AUTHOR=Chen Yongjian , Li Yanru , Su Huiling , Lyu Guorong TITLE=Comparison of the value of the GI-RADS and ADNEX models in the diagnosis of adnexal tumors by junior physicians JOURNAL=Frontiers in Oncology VOLUME=Volume 14 - 2024 YEAR=2024 URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2024.1435636 DOI=10.3389/fonc.2024.1435636 ISSN=2234-943X ABSTRACT=Objective To compare the diagnostic effectiveness of the Gynaecologic Imaging Reporting and Data System (GI-RADS) and Neoplasias in the Adnexa (ADNEX) model for the diagnosis of benign and malignant ovarian tumours by junior physicians. Methods The sonographic data of 634 patients with ovarian tumours confirmed by pathology in our hospital over 4 years were analysed retrospectively by junior doctors. The diagnostic efficacy of the GI-RADS and ADNEX models was compared based on pathology. Results (1)Regarding the diagnostic efficacy of the GI-RADS and ADNEX models, the sensitivity was 90.15% and 84.85%, the specificity was 87.65% and 85.86%, the accuracy rates were 88.17% and 85.65%, and the Youden Indices were 0.778 and 0.707, respectively. The areas under the receiver operating characteristic (ROC) curves were 0.924 (95% CI: 0.900-0.943) and 0.933 (95% CI: 0.911-0.951), respectively. The GI-RADS classification was equivalent to that of the ADNEX model in the diagnosis of adnexal tumours (P>0.05). These findings were highly consistent with the pathological results (Kappa values were 0.684 and 0.691, respectively). ( 2) When differentiating between different pathological types of adnexal tumours, the ADNEX model had the best diagnostic value for distinguishing between benign tumours and stage II-IV ovarian cancer (AUC=0.990, 95% CI: 0.978-0.997). Conclusions (1) The diagnostic efficacy of the GI-RADS and ADNEX models in the diagnosis of benign and malignant ovarian tumours by junior physicians is excellent and comparable. (2) The ADNEX model shows good value for differentiating ovarian tumours of different pathological types by junior physicians.