|
磁振影像在臨床上提供高解析度、高訊雜比的影像，使得研究人員能夠判斷生理及組織狀態細微的變化。然而，無所不在的生理運動將導致影像假影，這假影將降低影像的精細度且破壞組織特性的量化資訊。
本研究的目的是要使用巡弋回波校正運動及擴散作用所造成的磁振影像假影，巡弋回波可以紀錄物體投影值位置及相位資訊，藉由算術的逆運算將可補償物體因運動導致的位移及相位的改變，使其能夠得到如同物體靜止不動時所獲得的清晰影像。在運動假影校正方面，我們採用三種方法計算物體住移量並比較其優劣。在擴散權重影像假影方面，我們將擴散梯度磁場分別加在三個方向以求得該擴散梯度場對擴散權重影像的影響，並評估使用巡弋回波技術之校正成效。
實驗結果顯示，利用巡弋回波能夠有效地校正生理運動的假影。在校正擴散權重影像方面，Y方向擴散影像假影的校正結果要較X及Z方向為佳。未來的研究目標將是加入校正Y及Z方向運動假影的能力;此外，並研究改進X及Z方向擴散權重影像假影校正不佳的原因。
The efficacy and applicability of magnetic resonance imaging in clinical applications rely on its ability in producing high resolution, high signal to noise ratio (SNR) images that accurately reflect the subtle physiological or pathological condition changes. However, in many occasions, the ubiquitous physiological motions limit the MRTs ability in obtaining clean and sharp images. The motion induced image artifacts decrease lesion clarity and anatomical details and ruin the quantitative information to do tissue characterizations.
In this study, we use the Navigator Echo to correct the motion-induced image artifacts. Navigator Echo can record the object projected position and phase information. Adaptive correction of the object motion is achieved by reversing the effects caused by object displacements and phase shifts. The method not only reduces motion artifact , but also uniquely diminishes motion-related image unsharpness. For the correction, we use three different type of methods to compute the object displacements and compare the corrected image quality. In addition, we evaluate the results after correcting the diffusion-weighted MR image where diffusion gradients are added in each of the three directions.
The results show that Navigator Echo can remove motion-related image artifact effectively, especially in correcting the diffusion-weighted image artifact along the Y direction. Because we have not tried to correct motion-induced artifact other than the X direction. The future goals will be to include correction of motion-induced artifacts in Y and Z directions and to improve the correction results for diffusion-weighted images along the X and Z directions.
|