J Clin Ultrasound. 2025 Oct 11. doi: 10.1002/jcu.70092. Online ahead of print.
ABSTRACT
OBJECTIVE: This study aimed to quantify the O-RADS classification using the ultrasound scoring method (USM) and explore its clinical application value.
METHODS: A retrospective analysis was conducted on 205 patients with adnexal tumors (unilateral or bilateral), corresponding to 244 tumor cases, admitted to our hospital from 2018 to 2023. This cohort included 100 patients with malignant tumors (122 malignant lesions, comprising 34 borderline tumors, and 6 malignant tumors of the fallopian tubes) and 105 patients with benign tumors (122 benign lesions). All cases were confirmed by preoperative ultrasound examination and postoperative pathology. A senior ultrasound physician performed the O-RADS classification, while another ultrasound physician applied the USM. The scores were manually assigned, and the sensitivity, specificity, and area under the curve (AUC) of the USM and O-RADS classification in the differential diagnosis of adnexal tumors were determined through ROC curve analysis. The Delong test was used in R to compare the diagnostic performance of the two diagnostic methods. The optimal range of scores between different categories of O-RADS was calculated using by ROC curve.
RESULTS: The sensitivity, specificity, and AUC of the USM (when the optimal cut-off value was ≥ 14 points) in the differential diagnosis of adnexal tumors were 0.951, 0.877, and 0.975 (95% CI: 0.960-0.990). The sensitivity, specificity, and AUC of O-RADS classification (when the optimal cut-off value was ≥ 4 categories) in the differential diagnosis of adnexal tumors were 0.984, 0.887, and 0.973 (95% CI: 0.955-0.992). The AUC comparison between O-RADS classification and USM revealed no statistically significant difference (p = 0.87, p > 0.05).
CONCLUSIONS: Both the USM and the O-RADS classification exhibit high diagnostic efficacy in the differential diagnosis of adnexal tumors. The quantification of O-RADS categories using the USM provides a more intuitive and convenient approach, suitable for ultrasound physicians at all levels. This method demonstrates strong generalizability and can effectively guide clinical diagnosis and treatment.
PMID:41074610 | DOI:10.1002/jcu.70092
