimer's stage classification using hippocampal MRI segmentation, feature fusion, and nomogram modeling[J]. Eur J Med Res, 2025, 30(1): 910. doi: 10.1186/s40001-025-03204-z .
[48] HAIDER S A, BORNA S, GOMEZ-CABELLO C A, et al. The Algorithmic Divide: A Systematic Review on AI-Driven Racial Disparities in Healthcare[J]. J Racial Ethn Health Disparities, 2026, 13(1): 188-217. doi:10.1007/s40615-024-02237-0 .
[49] WANG C, WANG Z, HERRERO M T, et al. Early diagnosis of mild cognitive impairment and Alzheimer's disease using multimodal feature-based deep learning models in a Chinese elderly population[J]. Asian J Psychiatr, 2025, 111: 104632. doi:10.1016/j.ajp.2025.104632 .
[50] BEHESHTI I, ALBENSI B C, FREITAS A, et al. Advancements and challenges in using AI for biomarker detection in early Alzheimer's disease[J]. Drug Discov Today, 2025, 30(7): 104415. doi:10.1016/j.drudis.2025.104415 .
[51] KALE M, WANKHEDE N, PAWAR R, et al. AI-driven innovations in Alzheimer's disease: Integrating early diagnosis, personalized treatment, and prognostic modelling[J]. Ageing Res Rev, 2024, 101: 102497. doi:10.1016/j.arr.2024.102497 .
[52] 王立坤, 郝琦, 金炜涵, 等. 多组学与人工智能在预测和诊断结直肠癌肝转移中的应用[J]. 实用医学杂志, 2025, 41(7): 1070-1078. doi: 10.3969/j.issn.1006-5725.2025.07.021 .
[53] 胡高凯, 牛亚楠, 龚玉康, 等. 深度学习在腰椎疾病诊断、手术规划及术后预测中的应用研究进展[J]. 实用医学杂志, 2025, 41 (6): 921-928. doi : 10.3969/j.issn.1006-5725.2025.06.023 .
文章导航

/