Back to all papers

Quantifying generalization error in machine learning prediction of cognitive decline.

August 8, 2026pubmed logopapers

Authors

Hüppi RM,Langer N,Hebling Vieira B

Affiliations (3)

  • Methods of Plasticity Research, Department of Psychology, University of Zurich, 8050, Zurich, Zurich, Switzerland; Neuroscience Center Zurich (ZNZ), University of Zurich & ETH Zurich, 8057, Zurich, Zurich, Switzerland; Department of Adult Psychiatry and Psychotherapy, Psychiatric University Clinic Zurich and University of Zurich, 8032, Zurich, Zurich, Switzerland. Electronic address: [email protected].
  • Methods of Plasticity Research, Department of Psychology, University of Zurich, 8050, Zurich, Zurich, Switzerland; Neuroscience Center Zurich (ZNZ), University of Zurich & ETH Zurich, 8057, Zurich, Zurich, Switzerland.
  • Methods of Plasticity Research, Department of Psychology, University of Zurich, 8050, Zurich, Zurich, Switzerland; Neuroscience Center Zurich (ZNZ), University of Zurich & ETH Zurich, 8057, Zurich, Zurich, Switzerland. Electronic address: [email protected].

Abstract

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. 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. 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. Data from ADNI and OASIS-3 were used for this study. 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. 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. Including structural MRI data improved prediction of CDR-SOB and MMSE change, reaching respective R<sup>2</sup> values of .41 and .33 in ADNI and .42 and .33 in OASIS-3. Model performance for across-dataset predictions was reduced (R<sup>2</sup> 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<sup>2</sup> between .18 and .34), suggesting predictor redundancy. 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.

Topics

Journal Article

Ready to Sharpen Your Edge?

Subscribe to join 11k+ peers who rely on RadAI Slice. Get the essential weekly briefing that empowers you to navigate the future of radiology.

We respect your privacy. Unsubscribe at any time.