Automatic Melanoma Skin Cancer Detection and Segmentation using SnakeCut Algorithm

Dondapati Rajendra Dev; T. Sivaprakasam; K. Vijaya Kumar

doi:10.59796/jcst.V14N2.2024.35

Authors

Dondapati Rajendra Dev Department of Computer Science & Engineering, Annamalai University, Annamalai Nagar, Tamil Nadu 608002, India & Department of Computer Science & Engineering, Vignan’s Institute of Engineering for Women, Kapu Jaggaraju Peta, Vsez Post, Visakhapatnam 530046, India
T. Sivaprakasam Department of Computer Science & Engineering, Annamalai University, Annamalai Nagar, Tamil Nadu 608002, India
K. Vijaya Kumar Department of Computer Science & Engineering, Gitam University, Vishakhapatnam, Andhra Pradesh 530045, India

DOI:

https://doi.org/10.59796/jcst.V14N2.2024.35

Keywords:

CAD Systems, Active Contour, GrabCut, SnakeCut, Object segmentation

Abstract

Early detection of melanoma skin cancer is crucial for effective treatment, and computer-aided diagnostic technologies offer promising advancements for dermatologists to make faster, more precise diagnoses of skin lesions. Segmenting skin lesions is a crucial first step towards automated Computer-Aided Diagnosis for skin cancer. This paper aims to use SnakeCut, a foreground extraction approach, to automatically segment skin lesions in HSV color space with little human interaction. Active contour (otherwise called Snake) and Improved GrabCut are the two popular methods. By decreasing the energy function of the related contour, the active contour acts as a deformable segmentation contour. Improved GrabCut uses updated iterated graph cuts to store color attributes used as segmentation signals in order to achieve foreground segmentation from close-by pixel similarities in its foreground segmentation algorithm. The proposed integrated solution, which is predicated on a probabilistic framework, is termed “SnakeCut.” We removed the outer black border using preprocessing. Later feature extraction is done using HOG and HSV and classifies the benign or melanoma state using Naïve Bayes, Decision tree, and K-nearest neighbor classifiers. The efficiency of the segmentation strategy was measured using the Jaccard Index. We compared the classification results of our method with existing state-of-the-art approaches. The study demonstrates the efficacy of Automatic SnakeCut in accurately segmenting skin lesions, thereby enhancing the performance of subsequent classification tasks. The average F-score was 0.75 on the 2017 ISIC challenge training dataset of 100 images. Compared to other methods, this study’s findings reveal that the suggested method is highly accurate.

References

Al-Masni, M. A., Al-Antari, M. A., Choi, M. T., Han, S. M., & Kim, T. S. (2018). Skin lesion segmentation in dermoscopy images via deep full resolution convolutional networks. Computer Methods and Programs in Biomedicine, 162, 221-231. https://doi.org/10.1016/j.cmpb.2018.05.027

Celebi, M. E., Iyatomi, H., Schaefer, G., & Stoecker, W. V. (2009). Lesion border detection in dermoscopy images. Computerized Medical Imaging and Graphics, 33(2), 148-153. https://doi.org/10.1016/j.compmedimag.2008.11.002

Celebi, M. E., Kingravi, H. A., Iyatomi, H., Lee, J., Aslandogan, Y. A., Van Stoecker, W., ... & Marghoob, A. A. (2007, March 26). Fast and accurate border detection in dermoscopy images using statistical region merging [Conference Presentation]. Medical Imaging 2007: Image Processing, San Diego, CA, United States. https://doi.org/10.1117/12.709073

Celebi, M. E., Wen, Q. U. A. N., Iyatomi, H. I. T. O. S. H. I., Shimizu, K. O. U. H. E. I., Zhou, H., & Schaefer, G. (2015). A state-of-the-art survey on lesion border detection in dermoscopy images. Dermoscopy Image Analysis, 10(1), 97-129. https://doi.org/10.1201/b19107-5

das Chagas, J. V. S., Ivo, R. F., Guimarães, M. T., Rodrigues, D. D. A., Rebouças, E. D. S., & Rebouças Filho, P. P. (2020). Fast fully automatic skin lesions segmentation probabilistic with Parzen window. Computerized Medical Imaging and Graphics, 85, Article 101774. https://doi.org/10.1016/j.compmedimag.2020.101774

Ding, X., & Wang, S. (2021). Efficient Unet with depth-aware gated fusion for automatic skin lesion segmentation. Journal of Intelligent & Fuzzy Systems, 40(5), 9963-9975. https://doi.org/10.3233/JIFS-202566

Garnavi, R., Aldeen, M., Celebi, M. E., Varigos, G., & Finch, S. (2011). Border detection in dermoscopy images using hybrid thresholding on optimized color channels. Computerized Medical Imaging and Graphics, 35(2), 105-115. https://doi.org/10.1016/j.compmedimag.2010.08.001

Iyatomi, H., Celebi, M. E., Oka, H., & Tanaka, M. (2008, August 20-25). An Internet-based melanoma screening system with acral volar lesion support [Conference Presentation]. 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Vancouver, BC, Canada. https://doi.org/10.1109/IEMBS.2008.4650375

Jaworek-Korjakowska, J. (2016). Computer-aided diagnosis of micro-malignant melanoma lesions applying support vector machines. BioMed Research International, 2016, Article 4381972. https://doi.org/10.1155/2016/4381972

Katapadi, A. B., Celebi, M. E., Trotter, S. C., & Gurcan, M. N. (2018). Evolving strategies for the development and evaluation of a computerised melanoma image analysis system. Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 6(4), 465-472. https://doi.org/10.1080/21681163.2016.1277785

Mei, Z., Hua, W. J., & Xing, P. X. (2016). Segmentation for range image based on snake active contour model. International Journal of Signal Processing, Image Processing and Pattern Recognition, 9(8), 393-400. https://doi.org/10.14257/ijsip.2016.9.8.33

Nagaoka, T., Nakamura, A., Kiyohara, Y., & Sota, T. (2012, August 28). Melanoma screening system using hyperspectral imager attached to imaging fiberscope [Conference Presentation]. 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, CA, USA. https://doi.org/10.1109/EMBC.2012.6346777

Peng, Y., Wang, N., Wang, Y., & Wang, M. (2019). Segmentation of dermoscopy image using adversarial networks. Multimedia Tools and Applications, 78, 10965-10981. https://doi.org/10.1007/s11042-018-6523-2

Sayed, G. I., Khoriba, G., & Haggag, M. H. (2020). The novel multi-swarm coyote optimization algorithm for automatic skin lesion segmentation. Evolutionary Intelligence, 17, 679–711. https://doi.org/10.1007/s12065-020-00450-4

Shan, P., Wang, Y., Fu, C., Song, W., & Chen, J. (2020). Automatic skin lesion segmentation based on FC-DPN. Computers in Biology and Medicine, 123, Article 103762. https://doi.org/10.1016/j.compbiomed.2020.103762

Siriwath, J., Wiriyakulsit, N., Klomkleang, P., Inpad, C., Roytrakul, S., &Kaewkong, W. (2021). SPHINX31 suppresses splicing factor phosphorylation and inhibits melanoma cell growth and aggressiveness. Journal of Current Science and Technology, 11(3), 346-354. https://doi.org/10.14456/jcst.2021.35

Singh, A., Maurya, R., Yadav, R., & Srivastava, V. (2017). Innovative Technique of Segmentation and Feature Extraction for Melanoma Detection. International Journal of Computer Sciences and Engineering, 5(10), 100-104. https://doi.org/10.26438/ijcse/v5i10.100104

Sunitha, T. O., Rajalakshmi, R., & Sujatha, S. S. (2022). Fuzzy based dynamic histogram equalization for enhancing quality of registered medical image. Journal of Current Science and Technology, 12(2), 243-264.

Takruri, M., & Abubakar, A. (2017, November 21-23). Bayesian decision fusion for enhancing melanoma recognition accuracy [Conference Presentation]. 2017 International Conference on Electrical and Computing Technologies and Applications (ICECTA), Ras Al Khaimah, United Arab Emirates. https://doi.org/10.1109/ICECTA.2017.8252063

Titov, K. S., Stepanova, V. V., Krasnorutskiy, A. V., & Nagaeva, M. V. (2019). Rare forms of skin cancer. Russian Journal of Clinical Dermatology and Venereology, 18(6), 735–741. https://doi.org/10.17116/klinderma201918061735

Wang, L., Wang, S., Liao, Y., Ye, X., & Wang, T. (2018, August 5-8). Improved level set model for color image segmentation [Conference presentation]. 2018 IEEE International Conference on Mechatronics and Automation (ICMA). Changchun, China. https://doi.org/10.1109/ICMA.2018.8484379

Wei, Z., Guo, B., Li, C., & Chen, Z. (2019, November 28-30). Speeded-up Simple Linear Iterative Clustering Based on Region Homogeneity [Conference Presentation]. 2019 2nd International Conference on Safety Produce Informatization (IICSPI), Chongqing, China. https://doi.org/10.1109/IICSPI48186.2019.9096051

Wibowo, A., Purnama, S. R., Wirawan, P. W., & Rasyidi, H. (2021). Lightweight encoder-decoder model for automatic skin lesion segmentation. Informatics in Medicine Unlocked, 25, Article 100640. https://doi.org/10.1016/j.imu.2021.100640

Yuan, Y., Chao, M., & Lo, Y. C. (2017). Automatic skin lesion segmentation using deep fully convolutional networks with jaccard distance. IEEE Transactions on Medical Imaging, 36(9), 1876-1886. https://doi.org/10.1109/TMI.2017.2695227