Related Experiment Video
Updated: Feb 20, 2026
A Cognitive Fusion-guided Prostate Biopsy Using Multiparametric Magnetic Resonance Imaging and Transrectal Ultrasound
Published on: March 21, 2025
Multi-view collaborative segmentation for prostate MRI images
Abstract:
Prostate delineation from MRI images is a prolonged challenging issue partially due to appearance variations across patients and disease progression. To address these challenges, our proposed collaborative method takes into account the computed multiple label-relevance maps as multiple views for learning the optimal boundary delineation. In our method, we firstly extracted multiple label-relevance maps to represent the affinities between each unlabeled pixel to the pre-defined labels to avoid the selection of handcrafted features. Then these maps were incorporated in a collaborative clustering to learn the adaptive weights for an optimal segmentation which overcomes the seeds selection sensitivity problems. The segmentation results were evaluated over 22 prostate MRI patient studies with respect to dice similarity coefficient (DSC), absolute relative volume difference (ARVD) and average symmetric surface distance (ASSD) (mm). The results and t-Test demonstrated that the proposed method improved the segmentation accuracy and robustness and the improvement was statistically significant.
Insights
This study introduces a novel collaborative method for prostate MRI segmentation, enhancing accuracy by using multiple label-relevance maps. The approach significantly improves segmentation robustness and overcomes previous limitations.
Area of Science:
- Medical Imaging
- Computer Vision
- Machine Learning
Background:
- Prostate delineation from MRI is challenging due to patient variability and disease progression.
- Existing methods often rely on handcrafted features or are sensitive to initial seed selection.
Purpose of the Study:
- To develop an automated and robust prostate MRI segmentation method.
- To overcome the limitations of manual segmentation and improve accuracy.
Main Methods:
- Extracted multiple label-relevance maps to represent pixel affinities without handcrafted features.
- Employed collaborative clustering with adaptive weights for optimal segmentation.
- Evaluated performance using Dice Similarity Coefficient (DSC), Absolute Relative Volume Difference (ARVD), and Average Symmetric Surface Distance (ASSD) on 22 prostate MRI datasets.
Main Results:
- The proposed collaborative method demonstrated improved segmentation accuracy and robustness.
- Statistical analysis (t-Test) confirmed the significant improvement over existing techniques.
- The method effectively addressed seed selection sensitivity issues.
Conclusions:
- The novel collaborative approach offers a significant advancement in automated prostate MRI segmentation.
- This method provides a more accurate and reliable tool for clinical applications.
- Future work could explore its application to other anatomical structures or imaging modalities.
