Institute of Biomedical Engineering, Capital Medical University, Fengtai District, Beijing, PR China.
Acad Radiol. 2010 May;17(5):595-602. doi: 10.1016/j.acra.2009.12.009. Epub 2010 Feb 18.
To evaluate the diagnostic performance of a neural network ensemble-based computer-aided diagnosis (CAD) scheme for classifying lung nodules on thin-section computed tomography (CT).
Thirty-two CT images that depicted 19 malignant nodules and 13 benign nodules were used. One of three possible classifications (probably benign, uncertain, and probably malignant) for each nodule was determined by using a neural network ensemble-based CAD scheme. The images were presented to three senior radiologists (each with more than 10 years of thoracic radiology experience) who were asked to determine the classification for each nodule blindly. The radiologists made their diagnostic decisions solely based on images and excluded any external data. The performance of the CAD scheme and of the radiologists was evaluated with receiver operating characteristic (ROC) analysis and agreement analysis.
Areas under the ROC curve (Az values) for the CAD scheme and the radiologist group were 0.79 and 0.82, respectively, and the partial areas under the ROC curves at a range of sensitivity values greater than or equal to 90% were 0.051 and 0.020 (P = .203), respectively. The weighted Kappa coefficients between the CAD scheme and each radiologist were 0.657, 0.431, and 0.606, respectively. For the diagnosis of the 11 small nodules (with diameters not greater than 10 mm), areas under the ROC curves of the CAD scheme and the radiologist group were 0.915 and 0.683 (P = .227), respectively.
The diagnostic performance of the neural network ensemble-based CAD scheme is similar to that of senior radiologists for classifying lung nodules on thin-section CT. Furthermore, the CAD scheme has certain advantages in diagnosing small lung nodules.
评估基于神经网络集成的计算机辅助诊断（CAD）方案在对肺部结节进行薄层 CT 分类方面的诊断性能。
共使用了 32 张 CT 图像，其中包括 19 个恶性结节和 13 个良性结节。使用基于神经网络集成的 CAD 方案为每个结节确定了三个可能分类之一（可能良性、不确定和可能恶性）。将这些图像呈现给三位资深放射科医生（每位医生均有超过 10 年的胸部放射学经验），要求他们盲法为每个结节确定分类。放射科医生仅根据图像做出诊断决策，并排除任何外部数据。使用受试者工作特征（ROC）分析和一致性分析来评估 CAD 方案和放射科医生的性能。
CAD 方案和放射科医生组的 ROC 曲线下面积（Az 值）分别为 0.79 和 0.82，在灵敏度值大于或等于 90%的范围内，部分 ROC 曲线下面积分别为 0.051 和 0.020（P =.203）。CAD 方案与每位放射科医生之间的加权 Kappa 系数分别为 0.657、0.431 和 0.606。对于 11 个小结节（直径不大于 10mm）的诊断，CAD 方案和放射科医生组的 ROC 曲线下面积分别为 0.915 和 0.683（P =.227）。
基于神经网络集成的 CAD 方案在对肺部结节进行薄层 CT 分类方面的诊断性能与资深放射科医生相当。此外，该 CAD 方案在诊断小肺结节方面具有一定优势。
