Characterisation of black skin stratum corneum by digital macroscopic images analysis

Healthcare Technology Letters

article

Géraud M. Azehoun-Pazou¹  Kokou M. Assogba²  Hugues Adegbidi³  Antoine C. Vianou² 

[1] National University of Sciences, Engineering and Mathematics (UNSTIM);Laboratory of Electrical Engineering, University of Abomey-Calavi;Department of Dermatology and Venerology, Faculty of Health Sciences, University of Abomey-Calavi;Laboratory of Thermophysical Characterization and Energetic Appropriation (Lab-CTMAE), Polytechnic School of Abomey-Calavi

关键词: medical image processing;    biomedical optical imaging;    image colour analysis;    feature extraction;    skin;    multilayer perceptrons;    neural nets;    image texture;    image segmentation;    pattern clustering;    fuzzy set theory;    image classification;    sequential network construction algorithm-based method;    selected regions;    colour transformation;    lesion region;    created colour information;    named neural network-based fuzzy clustering;    black skin lesion images;    red colour channels;    blue colour channels;    600 images;    obtained results;    black skin stratum corneum;    digital macroscopic images analysis;    black skin medical images;    global initiative;    black skin horny layer;    digital images analysis;    four-step approach;    probable healthy skin regions;    automatic classification system;    multilayer perceptron artificial neural networks;    texture;    colour features;    features selection;   

DOI  :  10.1049/htl.2020.0057

学科分类：肠胃与肝脏病学

来源: Wiley

【 摘 要 】

Black skin medical images generally show very low contrast. Being in a global initiative of characterisation of black skin horny layer (stratum corneum) by digital images analysis, the authors in this study proposed a four-step approach. The first step consists of differentiation between probable healthy skin regions and those affected. For that, they used an automatic classification system based on multilayer perceptron artificial neural networks. The network has been trained with texture and colour features. Best features selection and network architecture definition were done using sequential network construction algorithm-based method. After classification, selected regions undergo a colour transformation, in order to increase the contrast with the lesion region. Thirdly, created colour information serves as the basis for a modified fuzzy c-mean clustering algorithm to perform segmentation. The proposed method, named neural network-based fuzzy clustering, was applied to many black skin lesion images and they obtained segmentation rates up to 94.67%. The last stage consists in calculating characteristics. Eight parameters are concerned: uniformity, standard deviation, skewness, kurtosis, smoothness, entropy, and average pixel values calculated for red and blue colour channels. All developed methods were tested with a database of 600 images and obtained results were discussed and compared with similar works.

【 授权许可】

CC BY|CC BY-ND|CC BY-NC|CC BY-NC-ND