Diagnosis support system based on clinical guidelines: comparison between case-based fuzzy cognitive maps and...

Area of Science:

• Medical Informatics
• Artificial Intelligence in Medicine
• Clinical Decision Support

Background:

• Diagnostic errors are a significant concern in healthcare, stemming from cognitive, educational, and training issues.
• Cognitive factors contributing to errors include flawed reasoning, incomplete knowledge, and heuristic biases.
• Existing diagnostic methods face challenges in accurately interpreting complex patient data.

Purpose of the Study:

• To introduce and evaluate case-based fuzzy cognitive maps (CFCMs) as an advanced approach for medical diagnosis.
• To compare the diagnostic performance of CFCMs against traditional Bayesian belief networks (BBNs).
• To develop a semantic web framework supporting both diagnostic reasoning methods.

Main Methods:

• Development of a semantic web framework integrating CFCMs and BBNs.
• Utilized a database of 174 anonymous European patient cases, including demographic and clinical data (signs, symptoms, age, sex).
• Employed a statistical approach to quantitatively compare the accuracy and performance of the two diagnostic methods.

Main Results:

• Case-based fuzzy cognitive maps demonstrated comparable or superior performance to Bayesian belief networks in diagnostic accuracy.
• The semantic web framework effectively supported the implementation and comparison of both reasoning models.
• Analysis highlighted the potential of CFCMs in handling complex medical diagnostic scenarios.

Conclusions:

• Case-based fuzzy cognitive maps offer a promising alternative for medical diagnosis, potentially reducing diagnostic errors.
• The integration of AI-driven methods like CFCMs within semantic web frameworks can enhance clinical decision support systems.
• Further research is warranted to validate CFCMs across diverse medical conditions and patient populations.