Skin cancer classification using a borderline-SMOTE enhanced neural network model on dermoscopic images

Key Points

• Classification accuracy reached 92% using the borderline-SMOTE method with neural network.
• The model processes dermoscopic images to identify skin cancer types effectively and consistently.
• A modified neural network architecture trained on enhanced datasets improves predictions significantly.
• These results may enhance dermatological analysis but require real-world validation beyond the dataset.