Related Experiment Video
Updated: Mar 8, 2026
04:48
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
Published on: July 5, 2024
852
Pancreas Segmentation in MRI using Graph-Based Decision Fusion on Convolutional Neural Networks
Jinzheng Cai1, Le Lu2, Zizhao Zhang3
1Department of Biomedical Engineering, University of Florida, Gainesville, FL 32611, USA.
Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|January 14, 2017
Summary
This study introduces a novel deep learning method for automated pancreas segmentation in MRI scans. The approach combines deep convolutional neural networks (CNNs) and a conditional random field (CRF) framework, achieving superior accuracy for clinical applications.
Area of Science:
- Medical Imaging Analysis
- Artificial Intelligence in Healthcare
- Radiology
Background:
- Accurate pancreas segmentation is crucial for diagnosing conditions like diabetes and pancreatic cancer, and for surgical planning.
- Current segmentation methods often struggle with the complexity and variability of pancreatic anatomy in medical images.
Purpose of the Study:
- To develop an automated and accurate method for pancreas segmentation in magnetic resonance imaging (MRI) scans.
- To improve the precision of pancreas segmentation for enhanced clinical decision-making.
Main Methods:
- A novel approach combining deep convolutional neural networks (CNNs) for tissue and boundary detection with a conditional random field (CRF) framework.
- Utilized two CNN models for differentiating pancreas tissue and segmenting its boundaries, fusing results for CRF initialization.
- Applied to a dataset of 78 abdominal MRI scans.
Main Results:
- Achieved a mean Dice Similarity Coefficient (DSC) of 76.1% with a standard deviation of 8.7%.
- The proposed algorithm demonstrated superior performance compared to existing state-of-the-art methods.
- Accurate segmentation of pancreatic tissue and boundaries was obtained.
Conclusions:
- The developed graph-based decision fusion method effectively integrates CNNs and CRF for robust pancreas segmentation in MRI.
- This automated approach shows significant potential for improving diagnostic accuracy and treatment planning in pancreatic diseases.
- The method offers a promising advancement in medical image analysis for gastroenterology and oncology.
