Biomechanically constrained non-rigid MR-TRUS prostate registration using deep learning based 3D point cloud matching.
Yabo FuYang LeiTonghe WangPretesh PatelAshesh B. JaniHui MaoWalter J. CurranTian LiuXiaofeng YangPublished in: Medical Image Anal. (2021)
Keyphrases
- deep learning
- prostate biopsy
- image guided
- prostate segmentation
- transrectal ultrasound
- magnetic resonance
- prostate cancer
- unsupervised learning
- registration accuracy
- prostate brachytherapy
- image registration
- patient specific
- mr imaging
- accurate registration
- machine learning
- mr images
- medical imaging
- target registration error
- medical images
- mental models
- deformable models
- weakly supervised
- minimally invasive
- imaging modalities
- automated segmentation
- rigid transformation
- registration errors
- intraoperative
- thin plate spline
- image data
- registration process
- pre operative
- object categories
- dimensionality reduction
- object recognition
- multiscale
- image segmentation
- computer vision