Analysis of Biomarkers from Visual Explorations in Individuals Suspected to Be in the Prodromal Phase of Parkinson's Disease
Abstract
Identifying prodromal Parkinson's Disease is a precondition for any future treatment that modifies the course of the disease, since such therapies would ideally act on the initial pathogenic events, early enough to delay the clinical onset of the disease rather than only slow its progression. In this work, a visual exploration model is retrained on an expanded corpus. Without any further training or threshold recalibration, it is then tested on an enriched cohort of patients with high risk of being in the prodromal phase of Parkinson's Disease, testing whether the learned signal follows a continuum of disease rather than only separating patients from controls. The resulting score distribution is analysed as an external evaluation of the hypothesis that the oculomotor signature targeted by the model is already partly present before clinical diagnosis. With every subject scored on data not used to train the model, the scores are ordered as controls < high-risk patients < Parkinson's Disease, with 42% of the high-risk patients exceeding the decision threshold. A SHAP decomposition of the scores into classifiers and gaze features shows that high scores in this cohort are driven mainly by reduced saccadic excursion, fewer fixations and longer fixations, the hypokinetic oculomotor profile that characterises Parkinson's Disease. The model score is also compared against the prodromal markers used for recruitment. The score is not correlated with hyposmia, constipation, depression or REM sleep behaviour disorder; these results, interpreted in the context of the selection induced by the recruitment criteria, are discussed as evidence that within the high-risk cohort the score is largely not explained by the available ancillary clinical variables. The results suggest that the visual biomarkers analysed are a complementary indicator to identify patients at risk of being in the prodromal stage.
Related
- Enhancement of gut-microbiome-based classification of prodromal Parkinson's Disease cases by hyperparameter tuning, integration of multimodal data, and meta-analysis.
- Modelling prodromal Parkinson's disease- gastrointestinal pathology and early motor deficits in mice after transient rotenone exposure
- State of the “Art” in Freezing of Gait: Drawing on Perspectives of Individuals With Parkinson's Disease
- Application of Bayesian-Derived Gene Interaction Network and Subnetwork for Classification of Idiopathic Parkinson Disease with Biomarkers
- EEG in Parkinson’s Disease: Mechanisms, Biomarkers and Translational Opportunities
BibTeX
@misc{alcaladurand2026analysis,
title = {{Analysis of Biomarkers from Visual Explorations in Individuals Suspected to Be in the Prodromal Phase of Parkinson's Disease}},
author = {Maria F. Alcala-Durand and J. Camilo Puerta-Acevedo and Valle Pérez del Olmo and Elisa Luque-Buzo and Julián D. Arias-Londoño and Francisco J. Grandas-Pérez and Juan I. Godino-Llorente},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.34524},
eprint = {2610.34524},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.34524},
}
RIS
TY - UNPB
TI - Analysis of Biomarkers from Visual Explorations in Individuals Suspected to Be in the Prodromal Phase of Parkinson's Disease
AU - Maria F. Alcala-Durand
AU - J. Camilo Puerta-Acevedo
AU - Valle Pérez del Olmo
AU - Elisa Luque-Buzo
AU - Julián D. Arias-Londoño
AU - Francisco J. Grandas-Pérez
AU - Juan I. Godino-Llorente
PY - 2026
DA - 2026/10/25
PB - arXiv
DO - 10.48550/arxiv.2610.34524
UR - https://arxiv.org/abs/2610.34524
AB - Identifying prodromal Parkinson's Disease is a precondition for any future treatment that modifies the course of the disease, since such therapies would ideally act on the initial pathogenic events, early enough to delay the clinical onset of the disease rather than only slow its progression. In this work, a visual exploration model is retrained on an expanded corpus. Without any further training or threshold recalibration, it is then tested on an enriched cohort of patients with high risk of being in the prodromal phase of Parkinson's Disease, testing whether the learned signal follows a continuum of disease rather than only separating patients from controls. The resulting score distribution is analysed as an external evaluation of the hypothesis that the oculomotor signature targeted by the model is already partly present before clinical diagnosis. With every subject scored on data not used to train the model, the scores are ordered as controls < high-risk patients < Parkinson's Disease, with 42% of the high-risk patients exceeding the decision threshold. A SHAP decomposition of the scores into classifiers and gaze features shows that high scores in this cohort are driven mainly by reduced saccadic excursion, fewer fixations and longer fixations, the hypokinetic oculomotor profile that characterises Parkinson's Disease. The model score is also compared against the prodromal markers used for recruitment. The score is not correlated with hyposmia, constipation, depression or REM sleep behaviour disorder; these results, interpreted in the context of the selection induced by the recruitment criteria, are discussed as evidence that within the high-risk cohort the score is largely not explained by the available ancillary clinical variables. The results suggest that the visual biomarkers analysed are a complementary indicator to identify patients at risk of being in the prodromal stage.
ER -