【正文】 cerebral aneurysm can be detected from the three kinds of branch structures by judging branch structure caused by cerebral aneurysm. In this paper, by doing depthfirst traversal search on skeleton tree, the pixel length of branches will be the basis of detection. The specific steps was conducted as follows. (1) Traverse the entire skeleton tree from the root of skeleton tree, and mark the pixels which have been visited。t visited. According to empirical value, T1 is set to 8 and T2 is set to 16, which can eliminate the burr to some extent and the interference to the detection of cerebral aneurysm from normal vessel branch. This can obtain the best effect. IV. EXPERIMENTS AND ANALYSIS The hardware environment used in the experiment is CPU: P (R) 4 。s a betterperforming algorithm, but the contrast in the experiment appears different effects of skeleton extraction, Figure 8(a) with less burrs of skeleton, while Figure 8(b) with more burrs in skeleton. After a large number of experiments, the reason is found that the emergence of burr is largely due to the inadequate binary effect of blood vessels, noise jamming and other reasons. Burr in the vascular skeleton structure also manifested as the same branch structure, so it affects on the precision of cerebral aneurysm detection greatly. As shown in Figure8, the position marked by pane is the detected cerebral aneurysm part. The detection effect in Figure8 (a) is good, while two wrong regions exist in Figure 8(b). In this paper, the cerebral aneurysm detection method based on skeleton features can basically determine the position of cerebral aneurysm. Although the length of branch used as a basis will encounter the interference of skeleton burr and exist incorrect detection, the significance of cerebral aneurysm detection is to reduce the numbers of detection objectives of cerebral aneurysm and provide data premise for recognition. How to do binarization effectively, provide best foundation to skeleton extraction, and reduce adverse effect on cerebral aneurysm detection caused by burr will be the future research emphasis. V. CONCLUSION This paper analyzes the morphological characteristics of cerebral aneurysm, researches OPTA and its improved algorithms deeply and applies them to the extraction of skeleton image successfully. A detection method of cerebral aneurysm based on skeleton is proposed, which detected cerebral aneurysm by the length of branch element. Experimental results show that 19 the regions of cerebral aneurysm can be detected by this method after the extraction of branch elements on skeleton map. Furthermore, the accuracy of detection is affected by the effect of cerebral vascular segmentation and the extraction of skeleton. How to overe these shortings and improve the success rate of detection is the focus of future works. ACKNOWLEDGMENT This research was partially supported by the National Natural Science Foundation of China(60673092) and the opening project of JiangSu Province Support Software Engineering Ramp。 display card: 128 M. The development environment is VC . Experimental data of cerebral vascular DSA images, whose format accords with DICOM standard, are provided by the First Affiliated Hospital of Suzhou University. By using DICOM software developed by our institute, each DSA image is divided into DSA image sequences, which are saved as BMP format. In the experiment, we do binarization segmentation on original images Figure 1(a) and Figure 1(b) firstly, then obtain its skeleton by using improved thinning algorithm, and detect the skeleton map using the cerebral aneurysm detection method based on the branches of skeleton map, lastly mark the suspected region of cerebral aneurysm with pane. The vascular structure of original cerebral vascular image in Figure 1(a) is relatively simple, and the image edge is smooth after binarization segmentation, which shown in Figure 7(a)。 (3) Retrieve an element from the stack as a beginning point, continue to traverse pixels of skeleton tree with the above method, until there is no point that can be visited. Simultaneously use variable S to count the pixels which have been visited from the beginning point to the end. If the S is more than threshold T1 and less than threshold T2, the location of the curve is the part of cerebral aneurysm, and it can be considered as burr if S is less than T1, normal blood 17 vessels branch if S is more than T2。 q10=0 amp。 q7=0 amp。 q10=1 amp。 q11=0) || (q8=1 amp。 q7=0 amp。 q14=0) || ( q8=0 amp。 q13=0 amp。 q6=1 amp。 q11=0) || (q2=0 amp。 q14=0) (1) (q2=1 amp。 q7=0) || (q5=1 amp。 pixel value, and it uses logical operation “amp。 聲明 這次研究由中國(guó)自然科學(xué)基金會(huì) (60673092)以及江蘇省現(xiàn)代信息技術(shù)應(yīng)用軟件工程研究中心 。 提出了一種