Related Experiment Video
Updated: Mar 9, 2026
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
Published on: January 7, 2019
A supervoxel-based segmentation method for prostate MR images
Zhiqiang Tian1, Lizhi Liu1,2, Zhenfeng Zhang2
1Department of Radiology and Imaging Sciences, School of Medicine, Emory University, 1841 Clifton Road NE, Atlanta, GA, 30329, USA.
This study introduces a novel supervoxel-based method for segmenting prostate MR images, achieving accurate results for prostate cancer management. The technique shows promise for diagnosis and therapy applications.
Area of Science:
- Medical Imaging
- Computer Vision
- Radiology
Background:
- Prostate segmentation on MR images is crucial for prostate cancer management.
- Accurate segmentation aids in diagnosis and therapy planning.
Purpose of the Study:
- To propose a supervoxel-based segmentation method for prostate MR images.
- To evaluate the method's performance on in-house and public datasets.
Main Methods:
- Utilized supervoxels, groups of pixels with similar characteristics, for segmentation.
- Employed a supervoxel-based energy function with data and smoothness terms.
- Applied 3D graph cut for energy minimization and a 3D active contour model for surface smoothing.
Main Results:
- Achieved mean Dice similarity coefficients of 87.2% on in-house data and 88.2% on the PROMISE12 dataset.
- Demonstrated satisfactory performance for prostate MR image segmentation.
Conclusions:
- The supervoxel-based method accurately segments prostate MR images.
- This technique has potential applications in prostate cancer diagnosis and therapy.
More Related Videos
11:38Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench
Published on: August 23, 2017
06:08A Cognitive Fusion-guided Prostate Biopsy Using Multiparametric Magnetic Resonance Imaging and Transrectal Ultrasound
Published on: March 21, 2025