Loading...

Modified Computed Tomographic Imaging Systems with Improved Reconstructed Image Quality

Atashbar, Hamed | 2017

578 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 49891 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Kavehvash, Zahra
  7. Abstract:
  8. Computerized tomography (CT) imaging is a powerful tool among the existing bio-imaging techniques for capturing bio-images. In a CT scan procedure, linear sensors receive x-ray radiations, passed through the patient’s body and the reconstruction algorithms provide the required image for physicians from the captured data. In spite of its great advantages, due to the use of x-ray radiations and its ionization effect, it will raise the risk of cancer in long times. Therefore, to take benefit from several advantages of CT such as low-cost and high speed, solving this problem is aimed by many physicians and scientists. A new compressed sensing-based algorithm in order to reduce the number of acquired samples and thus x-ray radiation dose in a computerized tomographic system, is proposed in this thesis. In this study, we propose to use a novel dictionary for defining the cost function in compressed sensing algorithm. Our dictionary is an optimum combination of Wavelet Transform (WT) coefficients, Discrete Cosine Transform (DCT) coefficients, and Total Variation (TV) of the image. We utilize three quality assessment metrics including mean square error (MSE), peak signal to noise ratio (PSNR) and Structural Similarity index (SSIM) to quantitatively evaluate the reconstructed images. The results show that the proposed method can generate high quality images with less artifacts while preserving edges when fewer number of view angles are used for reconstruction compared to previous compressed-sensing based computerized tomography systems. This is in comparison with those results obtained from other reconstruction algorithms in a computerized tomography system
  9. Keywords:
  10. Computerized Tomography ; Discrete Cosine Transform ; Wavelet Transform ; Total Variation ; Compressive Sensing ; Image Resolution ; Sparse Operator

 Digital Object List

 Bookmark

No TOC