Six skin diseases classification using deep convolutional neural network

<span>Smart imaging-based medical classification systems help the human diagnose diseases and make better decisions about patient health. Recently, computer-aided of skin has been a popular research area due to its importance in early detection diseases. This paper presents at core, system that exploits convolutional neural networks classify color images lesions. It relies on pre-trained deep network between six diseases: acne, athlete’s foot, chickenpox, eczema, cancer, vitiligo. Additionally, we constructed dataset 3000 colored from several online datasets Internet. Experimental results are encouraging, where proposed model achieved an accuracy 81.75%, which is higher than state art researches this field. was calculated using holdout method, 90% were used for training, 10% out-of-sample testing.</span>