使用體素內不同調運動磁共振影像這項技術,可以同時得到兩種擴散係數,第一種是組織間水分子的擴散係數 (D);另一個是在微血管網絡中,微血流的不同調流動的擴散係數 (D*)。然而,在相關的文獻中,所得到的肝組織D*數值彼此並不一致。此論文旨在探討在使用3T MR system的狀況下,健康的肝組織中D*數值的再現性。影響D*數值的因素可能包括: 影像訊雜比、b值的數目和感興趣區域的血管。在本論文中,我們提出修正版的曲線擬合方法,並搭配評估擬合錯誤的機制來進行分析。藉由這兩種方法,結果發現在平均六次的影像中,D*數值的再現性有相當的改善。未來相關研究可以使用此種分析方法,但仍需最佳化此技術的掃描時間。
Intravoxel incoherent motion diffusion-weighted magnetic resonance imaging (IVIM DW MRI) is a technique that can quantitatively assess both the water molecular diffusion in static tissues (true diffusion coefficient, D) and the incoherent motion of blood water molecules in capillary networks due to perfusion (pseudo-diffusion coefficient, D*). However, D* value in liver parenchyma appears to be of high uncertainty among literatures. The aim of this study is to determine the reproducibility of D* in normal liver at 3T, considering SNR, b value sampling numbers and the existence of large vessel in ROI. A modified fit approach combined with an error evaluation method was proposed. Through the proposed method, the reproducibility of D* in 6-time-averaged liver images improved. Future studies can utilize this procedure for IVIM liver investigation while the scanning time can still be further optimized.