The value of a nomogram model combining ultrasonic features and blood inflammatory indicators in differentiating benign and malignant thyroid nodules

This study aimed to investigate the value of clinical characteristics, conventional ultrasound features, and peripheral blood inflammatory markers in the preoperative differentiation of benign and malignant thyroid nodules and develop a nomogram-based predictive model. A total of 355 patients with thyroid nodules confirmed by surgical pathology or fine-needle aspiration cytology were retrospectively analyzed. Clinical data (gender and age), ultrasound characteristics, and peripheral blood inflammatory markers were collected. Univariate and multivariate logistic regression analyses were performed to identify independent predictors, which were used to construct a nomogram. The performance of the model was evaluated using receiver operating characteristic curves, calibration curves, and decision curve analysis. Multivariate logistic regression analysis showed that sex (male, odds ratio [OR] = 3.41), nodule shape (taller-than-wide, OR = 16.62), calcification (OR = 5.90), irregular margin (OR = 5.46), hypoechogenicity (OR = 4.75), hypervascularity on color doppler flow imaging (OR = 2.53), and neutrophil-to-lymphocyte ratio (OR = 4.23) were independent predictors of malignant thyroid nodules (all P < .05).The combined model (clinical data + ultrasound features + inflammatory marker) achieved an area under the curve (AUC) of 0.850, which was higher than that of the clinical-data-only model (AUC = 0.546) and the clinical data plus ultrasound model (AUC = 0.831). The nomogram also demonstrated better agreement and greater net benefit in calibration and decision curve analysis. A nomogram integrating routine clinical information, conventional ultrasound characteristics, and neutrophil-to-lymphocyte ratio offers improved preoperative risk stratification for thyroid nodules. This tool demonstrates favorable discriminatory ability and potential clinical utility, and may complement existing ultrasound-based risk assessment systems.