|
|
Pupil Ultrasound Images Segmentation and Diameter Measurements Based on Improved Graph Cuts |
1 Biomedical Engineering and Instrument Science, Zhejiang University, Hangzhou 310027, China
2 Emergency Department of Second Affiliated Hospital of Zhejiang University, Hangzhou 310027, China |
|
|
Abstract The traditional measurement of pupil diameter is manually measured, but for patients who have ocular trauma or loss of consciousness this method is inconvenience. Aiming to solve the problem of large manual interaction in pupil diameter measurement and the weak measuring robustness, we used improved graph cut segmentation algorithm to segment pupil ultrasound images and measure pupil diameters. In this paper, we improved the traditional graph cut algorithm in two aspects. One is using adaptive threshold region growing to take place of manually seeds selection, which ensures the segmentation results while reducing the amount of manual interaction. The other one is increasing gradient information of the image into data entry portion of the energy function, which reduces the small area in the segmentation results and enhances the weak edges of segmentation. The proposed method realized automatic segmentation of image and automatic measurement of the pupil diameter. By employing the method we acquired dynamic changes of patient pupil diameter and provide a basis for clinical diagnosis. To verify the validity of the algorithm, we used the method to measure the diameter of ten patients’ dynamic pupil ultrasound images, and compared the results with that obtained by manual measurements. It is shown that the absolute error is less than 0.2 mm, the correlation coefficient is at least 0.83. In conclusion, the modified graphcut algorithm improved segmentation results, achieved automatical measurements of pupil diameters using dynamic ultrasound pupil images, and can be expected to substitute manual measurements and reduce the amount of human interaction.
|
|
|
|
|
[1]Wingate M, Hood B, Shi H, et al. Automated pupil size determination for evaluating fluctuations in physiological arousal [C] //Yue Minzhu, Feng Yang, eds. Signal Processing Proceedings. Beijing: IEEE, 1998:1666-1669.
[2]Parnandi A, GutierrezOsuna R. Contactless measurement of heart rate variability from pupillary fluctuations [C] //Suter G, eds. Affective Computing and Intelligent Interaction. Geneva: IEEE, 2013:191-196.
[3]Yoshida H, Mizuta H, Gouhara T, et al. Statistical properties of simultaneously recorded fluctuations in pupil diameter and heart rate [C] //Hood L, eds. Engineering in Medicine and Biology Society. Montreal : IEEE, 1995: 165-166.
[4]Yamaji K, Hiratai Y, Usui S. The pupil as a possible monitor as the autonomic nervous system [C] //Hood L, eds. Engineering in Medicine and Biology Society. Chicago: IEEE, 1997: 2777-2781.
[5]汪育文,保金华,吕帆. 三种瞳孔直径测量方法在不同测量状态下测量结果的比较和分析 [J]. 眼视光学杂志, 2009, 11(3): 218-220.
[6]Starck T, Liu Yun, Prewett AL, et al. Comparison of scotopic pupil measurement with slitlampbased cobalt blue light and infrared videobased system [J]. Journal of Cataract and Refractive Surgery, 2002, 28(11): 1952-1956.
[7]杨亚波, 俞一波, 付清. Ⅱ型糖尿病患者瞳孔相关性自主神经病变的研究 [J]. 中华眼科杂志, 2006, 42(7): 616-619.
[8]Yang H, Lee M, Kim JB, et al. Burstshot infrared digital photography to determine scotopic pupil diameter [J]. Journal of Cataract and Refractive Surgery, 2006, 32(12): 2113-2117.
[9]Schnitzler EM, Baumeister M, Kohnen T. Scotopic measurement of normal pupils: colvard versus video vision analyzer infrared pupillometer [J]. Journal of Cataract and Refractive Surgery, 2000, 26(6): 859-866.
[10]Arthur C, Dennis L. Comparison of the colvard pupillometer and the zywave for measuring scotopic pupil diameter [J]. Journal of Refractive Surgery, 2004, 20(3): 248-252.
[11]Fotiou F, Fountoulakis KN, Goulas A, et al. Automated standardized pupillometry with optical method for purposes of clinical practice and research [J]. Clinical Physiology, 2000, 20(5): 336-347.
[12]Pascal SCJ, Rosen P. Ocular examination technique for the emergency department [J]. The Journal of Emergency Medicine, 1997, 15(6): 793-810.
[13]Harries A, Shah S, Teismann N, et al. Ultrasound assessment of extraocular movements and pupillary light reflex in ocular trauma [J]. American Journal of Emergency Medicine, 2010, 28(8): 956-959.
[14]Terrence D. Ultrasound physics and instrumentation [J]. Surgical Clinics of North America, 1998, 78(2): 197-217.
[15]汪一休. 关于交互式图像分割算法的研究 [D]. 合肥: 中国科学技术大学, 2009.
[16]Chang Huali, Chen Zhenping, Huang Qinghua, et al. Graphbased learning for segmentation of 3D ultrasound images [J]. Neurocomputing, 2015, 151(2): 632-644.
[17]Malcolm J, Rathi Y, Tannenbaum A. Graph cut segmentation with nonlinear shape priors [C] //Mahesh P, eds. Image Processing. San Antonio: IEEE, 2007: 365-368.
[18]Slabaugh G, Unal G. Graph cuts segmentation using an elliptical shape prior [C] //Mahesh P, eds. Image Processing. San Antonio: IEEE, 2005: 365-368.
[19]Veksler O. Star shape prior for graphcut image segmentation [C] //David F, Philip T, Andrew Z, eds. Computer Vision. Marseille: Springer Berlin Heidelberg, 2008:454-467.[20]胡焦, 陈辉. 基于像素相似性与GraphCut 的图像自动分割 [J]. 现代电子技术, 2004, 37(2): 45-47.
[21]徐昱琳, 张雷, 陈万米,等. 基于SIFT 特征和图割算法的图像分割方法研究 [J]. 计算机测量与控制, 2009, 17(12): 2491-2493.
[22]Kolmogorov V, Zabih R. What energy functions can be minimized via graph cuts? [J]. IEEE Transactions on Pattern Analysis and Machine Intelligenc, 2004, 26(2): 147-159.
[23]Boykov Y, Veksler O, Zabih R. Fast approxamate energy minimization via graph cuts [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(11): 1222-1239.
[24]Boykov Y, Kolmogorov V. An experimental comparison of mincut/maxflow algorithms for energy minimization in vision [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(9): 1124-1137.
|
|
|
|