Scan-IT: Domain segmentation of spatial transcriptomics images by graph neural network.
Zixuan CangXinyi NingAnnika NieMin XuJing ZhangPublished in: BMVC (2021)
Keyphrases
- segmentation algorithm
- segmentation method
- neural network
- test images
- spatial relationships
- segmentation errors
- image data
- image analysis
- ground truth
- edge detection
- spatial information
- accurate segmentation
- segmentation accuracy
- fully automatic
- image database
- segmented images
- unsupervised segmentation
- image regions
- microscopic images
- image retrieval
- input image
- region segmentation
- three dimensional
- segmented regions
- image classification
- automatically segmented
- spatial frequency
- image segmentation
- adaptive thresholding
- microscopy images
- image segmentation algorithms
- image features
- spatial correlation
- pixel wise
- tubular structures
- image slices
- pixel level
- automatic segmentation
- object segmentation
- low depth of field
- image segments
- neural network is trained
- object recognition
- spatial location
- foreground and background
- multiscale
- grey level
- texture segmentation
- intensity images
- range scans
- artificial neural networks
- pigmented skin lesions
- bounding box
- positron emission tomography
- intensity distribution
- multiple objects
- shape prior
- region growing
- spatial data
- energy function
- level set
- image processing