Related Experiment Video
Updated: Mar 28, 2026
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
3.7K
[Lesion Extraction from B-type Ultrasound Image Using Subordinate Degree Region Level Set Method]
Summary
This study introduces a new subordinate degree region level set model for segmenting lesions in B-type ultrasound images. The method enhances accuracy by focusing calculations locally, improving upon existing level set techniques.
Area of Science:
- Medical Imaging
- Image Processing
- Computational Biology
Background:
- B-type ultrasound images are crucial for medical diagnosis.
- Challenges in automatic lesion segmentation include inhomogeneity, low contrast, noise, and blurred edges.
Purpose of the Study:
- To develop an improved automatic lesion segmentation method for B-type ultrasound images.
- To address the limitations of existing segmentation techniques.
Main Methods:
- A subordinate degree region level set model was proposed, defining pixel subjection to target and background.
- Pixels were classified based on subordinate degree probabilities.
- Lesion segmentation was localized to a specific area, reducing computational complexity.
Main Results:
- The proposed method achieved improved lesion segmentation results on B-type ultrasound images.
- Performance was superior compared to popular level set methods.
Conclusions:
- The subordinate degree region level set model effectively enhances automatic lesion segmentation in B-type ultrasound images.
- Localizing calculations significantly reduces complexity while maintaining accuracy.
