Medical image segmentation based on a hybrid region-based active contour model

Area of Science:

• Medical imaging
• Computer vision
• Image processing

Background:

• Medical image segmentation is crucial for diagnosis and treatment planning.
• Intensity inhomogeneity presents a significant challenge in accurate medical image segmentation.
• Existing active contour models often struggle with complex intensity variations.

Purpose of the Study:

• To develop a novel hybrid region-based active contour model.
• To address the challenge of intensity inhomogeneity in medical image segmentation.
• To improve the efficiency and robustness of medical image segmentation techniques.

Main Methods:

• A hybrid region-based active contour model incorporating global and local energy terms.
• Level set formulation with regularization for curve evolution.
• Energy minimization through a derived curve evolution equation.

Main Results:

• The proposed model effectively segments medical images with intensity inhomogeneity.
• Experimental results show superior efficiency compared to the LRBAC method.
• The model demonstrates enhanced robustness over the Chan-Vese (C-V) active contour model.

Conclusions:

• The novel hybrid active contour model provides an effective solution for medical image segmentation.
• The model's ability to handle intensity inhomogeneity offers significant advantages.
• This approach enhances the reliability and performance of medical image analysis tools.