杜卜勒血流速度估測之研究

Abstract

本篇論文討論了彩色血流儀之系統架構中兩個重要的部份。第一乃是回波 抑制處理。在回波強大的臨床環境中，我們嘗試使用無限與有限脈衝響應 之高通濾波器，以過濾低頻的回波信號。實驗結果顯示，若在無限脈衝響 應濾波器的暫存器初始值上採取適當的前處理，即使在極短的輸入信號通 過時，也不會產生暫態效應。我們也討論血流平均速度的估測法，主要有 自相關及自迴歸兩種方法。自相關估測法的計算簡單，且優於傳統的快速 傅立葉轉換估測法，但無法有效分辨血流信號與回波信號。自迴歸估測法 將信號頻譜上的極點相位對應到信號的波峰位置，其優點是在有限回波的 情況下，不必使用回波抑制濾波器也可以進行血流速度估測。 Two significant building blocks in Color Flow Imaging systems have been discussed in this study. The first part is dedicated to clutter suppression. When the clutter is dominant as is normal in clinical environment, FIR or IIR high-pass filters are used to suppress the low-frequency clutter component. Simulation results show that the transient effect can be mitigated even when very short data sequence was available to the IIR filter whose registers have been well-initialized. Another topic is concerned with mean frequency estimation of Doppler blood flow. Emphasis is placed on the AC (AutoCorrelation) and AR (AutoRegressive) methods. The AC estimator is easy to implement and superior to the conventional FFT estimator, but it lacks the capability of distinguishing between blood flow and clutter. By mapping phases to the spectral peaks of the backscattered signal, an AR estimator has the advantage of reliable estimation of blood flow without clutter suppression.