An Improved Sparrow Search Algorithm and Its Application in HIFU Sound Field

Area of Science:

• Computational Intelligence
• Swarm Intelligence Optimization
• Medical Physics

Background:

• The Sparrow Search Algorithm (SSA) is a novel swarm intelligence method known for rapid convergence and global search capabilities.
• However, SSA suffers from unstable initial populations, susceptibility to local optima, and decreasing population diversity during iterations.

Purpose of the Study:

• To address the limitations of the standard SSA.
• To propose an improved Sparrow Search Algorithm (ISSA) with enhanced population initialization, local optima avoidance, and diversity maintenance.

Main Methods:

• ISSA employs a Chebyshev chaotic map and elite opposition-based learning for improved initial population quality.
• Dynamic weight factors and Levy flight strategies are integrated into the producer update process to prevent local convergence.
• A mutation strategy is applied to the scrounger update process to maintain population diversity.

Main Results:

• ISSA was validated on 23 benchmark functions, demonstrating superior convergence accuracy, speed, and stability compared to seven other algorithms.
• Application of ISSA to optimize high-intensity focused ultrasound (HIFU) sound fields showed significant improvements in focusing performance.
• ISSA effectively reduced sound field sidelobe influence in HIFU applications.

Conclusions:

• The proposed ISSA effectively overcomes the shortcomings of the original SSA.
• ISSA offers enhanced optimization capabilities, making it suitable for complex problems.
• ISSA shows significant potential for improving HIFU treatment by enhancing sound field focusing and reducing sidelobe effects.