MRI microstructure and morphology enable machine learning-based prediction of freezing of gait in Parkinson’s disease.

Freezing of gait (FOG) in Parkinson’s disease (PD) is a debilitating, often treatment-resistant symptom impairing mobility and quality of life. Early identification of high-risk patients may enable targeted intervention, yet few prognostic biomarkers exist. In this thesis, we developed a machine learning classifier to predict FOG onset from baseline structural neuroimaging of 106 de novo PD patients from the Parkinson’s Progression Marker Initiative. The trained model demonstrated high accuracy (AUC=0.91, Sensitivity=0.94, Specificity=0.80) on unseen data. Model explainability analysis revealed key microstructural and morphometric features within limbic, executive, and visual networks, supporting the idea that multi-network vulnerabilities may contribute to the onset of FOG. Our approach, which relied solely on accessible imaging data while managing confounds, enhances scalability and interpretability compared to previous prognostic frameworks. Future work should validate these biomarkers in larger, more diverse cohorts and explore the role of a prognostic model in enriching clinical trials for preventative interventions.

Name thesis:degree_name

M Sc

Discipline thesis:degree_discipline

Neuroscience

Grantor dc:publisher

The University of Western Ontario

Year dc:date.issued

2025

Author dc:creator

• Rothery, Nathan

Advisors dc:contributor.advisor

• MacDonald, Penny
• Daley, Mark

• Parkinson’s disease (PD)
• freezing of gait (FOG)
• prognosis
• biomarkers
• machine learning
• magnetic resonance imaging (MRI)
• diffusion tensor imaging (DTI)

Statement dc:rights

• Attribution-NonCommercial-NoDerivatives 4.0 International

Language dc:language.iso

en

Handle dc:identifier.uri

https://hdl.handle.net/20.500.14721/38599

OAI identifier oai:identifier

oai:uwo.scholaris.ca:20.500.14721/38599