AI-powered microscopy image analysis for parasitology: integrating human expertise

Area of Science:

• Parasitology
• Artificial Intelligence
• Microscopy Image Analysis

Background:

• Microscopy image analysis is crucial for parasitology.
• Traditional deep learning (DL) methods lack explainability and sufficient resources.
• General-purpose DL models are data-driven and often act as black boxes.

Purpose of the Study:

• To review recent advancements in knowledge-integrated DL models for parasitology.
• To address the limitations of traditional DL in microscopy image analysis.
• To highlight the benefits of incorporating expert knowledge into AI models.

Main Methods:

• Comprehensive literature review of knowledge-integrated DL models.
• Focus on applications in parasitology microscopy image analysis.
• Analysis of how expert knowledge enhances AI performance.

Main Results:

• Knowledge-integrated DL models offer improved accuracy and explainability.
• Human expert knowledge from parasitologists can significantly enhance AI decisions.
• These models overcome limitations of traditional black-box AI.

Conclusions:

• Knowledge-integrated DL presents a promising approach for parasitology.
• Enhanced AI models can improve parasite identification and analysis.
• This integration is expected to broaden AI applications in parasitology research.