Berg Wendie A, Gur David, Bandos Andriy I, Nair Bronwyn, Gizienski Terri-Ann, Tyma Cathy S, Abrams Gordon, Davis Katie M, Mehta Amar S, Rathfon Grace, Waheed Uzma X, Hakim Christiane M
University of Pittsburgh School of Medicine, Department of Radiology, Pittsburgh, PA,USA.
Magee-Womens Hospital of UPMC, Pittsburgh, PA,USA.
J Breast Imaging. 2021 May 21;3(3):301-311. doi: 10.1093/jbi/wbab013.
For breast US interpretation, to assess impact of computer-aided diagnosis (CADx) in original mode or with improved sensitivity or specificity.
In this IRB approved protocol, orthogonal-paired US images of 319 lesions identified on screening, including 88 (27.6%) cancers (median 7 mm, range 1-34 mm), were reviewed by 9 breast imaging radiologists. Each observer provided BI-RADS assessments (2, 3, 4A, 4B, 4C, 5) before and after CADx in a mode-balanced design: mode 1, original CADx (outputs benign, probably benign, suspicious, or malignant); mode 2, artificially-high-sensitivity CADx (benign or malignant); and mode 3, artificially-high-specificity CADx (benign or malignant). Area under the receiver operating characteristic curve (AUC) was estimated under each modality and for standalone CADx outputs. Multi-reader analysis accounted for inter-reader variability and correlation between same-lesion assessments.
AUC of standalone CADx was 0.77 (95% CI: 0.72-0.83). For mode 1, average reader AUC was 0.82 (range 0.76-0.84) without CADx and not significantly changed with CADx. In high-sensitivity mode, all observers' AUCs increased: average AUC 0.83 (range 0.78-0.86) before CADx increased to 0.88 (range 0.84-0.90), P < 0.001. In high-specificity mode, all observers' AUCs increased: average AUC 0.82 (range 0.76-0.84) before CADx increased to 0.89 (range 0.87-0.92), P < 0.0001. Radiologists responded more frequently to malignant CADx cues in high-specificity mode (42.7% vs 23.2% mode 1, and 27.0% mode 2, P = 0.008).
Original CADx did not substantially impact radiologists' interpretations. Radiologists showed improved performance and were more responsive when CADx produced fewer false-positive malignant cues.
对于乳腺超声解读，评估原始模式或具有更高敏感性或特异性的计算机辅助诊断（CADx）的影响。
在这项经机构审查委员会（IRB）批准的方案中，9名乳腺影像放射科医生对筛查时发现的319个病灶的正交配对超声图像进行了评估，其中包括88个（27.6%）癌症（中位数7毫米，范围1 - 34毫米）。在模式平衡设计中，每位观察者在CADx前后分别给出乳腺影像报告和数据系统（BI-RADS）评估（2、3、4A、4B、4C、5）：模式1为原始CADx（输出良性、可能良性、可疑或恶性）；模式2为人工高敏感性CADx（良性或恶性）；模式3为人工高特异性CADx（良性或恶性）。在每种模式下以及单独的CADx输出中估计受试者操作特征曲线（AUC）下的面积。多读者分析考虑了读者间的变异性以及同一病灶评估之间的相关性。
单独CADx的AUC为0.77（95%置信区间：0.72 - 0.83）。对于模式1，读者在无CADx时的平均AUC为0.82（范围0.76 - 0.84），使用CADx后无显著变化。在高敏感性模式下，所有观察者的AUC均增加：CADx前平均AUC为0.83（范围0.78 - 0.86），增加到0.88（范围0.84 - 0.90），P < 0.001。在高特异性模式下，所有观察者的AUC均增加：CADx前平均AUC为0.82（范围0.76 - 0.84），增加到0.89（范围0.87 - 0.92），P < 0.0001。放射科医生在高特异性模式下对恶性CADx提示的反应更频繁（模式1为42.7%，模式2为23.2%，模式3为27.0%，P = 0.008）。
原始CADx对放射科医生的解读影响不大。当CADx产生较少的假阳性恶性提示时，放射科医生的表现有所改善且反应更积极。
