DCAN: Deep contour-aware networks for object instance segmentation from histology images

Keyphrases

• Accurate Probability 33%
• Accurate Segmentation 66%
• Adenocarcinoma 33%
• Appearance Variation 66%
• Automated Detection 33%
• Automatic Segmentation 33%
• Auxiliary Supervision 33%
• Contextual Features 33%
• Contour-aware 100%
• Deep Network 33%
• Dependant 33%
• Detection Accuracy 66%
• Error-prone 33%
• Fully Convolutional Network 33%
• Gland Segmentation 33%
• Glands 33%
• Histological Analysis 33%
• Histological Structure 66%
• Histology Images 100%
• Instance Segmentation 100%
• Malignancy Grade 33%
• Manual Annotation 33%
• MICCAI 66%
• Multi-task Learning Framework 33%
• Nuclei Segmentation 33%
• Object Instance 33%
• Object of Interest 66%
• Object Segmentation 33%
• Pathologist 33%
• Probability Map 33%
• Quantitative Diagnosis 33%
• Segmentation Challenge 100%
• Segmentation Performance 33%
• Superior Performance 33%
• Supervision Mechanism 33%
• Vanishing Gradient 33%

Computer Science

• Annotation 50%
• Appearance Variation 100%
• Fully Convolutional Network 50%
• Histopathological Image 50%
• Image Analysis 50%
• Instance Segmentation 100%
• Learning Framework 50%
• Multitask Learning 50%
• Object Instance 100%
• Probability Map 50%
• Segmentation Performance 50%
• Superior Performance 50%

Chemical Engineering

• Auxiliaries 100%