Melanoma is one form of skin cancer which is one of the most hazardous types of cancer happened in people. Incidence of skin cancer has been increasing over decades due to excess exposure of radiations from sun causing erosion to skin melanin. The automatic detection of melanoma in dermatological images is a challenging task because of the diverse contrast of skin lesions, the magnitude of melanoma within the class, and the utmost optical similarity to melanoma and lesions other than melanoma and the beingness of many artifacts in the lesion pictures. In this work, the skin lesion analysis system to aid for the melanoma detection is proposed. Firstly, the skin lesion from dermoscopy images is automatically segmented with the use of texture filters. Then, the features according to the underlying ABCD dermatology rules are then extracted from the segmented skin lesion. Finally, the system is classified by random subspace ensemble classifier in order to determine the images as benign or malignant melanoma The performance of the study was experimented with their precision and it achieves with compromising results.
Published In:IJCSN Journal Volume 8, Issue 3
Date of Publication : June 2019
Pages : 200-206
Figures :05
Tables : 05
Nay Chi Lynn :
Image Processing Lab, University of Computer Studies, Mandalay Mandalay, Myanmar.
Nu War :
Faculty of Computer Systems and Technologies, University of Computer Studies, Mandalay Mandalay, Myanmar.
In this study, a model for the segmentation of skin lesions is proposed. Lesion areas are initially segmented
using proposed texture based method and various features underlying ABCD rules are extracted to represent segmented lesion areas. Then the classification is performed with different classifiers. Experiments are conducted on our on the ISBI 2016 challenging dataset and PH2 dataset. The proposed segmentation algorithm can provide accurate segmentation results in order to use in segmentation task of skin lesion and the proposed feature is robust according to the weight value resulted from feature selection algorithm. In classification process, random subspace ensemble algorithm with feature data optimized by relief algorithm give better classification results for benign or melanoma cancer classification. The current research work address only to the classification of pigmented skin lesion as either benign or malignant melanomas skin cancer. Therefore, classification of various pigmented skin lesions such as Melanomas, Nevus, BCC, Seborrheic Keratoses, etc would be studied as the future research work.
[1] David Gutman1y, Noel C. F. Codellay, Emre Celebi, Brian Helba, Michael Marchetti, Nabin Mishra, Allan Halpern, "Skin Lesion Analysis toward Melanoma Detection", Proceedings of the International Symposium on Biomedical Imaging (ISBI), arXiv:1605.01397v1 [cs.CV] 4 May 2016. [2] Lequan Yu, Hao Chen, Qi Dou, Jing Qin, and Pheng-Ann Heng, "Automated Melanoma Recognition in Dermoscopy Images via Very Deep Residual Networks", IEEE Transactions on Medical Imaging, DOI 10.1109, April 2016. [3] Khalid Eltayef, Yongmin Li and Xiaohui Liu, "Knowledge Transfer for Melanoma Screening with Deep Learning", 2016 International Conference on Communication, Image and Signal Processing (CCISP 2016).
[4] Oliveira R.B., Pereira A.S., Tavares J.M.R.S. (2018) "Pattern Recognition in Macroscopic and Dermoscopic Images for Skin Lesion Diagnosis". In: Tavares J., Natal Jorge R. (eds) VipIMAGE 2017. ECCOMAS 2017. Lecture Notes in Computational Vision and Biomechanics, vol 27. Springer, Cham. [5] Bi, Lei & Kim, Jinman & Ahn, Euijoon & Feng, David Dagan Feng & Fulham, Michael, "Semi-automatic skin lesion segmentation via fully convolutional networks", 561-564. 10.1109/ ISBI. 2017. 7950583. [6] D. Gutman, et al., "Skin Lesion Analysis toward Melanoma Detection: A Challenge at the International Symposium on Biomedical Imaging (ISBI) 2016, hosted by the International Skin Imaging Collaboration (ISIC)," arXiv preprint arXiv:1605.01397, 2016.Li, Yuexiang & Shen, Linlin., Skin Lesion Analysis Towards Melanoma Detection Using Deep Learning Network. 2017. [7] Capdehourat G., Corez A., Bazzano A., and Mus P., "Pigmented Skin Lesions Classification Using Dermatoscopic Images", CIARP 2009, LNCS 5856, pp. 537-544, Springer-Verlag Berlin Heidelberg 2009. [8] Lee T, Ng V, Gallagher R, Coldman A, McLean D. "DullRazor: A software approach to hair removal from images. Computers in Biology and Medicine", 1997;27:533-543. [9] Bi, Lei & Kim, Jinman & Ahn, Euijoon & Feng, David Dagan Feng & Fulham, Michael, "Semi-automatic skin lesion segmentation via fully convolutional networks", 561-564. 10.1109/ ISBI. 2017. 7950583. [10] D. Gutman, et al., "Skin Lesion Analysis toward Melanoma Detection: A Challenge at the International Symposium on Biomedical Imaging (ISBI) 2016, hosted by the International Skin Imaging Collaboration (ISIC)," arXiv preprint arXiv:1605.01397, 2016.Li, Yuexiang & Shen, Linlin., Skin Lesion Analysis Towards Melanoma Detection Using Deep Learning Network. 2017. [11] Capdehourat G., Corez A., Bazzano A., and Mus P., "Pigmented Skin Lesions Classification Using Dermatoscopic Images", CIARP 2009, LNCS 5856, pp. 537-544, Springer-Verlag Berlin Heidelberg 2009. [12] Lee T, Ng V, Gallagher R, Coldman A, McLean D. "DullRazor: A software approach to hair removal from images. Computers in Biology and Medicine", 1997;27:533-543. [13] Bi, Lei & Kim, Jinman & Ahn, Euijoon & Feng, David Dagan Feng & Fulham, Michael, "Semi-automatic skin lesion segmentation via fully convolutional networks", 561-564. 10.1109/ ISBI. 2017. 7950583. [14] D. Gutman, et al., "Skin Lesion Analysis toward Melanoma Detection: A Challenge at the International Symposium on Biomedical Imaging (ISBI) 2016, hosted by the International Skin Imaging Collaboration (ISIC)," arXiv preprint arXiv:1605.01397, 2016.Li, Yuexiang & Shen, Linlin., Skin Lesion Analysis Towards Melanoma Detection Using Deep Learning Network. 2017. [15] Capdehourat G., Corez A., Bazzano A., and Mus P., "Pigmented Skin Lesions Classification Using
Dermatoscopic Images", CIARP 2009, LNCS 5856, pp. 537-544, Springer-Verlag Berlin Heidelberg 2009. [16] Lee T, Ng V, Gallagher R, Coldman A, McLean D. "DullRazor: A software approach to hair removal from images. Computers in Biology and Medicine", 1997;27:533-543. [17] Ammara Masood, Adel Ali Al-Jumaily, "Computer Aided Diagnostic Support System for Skin Cancer: A Review of Techniques and Algorithms", International Journal of Biomedical Imaging, vol. 2013, Article ID 323268, 22 pages, 2013. https://doi.org/10.1155/2013/323268. [18] I. Maglogiannis and C. N. Doukas,, "Overview of Advanced Computer Vision Systems for Skin Lesions Characterization", IEEE Transactions on Information Technology in Biomedicine, vol. 13, no. 5, pp. 721-733, doi: 10.1109/TITB.2009.2017529, Sept. 2009. [19] Qaisar Abbas, M. Emre Celebi, Irene FondoŽn G. and Muhammad Rashid, "Lesion border detection in dermoscopy images using dynamic programming," 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 (ISSI). 17. 91-100. 10.1111/j.16000846.2010.00472.x. [20] Konstantin Korotkov, Rafael Garcia, "Computerized analysis of pigmented skin lesions: a review", Artificial Intelligence in Medicine, 56(2), 69-90. doi:10.1016/ j.artmed. 2012.08.002.
