Computer-aided, detection of lung nodules: False positive reduction using a 3D gradient field method and 3D ellipsoid fitting
Computer-aided, detection of lung nodules: False positive reduction using a 3D gradient field method and 3D ellipsoid fitting
复制标题
DOI：
10.1118/1.1944667
复制
发表时间：
2005-08-01
期刊：
影响因子：
3.8
通讯作者：
Zhou, CA
中科院分区：
文献类型：
--
作者：
Ge, ZY;Sahiner, B;Zhou, CA
我们正在开发一个计算机辅助检测系统，以协助放射科医生在胸部计算机断层扫描（CT）图像上检测肺结节。本研究的目的是提高假阳性（FP）减少我们的算法阶段开发的功能，提取三维（3D）的形状信息，在预筛选阶段确定的感兴趣的体积。我们制定了3D梯度场描述符，并从他们的统计数据中推导出19个梯度场特征。通过计算椭球的长度和主轴长度比，得到六个椭球特征。梯度场特征和椭圆体特征两者都被设计成将球形对象（诸如肺结节）与细长对象（诸如血管）区分开。在这个新的25维特征空间的FP减少性能进行了比较，在一个19维的空间，包括使用以前开发的方法提取的特征的性能。在44维组合特征空间的性能也进行了评估。采用逐步特征选择的线性判别分析进行分类。使用单纯形算法优化用于特征选择的参数。使用留一患者方案进行培训和测试。在不同的特征空间的FP减少性能进行了评估，通过使用面积A，在接收器工作特性曲线和FP的数量在给定的灵敏度作为准确性的措施，每个CT部分。我们的数据集包括56例患者的82次CT扫描（3551个轴向切片）。厚度范围从1.0到2.5 mm。我们的预筛选算法检测到116个由经验丰富的胸部放射科医生标记的实性结节（结节大小：3.0-30.6 mm）中的111个。在新的、先前的和组合的特征空间中，测试A（z）值分别为0.95 +/- 0.01、0.88 +/-0.02和0.94 +/- 0.01。在这三个特征空间中，80%时每个切片的FP数、灵敏度分别为0.37、1.61和0.34。与仅使用先前的19个特征相比，使用25个新特征的测试A中的改善具有统计学显著性（p < 0.0001）。(c)2005年美国医学物理学家协会。
We are developing a computer-aided detection system to assist radiologists in the detection of lung nodules on thoracic computed tomography (CT) images. The purpose of this study was to improve the false-positive (FP) reduction stage of our algorithm by developing features that extract threedimensional (3D) shape information from volumes of interest identified in the prescreening stage. We formulated 3D gradient field descriptors, and derived 19 gradient field features from-their statistics. Six ellipsoid features were obtained by computing the lengths and the length ratios of the principal axes of an ellipsoid fitted to a segmented object. Both the gradient field features and the ellipsoid features were designed to distinguish spherical objects such as lung nodules from elongated objects such as vessels. The FP reduction performance in this new 25-dimensional feature space was compared to the performance in a 19-dimensional space that consisted of features extracted using previously developed methods. The performance in the 44-dimensional combined feature space was also evaluated. Linear discriminant analysis with stepwise feature selection was used for classification. The parameters used for feature selection were optimized using the simplex algorithm. Training and testing were performed using a leave-one-patient-out scheme. The FP reduction performances in different feature spaces were evaluated by using the area A, under the receiver operating characteristic curve and the number of FPs per CT section at a given sensitivity as accuracy measures. Our data set consisted of 82 CT scans (3551 axial sections) from 56 patients with section. thickness ranging from 1.0 to 2.5 mm. Our prescreening algorithm detected I I I of the 116 solid nodules (nodule size: 3.0-30.6 mm) marked by experienced thoracic radiologists. The test A(z) values were 0.95 +/- 0.01, 0.88 +/- 0 02, and 0.94 +/- 0.01 in the new, previous, and combined feature spaces, respectively. The number of FPs per section at 80%, sensitivity in these three feature spaces were 0.37, 1.61, and 0.34, respectively. The improvement in the test A, with the 25 new features, was statistically significant (p < 0.0001) compared to that with the previous 19 features alone. (c) 2005 American Association of Physicists in Medicine.