Study Of Parameter Mapping In Magnetic Resonance Imaging Based On Enhanced Compressed Sensing

With the development of science and technology as well as the enhance of people livingstandard, the human health make the whole society has paid more and more attention, under thisbackground, the medical image become sound momentum of development, so the research onmedical image direction become popular and become a new generation of sunrise industry inChina.Current medical image has been widely applied in medical fields such as magneticresonance imaging (MRI), ultrasound, CT and other methods, especially magnetic resonanceimaging (MRI),that is because of its high imaging quality and no damage to the human body,the application of MRI has become more and more widely used. The paper proposed the newmethods based on magnetic resonance imaging parameters, according to the imaging speed is tooslow, the scan time is too long,base on this problems, improved compressed sensing algorithmused in parameter mapping of magnetic resonance imaging are proposed, aims to increase thespeed of magnetic resonance imaging parameter mapping as well as guarantee the quality ofimaging. In this paper, the improved algorithm based on compressed sensing is applied inmagnetic resonance imaging parameter mapping.In compressed sensing theory wavelettransform is used as the sparse transformation but in the process of MRI parameter mapping,thenewest method is that principal component analysis is used as the sparse transformation on thetimeline and is proved can get the better results than Fourier in this paper,then two improvedmethods are put forward as the sparse transform on the timeline in magnetic resonance imagingparameter mapping.First, using two-dimensional principal component analysis instead of onedimensional principal component analysis as the sparse transformation in the process ofparameter imaging on the timeline in order to achieve the goal of shortening the time of imaging,use this method to deal with the experimental data can get the ideal conclusion which is theimaging speed greatly is shortened and the quality of the reconstructed images also enhancedwith a small range. Second, after analyze the result of the new method, we found there is stillthe need to improve imaging quality, and based on two-dimensional principal componentanalysis this paper put forward a new improved method, which is used wavelet transformationcombined principal components analysis to replace two-dimensional principal componentanalysis as the sparse transformation, we called it the WAVAPCA algorithm, on the condition ofthe one dimensional principal component analysis as the sparse transformation,we useWAVEPCA at first before the processing of raw data,and we can get wavelet coefficients after wavelet transform, then we use one dimensional principal component analysis to processthe coefficients to get the sparser express. In the compressed sensing theory, the sparser thedata is, the better reconstructed images we can get, through the simulation processing ofexperimental data,the ideal result is showed that the imaging quality is improved. This twomethods proposed in this paper, one is to improve the imaging quality, another is to shorten theimaging time. In the future the study direction of magnetic resonance imaging parametermapping can be explored on the purpose to combine the aspects.