Immunological Mechanisms and Machine Learning Applications in Post-COVID-19 Syndrome: A Narrative Review

Area of Science:

• Immunology
• Computational Biology
• Medical Informatics

Background:

• Post-COVID-19 syndrome (PCS) presents as diverse persistent symptoms after SARS-CoV-2 infection.
• Key factors include immune dysregulation, autonomic imbalance, impaired thymic function, and potential viral persistence.

Purpose of the Study:

• To systematically review the immunopathogenesis of PCS.
• To evaluate the application of artificial intelligence (AI) and machine learning (ML) for PCS prediction and clinical stratification.

Main Methods:

• A PRISMA 2020-guided systematic review of multiple databases (Jan 2020-Dec 2025).
• Inclusion of studies on PCS immunopathology and AI/ML applications.
• Risk of bias assessed using PROBAST; qualitative synthesis due to heterogeneity.

Main Results:

• PCS may stem from sustained inflammation, cytokine dysregulation, autoimmunity, and delayed T-cell homeostasis restoration.
• Persistent thymic dysfunction contributes to immune imbalance and susceptibility to infections/reactivations.
• AI/ML models show promise but face limitations: small sample sizes, overfitting, lack of external validation, and outcome heterogeneity.

Conclusions:

• Combining immunopathological data with AI/ML offers potential for PCS risk stratification.
• Current AI models require further validation before clinical implementation.
• Future research must focus on standardized methodologies, external validation, and mechanistically informed biomarkers.