gammaSTAR: A framework for the development of dynamic, real-time capable MR sequences

Area of Science:

• Medical Imaging
• Biophysics
• Computer Science

Background:

• Magnetic Resonance (MR) imaging sequence development is crucial for advancing diagnostic capabilities.
• Existing vendor-specific platforms can limit flexibility and innovation in MR sequence design.
• The need for real-time feedback in MR imaging is critical for improving image quality and enabling advanced applications.

Purpose of the Study:

• To introduce and evaluate the gammaSTAR MR sequence development framework.
• To demonstrate the real-time feedback capabilities of gammaSTAR.
• To showcase the advanced MR sequence library and its performance compared to vendor-specific sequences.

Main Methods:

• gammaSTAR integrates a Python frontend for sequence generation and a Lua backend for hardware instructions.
• A vendor-specific driver translates instructions for scanner compatibility.
• Real-time feedback is enabled via a dedicated interface, with prospective motion correction integrated into an EPI sequence.

Main Results:

• gammaSTAR sequences achieved image contrasts and qualities comparable to a leading vendor (Siemens Healthineers) at 1.5 T and 3 T.
• Complex sequences like pCASL GRASE and 3D radial UTE imaging were successfully implemented without extended preparation.
• Prospective motion correction significantly reduced motion artifacts, improving mean voxel displacement from 1.63 mm to 0.37 mm.

Conclusions:

• The gammaSTAR platform facilitates efficient MR sequence development.
• Real-time feedback capability is a key feature, demonstrated across diverse MR applications.
• gammaSTAR offers a flexible and powerful tool for advancing MR imaging and spectroscopy research.