Skin Cancer Classification

• Tech Stack: Python, google colab notebook, MATLAB, Tensorflow, Keras 3 API
• Research Paper: Link
• Github URL: Project Link

A comprehensive analysis of the effectiveness has been done on the top seven pre-trained deep CNN models, including VGG16, Xception, NASNetLarge, EfficientNetB7, MobileNetV2,InceptionResNetV2, and DenseNet169, and optimization techniques have been analyzed for their impact. In the end, the best model has been found, giving researchers a starting point for creating a CNN-based method that is more effective for early skin disease infection identification.

multi-operation data augmentation approaches to address the imbalanced and limited data in publically available datasets, guaranteeing equitable representation of samples from eight different classes.

Finally, to improve classification results, the suggested customized fusion model with several dense layers (dense-1, dense-2, dense-3, and dense-4) has been proposed.