ESP: an expert system for poisoning diagnosis and management

Area of Science:

• Medical Informatics
• Toxicology
• Artificial Intelligence

Background:

• Clinical decision-making in poisoning cases requires rapid access to accurate information.
• Existing systems may lack the specialized knowledge or real-time support needed for effective poisoning management.
• Timely and relevant information is crucial for improving patient outcomes in toxicological emergencies.

Purpose of the Study:

• To describe a clinical decision support system (CDSS) designed for poisoning cases.
• To detail the architecture and components of the expert system, including its knowledge base and inference engine.
• To present the development of a knowledge acquisition tool (KAT) for expert updates.

Main Methods:

• Development of a rule-based expert system incorporating a knowledge base and inference engine.
• Utilization of the C Language Integrated Production System (CLIPS) for inferencing.
• Design of the CDSS and KAT as web applications.
• Input of patient signs and symptoms to generate differential diagnoses and management procedures.

Main Results:

• The system successfully accepts patient data and outputs potential poisoning types with management strategies.
• A functional knowledge acquisition tool (KAT) was developed, enabling toxicological experts to update the system's knowledge base.
• The system architecture, web application design, and CLIPS integration were detailed.

Conclusions:

• The developed CDSS provides timely, expert-driven recommendations for poisoning cases.
• The system aids clinicians in making informed decisions for immediate patient evaluation and diagnosis.
• Future work may involve further system enhancements and broader clinical implementation.