本文選題：呼吸 + 肝臟運動��； 參考：《浙江大學(xué)學(xué)報(工學(xué)版)》2014年09期

【摘要】：為補償肝臟運動,分析呼吸引起的肝臟空間運動和形變,并嘗試將其簡化為剛性運動模型進行補償.利用基于表面和基于互信息的3D-3D圖像剛性配準(zhǔn)方法分析4例肝臟空間運動的規(guī)律并進行補償,比較使用手工分割肝臟輪廓和肝臟灰度信息進行配準(zhǔn)的結(jié)果及其補償效果,評估2種思路的可靠性,并進一步評價剛體運動模型簡化的可行性.結(jié)果表明:剛性運動模型、簡化的平移運動模型和矢狀面2D運動模型在實際應(yīng)用中具有可行性,此3種模型都可以補償80%以上的呼吸影響;使用肝臟灰度信息比手工分割輪廓建模和補償?shù)慕Y(jié)果更可靠.上述結(jié)論對在超聲融合多模成像中快速建模和實時補償呼吸對肝臟運動的影響有一定的參考價值.
[Abstract]:In order to compensate for liver motion, the spatial movement and deformation of the liver caused by respiration were analyzed and simplified as rigid motion model. The method of 3D-3D image rigid registration based on surface and mutual information was used to analyze the law of liver spatial motion in 4 cases and to compensate for it. The results and the compensation effect of using manual segmentation of liver contour and liver gray level information were compared. The reliability of the two ideas is evaluated and the feasibility of simplification of rigid body motion model is further evaluated. The results show that the rigid motion model, the simplified translational motion model and the sagittal 2D motion model are feasible in practical application, and the three models can compensate more than 80% of the respiratory effects. Using liver gray level information is more reliable than manual segmentation contour modeling and compensation. These conclusions have some reference value for fast modeling and real time compensation of respiration on liver motion in multimode ultrasound fusion imaging.
【作者單位】： 清華大學(xué)生物醫(yī)學(xué)工程學(xué)系;北京大學(xué)腫瘤醫(yī)院;北京市腫瘤防治研究所介入科 惡性腫瘤發(fā)病機制及轉(zhuǎn)化研究教育部重點實驗室;
【基金】：國家自然科學(xué)基金資助項目(81271671,81271735)
【分類號】：R318

【相似文獻】

相關(guān)會議論文 前1條

1 李奇;徐之海;馮華君;陳躍庭;;幀間旋轉(zhuǎn)圖像序列的運動矢量探測方法[A];浙江省光學(xué)學(xué)會第九屆學(xué)術(shù)年會暨新型光電技術(shù)青年論壇論文集[C];2005年

，

本文編號：2002199