Developing an Advanced Skin Disease Detection System by Integrating Hybrid Neural Network with Yolo V8 Model

Skin conditions that are caused by viruses, allergies, and fungi are common all over the world. Although sophisticated technologies such as lasers provide accurate diagnosis, their high price points restrict accessibility. Image processing makes automated dermatology screening possible in order to address this. After features are classified using multiclass SVM, users are given comprehensive illness information. This technique remarkably achieves 100 percent accuracy in identifying three different skin conditions. When used with morphology-based segmentation, image segmentation methods like adaptive thresholding and K-means clustering are very helpful in identifying diseases. This model attempts to accurately detect and analyze skin illnesses in spite of obstacles such as low lesion-to-skin contrast and visu.al similarities between diseased and non-diseased areas. Through the application of filters to reduce noise and the conversion of images to grayscale for improved processing, it offers significant diagnostic evidence, supporting doctors in their decision-making and reducing the possibility of adverse effects.