Reducing Variability in Deep Gray Matter QSM Using Differential ROI Referencing: A Phantom and In Vivo Evaluation

Area of Science:

• Neuroimaging
• Biophysics
• Medical Physics

Background:

• Quantitative susceptibility mapping (QSM) measures tissue magnetic susceptibility.
• Susceptibility values are relative, necessitating referencing to mitigate variability.
• Deep gray matter analysis in neurodegenerative diseases requires robust QSM referencing.

Purpose of the Study:

• To validate a differential region of interest (ROI) reference method for QSM.
• To assess the method's feasibility and robustness in deep gray matter analysis.
• To evaluate its utility in detecting group differences in Parkinson's disease (PD).

Main Methods:

• Digital phantom and in vivo studies were conducted.
• Differential ROI referencing was applied to QSM data.
• Statistical analysis (ANOVA, FDR-BH) compared PD patients and controls.

Main Results:

• The differential ROI method showed stable susceptibility estimates in phantoms.
• In vivo, referenced QSM values revealed consistent, significant group differences between PD patients and controls.
• Unreferenced QSM values exhibited greater variability.

Conclusions:

• The differential ROI reference method reduces QSM variability from acquisition and processing.
• This approach offers a practical, pathology-resilient referencing strategy for deep brain analysis.
• It shows promise for neuroimaging studies of neurodegenerative diseases.