Deep Learning Predicts Prevalent and Incident Parkinson's Disease From UK Biobank Fundus Imaging

TL;DR

This study demonstrates that deep learning models applied to retinal fundus images can accurately identify both existing and future cases of Parkinson's disease, offering a potential non-invasive screening tool.

Contribution

The paper introduces a deep learning approach using fundus imaging to predict Parkinson's disease, including preclinical cases, with explainability features and robustness analysis.

Findings

Achieved AUC of 0.77 in classifying Parkinson's disease.

Effective prediction of both prevalent and incident Parkinson's cases.

Enhanced model trustworthiness through explainability and robustness metrics.

Abstract

Parkinson's disease is the world's fastest-growing neurological disorder. Research to elucidate the mechanisms of Parkinson's disease and automate diagnostics would greatly improve the treatment of patients with Parkinson's disease. Current diagnostic methods are expensive and have limited availability. Considering the insidious and preclinical onset and progression of the disease, a desirable screening should be diagnostically accurate even before the onset of symptoms to allow medical interventions. We highlight retinal fundus imaging, often termed a window to the brain, as a diagnostic screening modality for Parkinson's disease. We conducted a systematic evaluation of conventional machine learning and deep learning techniques to classify Parkinson's disease from UK Biobank fundus imaging. Our results show that Parkinson's disease individuals can be differentiated from age and gender-matched healthy subjects with an Area Under the Curve (AUC) of 0.77. This accuracy is maintained when predicting either prevalent or incident Parkinson's disease. Explainability and trustworthiness are enhanced by visual attribution maps of localized biomarkers and quantified metrics of model robustness to data perturbations.