Department of Radiology, Ajou University School of Medicine, Suwon 16499, Korea.
Department of Biostatistics, Ajou University School of Medicine, Suwon 16499, Korea.
Korean J Radiol. 2018 Jul-Aug;19(4):665-672. doi: 10.3348/kjr.2018.19.4.665. Epub 2018 Jun 14.
To prospectively evaluate the diagnostic performance of computer-aided diagnosis (CAD) for detection of thyroid cancers via ultrasonography (US).
This study included 50 consecutive patients with 117 thyroid nodules on US during the period between June 2016 and July 2016. A radiologist performed US examinations using real-time CAD integrated into a US scanner. We compared the diagnostic performance of radiologist, the CAD system, and the CAD-assisted radiologist for the detection of thyroid cancers.
The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of the CAD system were 80.0, 88.1, 83.3, 85.5, and 84.6%, respectively, and were not significantly different from those of the radiologist ( > 0.05). The CAD-assisted radiologist showed improved diagnostic sensitivity compared with the radiologist alone (92.0% vs. 84.0%, = 0.037), while the specificity and PPV were reduced (85.1% vs. 95.5%, = 0.005 and 82.1% vs. 93.3%, = 0.008). The radiologist assisted by the CAD system exhibited better diagnostic sensitivity and NPV than the CAD system alone (92.0% vs. 80.0%, = 0.009 and 93.4% vs. 88.9%, = 0.013), while the specificities and PPVs were not significantly different (88.1% vs. 85.1%, = 0.151 and 83.3% vs. 82.1%, = 0.613, respectively).
The CAD system may be an adjunct to radiological intervention in the diagnosis of thyroid cancer.
前瞻性评估计算机辅助诊断（CAD）在超声（US）检测甲状腺癌中的诊断性能。
本研究纳入了 2016 年 6 月至 7 月期间连续 50 例 117 个甲状腺结节的患者。一名放射科医生使用实时 CAD 系统对 US 进行检查，该系统集成在超声扫描仪中。我们比较了放射科医生、CAD 系统和 CAD 辅助的放射科医生对甲状腺癌的诊断性能。
CAD 系统的灵敏度、特异度、阳性预测值（PPV）、阴性预测值（NPV）和准确率分别为 80.0%、88.1%、83.3%、85.5%和 84.6%，与放射科医生无显著差异（>0.05）。与单独使用放射科医生相比，CAD 辅助的放射科医生显示出更高的诊断敏感性（92.0% vs. 84.0%，=0.037），而特异性和 PPV 则降低（85.1% vs. 95.5%，=0.005 和 82.1% vs. 93.3%，=0.008）。CAD 系统辅助的放射科医生的诊断敏感性和 NPV 优于单独使用 CAD 系统（92.0% vs. 80.0%，=0.009 和 93.4% vs. 88.9%，=0.013），而特异性和 PPV 则无显著差异（88.1% vs. 85.1%，=0.151 和 83.3% vs. 82.1%，=0.613）。
CAD 系统可能是甲状腺癌放射学干预诊断的辅助手段。
