Related Experiment Video
Updated: May 4, 2026
A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment
Published on: May 24, 2022
Vascular tree segmentation in medical images using Hessian-based multiscale filtering and level set method.
Jiaoying Jin1, Linjun Yang1, Xuming Zhang1
1Department of Biomedical Engineering, School of Life Science and Technology, Key Laboratory of Image Processing and Intelligent Control of Education Ministry of China, Huazhong University of Science and Technology, Wuhan 430074, China.
This study introduces a new method for automatically extracting vascular trees from medical images. The technique combines Hessian-based filtering and a modified level set method, achieving over 95% accuracy on synthetic data and effective results on real MRA scans.
Area of Science:
- Medical Image Analysis
- Computational Imaging
- Vascular Biology
Background:
- Vascular segmentation is crucial for medical image analysis.
- Accurate extraction of vascular structures aids in diagnosis and treatment planning.
Purpose of the Study:
- To develop a novel, automatic technique for vascular tree extraction from 2D medical images.
- To improve upon existing methods by enhancing accuracy and robustness.
Main Methods:
- A hybrid approach combining Hessian-based multiscale filtering and a modified level set method.
- Utilized morphological top-hat transformation for background attenuation.
- Introduced an improved level set method with a constrained term for accurate boundary detection.
Main Results:
- Achieved segmentation rates above 95% on synthetic images with vascular-like structures.
- Demonstrated successful and accurate extraction of most vascular structures from 2D abdomen MRA images.
- Visualizations confirmed the effectiveness of the proposed method.
Conclusions:
- The proposed method is effective for automatic vascular tree extraction in medical images.
- The combination of Hessian filtering and modified level sets offers a robust solution.
- This technique shows significant potential for clinical applications in vascular analysis.
