This paper presents an automatic detection method for thin boundaries of silver-stained endothelial cells (ECs) imaged using light microscopy of endothelium mono-layers from rabbit aortas. To achieve this, a segmentation technique was developed, which relies on a rich feature space to describe the spatial neighbourhood of each pixel and employs a Support Vector Machine (SVM) as a classifier. This segmentation approach is compared, using hand-labelled data, to a number of standard segmentation/thresholding methods commonly applied in microscopy. The importance of different features is also assessed using the method of minimum Redundancy, Maximum Relevance (mRMR), and the effect of different SVM kernels is also considered. The results show that the approach suggested in this paper attains much greater accuracy than standard techniques; in our comparisons with manually labelled data, our proposed technique is able to identify boundary pixels to an accuracy of 93%. More significantly, out of a set of 56 regions of image data, 43 regions were binarised to a useful level of accuracy. The results obtained from the image segmentation technique developed here may be used for the study of shape and alignment of ECs, and hence patterns of blood flow, around arterial branches. 1. Introduction Atherosclerosis is the major cause of cardiovascular morbidity and mortality, with underlying pathological processes that may begin during childhood [1, 2]. Although the exact causes of atherosclerosis are not clear, it is thought to involve lipoprotein influx into the wall, across the endothelium, and chronic inflammation. Over time, lipids accumulate in the inner wall and plaques develop, resulting in reduction or blockage of blood flow. Eventually, this condition can lead to heart attacks and strokes [3, 4]. A striking feature of the disease is its nonuniform distribution within the arterial system. This is most evident in regions of branching and curvature and has therefore been attributed to spatial variation in mechanical forces, particularly the haemodynamic wall shear stress exerted on the endothelium by the flow of blood. Near-wall blood velocity, on which shear stress depends, cannot be accurately measured by direct techniques. However, ECs form a monolayer between the blood and arterial wall [3] that is regulated by haemodynamic forces through flow-mediated signal transduction [3, 5]. Of relevance to the present study, endothelial cells and their nuclei align with the predominant flow direction and elongate in response to increased shear. Therefore, ECs can be
References
[1]
S. R. Daniels, F. R. Greer, and the Committee on Nutrition, “Lipid screening and cardiovascular health in childhood,” Pediatrics, vol. 122, no. 1, pp. 198–208, 2008.
[2]
B. W. McCrindle, E. M. Urbina, B. A. Dennison, et al., “Drug therapy of high-risk lipid abnormalities in children and adolescents: a scientific statement from the American Heart Association Atherosclerosis, Hypertension, and Obesity in Youth Committee, Council of Cardiovascular Disease in the Young, with the Council on Cardiovascular Nursing,” Circulation, vol. 115, no. 14, pp. 1948–1967, 2007.
[3]
P. F. Davies, “Flow-mediated endothelial mechanotransduction,” Physiological Reviews, vol. 75, no. 3, pp. 519–560, 1995.
[4]
P. Libby, P. M. Ridker, and A. Maseri, “Inflammation and atherosclerosis,” Circulation, vol. 105, no. 9, pp. 1135–1143, 2002.
[5]
A. M. Malek, S. L. Alper, and S. Izumo, “Hemodynamic shear stress and its role in atherosclerosis,” Journal of the American Medical Association, vol. 282, no. 21, pp. 2035–2042, 1999.
[6]
J. T. Flaherty, J. E. Pierce, V. J. Ferrans, et al., “Endothelial nuclear patterns in the canine arterial tree with particular reference to hemodynamic events,” Circulation Research, vol. 30, no. 1, pp. 23–33, 1972.
[7]
J. Kittler and J. Illingworth, “Minimum error thresholding,” Pattern Recognition, vol. 19, no. 1, pp. 41–47, 1986.
[8]
S. L. Al-Musawi, J. Bishton, J. Dean, et al., “Evidence for a reversal with age in the pattern of near-wall blood flow around aortic branches,” Atherosclerosis, vol. 172, no. 1, pp. 79–84, 2004.
[9]
A. R. Bond, S. Iftikhar, A. A. Bharath, et al., “Morphological evidence for a change in the pattern of aortic wall shear stress with age,” Arteriosclerosis, Thrombosis, and Vascular Biology, vol. 31, pp. 543–550, 2011.
[10]
C. J. Burges, “A tutorial on support vector machines for pattern recognition,” Data Mining and Knowledge Discovery, vol. 2, pp. 121–167, 1998.
[11]
E. Z. Hirsch, W. Martino, C. H. Orr, et al., “A simple rapid method for the preparation of en face endothelial (H？utchen) monolayers from rat and rabbit aortas,” Atherosclerosis, vol. 37, no. 4, pp. 539–548, 1980.
[12]
S. Iftikhar and A. A. Bharath, “Cell boundary analysis using radial search for dual staining techniques,” in Medical Imaging 2009: Image Processing, vol. 7259 of Proceedings of the SPIE, Lake Buena Vista, Fla, USA, February 2009.
[13]
C. Kotropoulos and I. Pitas, “Segmentation of ultrasonic images using support vector machines,” Pattern Recognition Letters, vol. 24, no. 4-5, pp. 715–727, 2003.
[14]
R. O. Duda, P. E. Hart, and D. G. Stork, Pattern Classification, vol. 24, John Wiley & Sons, New York, NY, USA, 2nd edition, 2007.
[15]
H. Peng, F. Long, and C. Ding, “Feature selection based on mutual information: criteria of max-dependency, max-relevance, and min-redundancy,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 27, no. 8, pp. 1226–1238, 2005.
[16]
J. Shawe-Taylor and N. Cristianini, Support Vector Machines and other Kernel-Based Learning Methods, Cambridge University Press, New York, NY, USA, 1st edition, 2000.
[17]
A. A. Bharath and J. Ng, “A steerable complex wavelet construction and its application to image denoising,” IEEE Transactions on Image Processing, vol. 14, no. 7, pp. 948–959, 2005.
[18]
D. Glotsos, P. Spyridonos, D. Cavouras, et al., “An image-analysis system based on support vector machines for automatic grade diagnosis of brain-tumour astrocytomas in clinical routine,” Informatics for Health and Social Care, vol. 30, no. 3, pp. 179–193, 2005.
[19]
K. K. Chin, Support Vector Machines Applied to Speech Pattern Classification, M.S. dissertation, University of Cambridge, UK, 1998.
[20]
C.-C. Chang and C.-J. Lin, “Libsvm: a library for Support Vector Machines,” 2011, http://www.csie.ntu.edu.tw/~cjlin/libsvm.
[21]
T. MathWorks. Matlab, Version 7.9.0 (R2009b) on PC Windows, http://www.mathworks.co.uk/.
[22]
J. Canny, “A computational approach to edge detection,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 8, no. 6, pp. 679–698, 1986.
[23]
M. Sezgin and B. Sankur, “Survey over image thresholding techniques and quantitative performance evaluation,” Journal of Electronic Imaging, vol. 13, no. 1, pp. 146–165, 2004.
[24]
N. Otsu, “A threshold selection method from gray-level histograms,” IEEE Transactions on Systems, Man and Cybernetic, vol. 9, no. 1, pp. 62–66, 1979.
[25]
J. F. Cornhill, M. J. Levesque, E. E. Herderick, et al., “Quantitative study of the rabbit aortic endothelium using vascular casts,” Atherosclerosis, vol. 35, no. 3, pp. 321–337, 1980.
[26]
Y. Gavet and J. C. Pinoli, “Visual perception based automatic recognition of cell mosaics in human corneal endothelium microscopy images,” Image Analysis and Stereology, vol. 27, no. 1, pp. 53–61, 2008.
[27]
S. C. Sekhar, F. Aguet, S. Romain, et al., “Parametric B-spline snakes on distance maps—application to segmentation of histology images,” in Proceedings of the16th European Signal Processing Conference (EUSIPCO '08), Lausanne, Switzerland, August 2008.
[28]
F. Lamberti and B. Montrucchio, “Segmentation of in-vitro endothelial cell networks,” in Proceedings of the 2nd IEEE International Symposium on Biomedical Imaging: Nano to Macro, vol. 1, pp. 129–132, April 2004.
[29]
L. M. Vincent and B. R. Masters, “Morphological image processing and network analysis of cornea endothelial cell images,” in Image Algebra and Morphological Image Processing III, vol. 1769 of Proceedings of the SPIE, San Diego, Calif, USA, July 1992.
[30]
C. Cortes, “Invited talk: can learning kernels help performance?” in Proceedings of the 26th Annual International Conference on Machine Learning, Association for Computing Machinery, New York, NY, USA, 2009.
