Comparison of conventional morphometry and image analysis for the solution of histomorphometric problems

Area of Science:

• Histopathology
• Biomedical image analysis
• Quantitative pathology

Background:

• Conventional morphometry in histopathology relies on contextual information and prior knowledge.
• Image analysis presents challenges in integrating contextual information and a priori knowledge.
• Artificial intelligence (AI) is being explored to address these challenges in image analysis.

Purpose of the Study:

• To delineate the key distinctions between conventional morphometry and advanced image analysis techniques in histopathology.
• To highlight the strengths and limitations of each approach for pathological assessments.

Main Methods:

• Comparative analysis of conventional morphometry and digital image analysis methodologies.
• Evaluation of feature extraction capabilities, including densitometrical and textural properties.
• Discussion of the role of artificial intelligence in enhancing image analysis.

Main Results:

• Conventional morphometry effectively utilizes contextual information and a priori knowledge.
• Image analysis excels in measuring densitometrical and textural features, quantiles, and moments.
• Image analysis demonstrates a significantly higher descriptive power compared to conventional morphometry.

Conclusions:

• Image analysis provides a more comprehensive characterization of histopathological samples due to its advanced measurement capabilities.
• The integration of AI is crucial for unlocking the full potential of image analysis in overcoming its inherent complexities.
• Advanced image analysis, potentially augmented by AI, represents a more powerful tool for histopathological research and diagnostics.