WSI-Net: Branch-Based and Hierarchy-Aware Network for Segmentation and Classification of Breast Histopathological Whole-Slide Images.
Haomiao NiHong LiuKuansong WangXiangdong WangXunjian ZhouYueliang QianPublished in: MLMI@MICCAI (2019)
Keyphrases
- image classification
- pigmented skin lesions
- segmentation algorithm
- segmentation method
- image analysis
- segmentation errors
- test images
- mammographic images
- accurate segmentation
- skin lesion
- segmented images
- edge detection
- ground truth
- image data
- segmentation accuracy
- input image
- fully automatic
- mammogram images
- image features
- adaptive thresholding
- grey level
- image regions
- breast ultrasound
- image segmentation
- breast cancer diagnosis
- image retrieval
- breast tissue
- image segments
- level set
- quantitative evaluation
- digital mammograms
- object recognition
- image segmentation algorithms
- pixel classification
- image registration
- lung nodules
- automatic segmentation
- multiscale
- benign and malignant
- feature extraction
- feature space
- pectoral muscle
- grey level co occurrence matrix
- feature selection
- optic disc
- feature vectors
- microscopy images
- watershed transform
- textural features
- medical images
- gray level
- object segmentation
- low contrast