Related Experiment Videos
[Texture analysis of B-scan image using fractal dimension]
1Department of Biomedical Engineering and Instrumentation, Xi'an Jiaotong University, Xi'an 710049.
Summary
Fractal Dimension (D) analysis of liver B-scan images can help differentiate diseases. A new texture parameter S, derived from fractal dimension and scale, shows high efficiency in disease classification.
Area of Science:
- Medical Imaging
- Biophysics
- Radiology
Background:
- Liver diseases like cancer and cirrhosis present diagnostic challenges.
- Texture analysis of medical images offers potential for objective disease assessment.
- Fractal analysis provides a method to quantify complex image patterns.
Purpose of the Study:
- To investigate the fractal characteristics of liver B-scan images.
- To determine if fractal dimension (D) can differentiate between normal liver, liver cancer, and cirrhosis.
- To develop a novel texture parameter (S) for improved disease classification.
Main Methods:
- Collected B-scan images from 10 normal individuals, 3 liver cancer patients, and 3 cirrhosis patients.
- Applied fractal modeling to analyze image texture and calculate fractal dimension (D).
- Introduced and evaluated a texture classified parameter S, based on D and scale (epsilon).
Main Results:
- Fractal dimension (D) was found to be a significant parameter for differentiating liver diseases.
- The parameter S demonstrated efficient capability in distinguishing between normal and diseased liver conditions.
- Quantitative fractal analysis provides a basis for objective liver disease assessment.
Conclusions:
- Fractal dimension (D) is a valuable metric for characterizing liver B-scan images.
- The novel texture parameter S effectively differentiates liver diseases, offering a promising tool for clinical application.
- This study highlights the potential of fractal-based texture analysis in medical diagnostics.