Active shape models—their training and application
Computer Vision and Image Understanding
3D Prostate Surface Detection from Ultrasound Images Based on Level Set Method
MICCAI '02 Proceedings of the 5th International Conference on Medical Image Computing and Computer-Assisted Intervention-Part II
Evolving Bubbles for Prostate Susrface Detection from TRUS Images
BIBE '03 Proceedings of the 3rd IEEE Symposium on BioInformatics and BioEngineering
IEEE Transactions on Information Technology in Biomedicine
Editorial: Medical image segmentation: Quo Vadis
Computer Methods and Programs in Biomedicine
Level Set Segmentation of Knee Bones Using Normal Profile Models
MIRAGE '09 Proceedings of the 4th International Conference on Computer Vision/Computer Graphics CollaborationTechniques
3D Prostate Segmentation in Ultrasound Images Based on Tapered and Deformed Ellipsoids
MICCAI '09 Proceedings of the 12th International Conference on Medical Image Computing and Computer-Assisted Intervention: Part II
Rough set approach in ultrasound biomicroscopy glaucoma analysis
AST/UCMA/ISA/ACN'10 Proceedings of the 2010 international conference on Advances in computer science and information technology
Incremental shape statistics learning for prostate tracking in TRUS
MICCAI'10 Proceedings of the 13th international conference on Medical image computing and computer-assisted intervention: Part II
Automatic prostate segmentation using fused ultrasound B- mode and elastography images
MICCAI'10 Proceedings of the 13th international conference on Medical image computing and computer-assisted intervention: Part II
A PET/CT directed, 3D ultrasound-guided biopsy system for prostate cancer
MICCAI'11 Proceedings of the 2011 international conference on Prostate cancer imaging: image analysis and image-guided interventions
Computer Methods and Programs in Biomedicine
Hi-index | 0.00 |
Boundary outlining, or segmentation, of the prostate is an important task in diagnosis and treatment planning for prostate cancer. This paper describes an algorithm based on two-dimensional (2D) active shape models (ASM) for semi-automatic segmentation of the prostate boundary from ultrasound images. Optimisation of the 2D ASM for prostatic ultrasound was done first by examining ASM construction and image search parameters. Extension of the algorithm to three-dimensional (3D) segmentation was then done using rotational-based slicing. Evaluation of the 3D segmentation algorithm used distance- and volume-based error metrics to compare algorithm generated boundary outlines to gold standard (manually generated) boundary outlines. Minimum description length landmark placement for ASM construction, and specific values for constraints and image search were found to be optimal. Evaluation of the algorithm versus gold standard boundaries found an average mean absolute distance of 1.09+/-0.49mm, an average percent absolute volume difference of 3.28+/-3.16%, and a 5x speed increase versus manual segmentation.