Generating Synthetic Radiological Images with PySynthMRI: An Open-Source Cross-Platform Tool

Area of Science:

• Medical Imaging
• Computational Imaging
• Open-Source Software

Background:

• Synthetic MR imaging reconstructs multiple contrasts from one scan, reducing acquisition time.
• Current commercial synthetic MRI tools use vendor-specific methods and lack customization for multiparametric imaging.
• There is a need for flexible, open-source solutions for generating synthetic MR contrasts.

Purpose of the Study:

• Introduce PySynthMRI, an open-source Python tool for generating synthetic MR images.
• Provide a user-friendly interface for creating diverse radiological contrasts via adjustable sequence parameters.
• Enable customization and extension of synthetic MR imaging techniques.

Main Methods:

• PySynthMRI utilizes input images (e.g., T1, T2 maps) and user-defined scanner parameters (TE, TR, TI) to calculate pixelwise signal intensity.
• The tool offers a graphical interface for parameter selection and includes default synthetic contrasts (T1w, T2w, FLAIR, DIR).
• Generated synthetic images can be exported in DICOM or NiFTI formats, supporting various research workflows.

Main Results:

• PySynthMRI successfully generates synthetic MR images with different contrasts from quantitative maps.
• The software is compatible with Linux, Windows, and MacOS, available as a script or executable.
• Its modular design allows for user customization and the addition of new features.

Conclusions:

• PySynthMRI offers a fast, flexible, and open-source solution for synthetic MR image generation.
• The tool empowers researchers to create custom contrasts and integrate multiparametric imaging techniques.
• PySynthMRI facilitates advanced MR imaging research through its accessibility and adaptability.