Construction of a predictive model for malignant risk of ovarian-adnexal tumors:based on ultrasound,contrast-enhanced ultrasound,clinical,and serological indicators
FAN Linxia
CAO Huimei
ZHOU Huili
Abstract：Objective A nomogram model incorporating ultrasound,contrast-enhanced ultrasound,clinical,and serological indicators was constructed to evaluate its predictive value for the malignant risk of ovarian-adnexal tumors.Methods Retrospective analysis of 220 patients diagnosed with ovarian-adnexal tumors at the First Affiliated Hospital of Xinjiang Medical University from January 2022 to January 2025.Among these,161 cases were benign and 59 were malignant.They were randomly divided into a training set(n=154)and a validation set(n=66)at a ratio of 7:3.The characteristics of ultrasound,contrast-enhanced ultrasound,clinical,and serological indicators between the training and validation datasets were compared.Subsequently,independent influencing factors were identified through Least Absolute Shrinkage and Selection Operator(LASSO)-logistic regression and construct a regression model and the receiver operating characteristic curve,calibration curve,and decision curve were plotted to evaluate the model's diagnostic performance.Results The training and validation sets showed significant differences in CA125 and tumor type(P<0.05),with no significant differences in other baseline characteristics(P>0.05).The LASSO-logistic regression analysis revealed that menopausal status,HE4,nipple,morphology,blood flow,enhancement level,and distribution characteristics are independent risk factors for diagnosing the benign or malignant nature of ovarian-adnexal tumors(P<0.05).A regression model constructed based on the above seven indicators.The model achieved an area under the curve(AUC)of 0.961(95%CI:0.931-0.991)on the training set and 0.973(95%CI:0.939-1.000)on the validation set,demonstrating high accuracy.Calibration curve analysis indicated good model calibration.Clinical decision curve analysis demonstrated that the model exhibits favorable net benefit in clinical applications.Conclusion A nomogram model incorporating ultrasound,contrast-enhanced ultrasound,clinical,and serological indicators serves as a non-invasive quantitative tool for predicting the malignant risk of ovarian-adnexal tumors,providing reliable diagnostic evidence for precision clinical diagnosis and treatment.
Keywords：ovarian-adnexal tumorscontrast-enhanced ultrasoundLASSO-logistic regressionnomogram
Publication Date：2026-01-20
Online Publishing Date：2026-02-04(First online date of this platform, not the publication date of the document)
Pages：8( 94-101 )