Date of Publication :5th April 2025

Abstract: Skin cancer is among the most prevalent types of illnesses, and early detection is key to better patient outcomes. Traditional diagnostic methods, such as biopsies and visual inspection, can be laborious and subjective. Convolutional neural networks (CNNs), one of the newest developments in artificial intelligence, have demonstrated promise in raising the accuracy of skin cancer detection automation. This project aims to use MATLAB and Python to develop a model that can recognize skin cancer using dermoscopic images. Pre-processing: the images ensure that CNN is aware of the many types of cancer while enhancing the image quality. Through the automated extraction of relevant data, the model improves diagnostic accuracy. Metrics like accuracy, specificity, and sensitivity are used to assess performance. Additionally, understanding hierarchical features from low-level edges is enabled by deep learning methods such as Deep Neural Networks (DNNs), which help to model complex data linkages by adding high-level lesion patterns, hence enhancing the model's classification performance for skin cancer.

Reference :