Advancing Skin Cancer Detection: Deep Learning Approaches for Enhanced Diagnostic Accuracy
Sandipan Chatterjee, Arpit Kumar Sharma, Pramod Singh Rathore
Abstract
Skin cancer, a common and potentially fatal condition, greatly benefits from early detection. This helps accurately identify whether lesions are malignant or benign, thereby aiding dermatologists in making well-informed treatment decisions. We utilize a dataset comprising a diverse range of skin lesion images, including melanomas, basal cell carcinoma, and benign lesions. Performance evaluation is conducted using standard metrics such as accuracy, sensitivity, and specificity, demonstrating the efficacy of our approach. Our results showcase promising performance levels, with our CNN model achieving high accuracy rates in distinguishing between malignant and benign lesions. Moreover, the system exhibits robustness in handling diverse skin types and lesion characteristics, highlighting its potential for real-world clinical deployment.