Automated Skin Disease Detection Using GoogLeNet and MobileNet to Enhance the Diagnostic Accuracy with Convolutional Neural Networks

Skin diseases affect human health through physical discomfort, emotional distress, and social stigma The high clinical variability and that the diagnostic procedures are subjective and may change over time make it difficult to detect accurately. This paper suggests applying CNN architectures, such as GoogLeNet and MobileNet, for automated detection. The stages involved include image gathering, preprocessing, selection and design of CNN models such as GoogLeNet and MobileNet, intensive training, evaluation, and deployment. The aim of this system is to help healthcare professionals make reliable diagnoses especially in areas where dermatologists are not accessible, thus allowing for more timely treatments.