医学
恶性肿瘤
对比度(视觉)
前瞻性队列研究
子宫附件疾病
接收机工作特性
放射科
危险分层
对比度增强
核医学
外科
内科学
磁共振成像
腹腔镜检查
人工智能
计算机科学
作者
Kun Yuan,Yun Peng Huang,Muyi Mao,Lu Tian,Song-Juan Wang,Dong Chang He,Wenfen Liu,Mengxiong Li,Xiao-Min Zhu,Xinyu Chen,Yunxiao Zhu
出处
期刊:Radiology
[Radiological Society of North America]
日期:2023-08-01
卷期号:308 (2)
被引量:2
标识
DOI:10.1148/radiol.223003
摘要
Background The Ovarian-Adnexal Reporting and Data System (O-RADS) has limited specificity for malignancy. Contrast-enhanced US can help distinguish malignant from benign lesions, but its added value to O-RADS has not yet been assessed. Purpose To establish a diagnostic model combining O-RADS and contrast-enhanced US and to validate whether O-RADS plus contrast-enhanced US has a better diagnostic performance than O-RADS alone. Materials and Methods This prospective study included participants from May 2018 to March 2021 who underwent contrast-enhanced US before surgery and had lesions categorized as O-RADS 3, 4, or 5 by US, with a histopathologic reference standard. From April 2021 to July 2022, participants with pathologically confirmed ovarian-adnexal lesions were recruited for the validation group. In the pilot group, the initial enhancement time and enhancement intensity in comparison with the uterine myometrium, contrast agent distribution pattern, and dynamic changes in enhancement of lesions were assessed. Contrast-enhanced US features were used to calculate contrast-enhanced US scores for benign (score ≤2) and malignant (score ≥4) lesions. Lesions were then re-rated according to O-RADS category plus contrast-enhanced US scores. Receiver operating characteristic curves were constructed and compared using the DeLong method. The combined system was validated in an independent group. Results The pilot group included 76 women (mean age, 44 years ± 13 [SD]), and the validation group included 46 women (mean age, 42 years ± 14). Differences in initial enhancement time (P < .001), enhancement intensity (P < .001), and dynamic changes in enhancement (P < .001) between benign and malignant lesions were observed in the pilot group. Contrast-enhanced US scores were calculated using these features. The O-RADS risk stratification was upgraded one level for contrast-enhanced US scores of 4 or more and downgraded one level for contrast-enhanced US scores of 2 or less. In the validation group, the diagnostic performance of O-RADS plus contrast-enhanced US score was higher (area under the receiver operating characteristic curve [AUC] = 0.93) than O-RADS (AUC = 0.71, P < .001). Conclusion Contrast-enhanced US improved the diagnostic performance for malignancy of the O-RADS categories 3–5. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Grant in this issue.
科研通智能强力驱动
Strongly Powered by AbleSci AI