AI outperformed every dermatologist: Improved dermoscopic melanoma diagnosis through customizing batch logic and loss function in an optimized Deep CNN architecture.
Tri-Cong PhamChi-Mai LuongVan-Dung HoangAntoine DoucetPublished in: CoRR (2020)
Keyphrases
- skin lesion
- loss function
- skin cancer
- malignant melanoma
- digital image processing
- computer aided
- pairwise
- early diagnosis
- early detection
- dermoscopy images
- learning to rank
- support vector
- boosting framework
- cellular neural networks
- regularization term
- expert systems
- logistic regression
- empirical risk
- reproducing kernel hilbert space
- hinge loss
- extracted features
- stochastic gradient descent
- risk minimization
- digital images
- fixed point
- boosting algorithms
- active learning
- image enhancement
- machine learning
- pigmented skin lesions
- convex loss functions