DLGNet: A dual-branch lesion-aware network with the supervised Gaussian Mixture model for colon lesions classification in colonoscopy images.
Kai-Ni WangShuaishuai ZhuangQi-Yong RanPing ZhouJie HuaGuang-Quan ZhouXiaopu HePublished in: Medical Image Anal. (2023)
Keyphrases
- skin lesion
- gaussian mixture model
- pigmented skin lesions
- feature vectors
- image classification
- input image
- computer aided
- supervised learning
- image features
- skin cancer
- pet images
- log gabor
- pattern recognition
- em algorithm
- mixture model
- mouth region
- breast tissue
- object recognition
- machine learning
- image retrieval
- feature selection
- dermoscopy images
- malignant melanoma
- unsupervised learning
- feature space
- gaussian mixture modeling
- lesion detection
- feature extraction
- training set
- background subtraction
- eye tracking
- computer aided detection
- feature representation
- lesion segmentation
- extracted features
- information retrieval
- expectation maximization
- segmentation algorithm