The Software Implementation Of DSP-based Pulse Programmer In MRI Spectrometer

The system of Magnetic Resonance Imaging (MRI), as a mainstream of medical imaging technology, has been widely used in clinical and medical research. The performance of MRI spectrometer, which has been working as the kernel part of MRI system, has a fateful influence upon the quality of the reconstructed image. While pulse programmer is the control kernel of MRI spectrometer, the main function of pulse programmer is to realize the operation of pulse sequence and to control each function module. The breakthrough in MRI, computer, digital processing technology (especially DSP) and other advanced technologies, has provided a good foundation for the further research and development in pulse sequence.In this paper, the software implementation of DSP-based pulse programmer in MRI spectrometer has been achieved, which is referred to the development of magnetic resonance imaging sequence, data communication, C programming and assembly language programming. The principal contents realized in this paper are listed as below:(1) The programming work of pulse programmerRealizing the handling of run-time information in the DSP procedure, which are using state-query-based respiratory gating during the selection of data acquisition time (reducing the artifacts in reconstructed image while improving the quality of the image), and interrupt processing for operation exceptions, such as the overflow of the status bits (realizing real-time monitoring for the status bit in each function module).Achieving the expansion of the existing timing function in pulse sequence (C language) in PC, realizing large dynamic range of precise timing, for simplifying the program complexity.(2) The programming work of data communicationRealizing the USB-based data communication in MRI spectrometer (the parameter downloading and data uploading between the PC and the kernel part of MRI spectroscopy-DSP), with which the circuit design and the program have been both simplified, and the miniaturization of MRI spectroscopy has also been realized.