Integrating neuroimaging and plasma biomarkers to predict preclinical Alzheimer's disease progression.

Objective: To develop and validate a multimodal model integrating neuroimaging and plasma biomarkers for predicting the risk of cognitive progression in preclinical Alzheimer's disease (AD). Methods: This retrospective study enrolled 320 patients with Aβ-positive preclinical AD or AD-related mild cognitive impairment. Participants were randomly allocated into training and validation sets at a 7:3 ratio. In the training set, univariable analysis, least absolute shrinkage and selection operator (LASSO) regression, and multivariable Logistic regression were employed to identify core predictive variables. Subsequently, four machine learning models were constructed based on these variables. Model performance was evaluated using the area under the receiver operating characteristic Curve (AUC), calibration curves, and decision curve analysis. Interpretability was assessed using SHapley Additive exPlanations (SHAP) values. Results: The baseline characteristics were balanced between the training and validation sets. LASSO regression identified five core variables: Mini-Mental State Examination total score, Rey Auditory Verbal Learning Test delayed recall, normalized hippocampal volume, plasma phosphorylated tau181, and apolipoprotein E ε4 allele status. Multivariable analysis confirmed these as independent predictors ( Conclusion: The developed multimodal model exhibits robust predictive performance and clinical utility. It may serve as a quantitative tool for individualized risk management in preclinical AD.