O-RADS Application Of Ultrasound Classification For Ultrasound Assessment Of Benign And Malignant Ovarian Adnexal Masses

Objective: The application value of O-RADS classification in the ultrasound assessment of benign and malignant ovarian adnexal masses was discussed,and the diagnostic efficacy of improved O-RADS was obtained by secondary classification of O-RADS class 4 and combined with ADNEX model,and an improved method was sought to improve the diagnostic efficacy of benign and malignant ovarian adnexal masses.Methods: A retrospective analysis was performed on 182 patients who underwent surgical treatment or needle biopsy due to ovarian adnexal masses in our hospital from September 2018 to April 2022.The incidence of O-RADS class 2～5 ovarian adnexal masses was calculated by traditional O-RADS classification according to the ultrasound image characteristics such as tumor size,morphology,composition,papillary bulge and color blood flow score,and univariate analysis of nine suspected malignant ultrasound signs was performed.At the same time,two sonographers of different seniority performed O-RADS classification on 122 randomly selected cases,and evaluated the consistency of the O-RADS classification results of the two.The cases classified as O-RADS category 4 were then classified in the form of scoring and classified by ADNEX model,and the pathological results were taken as the gold standard,and the traditional O-RADS classification system,the modified O-RADS secondary classification,the ADNEX model,the modified O-RADS secondary classification and the ADNEX model were combined,and the subject working characteristics(ROC)curves of the traditional O-RADS classification and the improved three methods were drawn respectively,and the area under the curve was obtained(AUC),calculated the Jordon index,obtained the best cut-off value,and compared the above four methods to diagnose the sensitivity,specificity,positive predictive value,negative predictive value,accuracy,misdiagnosis rate,missed diagnosis rate of ovarian adnexal mass,P<0.05 is statistically significant.Results: Postoperative pathological results of 182 patients: 50 cases of malignant lesions and 132 cases of benign lesions.Among them,there were 68 cases of O-RADS category 2,and 1 case was malignant(1/68);There were 28 cases in O-RADS category 3,and 1 case was malignant(1/28);There were 60 cases of O-RADS category 4,and 26 cases were malignant(26/60);There were26 cases in O-RADS category 5,and 22 cases were malignant(22/26).The malignant rates of O-RADS 2,3,4 and 5 were 1.5%,3.6%,43.3% and 84.6%,respectively.The diagnostic consistency assessment of O-RADS classification by different senior sonographers for adnexal ovarian tumors kappa coefficient was 0.826,P<0.001.The working characteristics(ROC)curve of O-RADS 2～5classified subjects was plotted,and the best cut-off value for identifying benign and malignant ovarian adnexal masses was O-RADS class 4.O-RADS 4 and O-RADS 5 were classified as malignant group,and O-RADS 2～3 were classified as benign group,and the sensitivity,specificity,positive predictive value,negative predictive value,accuracy,misdiagnosis rate and missed diagnosis rate of ovarian adnexal mass were 96.0%,71.2%,55.8%,97.9%,78.0%,27.7% and 4.0%,respectively.Among them,60 cases classified as O-RADS 4 were classified twice,namely O-RADS 4a and O-RADS 4b,O-RADS 4b～5 were classified as malignant group,and O-RADS 2～4a was classified as benign group.There were 29 cases in O-RADS 4a,21 were benign and 8 were malignant,with a malignant rate of 27.5%.There were 31 cases in O-RADS 4b,13 were benign and 18 were malignant,with a malignant rate of58.1%.The ADNEX model was also evaluated for 60 cases classified as O-RADS category 4,with 10% as the cut-off value,and the malignant risk ≥10% classified as malignant and < 10% as benign.Among them,there were 23 cases of ADNEX model malignant risk < 10% in O-RADS 4 cases,of which 19 cases were benign and 4 cases were malignant,with a malignant rate of 17.3%.The malignancy risk of ADNEX model in O-RADS class 4 cases was ≥ 10%,a total of 37 cases,of which 15 cases were benign and 22 were malignant,with a malignant rate of 59.4%.O-RADS 4 quadratic classification combined with ADNEX model,and 10% of the risk ≥ in accordance with O-RADS 4b and ADNEX models were classified as malignant group,and vice versa,a total of 37 cases were in the benign group,of which 28 cases were benign and 9 cases were malignant,with a malignant rate of 24.3%.There were 23 cases in the malignant group,of which 6 were benign and 17 were malignant,with a malignant rate of73.9%.The sensitivity,specificity,positive predictive value,negative predictive value,accuracy,misdiagnosis rate and missed diagnosis rate of O-RADS 4quadratic classification and O-RADS 4 type 4 ADNEX model combined with O-RADS 4 ADNEX model were 80.0%,88.0%,78.0%,87.1%,85.6%,92.4%,70.1%,69.8%,79.5%,92.0%,respectively.94.96%,91.7%,85.1%,86.2%,88.4%,12.8%,14.3% and 7.5%,the AUCs of the traditional O-RADS and the improved three methods were 0.836,0.836,0.868 and 0.852,and the Jordon index was 0.672,0.671,0.736 and 0.704,respectively,but the diagnostic performance of the above four methods was not statistically significant.Conclusion: The classification of O-RADS was strongly consistent among different senior physicians,and the classification of O-RADS could better distinguish and diagnose benign and malignant masses in the adnexa.O-RADS4 quadratic classification in the form of scores,O-RADS 4 ADNEX model evaluation,O-RADS 4 quadratic classification combined with O-RADS 4ADNEX model have a certain impact on the malignant rate of O-RADS 4subtypes,and have a certain help for malignant risk stratification,but the traditional O-RADS classification and the improved O-RADS 4 quadratic classification,O-RADS 4 ADNEX model and O-RADS 4 quadratic classification combined with O-RADS The four methods of the four types of ADNEX model have comparable diagnostic performance in predicting adnexal malignant tumors of the ovarians.