Research On MR Image Combination Method Based On Multiple-Channel Phased Array Coils

Magnetic resonance imaging (MRI) is one of the key methods in medical treatment and research, which has no injury characteristic. Receiving coil has a very important influence on the image quality because it is placed at the start of the receiving chain. Multi-channel phased array coils have much higher SNR, higher spatial resolution, faster imaging speed comparaed with other kinds of coils. Multi-channel imaging has become the important tasks in the field of MR. This work we research on MR image combination method based on multi-channel phased array coils.Coil sensitivity is an important issue in MRI, the knowledge of coil sensitivity is not generally available, so the accuracy of the estimation coil sensitivity has direct affect image quality. Based on the research, analysis and comparison on the current methods for image combination in MRI, it discovered that all of these methods application the virtual estimate to the calculation of coil sensitivity, they did not consider the influence of noise and MR signal and then reduced finally image quality. So this work proposes a new method that use smoothing filter to estimate coil sensitivity. It not only can reduce the noise, but also makes coil sensitivity information achieve maximized separation from the object information. So the coil sensitivity can better reflect its own attribute, image SNR and contrast have been greatly improved. The effectiveness and accuracy of this method are proved through the experimental comparison.Since the coil sensitivity estimated by the improved algorithm, some noise still exist in the process of image combination, a low resolution factor is utilized to minimize the influence of noise on the finally image. Experimental results show that this method can further improve image SNR and contrast and thus prove the effectiveness of the further improved method.However, with the influence of hardware conditions including multi-channel coils’own factor and the difference of coil coefficient, the finally image in MRI often appears inhomogeneities. It can not be resolved by the method above. An improved method based on PTH-NOMR combination is presented for inhomogeneity correction. Through the correction of the pathological sensitivities, image homogeneity has been great increased without compromising SNR and the finally images are more clearly. Experimental results prove the effectiveness of the improved method. Finally, the summary and the prospect on this research are given.