Design of Metaheuristic Optimization-Based Vascular Segmentation Techniques for Photoacoustic Images.

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Computational Biology

Background:

• Photoacoustic imaging (PAI) offers functional, anatomical, and molecular details for internal organs.
• PAI is increasingly used in brain and cancer imaging (prostate, breast, ovarian).
• Vessel segmentation in PAI is crucial for diagnosis but technically challenging.

Purpose of the Study:

• To develop and evaluate metaheuristic optimization-based techniques for accurate vascular segmentation in PAI.
• To enhance the diagnostic potential of PAI through improved vessel image analysis.

Main Methods:

• The study employed Shannon's entropy function (SEF) and multilevel Otsu thresholding (MLOT) for vessel segmentation.
• Three metaheuristic algorithms—cuckoo search (CS), equilibrium optimizer (EO), and harmony search (HS)—were used to optimize segmentation parameters.
• The model was validated using a benchmark PAI dataset.

Main Results:

• The proposed metaheuristic optimization approach significantly improved vascular segmentation accuracy.
• The model achieved a high accuracy rate of 98.71% on the benchmark PAI dataset.
• Experimental analysis confirmed the model's effectiveness under varying conditions.

Conclusions:

• Metaheuristic optimization-based vascular segmentation is a superior method for PAI.
• The developed techniques enhance the diagnostic utility of PAI for various medical applications.
• This approach offers a robust solution for challenging vessel segmentation tasks in biomedical imaging.