Quantifying generalization error in machine learning prediction of cognitive decline - 09/08/26
, Nicolas Langer a, b, Bruno Hebling Vieira a, b, ⁎ for the Alzheimer’s Disease Neuroimaging Initiative 1
Bienvenue sur EM-consulte, la référence des professionnels de santé.
Article gratuit.
Connectez-vous pour en bénéficier!
Highlights |
• | Adding structural MRI to non-brain data improves prediction of cognitive decline. |
• | Cross-cohort evaluation reveals limits of model generalizability. |
• | Key-feature models match full-model performance in predictions across datasets. |
• | Continuous decline prediction captures interindividual variation in change. |
Abstract |
Background |
Predicting cognitive decline as a continuum, from healthy age-related decline to mild cognitive impairment and dementia, enables more precise individual-level predictions. However, the practical value of such models for early intervention and prevention depends on their ability to generalize to independent cohorts, a property that is often not evaluated.
Objectives |
This study investigated whether adding structural magnetic resonance imaging (MRI) to non-brain data improved machine learning predictions of continuous cognitive decline and analyzed the models’ generalizability.
Design |
Multi-target random forest regression models predicted annual decline in the Clinical Dementia Rating Scale Sum of Boxes (CDR-SOB) and Mini-Mental State Examination (MMSE) using non-brain data, structural MRI data, or their combination from the Alzheimer's Disease Neuroimaging Initiative (ADNI; N = 1237) and Open Access Series of Imaging Studies (OASIS-3; N = 662) datasets. Cross-site generalizability was evaluated.
Setting |
Data from ADNI and OASIS-3 were used for this study.
Participants |
A total of 1899 participants who had demographic, clinical, and brain imaging data from a baseline session and clinical data from at least 2 follow-up sessions were included.
Measurements |
Baseline non-brain (demographics, clinical and neuropsychological scores, information on APOE genotype, cognitive diagnosis, health, and number of sessions before baseline) and/or structural MRI data were used to predict the yearly rate of change in CDR-SOB and MMSE scores.
Results |
Including structural MRI data improved prediction of CDR-SOB and MMSE change, reaching respective R 2 values of .41 and .33 in ADNI and .42 and .33 in OASIS-3. Model performance for across-dataset predictions was reduced ( R 2 between .18 and .35), unexplained by distributional shifts of target variables. Models using only top predictive features performed similarly to full models when tested externally ( R 2 between .18 and .34), suggesting predictor redundancy.
Conclusions |
Incorporating structural MRI data enhances within-dataset prediction of continuous cognitive decline, allowing for more precise individual-level prediction and advancing towards precision medicine. Even though external validation remains limited, quantifying the generalizability gap is a crucial step towards the responsible use of ML models in clinical intervention and prevention.
Le texte complet de cet article est disponible en PDF.Keywords : Cognitive decline, Structural MRI, Generalizability, Machine learning, Predictive modeling
Plan
Vol 13 - N° 9
Article 100646- novembre 2026 Retour au numéro
Article précédent
- Cost-effectiveness of the special care unit for persons with dementia
- Ron Handels, Bernhard Michalowsky, Aline Mendes, Sverre Bergh, Bruno Mario Cesana, Alfonso Ciccone, Emmanuel Cognat, Andrea Fabbo, Sara Fascendini, Giovanni B. Frisoni, Lutz Froelich, Patrizia Mecocci, Paola Merlo, Oliver Peters, Magdalini Tsolaki, Carlo Alberto Defanti, RECage consortium collaborators
- Risk of Dementia after initiation of GLP-1 RA versus long-acting insulin in patients with type 2 Diabetes mellitus
- Chien-Hung Lin, Peir-Haur Hung, Mu-Chi Chung, Laing-You Wu, Chi-Jung Chung
Bienvenue sur EM-consulte, la référence des professionnels de santé.