Fully automatic segmentation of paraspinal muscles from 3D torso CT images via multi-scale iterative random forest classifications.
Naoki KamiyaJing LiMasanori KumeHiroshi FujitaDinggang ShenGuoyan ZhengPublished in: Int. J. Comput. Assist. Radiol. Surg. (2018)
Keyphrases
- ct images
- random forest
- multiscale
- computed tomography
- medical images
- decision trees
- feature set
- medical imaging
- ensemble methods
- ct scans
- multi label
- region of interest
- fold cross validation
- image segmentation
- anatomical structures
- image processing
- edge detection
- coarse to fine
- base classifiers
- image analysis
- ct data
- contrast enhanced
- remote sensing
- image features
- manual segmentation