Early diagnosis of dementia is essential for enabling timely interventions that may slow disease progression, improve patient outcomes, and reduce healthcare costs. This study aims to develop machine learning models to predict dementia risk using longitudinal electronic health record (EHR) data.
Objective
This research aims to develop and evaluate machine learning models for dementia risk prediction using longitudinal EHR data from routine clinical care and to identify key clinical features associated with elevated dementia risks.
Methods
We conducted an incidence-based case-control study using EHR data from the UMass Memorial Health system (2017–2024) to develop a dementia risk prediction model.
Results
This study included 5622 dementia cases and 44,976 controls. The XGBoost model achieved the highest AUC (0.802), with top predictors included thyroid-stimulating hormone (TSH), vitamin B12, and HDL cholesterol. Model performance was consistent across sexes and remained robust in multiple sensitivity analyses.
Conclusions
Machine learning models that integrate comorbid conditions and longitudinal laboratory test patterns show their potential in predicting dementia risk. These findings highlight the promise of routinely collected EHR data as a scalable, low-cost resource for identifying individuals at elevated risk for dementia.
ArvanitakisZShahRCBennettDA. Diagnosis and management of dementia: review. JAMA2019; 322: 1589–1599.
3.
EckertMOstermannTEhlersJP, et al.Dementia and video games: systematic and bibliographic review. Healthcare (Basel)2024; 12: 2310.
4.
CoxCGBrushBLKobayashiLC, et al.Determinants of dementia diagnosis in U.S. primary care in the past decade: a scoping review. J Prev Alzheimers Dis2025; 12: 100035.
5.
BanerjeeDMuralidharanAHakim MohammedAR, et al.Neuroimaging in dementia: a brief review. Cureus2020; 12: e8682.
6.
RameshJGuruprasadP. Cognitive screening tools for dementia detection in primary healthcare centres in India: a scoping review. medRxiv 2024: 2024.2012. 2004.24318472.
7.
DoerrAJOrwigTAMcNultyM, et al.Digital assessment of cognitive health in outpatient primary care: usability study. JMIR Form Res2025; 9: e66695.
8.
RutkowskiTMAbeMSKomendzinskiT, et al.Machine learning approach for early onset dementia neurobiomarker using EEG network topology features. Front Hum Neurosci2023; 17: 1155194.
9.
ZhangX. Deep learning for medical prediction in electronic health records. In: Proceedings of the AAAI Conference on Artificial Intelligence, 2023, pp.16145–16146.
10.
RajayyanSMustafaSMM. Prediction of dementia using machine learning model and performance improvement with cuckoo algorithm. Int J Electr Comput Eng2023; 13: 4623–4632.
11.
NoshinKBolandMRHouB, et al.Integrating social determinants of health in a multi-modal deep clustering survival model for injury-risk in Alzheimer’s and related dementia patients. In: AAAI Bridge Program on AI for Medicine and Healthcare, 2025, pp.155–164.
12.
KimMKRouphaelCMcMichaelJ, et al.Challenges in and opportunities for electronic health record-based data analysis and interpretation. Gut Liver2024; 18: 201–208.
13.
FanPMirandaOQiX, et al.Unveiling the enigma: exploring risk factors and mechanisms for psychotic symptoms in Alzheimer’s disease through electronic medical records with deep learning models. Pharmaceuticals2023; 16: 11.
14.
KhosraviBWestonADNugenF, et al.Demystifying statistics and machine learning in analysis of structured tabular data. J Arthroplasty2023; 38: 1943–1947.
15.
FinneyCABrownDAShvetcovA. Developing multifactorial dementia prediction models using clinical variables from cohorts in the US and Australia. Transl Psychiatry2025; 15: 15.
16.
NoshinKBolandMRHouB, et al.Integrating social determinants of health in a multi-modal deep clustering survival model for injury-risk in Alzheimer’s and related dementia patients. Proc Mach Learn Res2025; 281: 155–164.
17.
LiQYangXXuJ, et al.Early prediction of Alzheimer's disease and related dementias using real-world electronic health records. Alzheimers Dement2023; 19: 3506–3518.
18.
AkterSLiuZSimoesEJ, et al.Using machine learning and electronic health record (EHR) data for the early prediction of Alzheimer’s disease and related dementias. J Prev Alzheimers Dis2025; 12: 100169.
19.
BrainJKafadarAHErringtonL, et al.What's new in dementia risk prediction modelling? An updated systematic review. Dement Geriatr Cogn Dis Extra2024; 14: 49–74.
20.
BullLMLuntMMartinGP, et al.Harnessing repeated measurements of predictor variables for clinical risk prediction: a review of existing methods. Diagn Progn Res2020; 4: 9.
21.
Noella RSNPriyadarshiniJ. Machine learning algorithms for the diagnosis of Alzheimer and Parkinson disease. J Med Eng Technol2023; 47: 35–43.
22.
Secondary Analysis of Electronic Health Records. Cham, Switerland, 2016.
23.
HallinanCMWardRHartGK, et al.Seamless EMR data access: integrated governance, digital health and the OMOP-CDM. BMJ Health Care Inform2024; 31: e100953.
24.
ParkKChoMSongM, et al.Exploring the potential of OMOP common data model for process mining in healthcare. PLoS One2023; 18: e0279641.
25.
WangLWenAFuS, et al.A scoping review of OMOP CDM adoption for cancer research using real world data. NPJ Digit Med2025; 8: 89.
26.
ExaltoLGBiesselsGJKarterAJ, et al.Risk score for prediction of 10 year dementia risk in individuals with type 2 diabetes: a cohort study. Lancet Diabetes Endocrinol2013; 1: 183–190.
27.
SaeedALopezOCohenA, et al.Cardiovascular disease and Alzheimer's disease: the heart-brain axis. J Am Heart Assoc2023; 12: e030780.
28.
BreijyehZKaramanR. Comprehensive review on Alzheimer's disease: causes and treatment. Molecules2020; 25: 5789.
29.
WellsBJChaginKMNowackiAS, et al.Strategies for handling missing data in electronic health record derived data. EGEMS (Wash DC)2013; 1: 1035.
30.
American Diabetes Association Professional Practice Committee. 2. Diagnosis and classification of diabetes: standards of care in diabetes—2025. Diabetes Care2025; 48: S27–S49.
31.
NianogoRARosenwohl-MackAYaffeK, et al.Risk factors associated with Alzheimer disease and related dementias by sex and race and ethnicity in the US. JAMA Neurol2022; 79: 584–591.
32.
ViraniSSNewbyLKArnoldSV, et al.2023 AHA/ACC/ACCP/ASPC/NLA/PCNA guideline for the management of patients with chronic coronary disease: a report of the American Heart Association/American College of Cardiology Joint Committee on Clinical Practice Guidelines. J Am Coll Cardiol2023; 82: 833–955.
33.
Institute of Medicine Standing Committee on the Scientific Evaluation of Dietary Reference Intakes. Vitamin B12. Dietary reference intakes for thiamin, riboflavin, niacin, vitamin B6, folate, vitamin B12, pantothenic acid, biotin, and choline. Washington, DC: National Academies Press (US), 1998.
34.
HaugenBRAlexanderEKBibleKC, et al.2015 American Thyroid Association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid2016; 26: 1–133.
35.
KnopmanDSPetersenRC. Mild cognitive impairment and mild dementia: a clinical perspective. Mayo Clin Proc2014; 89: 1452–1459.
36.
TanCCYuJTWangHF, et al.Efficacy and safety of donepezil, galantamine, rivastigmine, and memantine for the treatment of Alzheimer's disease: a systematic review and meta-analysis. J Alzheimers Dis2014; 41: 615–631.
37.
YaghmaeiELuHEhwerhemuephaL, et al.Combined use of donepezil and memantine increases the probability of five-year survival of Alzheimer's disease patients. Commun Med (Lond)2024; 4: 99.
38.
MilellaFFamigliniLBanfiG, et al.Application of machine learning to improve appropriateness of treatment in an orthopaedic setting of personalized medicine. J Pers Med2022; 12: 1706.
39.
YasarSSabayanB. Hypertension, antihypertensive medications, and dementia risk in old age: a complicated relationship. Neurology2024; 103: e209788.
BrückCCWoltersFJIkramMA, et al.Projected prevalence and incidence of dementia accounting for secular trends and birth cohort effects: a population-based microsimulation study. Eur J Epidemiol2022; 37: 807–814.
42.
NedelecTZaidiKMontaudC, et al.Machine learning prediction algorithms for 2-, 5-and 10-year risk of Alzheimer’s, Parkinson’s and dementia at age 65: a study using medical records from France and the UK General Practitioners. medRxiv 2025: 2025.2001. 2022.25320969.
43.
Carrasco-RibellesLACabrera-BeanMKhalidS, et al.Development of attention-based prediction models for all-cause mortality, home care need, and nursing home admission in ageing adults in Spain using longitudinal electronic health record data. J Med Syst2025; 49: 17.
44.
WilliamsOAAnYArmstrongNM, et al.Profiles of cognitive change in preclinical and prodromal Alzheimer's disease using change-point analysis. J Alzheimers Dis2020; 75: 1169–1180.
45.
TjandraDMigrinoRQGiordaniB, et al.Cohort discovery and risk stratification for Alzheimer's disease: an electronic health record-based approach. Alzheimers Dement (N Y)2020; 6: e12035.
46.
TanakaMKohjitaniHYamamotoE, et al.Development of interpretable machine learning models to predict in-hospital prognosis of acute heart failure patients. ESC Heart Fail2024; 11: 2798–2812.
47.
UddinSLuH. Confirming the statistically significant superiority of tree-based machine learning algorithms over their counterparts for tabular data. PLoS One2024; 19: e0301541.
48.
Savitz STSavitz LAFleming NS, et al.How much can we trust electronic health record data?. Healthcare2020; 8: 100444.
49.
FardoulyJCrosbyRDSukunesanS. Potential benefits and limitations of machine learning in the field of eating disorders: current research and future directions. J Eat Disord2022; 10: 66.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
