Quantification of Fuzzy Borders and Fuzzy Asymmetry of Malignant Melanomas

In this paper we have proposed scan line method and fuzzy relational method, for quantifying the border asymmetricity for melanoma skin lesions. The procedure for quantifying asymmetricity through scan line method consists of various steps which are image preprocessing, boundary detection, evaluation of the skull region of the boundary, conversion of boundary into several straight lines, formation of polygon tables and edge tables, evaluation of centroid, principle axis and image transformations, embed 2nd, 3rd and 4th quadrant images into the first quadrant. The fuzzy relational approach follows the same procedure except the last phase for finding the asymmetricity. Sample of 60 skin lesion images are considered for analysis and its performances are tested, the proposed approaches seem to perform better than other methods reported in the recent literature.

[1]  David I. McLean,et al.  Irregularity index: A new border irregularity measure for cutaneous melanocytic lesions , 2003, Medical Image Anal..

[2]  Z. She,et al.  Combination of features from skin pattern and ABCD analysis for lesion classification , 2007, Skin research and technology : official journal of International Society for Bioengineering and the Skin (ISBS) [and] International Society for Digital Imaging of Skin (ISDIS) [and] International Society for Skin Imaging.

[3]  Randy H. Moss,et al.  Fast and accurate border detection in dermoscopy images using statistical region merging , 2007, SPIE Medical Imaging.

[4]  Lorenzo Scalise,et al.  Fractal characterisation of boundary irregularity in skin pigmented lesions , 2005, Medical and Biological Engineering and Computing.

[5]  M. Pauline Baker,et al.  Computer graphics with OpenGL , 1986 .

[6]  Li Ma,et al.  Analysis of the contour structural irregularity of skin lesions using wavelet decomposition , 2013, Pattern Recognit..

[7]  Ela Claridge,et al.  A Border Irregularity Measure Using a Modified Conditional Entropy Method as a Malignant Melanoma Predictor , 2005, ICIAR.

[8]  Gerald Schaefer,et al.  An improved objective evaluation measure for border detection in dermoscopy images , 2009, Skin research and technology : official journal of International Society for Bioengineering and the Skin (ISBS) [and] International Society for Digital Imaging of Skin (ISDIS) [and] International Society for Skin Imaging.

[9]  M. Emin Yüksel,et al.  Accurate Segmentation of Dermoscopic Images by Image Thresholding Based on Type-2 Fuzzy Logic , 2009, IEEE Transactions on Fuzzy Systems.

[10]  Antonio Pietrosanto,et al.  ELM image processing for melanocytic skin lesion diagnosis based on 7-point checklist: a preliminary discussion , 2004 .

[11]  M. Emre Celebi,et al.  Unsupervised border detection of skin lesion images , 2005, International Conference on Information Technology: Coding and Computing (ITCC'05) - Volume II.

[12]  T. Ross.  Fuzzy Logic with Engineering Applications , 1994 .

[13]  Tim K. Lee,et al.  Determining the asymmetry of skin lesion with fuzzy borders , 2005, Comput. Biol. Medicine.