EUVIMEDEUVIMED entdecken
ÜbersichtsarbeitOpen Access · CC BY-NC-ND 4.0

AI‐Integrated multimodal neuroimaging in Alzheimer's disease: toward biologically informed precision diagnostics

Jun-Su Kim, Sang Sun Han, So-Hyeon Yoo, Dong-Hyun Lee, Jaeho Jang, Hyun-Ju Lee, Jeong-Heon Song, Hyang-Sook Hoe iD

Alzheimer's & Dementia · 2026-11 · 22 (11)

DOI: 10.1002/alz.71923 ↗

Kurz gefasst

Übersichtsarbeit in der Zeitschrift Alzheimer's & Dementia, online erschienen am 10.11.2026. Verfasst von Jun-Su Kim und 7 weiteren Autor:innen. Beteiligte Einrichtung: AI-based Neurodevelopmental Diseases Digital Therapeutics Group, Korea Brain Research Institute (KBRI), Daegu, Republic of Korea. Hauptthemen laut MEDLINE: Alzheimer Disease, Artificial Intelligence, Neuroimaging, Multimodal Imaging. Die Arbeit nennt 185 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: the KBRI basic research program through KBRI funded by the Ministry of Science.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Alzheimer's disease (AD), which is characterized by the progressive accumulation of amyloid beta plaques and neurofibrillary tangles (NFTs), metabolic dysfunction, and neuroinflammation, is the most prevalent neurodegenerative disorder. The distinct pathological features of AD can be characterized in vivo via neuroimaging: Structural magnetic resonance imaging captures brain atrophy, while positron emission tomography detects Aβ/NFT deposition. The application of artificial intelligence (AI) can enhance the interpretation of neuroimaging data, but single‐modality AI provides only limited biological information. This limitation has motivated the development of multimodal approaches that integrate complementary imaging, molecular, and clinical data for diagnosis and prognosis. However, there has been no comprehensive examination of how modality‐specific neuroimaging information is incorporated into single‐modality AI and subsequently integrated within multimodal frameworks. This review addresses this gap by evaluating conventional neuroimaging, single‐modality AI, and multimodal AI as part of a unified framework, with particular emphasis on the added diagnostic and prognostic value of multimodal integration. We further identify key methodological and translational bottlenecks in developing clinically robust multimodal AI, including cross‐modal and cross‐cohort heterogeneity, model complexity, missing modalities, interpretability, and generalizability. Finally, we propose a translational roadmap emphasizing data harmonization, independent external validation, clinically meaningful explainability, and prospective evaluation. Ultimately, the clinical value of multimodal AI will depend on whether its added predictive value is reproducible across independent populations and translates into meaningful improvements in clinical decision‐making and patient outcomes.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Alzheimer DiseaseArtificial IntelligenceNeuroimagingMultimodal ImagingPrecision MedicineHumansBrainMagnetic Resonance ImagingPositron-Emission Tomography

Stichwörter der Autor:innen

Alzheimer's diseaseartificial intelligencemolecular imagingmultimodal neuroimagingneurodegenerative diseases

Autor:innen und Institutionen

  • AI-based Neurodevelopmental Diseases Digital Therapeutics Group, Korea Brain Research Institute (KBRI), Daegu, Republic of Korea.
  • Department of Neural Development and Disease, Korea Brain Research Institute (KBRI), Daegu, Republic of Korea.
  • Department of Brain and Cognitive Sciences, Daegu Gyeongbuk Institute of Science & Technology, Daegu, Republic of Korea.

ORCID-Profile: Hyang-Sook Hoe

Förderung

  • the KBRI basic research program through KBRI funded by the Ministry of Science

Zitieren

APA

Kim, J. S., Han, S. S., Yoo, S. H., Lee, D. H., Jang, J., Lee, H. J., Song, J. H., & Hoe, H. S. (2026). AI‐Integrated multimodal neuroimaging in Alzheimer's disease: toward biologically informed precision diagnostics. Alzheimer's & Dementia, 22(11). https://doi.org/10.1002/alz.71923

Vancouver

Kim JS, Han SS, Yoo SH, Lee DH, Jang J, Lee HJ, et al. AI‐Integrated multimodal neuroimaging in Alzheimer's disease: toward biologically informed precision diagnostics. Alzheimer's & Dementia. 2026;22(11). doi:10.1002/alz.71923

BibTeX
@article{Kim2026,
  author = {Kim, Jun-Su and Han, Sang Sun and Yoo, So-Hyeon and Lee, Dong-Hyun and Jang, Jaeho and Lee, Hyun-Ju and Song, Jeong-Heon and Hoe, Hyang-Sook},
  title = {AI‐Integrated multimodal neuroimaging in Alzheimer's disease: toward biologically informed precision diagnostics},
  journal = {Alzheimer's & Dementia},
  year = {2026},
  volume = {22},
  number = {11},
  doi = {10.1002/alz.71923}
}
Kostenlos starten →