Machine Learning in Dementia Research: Trends, Biomarkers, Prediction Goals, and Data Modalities
4 affiliations
- 1Taipei Medical University Hospital
- 2Taipei Medical University
- 3University At Buffalo
- 4Buddhist Tzu Chi Medical Foundation
Abstract
Objective Machine learning (ML) is increasingly used in dementia research for diagnosis, early detection, prognosis, biomarker characterization, and disease phenotyping. This study aimed to provide a bibliometric and thematic overview of ML applications in dementia, emphasizing biological mechanisms, prediction goals, data modalities, collaborative networks, and emerging computational approaches. Methods A title‐field search of the Web of Science Core Collection identified 3,833 records related to ML and dementia. After restricting publications to 2000 onward and excluding non‐analytic document types, 3,479 records were included in the bibliometric analysis. A subset of 3,223 original articles was used for thematic analyses. Publication trends, authorship, institutional collaboration, keywords, biomarker themes, ML objectives, data modalities, methods, shared research resources, and large language model (LLM) applications were evaluated. Results ML‐related dementia research increased substantially over time. Diagnostic classification was the predominant ML objective (55.0%), followed by early detection or screening (10.3%), progression or cognitive‐decline prediction (6.8%), and biomarker or feature discovery (4.7%). Structural magnetic resonance imaging was the most frequently identified modality (42.6%), followed by clinical or cognitive data (22.4%), positron emission tomography/single photon emission computerized tomography (10.3%), and genetic or genomic data (10.2%). Prominent biological themes included hippocampal measures, amyloid/Aβ, electroencephalogram, tau/p‐tau, and cerebrospinal fluid biomarkers. Institutional networks showed distinct international collaboration clusters and frequent use of shared infrastructures. Transformer‐based and LLM approaches emerged in recent years. Conclusions ML research in dementia has expanded beyond diagnostic neuroimaging toward multimodal, biomarker, prognostic, and language model applications, although diagnostic classification and structural imaging remain dominant. Emphasis on biological integration, longitudinal validation, and clinical translation are needed.
Related
- Hemodynamic biomarkers in symptomatic intracranial atherosclerotic stenosis: evidence boundaries and research pathways from prognostic stratification to treatment-effect prediction
- Blood Biomarkers of Alzheimer's Disease Neuropathology, Cognition, and Response to a Multidomain Lifestyle Intervention: Secondary analysis of the FINGER Randomized Controlled Dementia Risk Reduction Trial
- Machine Learning in Amyloidogenic Polypeptide Research: From Sequence Prediction to Drug Discovery
- Machine Learning and Deep Learning for Early Disease Detection Across Modalities: A Comprehensive Review of Methods, Applications, and Clinical Translation
- Cognition in healthy aging, ages 36 to 108: Associations with dementia risk factors and neurodegenerative disease biomarkers
BibTeX
@article{wang2026machine,
title = {{Machine Learning in Dementia Research: Trends, Biomarkers, Prediction Goals, and Data Modalities}},
author = {Yao‐Chin Wang and Hsien Chang Yeh and Md. Mohaimenul Islam and Woon‐Man Kung},
journal = {Brain and behavior},
year = {2026},
month = oct,
doi = {10.1002/brb3.71847},
url = {https://europepmc.org/article/PMC/PMC13673231},
note = {CC BY},
}
RIS
TY - JOUR
TI - Machine Learning in Dementia Research: Trends, Biomarkers, Prediction Goals, and Data Modalities
AU - Yao‐Chin Wang
AU - Hsien Chang Yeh
AU - Md. Mohaimenul Islam
AU - Woon‐Man Kung
PY - 2026
DA - 2026/10/01
JO - Brain and behavior
DO - 10.1002/brb3.71847
UR - https://europepmc.org/article/PMC/PMC13673231
AB - Objective Machine learning (ML) is increasingly used in dementia research for diagnosis, early detection, prognosis, biomarker characterization, and disease phenotyping. This study aimed to provide a bibliometric and thematic overview of ML applications in dementia, emphasizing biological mechanisms, prediction goals, data modalities, collaborative networks, and emerging computational approaches. Methods A title‐field search of the Web of Science Core Collection identified 3,833 records related to ML and dementia. After restricting publications to 2000 onward and excluding non‐analytic document types, 3,479 records were included in the bibliometric analysis. A subset of 3,223 original articles was used for thematic analyses. Publication trends, authorship, institutional collaboration, keywords, biomarker themes, ML objectives, data modalities, methods, shared research resources, and large language model (LLM) applications were evaluated. Results ML‐related dementia research increased substantially over time. Diagnostic classification was the predominant ML objective (55.0%), followed by early detection or screening (10.3%), progression or cognitive‐decline prediction (6.8%), and biomarker or feature discovery (4.7%). Structural magnetic resonance imaging was the most frequently identified modality (42.6%), followed by clinical or cognitive data (22.4%), positron emission tomography/single photon emission computerized tomography (10.3%), and genetic or genomic data (10.2%). Prominent biological themes included hippocampal measures, amyloid/Aβ, electroencephalogram, tau/p‐tau, and cerebrospinal fluid biomarkers. Institutional networks showed distinct international collaboration clusters and frequent use of shared infrastructures. Transformer‐based and LLM approaches emerged in recent years. Conclusions ML research in dementia has expanded beyond diagnostic neuroimaging toward multimodal, biomarker, prognostic, and language model applications, although diagnostic classification and structural imaging remain dominant. Emphasis on biological integration, longitudinal validation, and clinical translation are needed.
SN - 2162-3279
ER -