Context-specific image quality assessment for virtual histologic staining: checklist and guideline

Area of Science:

• Computational pathology
• Digital pathology
• Medical imaging analysis

Background:

• Virtual staining computationally transforms optical images into histology-like representations.
• Current virtual staining evaluation methods often lack context-specific image quality assessment (IQA).

Purpose of the Study:

• To review existing image quality assessment (IQA) approaches for virtual staining.
• To propose a context-specific IQA checklist and guideline for virtual staining systems.

Main Methods:

• Review of current virtual staining image quality assessment (IQA) methods.
• Development of a context-specific IQA checklist and guideline.
• Linking IQA to imaging methodology and biological truth.

Main Results:

• A context-specific IQA checklist and guideline for virtually stained images were developed.
• The guideline defines a specific context of use (COU), explains underlying mechanisms, and provides COU-specific validation evidence.
• The checklist connects engineering evaluation with biomedical utility.

Conclusions:

• The developed checklist and guideline facilitate more reliable development and evaluation of virtual staining systems.
• This approach enhances the biomedical utility of virtual staining technologies.
• Context-specific IQA is crucial for accurate virtual staining assessment.