Correction: Predictive models of recurrent implantation failure in patients receiving ART treatment based on clinical features and routine laboratory data

The objective was to construct a model for predicting the probability of recurrent implantation failure (RIF) after assisted reproductive technology (ART) treatment based on the clinical characteristics and routine laboratory test data of infertile patients. A model was developed to predict RIF. The model showed high calibration in external validation, helped to identify risk factors for RIF, and improved the efficacy of ART therapy. Research on the influencing factors of RIF has focused mainly on embryonic factors, endometrial receptivity, and immune factors. However, there are many kinds of examinations regarding these aspects, and comprehensive screening is difficult because of the limited time and economic conditions. Therefore, we should try our best to analyse the results of routine infertility screenings to make general predictions regarding the occurrence of RIF. A retrospective study was conducted with 5212 patients at the Reproductive Center of the First Affiliated Hospital of USTC from January 2018 to June 2022. This study included 462 patients in the RIF group and 4750 patients in the control group. The patients’ basic characteristics, clinical treatment data, and laboratory test indices were compared. Logistic regression was used to analyse RIF-related risk factors, and the prediction model was evaluated by receiver operating characteristic (ROC) curves and the corresponding areas under the curve (AUCs). Further analysis of the influencing factors of live births in the first cycle of subsequent assisted reproduction treatment in RIF patients was performed, including the live birth subgroup (n = 116) and the no live birth subgroup (n = 200). (1) An increased duration of infertility (1.978; 95