The most common and fatal type of cancer that poses a serious risk to Cutaneous malignancy with melanoma being particularly deadly. Early detection is crucial for effective treatment, but traditional diagnostic methods face challenges due to poor image quality and the complexity of visual differentiation. This study proposes an advanced deep learning approach for optimal skin cancer segmentation and classification, with a focus on severity analysis. The proposed method utilizes advanced image pre-processing techniques to reduce noise while preserving critical features, ensuring high-quality images for precise diagnosis. By leveraging state-of-the-art feature extraction techniques, the model identifies intricate patterns and correlations within the data. To enhance accuracy, multiple advanced classification methods are integrated, addressing computational challenges often associated with deep learning models. The approach is designed to be efficient and effective, even in resource-limited settings. Additionally, the study includes a comprehensive severity assessment, analyzing various clinical parameters to provide a detailed understanding of the cancer’s progression and impact on the patient. This holistic approach, from early detection to severity evaluation, enhances diagnostic accuracy and patient outcomes. Overall, this research highlights the transformative potential of deep learning in skin cancer diagnosis. By improving early detection, classification precision, and severity assessment, the proposed method offers a robust and reliable solution that advances patient care and treatment strategies.