Skin Cancer Classification and Comparison of Pretrained Models Performance using Transfer Learning

Abstract

Skin cancer is an uncommon but serious malignancy. Dermoscopic images examination and biopsy are required for cancer detection. Deep learning (DL) is extremely effective in learning characteristics and predicting malignancies. However, DL requires a large number of images to train. Image augmentation and transferring learning were employed to overcome the lack of images issue. In this study we divided images into two categories: benign and malignant. To train and test our models, we used the public ISIC 2020 database. Melanoma is classified as malignant in the ISIC 2020 dataset. Along with categorization, the dataset was studied to demonstrate variation. The performance of three top pretrained models was then benchmarked in terms of training and validation accuracy. Three optimizers were employed to optimize the loss: RMSProp, SGD, and ADAM. Using ResNet, VGG16, and MobileNetV2, we obtained training accuracy of 98.73%, 99.12%, and 99.76%, respectively. Using these three pretrained models, we attained a validation accuracy of 98.39%.

The Power of Generative AI to Augment for Enhanced Skin Cancer Classification: A Deep Learning Approach

“…In this study, Subroto Singha et al conducted an analysis using the public ISIC 2020 database to categorize lesions. They benchmarked the performance of three top pretrained models (ResNet, VGG16, and MobileNetV2) demonstrated excellent training accuracy, ranging from 98.73% to 99.76%, and 98.39% [20]. In this research, Abilash Panja et al implemented a (CNNs) model to recognise and categorise the cancerous state of skin into malignant (melanoma) and benign (non-malignant) categories [14].…”