A Research On Algorithm Of Gray-scale Inhomogeneity Correction In Magnetic Resonance

Magnetic resonance imaging(MRI)has been widely used in disease diagnose and scientific research.Because of the inhomogeneous signal receiving of the surface coil,MR image intensity is usually inhomogeneous,particularly at high field MRI.Intensity inhomogeneity in MRI affects not only image post-processing,but also the accuracy of clinical diagnose.In recent 30 years,a number of image intensity correction methods have been proposed.However,there is a common problem in most of those methods,i.e while image intensity inhomogeneity can be well corrected,image contrast may not be kept at the same time.To solve this problem,in this thesis,two solution are proposed.1?Intensity inhomogeneity correction based on multi-coil image histogram matching of the local area.Compared to surface coil,body coil images have good homogeneity but low SNR.Therefore,one can use the low resolution body coil image and surface coil image of the local area.This method can not only correct the intensity inhomogeneity,but also keep the contrast of the image.2?Intensity inhomogeneity correction method based on surface fitting using multi-coil image.Since the biased field of surface coil can be seen as a smooth surface,one can use body coil image and surface coil image of the corresponding location to fit the biased field.The corrected image using this method will have the tissue contrast of the body coil image.Through subjective quantitative analysis,the proposed method not only has good image correction effect but also has high time efficiency compared to some classical intensity inhomogeneity correction methods.