Melanoma and nevus skin lesion classification using handcraft and deep learning feature fusion via mutual information measures

Mathematics

• Melanoma 100%
• Information Measure 88%
• Skin 88%
• Mutual Information 82%
• Fusion 74%
• Feature Extraction 68%
• Learning 62%
• Preprocessing 52%
• Color 37%
• Network Architecture 34%
• Fusion Rule 31%
• Region of Interest 30%
• Support Vector Machine 28%
• Texture 28%
• Specificity 27%
• Framework 26%
• Asymmetry 26%
• Training 24%
• Justify 22%
• Linear regression 22%
• Neural Networks 21%
• Evaluation 16%
• Metric 15%
• Demonstrate 13%
• Performance 13%

Physics & Astronomy

• learning 74%
• lesions 74%
• fusion 55%
• preprocessing 42%
• pattern recognition 41%
• computer aided design 40%
• color 24%
• borders 18%
• regression analysis 17%
• education 16%
• textures 14%
• adjusting 13%
• asymmetry 13%
• evaluation 10%
• sensitivity 10%
• performance 7%

Engineering & Materials Science

• Deep learning 61%
• Skin 59%
• Fusion reactions 55%
• Computer aided design 38%
• Feature extraction 35%
• Color 32%
• Linear regression 23%
• Network architecture 21%
• Convolutional neural networks 20%
• Support vector machines 18%
• Textures 18%