LI-RADS 版本 2018 培训是否能提高读者对专家共识和 MRI 解读中读者间一致性的认同？

Does Training in LI-RADS Version 2018 Improve Readers' Agreement with the Expert Consensus and Inter-reader Agreement in MRI Interpretation?

机构信息

Department of Radiology, Beijing Friendship Hospital, Capital Medical University, Beijing, China.

National Clinical Research Center of Digestive Diseases, Beijing, China.

出版信息

J Magn Reson Imaging. 2021 Dec;54(6):1922-1934. doi: 10.1002/jmri.27688. Epub 2021 May 8.

DOI:10.1002/jmri.27688

PMID:33963801

Abstract

BACKGROUND

The Liver Imaging Reporting and Data System (LI-RADS) was established for noninvasive diagnosis for hepatocellular carcinoma (HCC). However, whether training can improve readers' agreement with the expert consensus and inter-reader agreement for final categories is still unclear.

PURPOSE

To explore training effectiveness on readers' agreement with the expert consensus and inter-reader agreement.

STUDY TYPE

Prospective.

SUBJECTS

Seventy lesions in 61 patients at risk of HCC undergoing liver MRI; 20 visiting scholars.

FIELD STRENGTH/SEQUENCE: 1.5 T or 3 T, Dual-echo T WI, Fast spin-echo T WI, SE-EPI DWI, and Dynamic multiphase fast gradient-echo T WI.

ASSESSMENT

Seventy lesions assigned LI-RADS categories of LR1-LR5, LR-M, and LR-TIV by three radiologists in consensus were randomly selected, with 10 cases for each category. The consensus opinion was the standard reference. The third radiologist delivered the training. Twenty readers reviewed images independently and assigned each an LI-RADS category both before and after the training.

STATISTICAL TESTS

Accuracy, sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, receiver operating characteristic (ROC) analysis, simple and weighted kappa statistics, and Fleiss kappa statistics.

RESULTS

Before and after training: readers' AUC (areas under ROC) for LR-1-LR-5, LR-M, and LR-TIV were 0.898 vs. 0.913, 0.711 vs. 0.876, 0.747 vs. 0.860, 0.724 vs. 0.815, 0.844 vs. 0.895, 0.688 vs. 0.873, and 0.720 vs. 0.948, respectively, and all improved significantly (P < 0.05), except LR-1(P = 0.25). Inter-reader agreement between readers for LR-1-LR-5, LR-M, LR-TIV were 0.725 vs. 0.751, 0.325 vs. 0.607, 0.330 vs. 0.559, 0.284 vs. 0.488, 0.447 vs. 0.648, 0.229 vs. 0.589, and 0.362 vs. 0.852, respectively, and all increased significantly (P < 0.05). For training effectiveness on both AUC and inter-reader agreement, LR-TIV, LR-M, and LR-2 improved most, and LR-1 made the least.

DATA CONCLUSION

This study shows LI-RADS training could improve reader agreement with the expert consensus and inter-reader agreement for final categories.

LEVEL OF EVIDENCE

2 TECHNICAL EFFICACY STAGE: 2.

摘要

背景

肝脏影像报告和数据系统（LI-RADS）旨在为肝细胞癌（HCC）的非侵入性诊断提供支持。然而，培训是否能够提高读者对专家共识和最终类别中读者间一致性的一致性尚不清楚。

目的

探讨培训对读者与专家共识和读者间一致性的影响。

研究类型

前瞻性。

受试者

61 例 HCC 高危患者的 70 个病灶；20 名访问学者。

磁场强度/序列：1.5T 或 3T，双回波 TWI、快速自旋回波 TWI、SE-EPI DWI 和动态多期快速梯度回波 TWI。

评估

由三位放射科医生以共识方式对 70 个病灶进行 LI-RADS 类别 LR1-LR5、LR-M 和 LR-TIV 的分配，每个类别各有 10 例。共识意见为标准参考。第三位放射科医生提供培训。20 名读者独立地查看图像，并在培训前后为每个病灶分配 LI-RADS 类别。

统计学检验

准确性、敏感性、特异性、阳性预测值、阴性预测值、阳性似然比、阴性似然比、受试者工作特征（ROC）分析、简单和加权kappa 统计、Fleiss kappa 统计。

结果

培训前后：LR-1-LR-5、LR-M 和 LR-TIV 的读者 AUC（ROC 下面积）分别为 0.898 比 0.913、0.711 比 0.876、0.747 比 0.860、0.724 比 0.815、0.844 比 0.895、0.688 比 0.873 和 0.720 比 0.948，均显著提高（P<0.05），除 LR-1（P=0.25）外。LR-1-LR-5、LR-M 和 LR-TIV 的读者间一致性分别为 0.725 比 0.751、0.325 比 0.607、0.330 比 0.559、0.284 比 0.488、0.447 比 0.648、0.229 比 0.589 和 0.362 比 0.852，均显著提高（P<0.05）。在 AUC 和读者间一致性的培训效果方面，LR-TIV、LR-M 和 LR-2 提高最多，LR-1 提高最少。