【摘要】：針對分割之后的腦血管體數(shù)據(jù)場,提出基于球B樣條曲線結合樹狀結構的表示模型.該模型中主要分為3個步驟表示腦血管:首先采用Hessian矩陣結合彈性球的方法獲取腦血管的中心線和半徑信息,可提高中心線的求取精度并且在中心線的求取過程中即可獲得半徑;然后對獲得的中心線和半徑進行樹狀重構,得到符合血管的形態(tài)拓撲結構,便于對腦血管進行遍歷;最后對重構之后的每一分支血管采用球B樣條曲線的方式來描述,以對腦血管病變區(qū)域進行檢測,并能針對不同大小的顯示窗口采用多尺度方式繪制.實驗結果表明,文中模型在腦血管的顯示、病灶檢測等方面具有極大優(yōu)勢.
[Abstract]:A representation model based on spherical B-spline curve and tree structure is proposed for the segmented cerebrovascular body data field. The model is mainly divided into three steps: firstly, the information of the center line and radius of cerebral vessels is obtained by using Hessian matrix and elastic sphere. It can improve the accuracy of the central line and obtain the radius in the process of obtaining the center line, and then the obtained center line and radius are reconstructed in tree shape, and the morphological topological structure of the blood vessel is obtained, which is convenient to traverse the cerebral vessels. Finally, each branch of the reconstructed blood vessel is described by the B-spline curve of the ball, which detects the area of cerebrovascular disease, and can be plotted in multi-scale mode for the display window of different sizes. The experimental results show that the model has great advantages in the display of cerebral vessels and the detection of lesions.
【作者單位】： 北京師范大學信息科學與技術學院;
【基金】：國家自然科學基金(61003134,61170170) 北京市自然科學基金重點項目(4081002)
【分類號】：R312;TP391.41

【參考文獻】

相關期刊論文 前3條

1 丁頤,劉文予,鄭宇化;基于距離變換的多尺度連通骨架算法[J];紅外與毫米波學報;2005年04期

2 馬p，

本文編號：2228606