Objective:

To develop and validate a simple yet effective nomogram based on clinical and ultrasound factors for predicting the malignant risk of ovarian-adnexal masses, emphasizing its diagnostic performance and clinical utility.

Key Findings:

• Eight independent risk factors for malignancy were identified: menopausal status, internal echogenicity, internal septations, blood flow signals, CA125, HE4, platelet count, and platelet-to-hemoglobin ratio, highlighting their individual contributions.
• The nomogram demonstrated excellent discriminatory performance with an AUC of 0.97, indicating its reliability.
• Machine learning models also showed high AUCs, with Logistic Regression at 0.95 and Random Forest at 0.94 in the validation set, showcasing the robustness of the findings.
• Calibration curves indicated good consistency between predicted probabilities and actual observations, reinforcing the model's accuracy.
• Decision curve analysis confirmed significant clinical net benefits across a range of probability thresholds, underlining its practical value.

Interpretation:

The nomogram developed for predicting the malignant risk of ovarian-adnexal masses shows promising diagnostic performance and can assist in clinical differentiation of benign and malignant cases, potentially improving patient outcomes.

Limitations:

• The model is framed as a hypothesis-generating tool and requires further external validation studies to confirm its applicability in diverse clinical settings.

Conclusion:

The nomogram can provide a quantitative basis for formulating individualized treatment plans for patients with ovarian-adnexal masses.