Artificial Intelligence Applications in COVID-19-Associated Coagulopathy: Lessons Learned

Area of Science:

• * Medical research
• * Data science
• * Immunology

Background:

• * Coronavirus disease 2019 (COVID-19)-associated coagulopathy (CAC) involves endothelial injury and thrombosis.
• * D-dimer elevation is a key indicator of CAC severity.
• * Existing models for CAC prediction have limitations in validation and generalizability.

Purpose of the Study:

• * To review the application of artificial intelligence (AI) in predicting thrombotic and mortality outcomes in COVID-19 patients.
• * To elucidate the mechanisms underlying CAC using machine learning (ML) approaches.
• * To identify potential therapeutic targets and risk stratification strategies for CAC.

Main Methods:

• * Review of studies applying AI and ML to COVID-19 coagulopathy data.
• * Analysis of multivariable and time-aware models incorporating D-dimer.
• * Mechanistic ML analyses of proteomic and coagulation data to identify key pathways.

Main Results:

• * D-dimer is a highly informative marker for CAC outcomes, especially within sophisticated models.
• * ML identified an IL-6-centered immunothrombotic network linking cytokine signaling and complement activation.
• * AI models quantified residual thrombotic risk, suggesting potential for identifying high-risk patients.

Conclusions:

• * AI and ML offer powerful tools for understanding and predicting CAC.
• * Integrated data analysis reveals actionable pathways for intervention.
• * Future research requires prospectively validated AI models for risk-adapted anticoagulation and immunomodulation.