Reproducibility of a network atrophy biomarker for Parkinson's disease
A LivingPark paper replication
This project is a part of the LivingPark project led by Tristan Glatard (Concordia University), Jean-Baptiste Poline (McGill), and Madeleine Sharp (McGill). See here for more information about the broader project.
Parkinson’s disease (PD) is a severe neurodegenerative disorder characterized by both motor and non-motor symptoms. It is currently not possible to reliably predict a PD patient’s disease course from their current state. There have been several attempts to identify magnetic resonance imaging-based (MRI) biomarkers of disease progression in PD (Mitchell et al., 2021). Notably, Zeighami and colleagues used deformation-based morphometry (DBM) and independent component analysis (ICA) to produce a PD-network atrophy biomarker that predicts changes in clinical and cognitive outcomes over a 4.5-year time period (Zeighami et al., 2015, 2019). Although these results are promising, they have not yet been reproduced nor replicated in an independent dataset. Ensuring the robustness of biomarkers is important for any potential future clinical application. We attempt to reproduce the anatomical signature of this biomarker using the same dataset as the original study.
GitHub repository: zeighami-etal-2019.
People
Michelle Wang
