A simple "expert system" for morphometric evaluation of cells in pleural effusions

Area of Science:

• Digital Pathology and Medical Image Analysis
• Computational Pathology
• Oncology Diagnostics

Background:

• Diagnostic pathology relies on quantitation for improved reliability and tumor prognosis.
• Lack of affordable, user-friendly, and compatible morphometric systems hinders clinical application.
• Need for accessible tools for quantitative analysis in histopathology.

Purpose of the Study:

• To develop and describe a modular, cost-effective Apple II PC-based morphometric system.
• To implement an expert system for diagnostic assistance using quantitative pathological data.
• To evaluate the system's efficacy in differentiating between specific cell types in pleural effusions.

Main Methods:

• Development of a modular Apple II PC-based system for stereologic, planimetric, and digital image analysis.
• Calculation of secondary parameters and data transfer for advanced statistical analysis.
• Utilization of invariant moments for nucleus texture description within an expert system framework.

Main Results:

• The developed system successfully records and analyzes morphometric and stereologic data.
• Invariant moments effectively characterized nucleus textures for differential diagnosis.
• Cytological analysis of pleural effusions demonstrated accurate differentiation between normal mesothelium, mesothelioma, and adenocarcinoma metastases.

Conclusions:

• The in-house developed Apple II PC-based morphometric system provides a viable solution for quantitative pathology.
• The expert system, coupled with invariant moment analysis, shows promise for improving diagnostic accuracy in challenging cases.
• This approach offers a practical method for classifying pleural effusion cytology, aiding in patient prognosis and treatment decisions.