An Integrated Soft Computing Approach to Hughes Syndrome Risk Assessment

Area of Science:

• Autoimmune disorders
• Computational medicine
• Artificial intelligence in diagnostics

Background:

• Antiphospholipid Syndrome (APS), or Hughes syndrome, is an autoimmune disorder causing thrombosis and pregnancy complications due to antiphospholipid antibodies.
• Current diagnostic methods for APS are complex, necessitating improved risk assessment and diagnostic tools.
• Early and accurate diagnosis of APS is crucial for patient management and preventing severe outcomes like thrombotic events and death.

Purpose of the Study:

• To develop an advanced decision support system for diagnosing Antiphospholipid Syndrome (APS).
• To enhance the accuracy and reliability of APS diagnosis using a hybrid computational approach.
• To improve patient classification for both the presence and absence of APS.

Main Methods:

• Development of a diagnostic decision support system integrating Logic Programming for knowledge representation and reasoning.
• Implementation of a computational framework utilizing Artificial Neural Networks for enhanced diagnostic capabilities.
• Formalization of diagnostic criteria and patient data within a structured computational model.

Main Results:

• The developed system demonstrated improved diagnostic performance for Antiphospholipid Syndrome.
• Achieved a sensitivity exceeding 85% in correctly identifying patients with APS.
• Reached a specificity close to 95% in correctly identifying patients without APS.

Conclusions:

• The proposed AI-driven decision support system significantly enhances the accuracy of Antiphospholipid Syndrome diagnosis.
• This computational approach offers a promising tool for improving clinical risk assessment and patient stratification in APS.
• Further research and standardization of antibody quantification are supported by these findings for better APS management.