Proposing a novel unsupervised stack ensemble of deep and conventional image segmentation (SEDCIS) method for localizing vitiligo lesions in skin images.
depigmentation disorder
image segmentation
skin lesion localization
stacked ensemble method
Journal
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)
ISSN: 1600-0846
Titre abrégé: Skin Res Technol
Pays: England
ID NLM: 9504453
Informations de publication
Date de publication:
Mar 2021
Mar 2021
Historique:
received:
17
05
2020
accepted:
20
06
2020
pubmed:
15
7
2020
medline:
19
8
2021
entrez:
15
7
2020
Statut:
ppublish
Résumé
Vitiligo is an acquired pigmentary skin disorder characterized by depigmented macules and patches which brings many challenges for the patients suffering from. For vitiligo severity assessment, several scoring methods have been proposed based on morphometry and colorimetry. But, all methods suffer from much inter- and intra-observer variations for estimating the depigmented area. For all mentioned assessment methods of vitiligo disorder, accurate segmentation of the skin images for lesion detection and localization is required. The image segmentation for localizing vitiligo skin lesions has many challenges because of illumination variation, different shapes and sizes of vitiligo lesions, vague lesion boundaries and skin hairs and vignette effects. The manual image segmentation is a tedious and time-consuming task. Therefore, using automatic image segmentation methods for lesion detection is necessarily required. In this study, a novel unsupervised stack ensemble of deep and conventional image segmentation (SEDCIS) methods is proposed for localizing vitiligo lesions in skin images. Unsupervised segmentation methods do not require prior manual segmentation of vitiligo lesions which is a tedious and time-consuming task with intra- and inter-observer variations. Our collected dataset includes 877 images taken from 21 patients with the resolution of 5760*3840 pixels suffering from vitiligo disorder. Experimental results show that SEDCIS outperforms the compared methods with accuracy of 97%, sensitivity of 98%, specificity of 96%, area overlapping of 94%, and Dice index of 97%. The proposed method can segment vitiligo lesions with highly reasonable performance and can be used for assessing the vitiligo lesion surface.
Sections du résumé
BACKGROUND
BACKGROUND
Vitiligo is an acquired pigmentary skin disorder characterized by depigmented macules and patches which brings many challenges for the patients suffering from. For vitiligo severity assessment, several scoring methods have been proposed based on morphometry and colorimetry. But, all methods suffer from much inter- and intra-observer variations for estimating the depigmented area. For all mentioned assessment methods of vitiligo disorder, accurate segmentation of the skin images for lesion detection and localization is required. The image segmentation for localizing vitiligo skin lesions has many challenges because of illumination variation, different shapes and sizes of vitiligo lesions, vague lesion boundaries and skin hairs and vignette effects. The manual image segmentation is a tedious and time-consuming task. Therefore, using automatic image segmentation methods for lesion detection is necessarily required.
MATERIALS AND METHODS
METHODS
In this study, a novel unsupervised stack ensemble of deep and conventional image segmentation (SEDCIS) methods is proposed for localizing vitiligo lesions in skin images. Unsupervised segmentation methods do not require prior manual segmentation of vitiligo lesions which is a tedious and time-consuming task with intra- and inter-observer variations.
RESULTS
RESULTS
Our collected dataset includes 877 images taken from 21 patients with the resolution of 5760*3840 pixels suffering from vitiligo disorder. Experimental results show that SEDCIS outperforms the compared methods with accuracy of 97%, sensitivity of 98%, specificity of 96%, area overlapping of 94%, and Dice index of 97%.
CONCLUSION
CONCLUSIONS
The proposed method can segment vitiligo lesions with highly reasonable performance and can be used for assessing the vitiligo lesion surface.
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
126-137Informations de copyright
© 2020 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.
Références
Rzepecki AK, McLellan BN, Elbuluk N. Beyond traditional treatment: the importance of psychosocial therapy in Vitiligo. J Drugs Dermatol. 2018;17:688-691.
Bae JM, Lee SC, Kim TH, et al. Factors affecting quality of life in patients with vitiligo: a nationwide study. Br J Dermatol. 2018;178:238-244.
Geel NV, Lommerts J, Bekkenk M, et al. Development and validation of the Vitiligo extent score (VES): an International Collaborative Initiative. J Invest Dermatol. 2016;136:978-984.
Toh JJH, Bhoi S, Tan VWD, et al. Automated scoring of vitiligo using superpixel-generated computerized digital image analysis of clinical photographs: a novel and consistent way to score vitiligo. Br J Dermatol. 2018;179:220-221.
Alghamdi KM, Kumar A, Taieb A, Ezzedine K. Assessment methods for the evaluation of vitiligo. J Eur Acad Dermatol Venereol. 2012;26(12):1463-1471.
Sanders JE, Goldstein BS, Richards KA. Image processing techniques for quantitative analysis of skin structures. Comp Meth Prog Bio. 1999;59:167-180.
Paracchini M, Marcon M, Villa F, Tubaro S. Deep skin detection on low resolution grayscale images. Pattern Recogn Lett. 2020;131:322-328.
Wang H, Moss RH, Chen X, et al. Modified watershed technique and post-processing for segmentation of skin lesions in dermoscopy images. Comput Med Imaging Graph. 2011;35:116-120.
Kadampour M, Riyaee SA. Skin cancer detection: Applying a deep learning based model driven architecture in the cloud for classifying dermal cell images. Inform Med Unlocked. 2020;18:100282.
Das N, Pal A, Mazumder S, Sarkar S, Gangopadhyay D, Nasipuri M. An SVM Based Skin Disease Identification Using Local Binary Patterns. Third International Conference on Advances in Computing and Communications. Cochin, India: IEEE; 2013.
Dash M, Londhe ND, Ghosh S, Semwal A, Sonawane RS. PsLSNet: Automated psoriasis skin lesion segmentation using modified U-Net-based fully convolutional network. Biomed Signal Process Control. 2019;52:226-237.
Tschandl P, Sinz C, Kittler H. Domain-specific classification-pretrained fully convolutional network encoders for skin lesion segmentation. Comp Biol Med. 2019;104:111-116.
Peng T, Qiaokang L, Xintong Y, et al. Efficient skin lesion segmentation using separable-Unet with stochastic weight averaging. Comp Meth Prog Bio. 2019;178:289-301.
Vasconcelos FF, Medeiros AG, Peixoto SA, Filho PP. Automatic skin lesions segmentation based on a new morphological approach via geodesic active contour. Cogn Syst Res. 2019;55:44-59.
Nugroho H, Ahmad Fadzil MH, Shamsudin N, Hussein SH. Computerised image analysis of vitiligo lesion: evaluation using manually defined lesion areas. Skin Res Tech. 2013;19:72-77.
Kislal EE, Halasz C. Software for quantifying psoriasis and vitiligo from digital clinical photographs. J Dermatolog Treat. 2011;24:107-111.
Gupta M, Srinivasa S, Antony M. KL divergence based agglomerative clustering for automated vitiligo grading. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Boston, MA: IEEE.
Dhall A, Sharma G, Bhatt RB, Khan GM. Adaptive Digital Makeup. International Symposium on Visual Computing (ISVC). Berlin, Heidelberg: Springer; 2009:728-736.
Huang WS, Wang YW, Hung KC, et al. High correlation between skin color based on CIELAB color space, epidermal melanocyte ratio, and melanocyte melanin content. PeerJ; 2018;6:e4815.
Shapire RE, Freund Y. Boosting: Foundations and Algorithms. Cambridge, MA: MIT Press; 2012.
Sabzavari M, Martinez-Munoz G, Suarez A. Vote-boosting ensemble. Pattern Recogn. 2018;83:119-133.
Zhong C, Hu L, Yue X, Luo T, Fu Q, Xu H. Ensemble clustering based on evidence extracted from the co-association matrix. Pattern Recogn. 2019;92:93-106.
Wang SQ, Yang J, Chou KC. Using stacked generalization to predict membrane protein types based on pseudo-amino acid composition. J Theor Biol. 2006;242:941-946.
Ronneberger O, Fischer P, Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation. International conference on medical image computing and computer-assisted intervention MICCAI. Munich, Germany: Springer, Cham; 2015.
Loffe S, Szegedy CB. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. ICML'15: Proceedings of the 32nd International Conference on International Conference on Machine Learning. Lille, France: JMLR.org; 2015:448-456.
Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R. Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res. 2014;15:1929-1958.
Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation. IEEE Trans Pattern Anal Mach Intell. 2017;39:1.
Zheng Y, Zhao F, Liu H, Wang J. Adaptive multi-level thresholding segmentation based on multi-objective evolutionary algorithm. International Conference on Swarm Intelligence. Switzerland: Springer, Cham; 2016.
Korzynska A, Roszkowiak L, Lopez C, Bosch R, Witkowski L, Lejeune M. Validation of various adaptive threshold methods of segmentation applied to follicular lymphoma digital images stained with 3,3'-Diaminobenzidine & Haematoxylin. Diagn Pathol. 2013;8:48.
Adams R, Bischof L. Seeded region growing. IEEE Trans Pattern Anal Mach Intell. 1994;16:641-647.
Alqurashi T, Wang W. Clustering ensemble method. Mach Learn Cybernet. 2019;10:1227-1246.
Bhatt RB, Sharma G, Dhall A, Chaudhury S. Efficient skin region segmentation using low complexity fuzzy decision tree model. Gujarat, India: IEEE; 2009.
Breiman L. Bagging predictors. Mach Learn. 1996;24:123-140.
Chen S, Lin Y, Liu Y, Lu ZF. Image segmentation by multi-threshold based on Fisher function and histogram algorithm. 2010 International Conference on Computer Application and System Modeling (ICCASM 2010). Taiyuan, China: IEEE; 2010.
Homan MW, Wolkerstorfer A, Sprangers MAG, van der Veen JPW. Digital image analysis vs. clinical assessment to evaluate repigmentation after punch grafting in vitiligo. J Eur Acad Dermatol Venereol. 2012;27:e235-e238.
Kaya A. Cascaded classifiers and stacking methods for classification of pulmonary nodule characteristics. Comput Meth Prog Bio. 2018;166:77-89.
Lloyd SP. Least squares quantization in PCM. IEEE Trans Inf Theory. 1982;28:129-137.
Macqueen JB. Some Methods for classification and Analysis of Multivariate Observations. 5-th Berkeley Symposium on Mathematical Statistics and Probability. Berkeley, CA: University of California Press. 1967;281-297.
Pezhman Pour M, Seker H. Transform domain representation-driven convolutional neural networks for skin lesion segmentation. Expert Syst Appl, 2020, 144;113129.
Tan TY, Zhang L, Lim CP, Fielding B, Yu Y, Anderson E. Evolving Ensemble Models for Image Segmentation Using Enhanced Particle Swarm Optimization. IEEE Access. 2019;99:34004-34019.
Wolpert DH. Stacked generalization. Neural Networks. 1992;5:241-259.
Xie F, Yang J, Liu J, Jiang Z, Zheng Y, Wang Y. Skin lesion segmentation using high-resolution convolutional neural network. Comput Meth Prog Bio. 2020;186:105241.
Zuiderveld K. Contrast limited adaptive histogram equalization. Cambridge: Academic Press; 1994.