Automatic diagnosis of skin diseases using convolution neural network

Abstract Skin diseases are becoming a most common health issues among all the countries worldwide. The method proposed in this work detects four types of skin diseases using computer vision. The proposed approach involves Convolutional Neural Networks with specific focus on skin disease. The Convolutional Neural Network (CNN) used in this paper has utilized around 11 layers viz., Convolution Layer, Activation Layer, Pooling Layer, Fully Connected Layer and Soft-Max Classifier. Images from the DermNet database are used for validating the architecture. The database comprises all types of skin diseases out of which we have considered four different types of skin diseases like Acne, Keratosis, Eczema herpeticum, Urticaria with each class containing around 30 to 60 different samples. The challenges in automating the process includes the variation of skin tones, location of the disease, specifications of the image acquisition system etc., The proposed CNN Classifier results in an accuracy of 98.6% to 99.04%.

[1]  Sharath Pankanti,et al.  Deep learning ensembles for melanoma recognition in dermoscopy images , 2016, IBM J. Res. Dev..

[2]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[3]  Jeffrey Pennington,et al.  Dynamic Pooling and Unfolding Recursive Autoencoders for Paraphrase Detection , 2011, NIPS.

[4]  J. Mazurkiewicz,et al.  Detection of proopiomelanocortin-derived antigens in normal and pathologic human skin. , 1993, The Journal of laboratory and clinical medicine.

[5]  B. Guillot,et al.  Merkel cell polyomavirus DNA detection in lesional and nonlesional skin from patients with Merkel cell carcinoma or other skin diseases , 2010, The British journal of dermatology.

[6]  Geoffrey E. Hinton,et al.  On rectified linear units for speech processing , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[7]  W. Pichler,et al.  In vitro detection of cytotoxic T and NK cells in peripheral blood of patients with various drug‐induced skin diseases , 2010, Allergy.

[8]  R. S. Sabeenian,et al.  Modified Alexnet architecture for classification of diabetic retinopathy images , 2019, Comput. Electr. Eng..

[9]  J. Aravinth,et al.  An application of image processing techniques for detection of diseases on brinjal leaves using k-means clustering method , 2016, 2016 International Conference on Recent Trends in Information Technology (ICRTIT).

[10]  Abdesselam Bouzerdoum,et al.  Skin segmentation using color pixel classification: analysis and comparison , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Sven Behnke,et al.  Evaluation of Pooling Operations in Convolutional Architectures for Object Recognition , 2010, ICANN.

[12]  Ricardo Matsumura de Araújo,et al.  On the Performance of GoogLeNet and AlexNet Applied to Sketches , 2016, AAAI.

[13]  Ludmil Mikhailov,et al.  Evolving fuzzy medical diagnosis of Pima Indians diabetes and of dermatological diseases , 2010, Artif. Intell. Medicine.

[14]  Mai S. Mabrouk,et al.  Automatic Detection of Melanoma Skin Cancer using Texture Analysis , 2012 .

[15]  Ah Chung Tsoi,et al.  Face recognition: a convolutional neural-network approach , 1997, IEEE Trans. Neural Networks.

[16]  Sebastian Thrun,et al.  Dermatologist-level classification of skin cancer with deep neural networks , 2017, Nature.

[17]  T. Shanthi,et al.  Real time noisy dataset implementation of optical character identification using CNN , 2020 .