EUVIMEDOpdag EUVIMED
Systematische ÜbersichtOpen access

Multimodal AI for Alzheimer Disease Diagnosis: Systematic Review of Datasets, Models, and Modalities

Multimodale KI zur Alzheimer-Diagnose: systematische Übersicht zu Datensätzen, Modellen und Modalitäten

Ziwen Yu, Anthony Mulholland, Tianyan Huang, Qiang Liu

Journal of Medical Internet Research · 2026-03-25 · 28 · e85414-e85414

DOI: 10.2196/85414 ↗

Auf einen Blick

Die Übersicht von 66 Studien zeigt: Multimodale KI-Modelle schnitten bei AD-Diagnose, Prognose und Risikovorhersage durchweg besser ab als single-modale Baselines.

Fragestellung

Wie leistungsfähig sind multimodale KI-Modelle für die Diagnose, Prognose und Risikovorhersage bei Alzheimer-Krankheit?

Methode

Systematische Übersicht nach PRISMA 2020; Suche in PubMed, IEEE Xplore, Scopus, ACM Digital Library, Cochrane und arXiv; letzte Suche der Datensätze am 15. November 2025; Bewertung mit QUADAS-2.

Ergebnisse

66 Studien. ADNI-basierte Diagnose: mittlere Accuracy 92,5 % (SD 3,8 %). MCI-Conversion: mittlere AUC 0,922 (SD 0,045). UK Biobank: mittlere AUC 0,84 (SD 0,056). DementiaBank: mittlere AUC 0,813 (SD 0,042).

Einschränkungen

Hohe Heterogenität von Datensätzen, Outcome-Definitionen und Validierung; häufiges Verzerrungsrisiko; Generalisierbarkeit eingeschränkt, besonders bei kleinen Single-Center-Datensätzen.

Schlussfolgerung der Autor:innen

Die Autor:innen berichten, dass multimodale KI-Modelle multimodalen, biologischen, klinischen und Verhaltensinformationen für AD-Diagnose, Prognose und Risikovorhersage besser nutzen als single-modale Modelle.

Kernaussagen

  • 66 Studien wurden eingeschlossen.
  • Multimodale Modelle waren den single-modalen Baselines überlegen.
  • Generalisierbarkeit durch Heterogenität und Bias begrenzt.

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

At a glance

Systematische Übersicht in Journal of Medical Internet Research, published online on 2026-03-25. Written by Ziwen Yu and 3 co-authors. Institution: School of Engineering Mathematics and Technology, University of Bristol, Tankard's Close, Ada Lovelace Building, Bristol, BS8 1TW, United Kingdom, 44…. Major MEDLINE topics: Alzheimer Disease, Artificial Intelligence. Cited 12 times so far (Crossref); 165 references. The full text is freely available in PubMed Central.

Auf einen BlickAbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

Background

Early detection of Alzheimer disease (AD) is essential for timely intervention; yet, diagnostic performance varies widely across modalities and datasets. Recent multimodal artificial intelligence (AI) models have made significant progress, but the evidence base remains fragmented due to heterogeneous datasets, modeling frameworks, and reporting quality.

Objective

This systematic review aimed to analyze studies on multimodal AI models for AD diagnosis, prognosis, and risk prediction over 5 years. We evaluated dataset characteristics, modality combinations, modeling strategies, performance metrics, and methodological limitations. We further discuss real-world implications and translational pathways.

Methods

Following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 guidelines, we systematically searched PubMed, IEEE Xplore, Scopus, ACM Digital Library, Cochrane, and arXiv, with the final datasets last searched on November 15, 2025. Studies applying multimodal machine learning or deep learning to AD, mild cognitive impairment, and dementia outcomes were included, whereas studies using a single modality or lacking sufficient methodological detail were excluded. QUADAS-2 (Revised Quality Assessment of Diagnostic Accuracy Studies tool) assessed risk of bias. Extracted performance results were synthesized across 4 major multimodal dataset families.

Results

A total of 66 studies met the inclusion criteria. Across datasets, multimodal models consistently outperformed single-modal baselines. Alzheimer’s Disease Neuroimaging Initiative–based diagnosis achieved an average accuracy of 92.5% (SD 3.8%), while mild cognitive impairment–conversion models achieved an average area under the curve (AUC) of 0.922 (SD 0.045), and several fusion architectures reported AUCs above 0.95. In contrast, UK Biobank risk-prediction studies reported an average AUC of 0.84 (SD 0.056), and this reflects performance in large, population-based datasets. DementiaBank speech-language studies achieved an average AUC of 0.813 (SD 0.042), and cross-lingual AD detection achieved an accuracy of 77% (SD 6.5%). Self-collected multimodal datasets demonstrated average accuracies around 96% (SD 2.4%), but their generalizability is limited due to small sample sizes and single-center designs.

Conclusions

This systematic review demonstrates that multimodal AI models consistently outperform single-modal models for AD diagnosis, prognosis, and risk prediction by integrating complementary biological, clinical, and behavioral information. Unlike prior reviews, this review provides a unified synthesis across heterogeneous clinical, imaging, genetic, and linguistic datasets, enabling cross-domain comparison of modeling strategies and performance. However, the generalizability of reported performance was limited due to substantial heterogeneity in dataset composition, outcome definitions, and validation, and prevalent risks of bias. By evaluating these factors, this review clarifies where current evidence is robust and where caution is warranted. The findings highlight the need for standardized multimodal benchmarks, transparent evaluation protocols, and clinically grounded model design to enable reliable real-world deployment. Overall, this work advances the field by framing multimodal AI not only as a performance-driven tool but also as a translational framework for equitable, interpretable, and scalable AD diagnosis.

Subjects

MeSH terms (MEDLINE)

Alzheimer DiseaseArtificial IntelligenceHumansMachine Learning

Author keywords

alzheimer’s disease, neurodegenerative diseasecomputer-aided diagnosisdeep learningearly diagnosismachine learningmultimodal datasetmultimodal fusion

Authors and institutions

  • School of Engineering Mathematics and Technology, University of Bristol, Tankard's Close, Ada Lovelace Building, Bristol, BS8 1TW, United Kingdom, 44 01173746653.
  • Medical Physics and Biomedical Engineering, University College London, London, United Kingdom.

Cite

APA

Yu, Z., Mulholland, A., Huang, T., & Liu, Q. (2026). Multimodal AI for Alzheimer Disease Diagnosis: Systematic Review of Datasets, Models, and Modalities. Journal of Medical Internet Research, 28, e85414-e85414. https://doi.org/10.2196/85414

Vancouver

Yu Z, Mulholland A, Huang T, Liu Q. Multimodal AI for Alzheimer Disease Diagnosis: Systematic Review of Datasets, Models, and Modalities. Journal of Medical Internet Research. 2026;28:e85414-e85414. doi:10.2196/85414

BibTeX
@article{Yu2026,
  author = {Yu, Ziwen and Mulholland, Anthony and Huang, Tianyan and Liu, Qiang},
  title = {Multimodal AI for Alzheimer Disease Diagnosis: Systematic Review of Datasets, Models, and Modalities},
  journal = {Journal of Medical Internet Research},
  year = {2026},
  volume = {28},
  pages = {e85414--e85414},
  doi = {10.2196/85414}
}
Start for free →