Deep Learning and Machine Learning Techniques of Diagnosis Dermoscopy Images for Early Detection of Skin Diseases

With the increasing incidence of severe skin diseases, such as cancer, endoscopic medical imaging has become urgent for revealing internal and hidden tissues under skin. Diagnostic information to help doctors make an accurate diagnosis is provided by endoscopy devices. Nonetheless, most diseases have similar features, which it challenging dermatologists diagnose patients accurately. Therefore, machine deep learning techniques can a critical role in diagnosing dermatoscopy images early detection diseases. In this study, systems lesions were developed. The performance was evaluated on two datasets (e.g., International Skin Imaging Collaboration (ISIC 2018) Pedro Hispano (PH2)). First, proposed system based hybrid features that extracted three algorithms: local binary pattern (LBP), gray level co-occurrence matrix (GLCM), wavelet transform (DWT). Such then integrated into feature vector classified using artificial neural network (ANN) feedforward (FFNN) classifiers. FFNN ANN classifiers achieved superior results compared other methods. Accuracy rates 95.24% ISIC 2018 dataset 97.91% PH2 algorithm. Second, convolutional networks (CNNs) ResNet-50 AlexNet models) applied transfer method. It found model fared better than AlexNet. 90% 95.8% reached model.