Segmentation and Pore Structure Estimation in SEM Images of Tissue Engineering Scaffolds Using Genetic Algorithm.

Area of Science:

• Biomaterials Science
• Image Analysis
• Computational Biology

Background:

• Scaffold characterization is crucial for bone tissue engineering.
• Analyzing Scanning Electron Microscope (SEM) images of scaffolds is complex.
• Existing methods often require extensive labeled data.

Purpose of the Study:

• To develop a Python package for automated segmentation and analysis of SEM images of bone tissue engineering scaffolds.
• To enable accurate pore characteristic detection using minimal labeled training data.
• To provide a versatile tool applicable to various SEM image analysis tasks.

Main Methods:

• Image quality enhancement using histogram equalization.
• Global thresholding for image segmentation.
• Genetic Algorithm (GA) for optimizing threshold values.
• Analysis of pore size, elongation, and orientation.

Main Results:

• The developed Python package successfully segments and analyzes SEM images of scaffolds.
• The method accurately detects pore characteristics with limited training data.
• Results for chitosan-alginate scaffolds showed satisfactory agreement with experimental data.

Conclusions:

• The Python package offers an efficient and accurate method for SEM image analysis in bone tissue engineering.
• The algorithm's adaptability extends to diverse SEM imaging conditions and scaffold types.
• The image segmentation technique has broader applications beyond scaffold analysis, including cell and subcellular component identification.