Department of Computer Science, City University of New York, the Graduate Center, NY, USA.
Department of Computer Science, City University of New York at CSI, NY, USA.
J Xray Sci Technol. 2019;27(1):17-35. doi: 10.3233/XST-180426.
Computer aided detection (CADe) of pulmonary nodules from computed tomography (CT) is crucial for early diagnosis of lung cancer. Self-learned features obtained by training datasets via deep learning have facilitated CADe of the nodules. However, the complexity of CT lung images renders a challenge of extracting effective features by self-learning only. This condition is exacerbated for limited size of datasets. On the other hand, the engineered features have been widely studied.
We proposed a novel nodule CADe which aims to relieve the challenge by the use of available engineered features to prevent convolution neural networks (CNN) from overfitting under dataset limitation and reduce the running-time complexity of self-learning.
The CADe methodology infuses adequately the engineered features, particularly texture features, into the deep learning process.
The methodology was validated on 208 patients with at least one juxta-pleural nodule from the public LIDC-IDRI database. Results demonstrated that the methodology achieves a sensitivity of 88% with 1.9 false positives per scan and a sensitivity of 94.01% with 4.01 false positives per scan.
The methodology shows high performance compared with the state-of-the-art results, in terms of accuracy and efficiency, from both existing CNN-based approaches and engineered feature-based classifications.
计算机辅助检测（CADe）在 CT 扫描中检测肺结节对早期肺癌的诊断至关重要。通过深度学习训练数据集获得的自学习特征有助于 CADe 检测结节。然而，仅通过自学习提取有效特征是具有挑战性的，因为 CT 肺部图像的复杂性增加了这一难度。在数据集有限的情况下，这种情况会更加恶化。另一方面，工程化特征已得到广泛研究。
我们提出了一种新的结节 CADe，旨在通过使用可用的工程化特征来缓解这一挑战，防止卷积神经网络（CNN）在数据集有限的情况下过度拟合，并降低自学习的运行时复杂度。
该 CADe 方法充分融合了工程化特征，特别是纹理特征，到深度学习过程中。
该方法在来自公共 LIDC-IDRI 数据库的至少有一个贴壁结节的 208 名患者上进行了验证。结果表明，该方法的灵敏度为 88%，每扫描有 1.9 个假阳性，灵敏度为 94.01%，每扫描有 4.01 个假阳性。
该方法在准确性和效率方面均优于现有基于 CNN 的方法和基于工程化特征的分类方法的最新结果。
