The Role of Artificial Intelligence in Revolutionizing Endometriosis Diagnosis and Clinical Pathology: A Commentary Review of New Biomarkers (2020–2026) Volume 66- Issue 3

Raouf Roshdy¹* and Reham A Rashed²

• ¹Egypt Health Foundation, Egypt
• ²Clinical Pathology Department, National Cancer Institute, Cairo University, Egypt

Received: August 07, 2026; Published: August 18, 2026

*Corresponding author: Raouf Roshdy, Egypt Health Foundation, Egypt

DOI: 10.26717/BJSTR.2026.66.010350

Abstract PDF

The diagnosis of endometriosis has historically been hampered by an average diagnostic delay of seven to ten years, owing largely to the field’s reliance on invasive surgical laparoscopy as the “gold standard.” The integration of artificial intelligence (AI) and machine learning (ML) into clinical pathology, however, is now catalyzing a paradigm shift toward non-invasive, precision diagnostics. This report synthesizes research published between 2020 and 2026, highlighting the emergence of high-accuracy diagnostic tools built on microRNA (miRNA) signatures, plasma protein panels, and genomic markers. Key findings include the validation of blood- and saliva- based miRNA signatures achieving area under the curve (AUC) values exceeding 0.95, rivaling the accuracy of surgical diagnosis. AI methodologies including Random Forest, eXtreme Gradient Boosting (XGBoost), and support vector machines (SVM) have identified novel biomarkers such as PNMA2, LRP1, and WISP2/CCN5. These advances promise not only to shorten diagnostic latency but also to enable molecular stratification, supporting more personalized treatment strategies. Challenges remain in clinical standardization and in validating certain protein-based models, yet the accumulated evidence suggests that AI-driven clinical pathology will soon augment and in some cases replace traditional surgical diagnosis for many patients.

Abbreviations: AI: Artificial Intelligence; ML: Machine Learning; miRNA: microRNA; AUC: Area Under the Curve; SVM: Support Vector Machines; ANN: Artificial Neural Networks; CNN: Convolutional Neural Networks; WGCNA: Weighted Gene Co-Expression Network Analysis

Endometriosis is a complex, multifactorial inflammatory disease characterized by the presence of endometrial-like tissue outside the uterine cavity. It affects millions of individuals worldwide and is a leading cause of chronic pelvic pain and infertility [1]. Despite this prevalence, the current diagnostic framework remains significantly flawed: laparoscopy an invasive surgical procedure carrying real risks and high costs remains the only definitive diagnostic method [2,3]. This reliance on surgery contributes to substantial diagnostic delays and prolonged patient suffering. Clinical pathology is now undergoing a transformation driven by artificial intelligence. By leveraging large multi-omics datasets (transcriptomics, proteomics, methylomics) alongside high-resolution imaging, AI-based tools are uncovering patterns “crimson clues,” in the words of one recent review within peripheral blood, saliva, and tissue that are invisible to the human eye [4,5]. These technologies aim to shift endometriosis diagnosis away from a purely symptom-based or surgical approach and toward one grounded in precision medicine and molecular stratification [3,5]. This report examines the AI methodologies, novel biomarkers, and clinical impacts described in the recent literature, with the goal of providing a comprehensive overview of the current state of the art.

The literature reviewed here reveals a diverse array of AI and machine learning methodologies, each tailored to a particular type of biological data. Broadly, these approaches fall into two categories: supervised learning for classification, and unsupervised learning for subtype identification.

Supervised Learning and Ensemble Methods

The most common application of AI in the sources reviewed is supervised learning used to differentiate patients with endometriosis from healthy controls.

• Ensemble classifiers: “GenomeForest” is an advanced ensemble machine learning classifier that uses chromosomal partitioning to analyze transcriptomic and methylomic data, achieving F1 scores as high as 0.968 [6].

• Tree-based models: Random Forest, XGBoost, and Ada- Boost are frequently favored for their robustness in handling high-dimensional omics data. Random Forest algorithms, for example, have been used to validate salivary miRNA signatures with 96.2% sensitivity [7] and to identify driver biomarkers from RNA-seq data [8].

• Neural networks: Artificial neural networks (ANN) and convolutional neural networks (CNN) are increasingly applied to complex diagnostic models and imaging data [8,9].

Integrated Discovery Pipelines

Many studies rely on hybrid pipelines that combine bioinformatics with machine learning for biomarker screening.

• WGCNA and ML: Weighted gene co-expression network analysis (WGCNA) is often paired with machine learning to identify lipid metabolism-related biomarkers and molecular subtypes [10].

• Multi-omics integration: Markov et al. describe the integration of ML and deep learning with imaging, molecular biomarkers, and histopathology to build multi-omics diagnostic pathways [11].

• Stacked models: Stacked machine learning models have been developed to distinguish endometriosis from adenomyosis using peripheral blood and coagulation data [12].

The synthesis of recent research points to several high-potential biomarker categories, with miRNA signatures currently showing the greatest diagnostic accuracy.

miRNA Signatures and Non-Invasive Diagnostics

MicroRNAs have emerged as the most promising candidates for non-invasive testing.

• Blood-based signatures: Bendifallah et al. [13] developed an 86-miRNA signature using algorithms such as Random Forest and XGBoost, achieving an AUC of 0.984 [13] a robust, reproducible diagnostic that could potentially replace laparoscopy [13].

• Saliva-based testing: A 109-miRNA saliva signature has been validated with 96.2% sensitivity and 95.1% specificity [7]. This approach offers a genuinely non-invasive alternative that could be administered in a primary care setting, meaningfully shortening the path to diagnosis [14].

• Early-stage detection: Specific miRNA panels (e.g., miR- 199a, miR-122) combined with CA-125 have demonstrated an AUC of 0.939 for detecting early-stage disease [15].

Proteomic and Immunologic Biomarkers

While miRNAs dominate the literature, protein- and immune- based markers offer critical insight into disease pathophysiology.

• Plasma protein panels: Schoeman et al. [16] identified a panel of 10 plasma protein biomarkers. Using a Random Forest model, they achieved an AUC of 0.997 for severe endometriosis, though accuracy for milder cases was considerably lower (Model 2 AUC = 0.729) [16].

• Immune screening: Non-invasive immune screening tests combined with ML (logistic regression) have identified markers predictive of disease, including IL-10 in Tc cells, reduced NK cytotoxic activity, and CXCL1 [2].

• Challenges in reproducibility: Notably, Fishman [17] found that some cytokine-based ML models performed no better than chance, underscoring the need for rigorous standardization in protein-based discovery [17].

Genomic and Transcriptomic Drivers

Machine learning has also proven instrumental in identifying the molecular drivers underlying the disease.

• Hub genes: Integrated identification methods have pointed to five key diagnostic genes FOS, EPHX1, DLGAP5, PCSK5, and ADAT1 [18]. A separate study identified LRP1 as a potential diagnostic and immunomodulatory target, showing a strong correlation with M2 macrophage infiltration [19].

• Novel biomarkers: AI-driven screening has highlighted PNMA2 (linked to malignant growth and anti-apoptosis) [20], WISP2/CCN5 (a potential early-screening marker) [21], and SPP1 (a modulator of macrophage polarization) [22].

• Cuproptosis-related genes: ML algorithms have identified genes such as GLS and PDHA1 as being associated with endometriosis progression and immune infiltration [23].

AI in Imaging and Digital Pathology

AI’s impact extends to the visual characterization of the disease as well.

• Automated recognition: Proof-of-concept studies have demonstrated the automatic visual recognition of endometriosis lesions during laparoscopy using AI, a step that could help standardize surgical assessment [24].

• Tissue segmentation: AI-based tissue segmentation and cell identification in histological sections for instance, identifying CD138+ cells are improving the precision of pathological diagnosis in infertility-associated conditions [25,26].

• Imaging integration: Deep learning models are increasingly being integrated with ultrasound and MRI data to improve detection of deep infiltrating endometriosis and ovarian endometriomas [27,28].

The integration of AI into clinical pathology is reshaping the diagnostic landscape in several ways:

• Reducing diagnostic latency: By enabling high-accuracy blood or saliva tests, AI tools can identify candidate biomarker genes and signatures that shorten the years-long wait many patients face for a diagnosis [6,21].

• Molecular stratification: AI enables identification of disease subtypes, moving the field away from a one-size-fits-all approach and toward models informed by molecular, epidemiologic, and cell-specific data [5].

• Precision medicine: The development of diagnostic models that predict treatment response for example, PNMA2 as a therapeutic target is helping guide more personalized therapeutic strategies [3,20].

• Augmenting the gold standard: While some sources suggest AI could ultimately replace laparoscopy [13,15], others view it instead as a powerful complement that improves the efficiency and accuracy of existing histopathological and surgical workflows [18,29,30].

Despite rapid progress in the field, several limitations must be acknowledged:

• Validation gaps: Many of the biomarkers identified so far remain in the discovery or interim-validation stage [4,14]. Large-scale, multi-center prospective trials will be necessary to confirm their clinical utility.

• Data heterogeneity: AI model performance varies considerably depending on the dataset and disease stage, as reflected in the gap between severe- and mild-endometriosis detection in protein-based models [16].

• Interpretability and fallibility: The “black box” nature of some ML models remains a challenge in diagnostic pathology, where clinical decisions demand transparent reasoning [31,32].

• Standardization: Standardized protocols for biomarker collection and processing are urgently needed to ensure that AI-driven findings remain reproducible across different clinical settings [4].

Future research in this area should focus on:

• Longitudinal studies: Understanding how AI-identified biomarkers change over the course of the disease and in response to treatment.

• Integration with clinical data: Combining molecular signatures with symptom clustering and imaging data to build a holistic diagnostic “score” [33].

• Cost-effectiveness analysis: Evaluating the economic impact of replacing laparoscopy with AI-driven non-invasive tests.

• Pathologist training: Developing educational frameworks that help pathologists integrate AI assistance into daily practice [34].

Artificial intelligence is no longer a peripheral technology in the study of endometriosis it is the engine driving the next generation of clinical pathology. Through the identification of highly accurate miRNA signatures, novel genomic drivers such as LRP1 and PNMA2, and the automation of digital pathology, AI is supplying the tools needed to overcome the field’s historical challenges of delayed diagnosis and invasive surgery. Rigorous clinical validation remains the final hurdle, but the evidence accumulated between 2020 and 2026 strongly suggests that AI-powered non-invasive diagnostics will soon become a cornerstone of endometriosis management, ushering in an era of precision medicine for millions of patients.

1. A Ashish, K Kusum, S Rai, R Singh (2020) Endometriosis a brief review: evaluation of crucial risk factors and current treatment regimes. International Journal of Advances in Medicine 7(12): 1896-1905.
2. N Attia, M Franzoi, M P McHale, A R Gargiulo, M S Abrão, et al. (2023) Identification of immunologic predictive biomarkers of endometriosis using a non-invasive immune screening test in association with machine learning modeling. Fertility and Sterility 120(4): e226-e227.
3. L N Onah, M N Ezenwaeze, C N Nwankwo (2026) Endometriosis: From delayed diagnosis to precision medicine. International Journal of Science and Research Archive 18(1): 306-314.
4. T Gibbons, N Rahmioglu, K T Zondervan, C M Becker (2024) Crimson clues: advancing endometriosis detection and management with novel blood biomarkers. Fertility and Sterility 121(2): 145-163.
5. C R Nezhat, Tomiko T Oskotsky, Joshua F R, Susan J F, Angie T, et al. (2025) Real World Perspectives on Endometriosis Disease Phenotyping Through Surgery, Omics, Health Data, and Artificial Intelligence. npj Women’s Health 3(1): 8.
6. S Akter, D Xu, S C Nagel, J J Bromfield, K E Pelch, et al. (2020) GenomeForest: An Ensemble Machine Learning Classifier for Endometriosis, p. 33-42.
7. S Bendifallah, Y Dabi, S Suisse, L Delbos, A Spiers, et al. (2023) Validation of a Salivary miRNA Signature of Endometriosis — Interim Data. NEJM Evidence 2(7).
8. Zhen G, T Chu, J She, S Chen, P Wang, et al. (2026) Exploration of driver biomarkers and immune microenvironment in patients with endometriosis: Evidence from RNA-seq and machine learning. Journal of Reproductive Immunology 175: 104872.
9. H Malvezzi, E B Marengo, S Podgaec, C de A Piccinato (2020) Endometriosis: current challenges in modeling a multifactorial disease of unknown etiology. Journal of Translational Medicine 18(1): 1-21.
10. Y Guo, Y Hou, J Wu, N Lou, D Yang, et al. (2025) Identification and Subtype Analysis of Lipid Metabolism‐Related Diagnostic Biomarkers for Endometriosis Based on WGCNA and Machine Learning. American Journal of Reproductive Immunology, p. 94.
11. D Markov, J Gurung, U Khalid, K Bechev, V Aleksiev, et al. (2026) Emerging Pathways to Non-Invasive Diagnosis in Endometriosis: Integrating Machine Learning, Deep Learning and Multi-Omics Biomarkers. Diagnostics 16(12): 1823.
12. W Wang, W Zeng, S Yang (2024) A stacked machine learning-based classification model for endometriosis and adenomyosis: a retrospective cohort study utilizing peripheral blood and coagulation markers. Frontiers in Digital Health 6: 1463419.
13. S Bendifallah, Y Dabi, S Suisse, L Jornea, D Bouteiller, et al. (2022) MicroRNome analysis generates a blood-based signature for endometriosis. Scientific Reports 12(1).
14. Gaia-Oltean AI, Boitor D, Laura-Ancuta P, Geanina G, Teodora T, et al. (2025) Non-Invasive Methods for Early Diagnosis of Endometriosis-A Comprehensive Narrative Literature Review. Healthcare (Basel, Switzerland) 13(24): 3276.
15. M K Gałuszka, K K Figurny, P P Jagodziński, A Pławski (2025) Potential biomarkers for early detection of endometriosis: current state of art (what we know so far). Journal of Applied Genetics.
16. E M Schoeman, S Bringans, K Peters, T Casey, C Andronis, et al. (2024) Identification of plasma protein biomarkers for endometriosis and the development of statistical models for disease diagnosis. Human Reproduction 40(2): 270-279.
17. D Fishman (2022) Developing a data analysis pipeline for automated protein profiling in immunology.
18. H Zhang, H Zhang, H Yang, A N Shuid, D Sandai, et al. (2023) Machine learning-based integrated identification of predictive combined diagnostic biomarkers for endometriosis. Frontiers in Genetics 14: 1290036.
19. Xie Liu (2026) LRP1 as a potential diagnostic and immunomodulatory target in endometriosis: evidence from multi-omics and single-cell analyses. Frontiers in Immunology p: 17.
20. Mengjun Zhang, Zidi Zhang, Jialin Wang, Haodi Yue, Xueling Lou, et al. (2025) Unveiling the Role of PNMA2 in Endometriosis: From Proliferation and Apoptosis to Immunomodulation. Journal of Cellular and Molecular Medicine 29(9): e70576.
21. Sheng Dou, Shaohua Ling, Weihua Nong, Bixiao Wei, Yuehua Huang, et al. (2025) WISP2/CCN5 revealed as a potential diagnostic biomarker for endometriosis based on machine learning and single-cell transcriptomic analysis. Functional & Integrative Genomics 25(1): 131.
22. Sijia Zhang, Yi Yang, Chen Zhang, Xiaoyao Che, Fawei Li, et al. (2025) SPP1 as a key modulator of M2 macrophage polarization promotes endometriosis progression via activation of the FAK/PI3K/AKT pathway: A bioinformatics and experimental study. International Immunopharmacology 166: 115563.
23. Y Lu, J Wang, G Sun, Z Fang, Z Xing, et al. (2023) Machine learning algorithms for a novel cuproptosis-related gene signature of diagnostic and immune infiltration in endometriosis. Scientific Reports 13(1): 21603.
24. A Netter, S Noorzadeh, F Duchateau, H Abrao, J Desternes, et al. (2025) Initial results in the automatic visual recognition of endometriosis lesions by artificial intelligence during laparoscopy: a proof-of-concept study. Journal of Minimally Invasive Gynecology 32(12): 1118-1125.
25. S E Korman, G Vissers, M A J Gorris, K Verrijp (2024) Artificial intelligence-based tissue segmentation and cell identification in multiplex-stained histological endometriosis sections. Human Reproduction 40(3): 450-460.
26. S Lee, R K Arffman, E K Komsi, O Lindgren, J A Kemppainen, et al. (2024) AI-algorithm training and validation for identification of endometrial CD138+ cells in infertility-associated conditions; polycystic ovary syndrome (PCOS) and recurrent implantation failure (RIF). Journal of Pathology Informatics 15: 100380-100380.
27. N Jiang, H Xie, J Lin, Y Wang, Y Yin, et al. (2022) Diagnosis and Nursing Intervention of Gynecological Ovarian Endometriosis with Magnetic Resonance Imaging under Artificial Intelligence Algorithm. Computational Intelligence and Neuroscience, p. 1-10.
28. J Avery, A Deslandes, S M Freger, M Leonardi, G Lo, et al. (2023) Non-invasive Diagnostic Imaging for Endometriosis Part 1: A Systematic review of recent developments in Ultrasound. Combination Imaging and Artificial Intelligence. Fertility and Sterility 121(2): 164-188.
29. Y Yu, W Hing W, Y Hu, Y Shen, X Xu, et al. (2025) Translation of miRNA blood-based discovery to molecular testing for clinical diagnosis of endometriosis. medRxiv.
30. M J Iqbal, Z Javed, H Sadia, I A Qureshi, A Irshad, et al. (2021) Clinical applications of artificial intelligence and machine learning in cancer diagnosis: looking into the future. Cancer Cell International 21(1): 270-270.
31. P Mishra, A Panda, M Mahapatra, P Dakshinakabat, A Mohanty, et al. (2024) Application and fallibility of Artificial Intelligence and machine learning in Diagnostic Pathology. Bangladesh Journal of Medical Science 23(10): S32-S37.
32. J H Harrison, John R Gilbertson, M G Hanna, N H Olson, J N Seheult, et al. (2021) Introduction to Artificial Intelligence and Machine Learning for Pathology. Archives of Pathology & Laboratory Medicine 145(10): 1228-1254.
33. S M Mouazen, S Khan, N S Farag (2025) Machine learning in the early detection of endometriosis: a literature review on symptom clustering and imaging integration. Precision and Future Medicine 9(3): 117-128.
34. Z Ahmad, S Rahim, M Zubair, J A Ghafar (2021) Artificial intelligence (AI) in medicine, current applications and future role with special emphasis on its potential and promise in pathology: present and future impact, obstacles including costs and acceptance among pathologists, practical and philosophical considerations. A comprehensive review. Diagnostic Pathology 16(1): 24-24.