Developing and validating a prediction model of live birth in patients with moderate-to-severe intrauterine adhesions: a new approach with endometrial morphology measurement by 3D transvaginal ultrasound

Abstract

Background There is no reliable method to predict the live birth rate among patients with moderate-to-severe intrauterine adhesions (IUA) after second-look hysteroscopy. Therefore, we aimed to construct a practical prediction model mainly based on the features of 3D transvaginal ultrasound (3D-TVUS). and other clinical characteristics. Methods From January 2018 to February 2020, a total of 870 IUA patients with fertility requirements were retrospectively enrolled based on the same method. First, the predictors were screened by logistic regression analysis. A nomogram was constructed based on the screened predictive factors in the derivation cohort. Next, receiver operating characteristic (ROC), calibration curve, and decision curve analysis (DCA) were used to assess the predictive accuracy and discriminability of the model. Finally, correlation analysis was performed to analyze the correlation between the results of 3D-TVUS and second-look hysteroscopy. Results A total of 558 (64.14%) participants had live births. Age, endometrial thickness, assisted reproductive technology, a homogeneous endometrial echo, a lower segment of scar contraction, and upper segmentation of the endometrial absence were included in the model. The predictive model showed good predictive performance in the derivation cohort (area under the curve, 0.837) and validation cohort (0.857). DCA demonstrated its clinical utility. A homogeneous endometrial echo was related to no segmentation of scar contraction (r=0.219; P<0.001) or no segmentation of the endometrial absence (r=0.226; P<0.001). Thicker endometrium was associated with no segmentation of the endometrial absence (r=−0.145; P=0.007). Conclusions The proposed method can effectively predict live birth. 3D-TVUS should be an important means for evaluating the endometrium of moderate-to-severe patients with IUA preparing for pregnancy after operation.

Artificial intelligence-driven prognostic system for conception prediction and management in intrauterine adhesions following hysteroscopic adhesiolysis: a diagnostic study using hysteroscopic images

“…However, these studies also acknowledged the limitations of their research, such as the lack of consensus and consistency on the clinical indicators, such as the IUA scoring system. Sun et al (2024) proposed a logistic prediction model based on various evaluation parameters of 3D transvaginal ultrasound, which also showed promising results, but suffered from the same drawbacks of subjectivity and variability. The subjective differences in judgment and evaluation among different doctors are important factors that hinder the wide application of these models in clinical practice.…”

“…In our previous study, we used machine learning algorithms such as decision trees and XGboost to predict the pregnancy outcomes of IUA patients based on clinical data ( Zhu et al, 2022 ; Li et al, 2023a ). Recently, some studies have also used 3D ultrasound parameters to predict the outcomes ( Sun et al, 2024 ). These studies have achieved good prediction results, but they have some limitations.…”

“…The data they used are based on the clinical scores and manual measurements by doctors, which may have subjectivity and heterogeneity issues. Moreover, the lack of external validation makes these algorithms challenging to apply in clinical practice ( Zhu et al, 2022 ; Sun et al, 2024 ).…”