{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T14:12:30Z","timestamp":1785161550604,"version":"3.55.0"},"reference-count":43,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,5,25]],"date-time":"2026-05-25T00:00:00Z","timestamp":1779667200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurosci."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Timely identification of individuals at risk for Alzheimer\u2019s disease (AD) progression remains a major clinical challenge. Traditional cognitive assessments provide limited prognostic insight, while many machine learning (ML) models rely on costly biomarkers or poorly interpretable algorithms that limit clinical scalability. This study evaluated whether widely available baseline demographic, clinical, and cognitive measures could support short-term progression prediction using interpretable ML methods under extreme class imbalance.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We analyzed 3,240 participants from the Alzheimer\u2019s Disease Neuroimaging Initiative (ADNI), of whom 2,423 had valid 24-month follow-up data. The primary outcome was strict unidirectional diagnostic worsening within 24\u202fmonths (13 events; 0.5%). Baseline demographic, clinical, and cognitive variables were used to train XGBoost and logistic regression models under natural class imbalance using stratified k-fold cross-validation with out-of-fold predictions. Model performance was evaluated using AUROC, area under the precision-recall curve (AUPRC), calibration analyses, and bootstrap confidence intervals. Sensitivity analyses evaluated cost-sensitive learning, threshold optimization, and alternative imputation strategies (KNN and MICE). Longitudinal mixed-effects modeling was conducted separately to characterize cognitive decline and was not used as input to the predictive models. SHAP (Shapley Additive Explanations) quantified feature contributions.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Under natural class imbalance, XGBoost achieved AUROC\u202f=\u202f0.912 and AUPRC\u202f=\u202f0.051, while logistic regression achieved AUROC\u202f=\u202f0.787 and AUPRC\u202f=\u202f0.038. Although discrimination exceeded baseline prevalence, precision remained low and threshold optimization produced substantial false-positive burdens, limiting immediate clinical applicability. Cost-sensitive learning did not materially improve performance. MICE imputation produced results comparable to median imputation, whereas KNN imputation reduced performance. SHAP analyses identified baseline cognitive severity, functional measures, and diagnostic status as dominant predictors. Mixed-effects modeling confirmed significant cognitive decline over time (\n                      <jats:italic>\u03b2<\/jats:italic>\n                      =\u202f\u22120.027 points\/month,\n                      <jats:italic>p<\/jats:italic>\n                      &amp;lt;\u202f0.001).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Accessible baseline clinical and cognitive variables contain measurable but limited predictive signal for short-term AD progression under extreme event scarcity. These findings should be interpreted as an early-stage proof-of-concept rather than a clinically deployable decision-support tool. External validation remains necessary before clinical translation.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fnins.2026.1736992","type":"journal-article","created":{"date-parts":[[2026,5,25]],"date-time":"2026-05-25T06:04:01Z","timestamp":1779689041000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Machine learning for Alzheimer\u2019s disease progression under extreme class imbalance"],"prefix":"10.3389","volume":"20","author":[{"given":"Patrick O.","family":"Akinwumi","sequence":"first","affiliation":[{"name":"College of Education, Clemson University","place":["Clemson, SC, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Meihua","family":"Qian","sequence":"additional","affiliation":[{"name":"College of Education, Clemson University","place":["Clemson, SC, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Taiwo A.","family":"Olorunsogbon","sequence":"additional","affiliation":[{"name":"University of Bradford","place":["Bradford, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chaoyi","family":"Zhou","sequence":"additional","affiliation":[{"name":"School of Computing, Clemson University","place":["Clemson, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Siyu","family":"Huang","sequence":"additional","affiliation":[{"name":"School of Computing, Clemson University","place":["Clemson, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,5,25]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"6726","DOI":"10.1038\/s41467-025-61650-z","article-title":"Machine learning in Alzheimer\u2019s disease genetics","volume":"16","author":"Bracher-Smith","year":"2025","journal-title":"Nat. Commun."},{"key":"ref2","doi-asserted-by":"publisher","first-page":"102712","DOI":"10.1016\/j.nicl.2021.102712","article-title":"Cross-cohort generalizability of deep and conventional machine learning for MRI-based diagnosis and prediction of Alzheimer\u2019s disease","volume":"31","author":"Bron","year":"2021","journal-title":"NeuroImage: Clin."},{"key":"ref3","doi-asserted-by":"publisher","first-page":"86","DOI":"10.3934\/Neuroscience.2021005","article-title":"The biological pathways of Alzheimer disease: a review","volume":"8","author":"Calabr\u00f2","year":"2020","journal-title":"AIMS Neurosci"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"1191","DOI":"10.1111\/ene.13964","article-title":"Estimating dementia cases amongst migrants living in Europe","volume":"26","author":"Canevelli","year":"2019","journal-title":"Eur. J. Neurol."},{"key":"ref5","doi-asserted-by":"crossref","DOI":"10.1145\/2939672.2939785","volume-title":"Xgboost: A Scalable tree Boosting system","author":"Chen","year":"2016"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"103","DOI":"10.1016\/j.bbr.2018.02.017","article-title":"Predicting cognitive decline with deep learning of brain metabolism and amyloid imaging","volume":"344","author":"Choi","year":"2018","journal-title":"Behav. Brain Res."},{"key":"ref7","doi-asserted-by":"crossref","first-page":"494","DOI":"10.1037\/0033-2909.114.3.494","article-title":"Dominance statistics: ordinal analyses to answer ordinal questions","volume":"114","author":"Cliff","year":"1993","journal-title":"Psychol. Bull."},{"key":"ref8","doi-asserted-by":"publisher","first-page":"3555","DOI":"10.1038\/s41467-021-23746-0","article-title":"Plasma biomarkers of Alzheimer\u2019s disease improve prediction of cognitive decline in cognitively unimpaired elderly populations","volume":"12","author":"Cullen","year":"2021","journal-title":"Nat. Commun."},{"key":"ref9","doi-asserted-by":"crossref","first-page":"837","DOI":"10.2307\/2531595","article-title":"Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach","volume":"44","author":"DeLong","year":"1988","journal-title":"Biometrics"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"265","DOI":"10.3389\/fnagi.2020.00265","article-title":"Mechanisms of pathogenic tau and a\u03b2 protein spreading in Alzheimer's disease","volume":"12","author":"d'Errico","year":"2020","journal-title":"Front. Aging Neurosci."},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1038\/s41514-026-00390-w","article-title":"Mild cognitive impairment cases affect the predictive power of Alzheimer\u2019s disease diagnostic models using routine clinical variables","author":"Finney","year":"2025","journal-title":"medRxiv"},{"key":"ref12","first-page":"1","article-title":"All models are wrong, but many are useful: learning a variable's importance by studying an entire class of prediction models simultaneously","volume":"20","author":"Fisher","year":"2019","journal-title":"J. Mach. Learn. Res."},{"key":"ref13","first-page":"1","article-title":"Verification of forecasts expressed in terms of probability","volume":"78","author":"Glenn","year":"1950","journal-title":"Mon. Weather Rev."},{"key":"ref14","doi-asserted-by":"publisher","first-page":"5481","DOI":"10.1038\/s41380-021-01249-0","article-title":"The amyloid-\u03b2 pathway in Alzheimer\u2019s disease","volume":"26","author":"Hampel","year":"2021","journal-title":"Mol. Psychiatry"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"685","DOI":"10.1002\/jmri.21049","article-title":"The Alzheimer's disease neuroimaging initiative (ADNI): MRI methods","volume":"27","author":"Jack","year":"2008","journal-title":"J. Magn. Reson. Imaging"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"535","DOI":"10.1016\/j.jalz.2018.02.018","article-title":"NIA-AA research framework: toward a biological definition of Alzheimer's disease","volume":"14","author":"Jack","year":"2018","journal-title":"Alzheimers Dement."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"e0294253","DOI":"10.1371\/journal.pone.0294253","article-title":"Explainable AI-based Alzheimer\u2019s prediction and management using multimodal data","volume":"18","author":"Jahan","year":"2023","journal-title":"PLoS One"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"7407","DOI":"10.1038\/s41467-025-62590-4","article-title":"AI-driven fusion of multimodal data for Alzheimer\u2019s disease biomarker assessment","volume":"16","author":"Jasodanand","year":"2025","journal-title":"Nat. Commun."},{"key":"ref19","doi-asserted-by":"crossref","first-page":"2022","DOI":"10.3389\/fnagi.2022.937486","article-title":"Global, regional, and national burden of Alzheimer's disease and other dementias, 1990\u20132019 [original research]","volume":"14","author":"Li","year":"2022","journal-title":"Front. Aging Neurosci."},{"key":"ref20","doi-asserted-by":"publisher","first-page":"1992","DOI":"10.1109\/TCYB.2020.3005859","article-title":"Attention-guided hybrid network for dementia diagnosis with structural MR images","volume":"52","author":"Lian","year":"2020","journal-title":"IEEE Transactions Cybernetics"},{"key":"ref21","article-title":"A unified approach to interpreting model predictions","volume":"30","author":"Lundberg","year":"2017","journal-title":"Adv. Neural Inf. Proces. Syst."},{"key":"ref22","doi-asserted-by":"publisher","first-page":"412","DOI":"10.1016\/j.dadm.2015.09.001","article-title":"Revolutionizing Alzheimer's disease and clinical trials through biomarkers","volume":"1","author":"Mattsson","year":"2015","journal-title":"Alzheimer's & Dementia: Diagnosis, Assess Dis Monitor"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"1345","DOI":"10.3233\/jad-220021","article-title":"Machine learning based multimodal neuroimaging genomics dementia score for predicting future conversion to Alzheimer's disease","volume":"87","author":"Mirabnahrazam","year":"2022","journal-title":"J. Alzheimer's Dis"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"2903","DOI":"10.1007\/s10618-022-00901-9","article-title":"Model-agnostic feature importance and effects with dependent features: a conditional subgroup approach","volume":"38","author":"Molnar","year":"2024","journal-title":"Data Min. Knowl. Disc."},{"key":"ref25","doi-asserted-by":"crossref","DOI":"10.1145\/1102351.1102430","volume-title":"Predicting good Probabilities with Supervised Learning","author":"Niculescu-Mizil","year":"2005"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"106584","DOI":"10.1016\/j.cmpb.2021.106584","article-title":"Explanation of machine learning models using shapley additive explanation and application for real data in hospital","volume":"214","author":"Nohara","year":"2022","journal-title":"Comput. Methods Prog. Biomed."},{"key":"ref27","doi-asserted-by":"publisher","first-page":"772","DOI":"10.1001\/jama.2020.12134","article-title":"Discriminative accuracy of plasma phospho-tau217 for Alzheimer disease vs other neurodegenerative disorders","volume":"324","author":"Palmqvist","year":"2020","journal-title":"JAMA"},{"key":"ref28","article-title":"World Alzheimer report 2015","author":"Prince","year":"2015","journal-title":"The Global Impact of Dementia: an Analysis of Prevalence, Incidence, Cost and Trends."},{"key":"ref29","doi-asserted-by":"publisher","first-page":"e67437","DOI":"10.2196\/67437","article-title":"Using machine learning to predict cognitive decline in older adults from the Chinese longitudinal healthy longevity survey: model development and validation study","volume":"8","author":"Ren","year":"2025","journal-title":"JMIR Aging"},{"key":"ref30","doi-asserted-by":"publisher","first-page":"3140","DOI":"10.1136\/bmj.i3140","article-title":"External validation of clinical prediction models using big datasets from e-health records or IPD meta-analysis: opportunities and challenges","volume":"353","author":"Riley","year":"2016","journal-title":"BMJ"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"e0118432","DOI":"10.1371\/journal.pone.0118432","article-title":"The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets","volume":"10","author":"Saito","year":"2015","journal-title":"PLoS One"},{"key":"ref32","doi-asserted-by":"publisher","first-page":"e1647","DOI":"10.1212\/WNL.0000000000008081","article-title":"High-precision plasma \u03b2-amyloid 42\/40 predicts current and future brain amyloidosis","volume":"93","author":"Schindler","year":"2019","journal-title":"Neurology"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"S188","DOI":"10.37765\/ajmc.2022.89236","article-title":"The economic and societal burden of Alzheimer disease: managed care considerations","volume":"28","author":"Skaria","year":"2022","journal-title":"Am. J. Manag. Care"},{"key":"ref34","doi-asserted-by":"publisher","first-page":"387","DOI":"10.1038\/s41591-020-0762-2","article-title":"Diagnostic value of plasma phosphorylated tau181 in Alzheimer\u2019s disease and frontotemporal lobar degeneration","volume":"26","author":"Thijssen","year":"2020","journal-title":"Nat. Med."},{"key":"ref35","doi-asserted-by":"publisher","first-page":"230","DOI":"10.1186\/s12916-019-1466-7","article-title":"Calibration: the Achilles heel of predictive analytics","volume":"17","author":"Van Calster","year":"2019","journal-title":"BMC Med."},{"key":"ref36","first-page":"101","article-title":"A critique and improvement of the CL common language effect size statistics of McGraw and Wong","volume":"25","author":"Vargha","year":"2000","journal-title":"J. Educ. Behav. Stat."},{"key":"ref37","doi-asserted-by":"publisher","first-page":"48","DOI":"10.1038\/s41746-022-00592-y","article-title":"Machine learning for medical imaging: methodological failures and recommendations for the future","volume":"5","author":"Varoquaux","year":"2022","journal-title":"NPJ Digital Med"},{"key":"ref38","doi-asserted-by":"publisher","first-page":"23","DOI":"10.1186\/alzrt47","article-title":"Role of structural MRI in Alzheimer's disease","volume":"2","author":"Vemuri","year":"2010","journal-title":"Alzheimer's Res. Ther."},{"key":"ref39","doi-asserted-by":"publisher","first-page":"871","DOI":"10.1038\/s41591-021-01309-6","article-title":"Four distinct trajectories of tau deposition identified in Alzheimer\u2019s disease","volume":"27","author":"Vogel","year":"2021","journal-title":"Nat. Med."},{"key":"ref40","doi-asserted-by":"publisher","first-page":"265","DOI":"10.1186\/s12967-024-05025-w","article-title":"Predicting long-term progression of Alzheimer\u2019s disease using a multimodal deep learning model incorporating interaction effects","volume":"22","author":"Wang","year":"2024","journal-title":"J. Transl. Med."},{"key":"ref41","doi-asserted-by":"publisher","first-page":"137","DOI":"10.1186\/s12911-023-02238-9","article-title":"XGBoost-SHAP-based interpretable diagnostic framework for alzheimer\u2019s disease","volume":"23","author":"Yi","year":"2023","journal-title":"BMC Med. Inform. Decis. Mak."},{"key":"ref42","doi-asserted-by":"publisher","first-page":"592","DOI":"10.1001\/jamaneurol.2022.0676","article-title":"Divergent cortical tau positron emission tomography patterns among patients with preclinical Alzheimer disease","volume":"79","author":"Young","year":"2022","journal-title":"JAMA Neurol."},{"key":"ref43","doi-asserted-by":"publisher","first-page":"2354","DOI":"10.1109\/TMI.2021.3077079","article-title":"Dual attention multi-instance deep learning for Alzheimer\u2019s disease diagnosis with structural MRI","volume":"40","author":"Zhu","year":"2021","journal-title":"IEEE Trans. Med. Imaging"}],"container-title":["Frontiers in Neuroscience"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnins.2026.1736992\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,25]],"date-time":"2026-05-25T06:04:03Z","timestamp":1779689043000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnins.2026.1736992\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,25]]},"references-count":43,"alternative-id":["10.3389\/fnins.2026.1736992"],"URL":"https:\/\/doi.org\/10.3389\/fnins.2026.1736992","relation":{},"ISSN":["1662-453X"],"issn-type":[{"value":"1662-453X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,25]]},"article-number":"1736992"}}