Skin cancer classification using EfficientNet architecture

Skin cancer is one of the most common deadly diseases worldwide. Hence, 
skin cancer classification is becoming increasingly important because 
treatment in the early stages of skin cancer is much more effective and 
efficient. This study focuses on the classification of three common types of 
skin cancer, namely basal cell carcinoma (BCC), squamous cell carcinoma 
(SCC), and melanoma using EfficientNet architecture. The dataset is 
preprocessed and each image in the dataset is resized to 256×256 pixels 
prior to incorporation in later stages. We then train all types of EfficientNet 
starting from EfficientNet-B0 to EfficientNet-B7 and compare their 
performances. Based on the test results, all trained EfficientNet models are 
capable of producing good accuracy, precision, recall, and F1-score in skin 
cancer classification. Particularly, our designed EfficientNet-B4 model 
achieves 79.69% accuracy, 81.67% precision, 76.56% recall, and 79.03% 
F1-score as the highest among others. These results confirm that 
EfficientNet architecture can be utilized to classify skin cancer properly.