Doctor of Philosophy
Parkinson’s disease (PD) is a progressive neurodegenerative disorder that primarily affects the elderly population, characterized by motor symptoms and non-motor symptoms. Unfortunately, there are currently no curative treatments for PD, largely due to the absence of reliable diagnostic biomarkers, particularly in the early stages of the disease. Early detection plays a crucial role in timely intervention and enrollment in clinical trials. This thesis project focuses on the development of robust biomarkers for PD using brain structural morphometry, with a specific emphasis on the striatum and ventral tegmental area/substantia nigra compacta (VTA/SNc), which are prominently affected in PD. The first chapter provides an in-depth back- ground on PD, existing biomarkers, and their limitations. Furthermore, it presents the methodology employed for the proposed analysis.
A novel automated parcellation pipeline called is introduced in the second chapter, which subdivides subcortical structures into distinct regions based on their connectivity to the cortex. This pipeline enables the extraction of various features related to volume, surface characteristics, and connectivity, enabling the identification of abnormalities in patient populations. This chapter evaluates the reliability and diagnostic capability of these features.
In the third chapter, machine learning models are trained to detect early-stage PD using features extracted from subregions of the striatum and VTA/SNc derived from multicentered MRI data. These models exhibit high accuracy and demonstrate that specific features, such as surface-based and bundle-based fractional anisotropy and mean diffusivity, play a crucial role in the classification of PD. The parcellation of subregions leads to an improvement in PD detection compared to unparcellated measures. The developed diagnostic test integrates shape- based measures and machine learning techniques, and ongoing validation is being conducted through a multicentered trial.
The final chapter introduces a novel approach to predict disease prognosis in PD by developing subject-specific models utilizing baseline MRI and clinical data. The proposed model, leveraging MRI data, outperforms ones that exists, demonstrating robustness in predicting severity scores and addressing assessment challenges.
This thesis advances PD research by developing reliable biomarkers, enhancing diagnostic accuracy, and predicting disease prognosis.
Summary for Lay Audience
Parkinson’s disease (PD) is a debilitating neurological disorder affecting the elderly population, characterized by motor symptoms like tremors and rigidity, as well as non-motor symptoms such as sleep disturbances and autonomic dysfunction. Unfortunately, there are currently no curative treatments for PD, partly due to the lack of reliable diagnostic biomarkers, especially in the early stages of the disease. Early detection is crucial for timely intervention and participation in clinical trials.
This thesis project focuses on developing reliable biomarkers for PD using structural morphometry of the brain, with a particular emphasis on the striatum and ventral tegmental area and substantia nigra compacta (VTA/SNc), which are highly affected in PD. Chapter 1 provides a comprehensive background on the disease, existing biomarkers, and their potential drawbacks, offering valuable insights for our analysis. Additionally, this chapter familiarizes the reader with the methods employed in our proposed research. To extract meaningful features, we utilize an automated parcellation and feature extraction pipeline called diffparc-surf. The second chapter evaluates the reliability, variability, and diagnostic capability of the features extracted from this pipeline, taking into account important factors such as test-retest reliability, variations across different sites, and age and sex differences.
Building upon the extracted features from the striatum and VTA/SNc subregions, chapter 3 focuses on training machine learning models to accurately detect early-stage PD. Through cross-validation, our models achieve an impressive accuracy of 86%, with microstructural changes emerging as crucial features for early-stage detection. Furthermore, the development of gender-specific models enhances diagnostic accuracy, highlighting the importance of personalized approaches in PD diagnosis.
In chapter 4, we shift our focus to predicting disease prognosis in PD by developing subject- specific models using baseline MRI and clinical data. The superiority of a logarithmic model over linear models in predicting late-stage severity is demonstrated, and a regression chain model achieves an average accuracy of over 85%. Notably, microstructural integrity measures play a significant role in predicting disease progression. The proposed model, leveraging MRI data, exhibits robustness and outperforms alternative imaging techniques.
The final chapter provides a comprehensive summary of the thesis findings and outlines potential future directions. By developing reliable biomarkers, improving diagnostic accuracy, and predicting disease progression, this thesis contributes to the advancement of PD research. The tools developed hold the potential to enhance personalized treatment planning, facilitate clinical decision-making, and aid in the enrollment of individuals with PD into clinical trials. Ultimately, this research endeavors to improve the lives of those affected by PD and drive progress in finding effective treatments for this debilitating disorder.
Henadeerage Don, Dimuthu, "MAGNETIC RESONANCE IMAGING BIOMARKERS FOR PARKINSON’S DISEASE: A MACHINE LEARNING APPROACH" (2023). Electronic Thesis and Dissertation Repository. 9624.
Available for download on Monday, January 01, 2024