Deep Learning-powered Skin Cancer Screening CNN and GRAD-CAM Insights

Deep Learning-powered Skin Cancer Screening CNN and GRAD-CAM Insights

Sathiyan A

DepartmentOf Electronics and Communication Engineering  PanimalarInstituteOfTechnologyChennai, India

sathiyansurya2021@gmail.com

Nishanth N

Department Of  Electronics and Communication Engineering

Panimalar Institute Of Technology  Chennai,India starnishanthn@gmail.com

PrasannaKumar Reddy A

Department Of  Electronics and Communication Engineering

Panimalar Institute Of Technology Chennai, India

prasannaankinapalli@gmail.com

Dr  S .Sathiya Priya

Professor and Head

Department of  Electronics and Communication Engineering

Panimalar Institute OfTechnology  Chennai, India ecehodpit@gmail.com

Abstract -Skin cancer is one of the most commonly diagnosed cancers worldwide, making early detection essential for effective treatment. Recent advancements in deep learning, particularly Convolutional Neural Networks (CNNs), have significantly improved medical image analysis. This study explores an innovative CNN-based approach for skin cancer detection, utilizing a multi-stage model trained on a comprehensive dataset of dermatoscopic images. By integrating transfer learning and data augmentation, the proposed method enhances model accuracy and generalization. Experimental evaluations demonstrate its effectiveness in accurately classifying malignant and benign skin lesions. This automated system offers a valuable tool for dermatologists, aiding in early diagnosis, improving patient care, and potentially reducing healthcare burdens.

Keywords - Skin cancer, Early detection, Deep learning, Convolutional Neural Networks (CNNs), Medical image analysis, Dermatoscopic images, Multi-stage model, Transfer learning, Data augmentation, Model accuracy, Generalization, Malignant lesions, Benign lesions, Automated system, Dermatologists, Early diagnosis, Patient care, Healthcare burden