The application of super-resolution ultrasound radiomics models in predicting the failure of conservative treatment for ectopic pregnancy

Abstract

Background: Conservative treatment remains a viable option for selected patients with ectopic pregnancy (EP), but failure may lead to rupture and serious complications. Currently, serum β-hCG is the main predictor for treatment outcomes, yet its accuracy is limited. This study aimed to develop and validate a predictive model that integrates radiomic features derived from super-resolution (SR) ultrasound images with clinical biomarkers to improve risk stratification.

Methods: A total of 228 patients with EP receiving conservative treatment were retrospectively included, with 169 classified as treatment success and 59 as failure. SR images were generated using a deep learning-based generative adversarial network (GAN). Radiomic features were extracted from both normal-resolution (NR) and SR ultrasound images. Features with intraclass correlation coefficient (ICC) ≥ 0.75 were retained after intra- and inter-observer evaluation. Feature selection involved statistical testing and Least Absolute Shrinkage and Selection Operator (LASSO) regression. Random forest algorithms were used to construct NR and SR models. A clinical model based on serum β-hCG was also developed. The Clin-SR model was constructed by fusing SR radiomics with β-hCG values. Model performance was evaluated using area under the curve (AUC), calibration, and decision curve analysis (DCA). An independent temporal validation cohort (n = 40; 20 failures, 20 successes) was used to validation of the nomogram derived from the Clin-SR model.

Results: The SR model significantly outperformed the NR model in the test cohort (AUC: 0.791 ± 0.015 vs. 0.629 ± 0.083). In a representative iteration, the Clin-SR fusion model achieved an AUC of 0.870 ± 0.015, with good calibration and net clinical benefit, suggesting reliable performance in predicting conservative treatment failure. In the independent validation cohort, the nomogram demonstrated good generalizability with an AUC of 0.808 and consistent calibration across risk thresholds. Key contributing radiomic features included Gray Level Variance and Voxel Volume, reflecting lesion heterogeneity and size.

Conclusions: The Clin-SR model, which integrates deep learning-enhanced SR ultrasound radiomics with serum β-hCG, offers a robust and non-invasive tool for predicting conservative treatment failure in ectopic pregnancy. This multimodal approach enhances early risk stratification and supports personalized clinical decision-making, potentially reducing overtreatment and emergency interventions.

Keywords: Conservative treatment failure prediction; Ectopic pregnancy; Machine learning; Radiomics; Super-resolution ultrasound.

Fig. 1

Schematic of the GAN-based super-resolution (SR) reconstruction pipeline. NR ultrasound images are enhanced using a GAN to produce SR outputs Abbreviations: NR, Normal-resolution; SR, Super-resolution; GAN, Generative Adversarial Network

Fig. 2

Comparison of model performance between the NR and SR radiomic models. (A–B) ROC curves, (C–D) calibration curves, and (E–F) decision curve analyses for the training and test cohorts, respectively Abbreviations: SR: Super-resolution; NR: Normal-resolution; AUC: Area under the curve; DCA: Decision curve analysis. ROC: receiver operating characteristic

Fig. 3

ROC curves of the clinical model, SR model, and Clin-SR fusion model in the training (A) and test (B) cohorts. The Clin-SR model demonstrated the highest AUC in both cohorts, indicating superior discriminatory performance Abbreviations: SR, Super-resolution; Clin-SR, Clinical–Super-resolution fusion model; AUC, Area under the curve. ROC: receiver operating characteristic

Fig. 4

Calibration curves of the clinical model, SR model, and Clin-SR fusion model in the training (A) and test (B) cohorts. The plots show the agreement between predicted probabilities and actual outcomes. The Clin-SR model demonstrated favorable calibration in both cohorts

Fig. 5

Decision curve analysis (DCA) of the clinical model, SR model, and Clin-SR fusion model in the training (A) and test (B) cohorts. The Clin-SR model showed superior net clinical benefit across a wide range of threshold probabilities, supporting its potential value in clinical decision-making

Fig. 6

Nomogram integrating radiomic features and serum β-hCG for predicting conservative treatment failure in ectopic pregnancy

Fig. 7

ROC curve, calibration plot, and decision curve analysis of the nomogram model in the independent temporal validation cohort. The nomogram demonstrated good discrimination (AUC = 0.807), calibration, and net clinical benefit in the temporal cohort