雙頻血管內超音波探頭於血管狹窄評估之應用

Abstract

冠狀動脈疾病是現代社會中致死率極高的心血管疾病，其主因為血管內斑塊堆積，導致狹窄進而阻礙心肌血流供應。臨床上常以血流儲備分數(Fractional Flow Reserve, FFR)作為評估狹窄的黃金指標，透過狹窄前後的壓力反映血流動力學之功能性影響，並由管壁剪應力(Wall Shear Stress, WSS)評估斑塊穩定性與進展。然而，傳統FFR量測需仰賴壓力導絲，在檢查時間、成本及風險上，皆有一定之限制。本研究提出一套以雙頻血管內超音波(Dual-frequency Intravascular Ultrasound, IVUS)為核心的整合系統，旨在透過體外定位及血管內影像，重建血管幾何、估算血流速度，並經計算流體力學(Computational Fluid Dynamics, CFD)模擬推估FFR與WSS。本系統採用之雙頻探頭為單一壓電元件，經由厚度振動模式產生約45 MHz的高頻訊號以進行血管成像，並同時由輪廓振動模式產生約10 MHz的低頻訊號，搭配一維線性陣列體外接收，結合電磁追蹤系統(Electromagnetic Tracking System, EMTS)，將切面相對座標轉換為三維空間資訊，以建立血管模型。此外，為了突破側視型IVUS無法量測血液流速對應之都卜勒頻移的限制，本研究基於去相關性處理，以訊號的二階矩(second-order moment)作為流速估算子。定位驗證方面，EMTS於三軸方向定位的均方根誤差為(0.037, 0.037, 0.063) mm，具備高精度潛力。血管重建結果方面，在不同彎曲度(Tortuosity Index)與狹窄程度(Diameter Stenosis)下，隨狹窄程度上升：(1)彎曲度變化不顯著，但彎曲效應改變截面形狀趨於橢圓，(2)平均半徑下降，反映幾何收縮。流速估算方面，在血管入口處的平均流速估算值與參考值差距6至59 mm/s，初步支持估算模型之可行性。CFD模擬方面，FFR隨狹窄加劇而下降，WSS於狹窄入口處的峰值則上升；並且彎曲導致WSS出現方向性調變與局部渦旋現象。分析幾何誤差對FFR的影響，發現當狹窄程度低於70%時，小幅度的管徑或彎曲誤差對其影響有限；惟於高度狹窄模型中，流速估算誤差可能造成FFR值明顯變化。總結而言，本論文整合雙頻IVUS於血管重建、流速估算與CFD模擬之整體流程，並成功推估FFR與WSS兩項重要指標，能有效降低傳統FFR量測風險與成本，省去術中X光血管攝影及壓力導絲的使用，為血管狹窄評估提供創新的可行方案。

Coronary artery disease (CAD) is a leading cause of death in modern society, primarily caused by plaque accumulation that leads to arterial stenosis and impairs myocardial perfusion. Clinically, Fractional Flow Reserve (FFR) is regarded as the gold standard for assessing the functional severity of stenosis. Additionally, Wall Shear Stress (WSS) is used to evaluate plaque stability and progression. However, traditional FFR measurement relies on pressure guidewires, which increases the procedure duration, cost, and risk. This study proposes an integrated system based on dual-frequency intravascular ultrasound (IVUS), using extracorporeal localizing and intravascular imaging to reconstruct vascular geometry, estimate blood flow velocity, and simulate FFR and WSS via computational fluid dynamics (CFD). The dual-frequency transducer employed in this system utilizes a single piezoelectric element: high-frequency signals (~45 MHz) generated by thickness mode for imaging vascular walls and low-frequency signals (~10 MHz) simultaneously generated by contour mode for external localization of the IVUS transducer. Integrated with an electromagnetic tracking system (EMTS), 3D spatial information is obtained for vascular model reconstruction. In localization validation, the EMTS achieved root-mean-square errors of (0.037, 0.037, 0.063) mm across three axes, indicating high spatial accuracy. In vascular reconstruction, increasing stenosis led to two effects observed under varying tortuosity index and diameter stenosis: (1) tortuosity is not significantly affected, but the curvature effects led to cross-sectional deformation, and (2) the average radius decreased, reflecting geometric contraction. In CFD simulations, FFR decreased with increasing stenosis, while WSS peaks rose at the stenosis inlet and exhibited directional modulation and local vortices. Sensitivity analysis revealed that minor geometric errors in diameter or tortuosity had a limited impact on FFR when stenosis was below 70%, but in high-stenosis models, flow velocity estimation errors could significantly affect FFR values. In summary, this research presents a comprehensive workflow integrating dual-frequency IVUS for FFR and WSS calculation. The proposed system successfully estimated the key functional indicators without the need for intraoperative X-ray angiography or pressure guidewires, offering an innovative tool and feasible solution for stenosis assessment with reduced risk and cost.