Loading...

Acceleration of Image Reconstruction on the Magnetic Resonance Imaging System

Rafiei, Ali | 2014

541 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 46795 (46)
  4. University: Sharif University of Technology
  5. Department: Energy Engineering
  6. Advisor(s): Vossoughi, Nasser; Vosughi Vahdat, Bijan; Zamani, Pooria
  7. Abstract:
  8. Magnetic Resonance Imaging ,MRI, although as one of the most advanced medical imaging equipment is well known but compared to other medical imaging systems needs more time to imaging, so it has caused the patient dissatisfaction. According to the recently researches, filling the image data in a space of information that called ‘k-space’, in the process of image reconstruction, is the main reason of slow MRI. Image reconstruction methods for accelerating like half scan based on image reconstruction by using only a part of k-space is used in MRI systems today, but this method is associated with lower SNR image. Recent research on the theory of compressed sensing has been opened a new approach in the MRI image reconstruction. Based on this theory can be recover the signal with sampling rate less than is expressed in nyquist theorem. In this study, after introducing procedures to accelerate the MRI image reconstruction, with an emphasis on compressed sensing and half-scan methods as the two important method based on image reconstruction with undersampled k-space, the proposed method is expressed. This method is a combination of half scan and compressive sensing method. Simulation show that to add compressed sensing to half-scan method, while maintaining acceleration of image reconstruction with 62.5%, could be increased SNR as much as 0.42 dB and reduce the relative error of the image as much as % 0.13 compared to use compressive sensing alone. In the half-scan method is also after the introduction of step merging filter and ramp filter, proposed merging filter with nonlinear region,Tan(x), that according to simulation is partly better than the two previous filter
  9. Keywords:
  10. Magnetic Resonance Imagin (MRI) ; Medical Images ; Acceleration ; Image Reconstruction ; Compressive Sensing ; Half Scan

 Digital Object List

 Bookmark

No TOC