Related Experiment Video
Updated: Jul 28, 2026
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
2.6K
Few-shot small vessel segmentation using a detail-preserving network enhanced by discriminator
Yan Huang1,2, Jinzhu Yang3,4,5, Qi Sun1,2
1Key Laboratory of Intelligent Computing in Medical image, Ministry of Education, Northeastern University, Shenyang, 110819, China.
Medical & Biological Engineering & Computing
|May 12, 2025
Summary
This study introduces a new AI method for segmenting small blood vessels, improving early disease detection. The approach enhances accuracy and generalization for coronary and pulmonary artery imaging.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Vascular Biology
Background:
- Accurate segmentation of small vessels like coronary and pulmonary arteries is vital for diagnosing vascular diseases.
- Challenges include small vessel size, complex anatomy, and limited annotated training data.
Purpose of the Study:
- To develop an improved few-shot segmentation method for small vessels.
- To enhance the accuracy and robustness of vessel segmentation models.
Main Methods:
- A detail-preserving network featuring multi-residual hybrid dilated convolution to capture fine vessel structures.
- Adversarial learning with a discriminator to leverage unlabeled data and improve model generalization.
Main Results:
- The proposed method significantly improves segmentation accuracy for small vessels.
- Achieved higher accuracy and lower false positive rates compared to state-of-the-art methods.
- Demonstrated superior generalization capability on coronary and pulmonary artery datasets.
Conclusions:
- The novel approach effectively addresses challenges in small vessel segmentation.
- This method offers a promising tool for assisting clinical diagnosis of vascular diseases.
