global-journal-of-management-and-business-research-c-finance Global Journal of Management and Business Research - C: Finance 0975-5853 2249-4588 Global Journals Publishing Group Incorporated 10.34257/GJMBRC257949 257949 A Comparative Study of Convolutional Neural Networks for Automated Skin Cancer Detection CNN-Based Automated Skin Cancer Detection BENEMarius DIKWEGaston NJOUPOUOGNIGNIMoussa CAMEROON, Advanced Technical Teacher Training College 26 06 2026 26 1 1 20

This study evaluates the performance of various Convolutional Neural Network (CNN) architectures, including ResNet50, VGG16, and InceptionV3, in the automated classification of malignant melanoma vs. benign nevi. Using the ISIC 2019 dataset, we implemented data augmentation and transfer learning techniques to overcome class imbalance. Results indicate that ResNet50 achieved the highest accuracy of 94.2%, suggesting that deep residual learning is highly effective for dermatological image analysis. The findings support the integration of AI tools in clinical workflows to aid early diagnosis.

 Convolutional Neural Networks Skin Cancer Detection Deep Learning Medical Imaging Melanoma Computer-Aided Diagnosis. Credit scoring contextual variables logit probit NN and SVM models. Full Text