A nomogram prediction model for ovarian endometrioma in patients with endometriosis: a retrospective study based on clinical indicators

Abstract Background Among the subtypes of endometriosis (EMs), ovarian endometrioma (OE) causes the most direct and progressive damage to ovarian reserve function. This study aimed to identify independent risk factors for OE in patients with EMs, and to develop and validate a clinical prediction model to support early identification and stratified intervention in high-risk patients. Methods A retrospective study was conducted on 342 patients with pathologically confirmed EMs admitted to the First Affiliated Hospital of Guangxi University of Chinese Medicine from January 2021 to December 2025. Among them, 103 patients had OE (OE group) and 239 had other types of EMs (non-OE group). Patients were randomly divided into training and validation sets at a 7:3 ratio. Least absolute shrinkage and selection operator (LASSO) regression was used for preliminary feature selection, followed by univariate and multivariate logistic regression analyses to identify independent predictors of OE. A nomogram prediction model was subsequently constructed. Model discrimination was assessed using the area under the receiver operating characteristic curve (AUC), calibration was evaluated using calibration curves comparing predicted probabilities with observed outcomes, and clinical utility was assessed using decision curve analysis (DCA). Results Multivariate logistic regression analysis revealed that history of dysmenorrhea (OR = 60.44, 95%CI: 20.95–213.90), infertility (OR = 13.10, 95%CI: 4.85–40.21), elevated fibrinogen levels (OR = 1.84, 95%CI: 1.13–3.01), and decreased lymphocyte count (OR = 0.42, 95%CI: 0.16–0.99) were independent predictors of OE in patients with EMs (all P < 0.05). The nomogram model constructed based on these factors demonstrated excellent discrimination in both the training and validation sets (AUC = 0.945 and 0.948, respectively), good calibration, and positive clinical net benefit across a wide range of threshold probabilities as shown by DCA. Conclusion This study successfully developed and validated a nomogram model integrating clinical symptoms (history of dysmenorrhea, infertility) and routine laboratory indicators (fibrinogen, lymphocyte count), which can effectively predict the individualized risk of OE in patients with EMs. This model provides a simple and practical quantitative tool for clinicians to identify high-risk OE populations, implement risk-stratified management, and formulate individualized intervention strategies.