Development of deep learning models for microglia analyses in brain tissue using DeePathology™ STUDIO.

Area of Science:

• Medical image analysis
• Artificial intelligence in pathology
• Computational neuroscience

Background:

• Growing interest in artificial intelligence (AI) for medical image analysis.
• Need for accessible AI tools for researchers and clinicians.
• Limitations of traditional histological image analysis methods.

Purpose of the Study:

• To promote and facilitate the use of AI in medical image analysis.
• To present custom deep learning models for histological image analysis.
• To provide practical guidance for implementing AI tools.

Main Methods:

• Development of custom deep learning models using DeePathology™ STUDIO.
• Workflow description for the commercially available software.
• Application of models to detect microglia in IBA1-stained mouse brain sections.

Main Results:

• Demonstration of AI models for microglia detection in mouse brain histology.
• Validation results and analysis of sample slides.
• Comparison of AI-assisted analysis with traditional methods.

Conclusions:

• Deep learning offers faster analysis than classical methods.
• AI provides extended detection possibilities beyond staining intensity.
• AI-assisted analysis reduces researcher bias and increases speed, outperforming traditional methods.