Abstract
In this paper we describe a novel image analysis method for the identification of region of interests and segmentation. The method was developed for the segmentation of the breast cancer cytological images, but it is also applicable to other image types. The primary technique of the method is a classification of the image pixels based on a spatial analysis of a feature variance in the pixel neighbourhood. The result of the analysis is association of a vector of values bounded to [0..1] to each pixel. Each of the vector values depends on a set of the feature differences between pairs of subsequent regions situated along symmetrical neighbourhood bearings. The vectors are then matched with a template vectors prepared for each of the types of searched regions of interest. The method allows for accurate localization of artefacts like edges, and subsequent segmentation of the image.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Marciniak A, Monczak R, Kolodzinski M, Pretki P, Obuchowicz A (2004) Fine Needle Biopsy Breast Cancer Diagnostic Benchmark (in Polish). Sztuczna inteligencja w inzynierii biomedycznej — SUB 2004, Wyd. PreTEXT, Krakow
Kimmel M, Lachowicz M, Świerniak A (Eds.) (2003) Cancer Growth and Progression, Mathematical Problems and Computer Simulations, Int. J. of Appl. Math, and Comput. Sci., Vol. 13, No. 3, Special Issue
Gonzalez R, Woods R (2002) Digital Image Processing, Prentice Hall
Pratt W (2001) Digital Image Processing, John Wiley & Sons
Russ J (1999) The Image Processing Handbook, CRC Press
Marciniak A, Obuchowicz A, Monczak R, Kolodzinski M (2005) Cytomorphometry of Fine Needle Biopsy Material from the Breast Cancer, Proc. of the 4th Int. Conf. on Comp. Recogn. Systems CORES’05, Adv. in Soft Computing, pp. 603–609, Springer
Zhang Y J (1996) A survey on evaluation methods for image segmentation. Pattern Recogition, Vol. 29. No. 8, 1335–1346, Elsevier Science Ltd
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Nieczkowski, T., Obuchowicz, A. (2007). ’Sonar’ — Region of Interest Identification and Segmentation Method for Cytological Breast Cancer Images. In: Kurzynski, M., Puchala, E., Wozniak, M., Zolnierek, A. (eds) Computer Recognition Systems 2. Advances in Soft Computing, vol 45. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-75175-5_71
Download citation
DOI: https://doi.org/10.1007/978-3-540-75175-5_71
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-75174-8
Online ISBN: 978-3-540-75175-5
eBook Packages: EngineeringEngineering (R0)