Diagnostic Labels and Measurement Timing Drive Systematic Inconsistency in ADNI Neuroimaging Data
- 1University of Glasgow
Abstract
The Alzheimer's Disease Neuroimaging Initiative (ADNI) is widely used to train machine learning models for Alzheimer's disease, yet whether it provides consistent ground truth for predictive modeling has not been systematically tested. In this paper, we showed that ADNI contains three interacting sources of bias with direct implications for machine learning: (a) diagnostic label inconsistency, (b) technical measurement drift, and (c) longitudinal survivor bias. A substantial proportion of cases, particularly within intermediate stages, fall outside ADNI's own diagnostic thresholds. MRI field strength and evolving processing pipelines introduce significant technical variability in hippocampal volume, while cohort survivor bias arising from differential retention of participants across phases further distorts longitudinal estimates of disease progression. These findings indicated that ADNI does not provide the stable, internally consistent labels often required in machine learning applications. We proposed a practical framework for diagnostic validation, feature harmonization, and cohort accounting, offering guidance for building more robust and biologically meaningful predictive models from large-scale neuroimaging cohorts.
Related
- Central nervous system effects of GLP-1 receptor agonists: A systematic review of human neuroimaging studies
- Assessing measurement uncertainty at laboratory network scale and its impact on diagnostic performance in the absence of a gold standard: application to ELISA tests for Coxiella burnetii in ruminants
- Recurrent Acute Liver Failure in Children: A Systematic Review of Genetic Etiologies, Diagnostic Clues, and Liver Transplantation
- Multiplex molecular respiratory diagnostic panels and antimicrobial use in paediatric respiratory infections: a systematic review and meta-analysis protocol.
- Diagnostic accuracy of Mycobacterium tuberculosis antigen-induced interferon-γ-inducible protein 10 for latent tuberculosis infection among high-risk groups: a systematic review and meta-analysis
BibTeX
@misc{smith2026diagnostic,
title = {{Diagnostic Labels and Measurement Timing Drive Systematic Inconsistency in ADNI Neuroimaging Data}},
author = {P. Smith and H. Richardson and K. Robertson and A. J. Dibble and C. Dalby and M. Svanera},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.16.752036},
url = {https://www.biorxiv.org/content/10.64898/2026.09.16.752036v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Diagnostic Labels and Measurement Timing Drive Systematic Inconsistency in ADNI Neuroimaging Data
AU - P. Smith
AU - H. Richardson
AU - K. Robertson
AU - A. J. Dibble
AU - C. Dalby
AU - M. Svanera
PY - 2026
DA - 2026/09/16
PB - bioRxiv
DO - 10.64898/2026.09.16.752036
UR - https://www.biorxiv.org/content/10.64898/2026.09.16.752036v1
AB - The Alzheimer's Disease Neuroimaging Initiative (ADNI) is widely used to train machine learning models for Alzheimer's disease, yet whether it provides consistent ground truth for predictive modeling has not been systematically tested. In this paper, we showed that ADNI contains three interacting sources of bias with direct implications for machine learning: (a) diagnostic label inconsistency, (b) technical measurement drift, and (c) longitudinal survivor bias. A substantial proportion of cases, particularly within intermediate stages, fall outside ADNI's own diagnostic thresholds. MRI field strength and evolving processing pipelines introduce significant technical variability in hippocampal volume, while cohort survivor bias arising from differential retention of participants across phases further distorts longitudinal estimates of disease progression. These findings indicated that ADNI does not provide the stable, internally consistent labels often required in machine learning applications. We proposed a practical framework for diagnostic validation, feature harmonization, and cohort accounting, offering guidance for building more robust and biologically meaningful predictive models from large-scale neuroimaging cohorts.
ER -