Classification of Skin Disease Using Deep Learning Neural Networks with MobileNet V2 and LSTM

Abstract

Deep learning models are efficient in learning the features that assist in understanding complex patterns precisely. This study proposed a computerized process of classifying skin disease through deep learning based MobileNet V2 and Long Short Term Memory (LSTM). The MobileNet V2 model proved to be efficient with a better accuracy that can work on lightweight computational devices. The proposed model is efficient in maintaining stateful information for precise predictions. A grey-level co-occurrence matrix is used for assessing the progress of diseased growth. The performance has been compared against other state-of-the-art models such as Fine-Tuned Neural Networks (FTNN), Convolutional Neural Network (CNN), Very Deep Convolutional Networks for Large-Scale Image Recognition developed by Visual Geometry Group (VGG), and convolutional neural network architecture that expanded with few changes. The HAM10000 dataset is used and the proposed method has outperformed other methods with more than 85% accuracy. Its robustness in recognizing the affected region much faster with almost 2× lesser computations than the conventional MobileNet model results in minimal computational efforts. Furthermore, a mobile application is designed for instant and proper action. It helps the patient and dermatologists identify the type of disease from the affected region’s image at the initial stage of the skin disease. These findings suggest that the proposed system can help general practitioners efficiently and effectively diagnose skin conditions, thereby reducing further complications and morbidity.

Identifying Patient–Ventilator Asynchrony on a Small Dataset Using Image-Based Transfer Learning

“…We adopted three types of CNN models (i.e., MobileNet, VGG 16, and Inception–ResNetV2) for feature extraction, as they represent light-weighted, moderate-weighted, and heavy-weighted feature extractors, respectively. MobileNet is a light-weight model designed to run DL models on mobile devices [ 25 ]. It uses depth-wise separable convolution to increase the computing efficiency with only a small reduction in accuracy.…”

Deep Learning with Neuroimaging and Genomics in Alzheimer’s Disease

“…Due to the improvements of artificial intelligence (AI) technologies, researchers have continued making notable contributions in the multi-disciplinary fields of machine learning, deep learning, neuroimaging, genomics, and the diagnosis and prediction of AD [ 4 , 5 ]. Up-to-date advances in AI technologies, in particular deep learning techniques, have exhibited their advantageous impacts in connection with health-related and genomic medicine applications [ 6 , 7 , 8 , 9 , 10 ]. For the tasks of the diagnosis and prediction of AD, the goal of computer-aided AI methods such as deep learning models is to facilitate data-driven algorithms that can on the whole help improve the diagnosis accuracy of AD using neuroimaging and/or genomics data [ 4 , 5 ].…”

Multiclassification of Endoscopic Colonoscopy Images Based on Deep Transfer Learning

“…Medical image classification refers to taking one or more examination images as input, predicting them through the trained model, and outputting a diagnostic result indicating whether a certain disease is suffering or whether the severity is graded. At present, it has been widely used in epidemic prevention and diagnosis of benign tumors and cancer and to distinguish between different categories of the same disease and other important clinical events [ 5 ]. The object of medical image classification is the image obtained by patients through various kinds of examination equipment, mainly including Computed Tomography (CT), X-ray, Magnetic Resonance Imaging (MRI), and ultrasound image (UI) [ 6 ].…”