A Brain Tumor Image Segmentation Method Based on Quantum Entanglement and Wormhole Behaved Particle Swarm...

Area of Science:

• Medical Image Analysis
• Computational Intelligence
• Artificial Intelligence

Background:

• Classical image segmentation techniques often fail with complex medical image features.
• Reliable segmentation of brain tumors, especially those with "bottle-neck" or "dual tail" shapes, remains a significant challenge.
• Existing methods struggle with indistinct regions and complex tumor morphologies.

Purpose of the Study:

• To develop an improved image segmentation algorithm for challenging medical imaging tasks.
• To address the limitations of current methods in segmenting complex brain tumor shapes.
• To enhance the adaptability and accuracy of segmentation for indistinct and tapering regions.

Main Methods:

• Analysis of existing image segmentation research based on wormhole and entanglement theory.
• Improvement of a quantum-behaved particle swarm optimization (QPSO) approach using hyperbolic wormhole path measures.
• Proposal of a novel Quantum and Wormhole-behaved Particle Swarm Optimization (QWPSO) algorithm.

Main Results:

• The proposed QWPSO algorithm demonstrates superior clustering adaptability for complex "dual tail" regions compared to conventional QPSO.
• Experimental results indicate improved operational efficiency and segmentation accuracy over competing reference methods.
• QWPSO effectively segments challenging "bottle-neck" and "dual tail" shapes in medical images.

Conclusions:

• The QWPSO method shows significant promise for segmenting smeared and indistinct regions in medical images.
• This technique offers particular advantages for segmenting complex shapes common in brain tumor imaging.
• QWPSO provides a more robust solution for challenging medical image segmentation tasks.