《厦门大学学报（自然科学版）》

颈椎椎体的分割在颈椎图像处理中起着关键的作用,是颈椎病灶确定和辅助诊断的重要基础.针对颈椎椎体边缘特征复杂的特点,提出一种基于水平集和最大稳定极值区域(maximally stable extremal regions,MSER)融合的颈椎椎体分割方法.首先采用基于图像密集度分布的图像分割方法对图像进行粗分割,自动提取颈椎区域; 然后采用改进的水平集方法提取出颈椎椎体的前缘轮廓; 根据颈椎椎体后缘的局部稳定特征,采用改进的MSER方法提取出椎体的后缘高亮区域,并结合椎体结构特征,采用最小二乘法拟合出椎体的后缘曲线; 最后融合颈椎椎体前缘轮廓与后缘曲线,从而提取完整的颈椎椎体.实验结果表明,该方法能有效地分割和提取颈椎椎体,提取的后缘曲线接近专家手工提取的结果,可以为颈椎病的临床诊断提供更客观的诊断依据.

Segmentation of cervical centrum plays a key part in cervical image processing,and is an important basis for confirming cervical lesions and auxiliary diagnosis.Targeting on the complexity of cervical centrum margin,this paper presents cervical centrum segmentation based on level set and combination of maximally stable extremal regions(MSER).First,rough segmentation of image will be undergone based on image distribution density to extract cervical regions automatically.Then,the improved level set is adopted to extract lip sketch of the cervical centrum.In line with partial stable features of cervical centrum trailing edge,the improved MSER is adopted to extract highlighted areas in rip of centrum.At the same time,with combination of structure features of the centrum,least square method is taken for centrum rip curve fitting.Finally,cervical centrum lip sketch and trailing curve are combined for extraction of completed cervical centrum.Experiment results show that methods used in this paper can efficiently segment and extract cervical centrum,and the result of trailing curve is similar to that by manual segmentation,and can offer more objective diagnostic basis for clinical diagnosis of cervical spondylosis.

引言
1 算法流程及原理
2 实验结果与分析
3 结论

图2 自动提取颈椎区域<br/>Fig.2 Automatic extraction of cervical regions

图2 自动提取颈椎区域
Fig.2 Automatic extraction of cervical regions

图1 分割流程图<br/>Fig.1 The flow chart of segmentation

图1 分割流程图
Fig.1 The flow chart of segmentation

图3 噪声区域剔除策略示意图<br/>Fig.3 Schematic diagram of the noise regions excluding strategy

图3 噪声区域剔除策略示意图
Fig.3 Schematic diagram of the noise regions excluding strategy

图4 CV-GAC方法与手工分割方法提取颈椎前缘轮廓<br/>Fig.4 Front edge of cervical contour extracted by CV-GAC and manual segmentation method

图4 CV-GAC方法与手工分割方法提取颈椎前缘轮廓
Fig.4 Front edge of cervical contour extracted by CV-GAC and manual segmentation method

图5 颈椎区域的MSER提取与噪声区域的剔除<br/>Fig.5 MSER extraction of cervical reguons and exclusion of the noise regions

图5 颈椎区域的MSER提取与噪声区域的剔除
Fig.5 MSER extraction of cervical reguons and exclusion of the noise regions

图6 本文算法与手工分割结果比较<br/>Fig.6 Results comparison of proposed segmentation and the manual segmentation

图6 本文算法与手工分割结果比较
Fig.6 Results comparison of proposed segmentation and the manual segmentation

表1 本文算法和手动分割测得的颈椎曲度<br/>Tab.1 Cervical curvature measured with proposed algorithm and manual segmentation

表1 本文算法和手动分割测得的颈椎曲度
Tab.1 Cervical curvature measured with proposed algorithm and manual segmentation

表2 3种类型颈椎不同方法测得的颈椎曲度<br/>Tab.2 Cervical curvature for three types cervical measured with different methods

表2 3种类型颈椎不同方法测得的颈椎曲度
Tab.2 Cervical curvature for three types cervical measured with different methods

[1] 张少群,李义凯.颈椎病研究的历史沿革[J].中国康复医学杂志,2016,31(11):1273-1276.
[2] 陈超.探讨不同的影像学方法诊断颈椎病的临床价值[J].检验医学与临床,2013,10(20):2694-2695.
[3] CHEN L L,LIN J X,XU T M,et al.The longitudinal sagittal growth changes of maxilla and mandible according to quantitative cervical vertebral maturation[J].Journal of Huazhong University of Science and Technology(Medical Sciences),2009,29(2):251-256.
[4] 蓝智聪,陈莉莉,许向阳,等.计算机辅助颈椎分析系统识别颈椎标志点的准确性和重复性探讨[J].临床口腔医学杂志,2010,26(9):526-530.
[5] 赵晓光,林久祥,王晴竹.颈椎影像计算机自动识别骨龄诊断系统的建立[EB/OL].(2013-09-03)[2017-06-17].http:∥www.paper.edu.cn/releasepaper/content/201309-17.
[6] 梁礼明,黄朝林,石霏,等.融合形状先验的水平集眼底图像血管分割[J/OL].计算机学报,2016.(2016-11-25).[2017-06-17].http:∥www.cnki.net/kcms/detail/11.1826.TP.20161125.2311.002.html.
[7] JEFFEE A I B,PAHL C,ABDULJABBAR H N,et al.Cervical segmentation in ultrasound image using level-set algorithm[C]∥Advances in Biomedicine and Health Science.Johor:[s.n.],2013:25-30.
[8] LU Z,CARNEIRO G,BRADLEY A P.An improved joint optimization of multiple level set functions for the segmentation of overlapping cervical cells[J].IEEE Transactions on Image Processing,2015,24(4):1261-1272.
[9] WANG J K,CHENG Y Z,GUO C Y,et al.Shape-intensity prior level set combining probabilistic atlas and probability map constrains for automatic liver segmentation from abdominal CT images[J].International Journal for Computer Assisted Radiology and Surgery,2016,11(5):817-826.
[10] CHENG Y Z,HU X,WANG J,et al.Accurate vessel segmentation with constrained B-Snake[J].IEEE Transactions on Image Processing,2015,24(8):2440-2455.
[11] 王阳萍,杜晓刚,赵庶旭,等.医学影像图像处理[M].北京:清华大学出版社,2012:69-109.
[12] MIKOLAJCZYK K,TUYTELAARS T,SCHMID C,et al.A comparison of affine region detectors[J].International Journal of Computer Vision,2005,65(1/2):43-72.
[13] 闻莎,游志胜.性能优化的同态滤波空域算法[J].计算机应用研究,2000,17(3):62-65.
[14] CHAN T F,VESE L.Active contours without edges[J].IEEE Transactions on Image Processing,2001,10(2):266-277.
[15] CASELLES V,KIMMEL R,SAPIRO G.Geodesic active contours[J].International Journal of Computer Vision,1997,22(1):61-79.
[16] TAO W B,TAI X C.Multiple piecewise constant with geodesic active contours(MPC-GAC)framework for interactive image segmentation using graph cuts optimization[J].Image and Vision Computing,2011,29(8):499-508.
[17] MIKOLAJCZYK K,TUYTELAARS T,SCHMID C,et al.A comparision of affineregion detectors[J].International Journal of Computer Vision,2005,65(1):43-72.
[18] MATAS J,CHUM O,URBAN M,et al.Robust wide baseline stereo from maximally stable extremal regions [C]∥Proceedings of the British Machine Vision Conference.Cardiff:[s.n.],2002:384-393.
[19] 王涛,周理乾,孙孟锟,等.6种颈椎曲度测量方法的可信度及可重复性比较[J].中国脊柱脊髓杂志,2015,25(4):323-327.
[20] 陈银海,姚红华,杨忠.颈椎曲度的X线测量在颈椎病康复评定中的应用价值[J].中国康复,2007,22(3):156-158.

备注

引言

1 算法流程及原理

2 实验结果与分析

3 结论

学报简介

备注

引言

1 算法流程及原理

2 实验结果与分析

3 结 论

学报简介

3 结论