Performance Evaluation of Total Variation based Compressed Sensing MRI for Different Sampling Patterns
T. Surya Kavitha1, K. Satya Prasad2
1T.Surya Kavita, Research Scholar, Department of Electronics and Communication Engineering, JNTUK, Kakinada, Andhra Pradesh, India,
2Dr. K. Satya Prasad, Rector, Vignan’s Foundation for Science, Technology and Research, Guntur, Andhra Pradesh, India.
Manuscript received on 10 September 2019. | Revised Manuscript received on 24 September 2019. | Manuscript published on 30 September 2019. | PP: 148-154 | Volume-8 Issue-11, September 2019. | Retrieval Number: K12620981119/2019©BEIESP | DOI: 10.35940/ijitee.K1262.0981119
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Magnetic Resonance Imaging (MRI) has been utilized broadly for clinical purposes to portray human anatomy due to its non-intrusive nature. The information acquisition method in MRI naturally picks up encoded signals (Fourier transformed) instead of pixel values and is called k-space information. Sparse reconstruction techniques can be executed in MRI for producing an image from fewer measurements. Compressive sensing (CS) technique samples the signals at a rate lower than traditional Nyquist’s rate and thereby reduces the data acquisition time in MRI. This paper investigates a new proposed sampling scheme along with radial sampling and 1D Cartesian variable density sampling. For various sampling percentages, subjective and quantitative analyses are carried out on the reconstructed Magnetic Resonance image. Experimental results depicts that the high sampling density near the center of k-space gives a better reconstruction of compressing sensing MRI.
Keywords: Cartesian Sampling, Compressive Sensing, k-space, MRI, Radial Sampling
Scope of the Article: Performance Evaluation of Networks