A Digital Phantom for 3D MR Spectroscopy Data Simulation

Motivation: Simulating MRS and MRSI data is essential for validating new methodologies, yet optimal practices remain undefined due to varied signal modeling and validation approaches. Goal(s): This study proposes a comprehensive, modular simulation framework that provides a high-quality, digital MRS phantom as an open-source Python tool, enhancing data reproducibility and validation flexibility. Approach: A modular Python framework combines anatomical brain models with metabolic data from a curated literature database, allowing customization of simulation parameters for flexible application. Results: The resulting framework efficiently simulates MRS(I) data in a batch-wise format, producing spectra ready for further analysis. Impact: This MRS phantom framework provides a flexible, tissue-specific model for realistic MRS and MRSI dataset simulation. Its modular design allows for precise control over the parameters and the simulation process to support a wide range of research and development applications.