文献速递泌尿生殖妇产影像US

超声深度学习预测卵巢恶性肿瘤与O-RADS及专家评估的比较

Deep Learning Prediction of Ovarian Malignancy at US Compared with O-RADS and Expert Assessment.
PubMed 文献 · Radiology · 2022-04-12

本回顾性研究纳入2019年1月至11月连续就诊的422例卵巢肿瘤女性,开发基于残差网络(ResNet)的深度学习算法(特征融合DLfeature与决策融合DLdecision),比较其与O-RADS及专家主观评估鉴别良恶性的效能。测试集中DLfeature的AUC为0.93,与O-RADS(AUC 0.92,P=0.88)及专家评估(AUC 0.97,P=0.07)相当,与DLdecision(AUC 0.90,P=0.29)相似。

要点

  • 测试集中DLfeature鉴别卵巢肿瘤良恶性的AUC为0.93(95%CI 0.85-0.97),与O-RADS的0.92(P=0.88)及专家评估的0.97(P=0.07)无显著差异。
  • 测试集中DLdecision、DLfeature、O-RADS及专家评估的敏感度分别为92%、92%、92%、96%,特异度分别为80%、85%、89%、87%。
  • 基于多模态超声图像的深度学习算法可能以与专家主观评估及O-RADS相当的诊断效能区分卵巢肿瘤良恶性。

核心结论

  • 基于多模态超声图像开发的深度学习算法可能以与专家主观评估及O-RADS相当的诊断效能区分卵巢肿瘤良恶性。

研究设计与样本

  • 回顾性研究,纳入2019年1月至2019年11月连续就诊的卵巢肿瘤女性422例(平均年龄46.4岁±14.8[SD]),其中良性肿瘤304例、恶性肿瘤118例;训练集与验证集共337例,测试集85例;采用灰阶与彩色多普勒超声,以组织病理学分析为参考标准;开发基于残差网络(ResNet)的两种融合策略算法(特征融合DLfeature与决策融合DLdecision),在测试集中以受试者工作特征曲线下面积(AUC)分析比较深度学习预测恶性与O-RADS风险分类及专家评估。

诊断标准与关键数值

  • 测试集中DLfeature鉴别卵巢恶性肿瘤与良性肿瘤的AUC为0.93(95%CI 0.85-0.97),与O-RADS(AUC 0.92,95%CI 0.85-0.97,P=0.88)及专家评估(AUC 0.97,95%CI 0.91-0.99,P=0.07)相当,与DLdecision(AUC 0.90,95%CI 0.82-0.96,P=0.29)相似;DLdecision、DLfeature、O-RADS及专家评估的敏感度分别为92%、92%、92%、96%,特异度分别为80%、85%、89%、87%。

临床应用与阅片要点

  • 原文未据此给出推荐。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

Background Deep learning (DL) algorithms could improve the classification of ovarian tumors assessed with multimodal US. Purpose To develop DL algorithms for the automated classification of benign versus malignant ovarian tumors assessed with US and to compare algorithm performance to Ovarian-Adnexal Reporting and Data System (O-RADS) and subjective expert assessment for malignancy. Materials and Methods This retrospective study included consecutive women with ovarian tumors undergoing gray scale and color Doppler US from January 2019 to November 2019. Histopathologic analysis was the reference standard. The data set was divided into training (70%), validation (10%), and test (20%) sets. Algorithms modified from residual network (ResNet) with two fusion strategies (feature fusion [hereafter, DLfeature] or decision fusion [hereafter, DLdecision]) were developed. DL prediction of malignancy was compared with O-RADS risk categorization and expert assessment by area under the receiver operating characteristic curve (AUC) analysis in the test set. Results A total of 422 women (mean age, 46.4 years ± 14.8 [SD]) with 304 benign and 118 malignant tumors were included; there were 337 women in the training and validation data set and 85 women in the test data set. DLfeature had an AUC of 0.93 (95% CI: 0.85, 0.97) for classifying malignant from benign ovarian tumors, comparable with O-RADS (AUC, 0.92; 95% CI: 0.85, 0.97; P = .88) and expert assessment (AUC, 0.97; 95% CI: 0.91, 0.99; P = .07), and similar to DLdecision (AUC, 0.90; 95% CI: 0.82, 0.96; P = .29). DLdecision, DLfeature, O-RADS, and expert assessment achieved sensitivities of 92%, 92%, 92%, and 96%, respectively, and specificities of 80%, 85%, 89%, and 87%, respectively, for malignancy. Conclusion Deep learning algorithms developed by using multimodal US images may distinguish malignant from benign ovarian tumors with diagnostic performance comparable to expert subjective and Ovarian-Adnexal Reporting and Data System assessment. © RSNA, 2022 Online supplemental material is available for this article.

来源:Radiology,查看原文。版权归原作者所有。

引用

  1. Chen, Yang, Qian, et al. Deep Learning Prediction of Ovarian Malignancy at US Compared with O-RADS and Expert Assessment. Radiology. 2022. PMID: 35412367. doi:10.1148/radiol.211367

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。