Introduction: Endometriosis, a chronic and estrogen-dependent inflammatory condition, affects millions worldwide, frequently causing pain, infertility, and a diminished quality of life. Delayed diagnosis remains a major challenge due to the lack of sensitive non-invasive biomarkers. Emerging evidence suggests that alterations in the gut and reproductive tract microbiomes contribute to disease pathophysiology through immune and hormonal dysregulation. Purpose of the Work: This review aims to synthesize current knowledge on microbiome changes in endometriosis and explore the potential applications of artificial intelligence (AI) and machine learning (ML) for identifying microbiome-derived biomarkers and improving early diagnosis. Material and Methods: A narrative review of peer-reviewed literature from 2015–2025 was conducted using PubMed, Scopus, and Web of Science. Keywords included “endometriosis,” “microbiome,” “artificial intelligence,” and “machine learning.” Studies were assessed for relevance, methodological quality, and contributions to understanding microbiome alterations and AI applications in endometriosis. Results: Gut dysbiosis appears to influence estrogen metabolism, immune responses, and inflammation, while reproductive tract microbiota contribute to local immune modulation. AI and ML approaches, including Random Forest, Gradient Boosting, and logistic regression, have shown promise in predicting disease and identifying potential microbial biomarkers. Interventions such as probiotics, prebiotics, and fecal microbiota transplantation, coupled with multi-omics analyses, represent potential avenues for personalized treatment. Conclusion: Integrating microbiome profiling with AI-driven models may enable non-invasive diagnosis, improved disease classification, and precision therapeutic strategies. Further large-scale, multicenter studies are needed to validate these approaches and support their translation into clinical practice.
Borucińska et al. (Fri,) studied this question.