摘要
目的:近些年附件肿块的发病率呈上升趋势,合理的诊断与治疗方案有助于提高患者的生活质量以及避免一些不必要的手术.超声检查是诊断附件肿块的良好方法.因此,本研究的目的是评估经验丰富、经验中等及缺乏经验的 3 组医师对于附件肿块特征的识别及O-RADS分类之间的一致性.方法:回顾性纳入了一家综合性医院 2021 年 9 月—2022 年 2 月 1234 名经病理诊断为附件肿块的患者.所有观察人员都被分配了附件肿块(n=100)的相同图像并进行离线评估,回答以下 3 个问题:①能否看到积液(盆腔或者腹腔)?②能否看到网膜增厚?③O-RADS分类为几类?使用Kappa值评估经验不足组、经验中等组以及经验丰富组观察者间一致性.结果:1234 个附件肿块中,1026 个良性、139 个恶性、69 个交界性肿瘤.受试者工作特征曲线下面积为 0.916(95%CI为 0.894～0.938).预测恶性肿瘤的最佳临界值为O-RADS分类>3 类.在是否观察到积液的问题上,经验丰富组和经验中等组均表现出中等的一致性(Kappa值分别为 0.574 和 0.521);经验不足组表现出较好的一致性(Kappa值为 0.663).在是否观察到网膜增厚的问题上,经验丰富组、经验不足组表现出中等的一致性(Kappa值分别为 0.402 和 0.427);经验中等组表现出较好的一致性(Kappa值为 0.795).在O-RADS分类问题上,经验丰富组、经验中等组、经验不足组均表现出中等的一致性(Kappa值分别为 0.482、0.447 和 0.493).结论:本研究发现 3 组医师在O-RADS分类有相似的一致性,说明O-RADS超声风险分层与管理系统的共识指南提供的附件病风险分层在具有不同经验的超声科医生之间具有广泛的应用空间和较好的适用性.
Abstract
Objective:To evaluate the consistency of identification of adnexal mass characteristics and ovarian-adnexal re-porting and data system(O-RADS)classification among three groups of physicians with varying levels of experience.Methods:A retrospective study was conducted from September 2021 to February 2022,which included 1 234 patients diagnosed with adnexal masses.All observers were assigned the same set of images(n=100)of adnexal masses for offline evaluation and were asked three questions:①Can you see any fluid(pelvic or abdominal)?②Can you see any thickening of the peritoneum?③What is your O-RADS classification?Kappa values were used to evaluate the inter-observer agreement among observers.Re-sults:Among the 1234 ovarian masses,1026 were benign,139 were malignant,and 69 were borderline tumors.The area un-der the working characteristic curve of the subjects was 0.916(95%CI:0.894～0.938).The optimal critical value for predicting malignant tumors was O-RADS classification>3.In terms of whether fluid accumulation was observed,both the experienced and moderate groups showed moderate consistency(Kappa values of 0.574 and 0.521,respectively).Inexperienced observers ex-hibited good consistency(Kappa value of 0.663).Experienced and inexperienced observers showed moderate inter observer con-sistency(Kappa values of 0.402 and 0.427,respectively)in the observation of omental thickening.The moderate experience group showed good consistency(Kappa value of 0.795).Experienced and inexperienced observers showed moderate inter observ-er consistency(Kappa values of 0.402 and 0.427,respectively)in the observation of omental thickening.The moderate experi-ence group showed good consistency(Kappa value of 0.795).On the O-RADS classification issue,observers from experienced,moderate,and inexperienced groups all showed moderate consistency(Kappa values of 0.482,0.447,and 0.493,respectively).Conclusion:Our study finds that the three groups of physicians have similar consistency in O-RADS classification,indicating that the accessory disease risk stratification provided by the consensus guidelines for O-RADS ultrasound risk stratification and management system has a wide range of application space and good applicability among radiologists with different experiences.
基金项目
辽宁省"兴辽英才计划"医学名家项目(YXMJ-LJ-10)
辽宁省科技计划联合计划(重点研发计划)()
维普
万方数据