Automated segmentation refinement of small lung nodules in CT scans by local shape analysis

One of the most important problems in the segmentation of lung nodules in CT imaging arises from possible attachments occurring between nodules and other lung structures, such as vessels or pleura. In this report, we address the problem of vessels attachments by proposing an automated correction method applied to an initial rough segmentation of the lung nodule. The method is based on a local shape analysis of the initial segmentation making use of 3-D geodesic distance map representations. The correction method has the advantage that it locally refines the nodule segmentation along recognized vessel attachments only, without modifying the nodule boundary elsewhere. The method was tested using a simple initial rough segmentation, obtained by a fixed image thresholding. The validation of the complete segmentation algorithm was carried out on small lung nodules, identified in the ITALUNG screening trial and on small nodules of the lung image database consortium (LIDC) dataset. In fully automated mode, 217/256 (84.8%) lung nodules of ITALUNG and 139/157 (88.5%) individual marks of lung nodules of LIDC were correctly outlined and an excellent reproducibility was also observed. By using an additional interactive mode, based on a controlled manual interaction, 233/256 (91.0%) lung nodules of ITALUNG and 144/157 (91.7%) individual marks of lung nodules of LIDC were overall correctly segmented. The proposed correction method could also be usefully applied to any existent nodule segmentation algorithm for improving the segmentation quality of juxta-vascular nodules.     

基于语义属性的肺结节良恶性分类

现有肺结节良恶性计算机辅助诊断的依据通常为肺部CT图像的底层特征,而临床医生的诊断依据为高级语义特征.为克服这种图像底层特征和高级语义特征之间的不一致性,提出一种基于语义属性的肺结节良恶性判别方法.首先,利用阈值概率图方法提取肺结节图像;其次,一方面提取肺结节图像的形状、灰度、纹理、大小和位置等底层特征,组成样本特征集.另一方面,根据专家对肺结节属性的标注,提取结节属性集;然后,根据特征集和属性集建立属性预测模型,实现两者之间的映射;最后,利用预测的属性进行肺结节的良恶性分类.LIDC数据库上的实验结果表明所提方法具有较高的分类精度和AUC值.     

Surface normal overlap: a computer-aided detection algorithm with application to colonic polyps and lung nodules in helical CT

We developed a novel computer-aided detection (CAD) algorithm called the surface normal overlap method that we applied to colonic polyp detection and lung nodule detection in helical computed tomography (CT) images. We demonstrate some of the theoretical aspects of this algorithm using a statistical shape model. The algorithm was then optimized on simulated CT data and evaluated using a per-lesion cross-validation on 8 CT colonography datasets and on 8 chest CT datasets. It is able to achieve 100% sensitivity for colonic polyps 10 mm and larger at 7.0 false positives (FPs)/dataset and 90% sensitivity for solid lung nodules 6 mm and larger at 5.6 FP/dataset.     

一种新策略的肺结节检测算法

针对计算机在肺部CT肺结节辅助检测过程中,二维CT图像序列血管横截面与肺结节成像特征类似,导致无法有效检测的问题,提出新策略的肺结节检测算法。以格式塔心理学为基础,以去除血管的新策略间接的对肺结节进行检测。实验结果表明,本算法可有效降低血管对肺结节检测的影响,从而提高肺结节的检测精度。     

Computer-aided diagnostic scheme for distinction between benign and malignant nodules in thoracic low-dose CT by use of massive training artificial neural network

Low-dose helical computed tomography (LDCT) is being applied as a modality for lung cancer screening. It may be difficult, however, for radiologists to distinguish malignant from benign nodules in LDCT. Our purpose in this study was to develop a computer-aided diagnostic (CAD) scheme for distinction between benign and malignant nodules in LDCT scans by use of a massive training artificial neural network (MTANN). The MTANN is a trainable, highly nonlinear filter based on an artificial neural network. To distinguish malignant nodules from six different types of benign nodules, we developed multiple MTANNs (multi-MTANN) consisting of six expert MTANNs that are arranged in parallel. Each of the MTANNs was trained by use of input CT images and teaching images containing the estimate of the distribution for the "likelihood of being a malignant nodule," i.e., the teaching image for a malignant nodule contains a two-dimensional Gaussian distribution and that for a benign nodule contains zero. Each MTANN was trained independently with ten typical malignant nodules and ten benign nodules from each of the six types. The outputs of the six MTANNs were combined by use of an integration ANN such that the six types of benign nodules could be distinguished from malignant nodules. After training of the integration ANN, our scheme provided a value related to the "likelihood of malignancy" of a nodule, i.e., a higher value indicates a malignant nodule, and a lower value indicates a benign nodule. Our database consisted of 76 primary lung cancers in 73 patients and 413 benign nodules in 342 patients, which were obtained from a lung cancer screening program on 7847 screenees with LDCT for three years in Nagano, Japan. The performance of our scheme for distinction between benign and malignant nodules was evaluated by use of receiver operating characteristic (ROC) analysis. Our scheme achieved an Az (area under the ROC curve) value of 0.882 in a round-robin test. Our scheme correctly identified 100% (76/76) of malignant nodules as malignant, whereas 48% (200/413) of benign nodules were identified correctly as benign. Therefore, our scheme may be useful in assisting radiologists in the diagnosis of lung nodules in LDCT.     

基于粗糙集特征级融合的肺结节检测算法

以肺结节的检测为研究目标,针对肺结节特征级融合检测算法中存在特征结构不合理和特征表达不紧致两个问题,提出了一种基于粗糙集特征级融合的肺结节检测算法,该算法首先分析肺部CT影像的医学征象,提出了六个新的三维特征,并综合其他二维和三维特征共42维特征分量共同量化ROI;然后基于粗糙集对提取的特征集合进行5次特征级融合实验;最后利用网格寻优算法优化核函数的SVM作为分类器进行肺结节识别.以70例肺结节患者的肺部CT影像为原始数据,通过4组对比实验验证算法的有效性和稳定性,实验结果表明,经过粗糙集特征级融合的肺结节检测算法识别肺结节的能力得到了有效提升.     

基于多模型融合方法的肺结节良恶性分类

针对CT图像中肺结节因边缘模糊、特征不明显造成的分类效果有偏差的问题,本文提出一种嵌入注意力机制的多模型融合方法(简称MSMA-Net).该方法先将原始CT图像进行肺实质分割和裁剪操作后得到两种不同尺寸的图像,然后分别输入到空间注意力模型和通道注意力模型进行训练,其中,空间注意力模型着重于提取肺结节在CT图像中的空间位...     

Patient-specific models for lung nodule detection and surveillancein CT images

The purpose of this work is to develop patient-specific models for automatically detecting lung nodules in computed tomography (CT) images. It is motivated by significant developments in CT scanner technology and the burden that lung cancer screening and surveillance imposes on radiologists. We propose a new method that uses a patient's baseline image data to assist in the segmentation of subsequent images so that changes in size and/or shape of nodules can be measured automatically. The system uses a generic, a priori model to detect candidate nodules on the baseline scan of a previously unseen patient. A user then confirms or rejects nodule candidates to establish baseline results. For analysis of follow-up scans of that particular patient, a patient-specific model is derived from these baseline results. This model describes expected features (location, volume and shape) of previously segmented nodules so that the system can relocalize them automatically on follow-up. On the baseline scans of 17 subjects, a radiologist identified a total of 36 nodules, of which 31 (86%) were detected automatically by the system with an average of 11 false positives (FPs) per case. In follow-up scans 27 of the 31 nodules were still present and, using patient-specific models, 22 (81%) were correctly relocalized by the system. The system automatically detected 16 out of a possible 20 (80%) of new nodules on follow-up scans with ten FPs per case.     

基于空间分布的三维自动化肺结节分割算法

针对肺结节分割中存在的自动化程度低、较少考虑空间结构以及粘附型肺结节分割不充分问题,提出了一种基于空间分布的三维自动化肺结节分割算法.该算法首先利用C-means聚类算法分割出肺实质,然后根据肺结节空间分布的差异性将其分为3类:孤立性肺结节、胸膜粘附性肺结节、血管粘附性肺结节,并对3种不同类型的肺结节分别采用基于连通性、灰度下降和散度差异的分割算法进行分割,70个肺结节(其中孤立性肺结节38个,血管粘附性肺结节17个,胸膜粘附性肺结节15个)CT图像的实验结果表明,算法能够准确、自动地分割出3种不同部位的肺结节.     

Lung No dules Detection in CT Images Using Gestalt-Based Algorithm

To overcome low accuracy and high false positive of existing computer-aided lung nodules detec-tion. We propose a novel lung nodule detection scheme based on the Gestalt visual cognition theory. The pro-posed scheme involves two parts which simulate human eyes cognition features such as simplicity, integrity and classification. Firstly, lung region was segmented from lung Computed tomography (CT) sequences. Then local three-dimensional information was integrated into the Maximum intensity projection (MIP) images from axial, coronal and sagittal profiles. In this way, lung nodules and vascular are strengthened and discriminated based on pathologic image characteristics of lung nodules. The experimental database includes fifty-three high resolution CT images contained lung nodules, which had been confirmed by biopsy. The experimental results show that, the accuracy rate of the proposed algorithm achieves 91.29%. The proposed frame-work improves performance and computation speed for computer aided nodules detection.     

GGO nodule volume-preserving nonrigid lung registration using GLCM texture analysis

In lung cancer screening, benign and malignant nodules can be classified through nodule growth assessment by the registration and, then, subtraction between follow-up computed tomography scans. During the registration, the volume of nodule regions in the floating image should be preserved, whereas the volume of other regions in the floating image should be aligned to that in the reference image. However, ground glass opacity (GGO) nodules are very elusive to automatically segment due to their inhomogeneous interior. In other words, it is difficult to automatically define the volume-preserving regions of GGO nodules. In this paper, we propose an accurate and fast nonrigid registration method. It applies the volume-preserving constraint to candidate regions of GGO nodules, which are automatically detected by gray-level cooccurrence matrix (GLCM) texture analysis. Considering that GGO nodules can be characterized by their inner inhomogeneity and high intensity, we identify the candidate regions of GGO nodules based on the homogeneity values calculated by the GLCM and the intensity values. Furthermore, we accelerate our nonrigid registration by using Compute Unified Device Architecture (CUDA). In the nonrigid registration process, the computationally expensive procedures of the floating-image transformation and the cost-function calculation are accelerated by using CUDA. The experimental results demonstrated that our method almost perfectly preserves the volume of GGO nodules in the floating image as well as effectively aligns the lung between the reference and floating images. Regarding the computational performance, our CUDA-based method delivers about 20× faster registration than the conventional method. Our method can be successfully applied to a GGO nodule follow-up study and can be extended to the volume-preserving registration and subtraction of specific diseases in other organs (e.g., liver cancer).     

On measuring the change in size of pulmonary nodules

The pulmonary nodule is the most common manifestation of lung cancer, the most deadly of all cancers. Most small pulmonary nodules are benign, however, and currently the growth rate of the nodule provides for one of the most accurate noninvasive methods of determining malignancy. In this paper, we present methods for measuring the change in nodule size from two computed tomography image scans recorded at different times; from this size change the growth rate may be established. The impact of partial voxels for small nodules is evaluated and isotropic resampling is shown to improve measurement accuracy. Methods for nodule location and sizing, pleural segmentation, adaptive thresholding, image registration, and knowledge-based shape matching are presented. The latter three techniques provide for a significant improvement in volume change measurement accuracy by considering both image scans simultaneously. Improvements in segmentation are evaluated by measuring volume changes in benign or slow growing nodules. In the analysis of 50 nodules, the variance in percent volume change was reduced from 11.54% to 9.35% (p = 0.03) through the use of registration, adaptive thresholding, and knowledge-based shape matching.     

基于深度卷积神经网络的低剂量CT肺部去噪

为了降低低剂量CT肺部噪声对肺癌筛查后期诊断的影响,该文提出一种基于深度卷积神经网络的低剂量CT肺部去噪算法。以完整的CT肺部图像作为输入,池化层对输入图像进行降维处理;批规范化解决随着网络深度的增加性能降低的问题;引入残差学习,学习模型中每一层的残差,最后输出去噪图像。与经典去噪算法实验结果对比,所提方法在解决去噪方面达到了很好的滤波效果,同时也较好地保留了肺部图像的细节信息,大大优于传统的去噪算法。     

Three-dimensional segmentation and growth-rate estimation of small pulmonary nodules in helical CT images

Small pulmonary nodules are a common radiographic finding that presents an important diagnostic challenge in contemporary medicine. While pulmonary nodules are the major radiographic indicator of lung cancer, they may also be signs of a variety of benign conditions. Measurement of nodule growth rate over time has been shown to be the most promising tool in distinguishing malignant from nonmalignant pulmonary nodules. In this paper, we describe three-dimensional (3-D) methods for the segmentation, analysis, and characterization of small pulmonary nodules imaged using computed tomography (CT). Methods for the isotropic resampling of anisotropic CT data are discussed. 3-D intensity and morphology-based segmentation algorithms are discussed for several classes of nodules. New models and methods for volumetric growth characterization based on longitudinal CT studies are developed. The results of segmentation and growth characterization methods based on in vivo studies are described. The methods presented are promising in their ability to distinguish malignant from nonmalignant pulmonary nodules and represent the first such system in clinical use.     

Feature subset selection for improving the performance of false positive reduction in lung nodule CAD.

We propose a feature subset selection method based on genetic algorithms to improve the performance of false positive reduction in lung nodule computer-aided detection (CAD). It is coupled with a classifier based on support vector machines. The proposed approach determines automatically the optimal size of the feature set, and chooses the most relevant features from a feature pool. Its performance was tested using a lung nodule database (52 true nodules and 443 false ones) acquired by multislice CT scans. From 23 features calculated for each detected structure, the suggested method determined ten to be the optimal feature subset size, and selected the most relevant ten features. A support vector machine classifier trained with the optimal feature subset resulted in 100% sensitivity and 56.4% specificity using an independent validation set. Experiments show significant improvement achieved by a system incorporating the proposed method over a system without it. This approach can be also applied to other machine learning problems; e.g. computer-aided diagnosis of lung nodules.     

Vessel tree reconstruction in thoracic CT scans with application to nodule detection

Vessel tree reconstruction in volumetric data is a necessary prerequisite in various medical imaging applications. Specifically, when considering the application of automated lung nodule detection in thoracic computed tomography (CT) scans, vessel trees can be used to resolve local ambiguities based on global considerations and so improve the performance of nodule detection algorithms. In this study, a novel approach to vessel tree reconstruction and its application to nodule detection in thoracic CT scans was developed by using correlation-based enhancement filters and a fuzzy shape representation of the data. The proposed correlation-based enhancement filters depend on first-order partial derivatives and so are less sensitive to noise compared with Hessian-based filters. Additionally, multiple sets of eigenvalues are used so that a distinction between nodules and vessel junctions becomes possible. The proposed fuzzy shape representation is based on regulated morphological operations that are less sensitive to noise. Consequently, the vessel tree reconstruction algorithm can accommodate vessel bifurcation and discontinuities. A quantitative performance evaluation of the enhancement filters and of the vessel tree reconstruction algorithm was performed. Moreover, the proposed vessel tree reconstruction algorithm reduced the number of false positives generated by an existing nodule detection algorithm by 38%.     

A Computational Geometry Approach to Automated Pulmonary Fissure Segmentation in CT Examinations

Identification of pulmonary fissures, which form the boundaries between the lobes in the lungs, may be useful during clinical interpretation of computed tomography (CT) examinations to assess the early presence and characterization of manifestation of several lung diseases. Motivated by the unique nature of the surface shape of pulmonary fissures in 3-D space, we developed a new automated scheme using computational geometry methods to detect and segment fissures depicted on CT images. After a geometric modeling of the lung volume using the marching cubes algorithm, Laplacian smoothing is applied iteratively to enhance pulmonary fissures by depressing nonfissure structures while smoothing the surfaces of lung fissures. Next, an extended Gaussian image based procedure is used to locate the fissures in a statistical manner that approximates the fissures using a set of plane ldquopatchesrdquo. This approach has several advantages such as independence of anatomic knowledge of the lung structure except the surface shape of fissures, limited sensitivity to other lung structures, and ease of implementation. The scheme performance was evaluated by two experienced thoracic radiologists using a set of 100 images (slices) randomly selected from 10 screening CT examinations. In this preliminary evaluation 98.7% and 94.9% of scheme segmented fissure voxels are within 2 mm of the fissures marked independently by two radiologists in the testing image dataset. Using the scheme detected fissures as reference, 89.4% and 90.1% of manually marked fissure points have distance les2 mm to the reference suggesting a possible under-segmentation of the scheme. The case-based root mean square (rms) distances (ldquoerrorsrdquo) between our scheme and the radiologist ranged from 1.48plusmn0.92 to 2.04plusmn3.88 mm. The discrepancy of fissure detection results between the automated scheme and either radiologist is smaller in this dataset than the interreader variability.     

Automated detection of pulmonary nodules in helical CT images based on an improved template-matching technique

The purpose of this study is to develop a technique for computer-aided diagnosis (CAD) systems to detect lung nodules in helical X-ray pulmonary computed tomography CT) images. We propose a novel template-matching technique based on a genetic algorithm (GA) template matching (GATM) for detecting nodules existing within the lung area; the GA was used to determine the target position in the observed image efficiently and to select an adequate template image from several reference patterns for quick template matching. In addition, a conventional template matching was employed to detect nodules existing on the lung wall area, lung wall template matching (LWTM), where semicircular models were used as reference patterns; the semicircular models were rotated according to the angle of the target point on the contour of the lung wall. After initial detecting candidates using the two template-matching methods, we extracted a total of 13 feature values and used them to eliminate false-positive findings. Twenty clinical cases involving a total of 557 sectional images were used in this study. 71 nodules out of 98 were correctly detected by our scheme (i.e., a detection rate of about 72%), with the number of false positives at approximately 1.1/sectional image. Our present results show that our scheme can be regarded as a technique for CAD systems to detect nodules in helical CT pulmonary images.     

Confocal microwave imaging for breast cancer detection: delay-multiply-and-sum image reconstruction algorithm

A new image reconstruction algorithm, termed as delay-multiply-and-sum (DMAS), for breast cancer detection using an ultra-wideband confocal microwave imaging technique is proposed. In DMAS algorithm, the backscattered signals received from numerical breast phantoms simulated using the finite-difference time-domain method are time shifted, multiplied in pair, and the products are summed to form a synthetic focal point. The effectiveness of the DMAS algorithm is shown by applying it to backscattered signals received from a variety of numerical breast phantoms. The reconstructed images illustrate improvement in identification of embedded malignant tumors over the delay-and-sum algorithm. Successful detection and localization of tumors as small as 2 mm in diameter are also demonstrated.     

Segmentation of Lung Lobes in High-Resolution Isotropic CT Images

Modern multislice computed tomography (CT) scanners produce isotropic CT images with a thickness of 0.6 mm. These CT images offer detailed information of lung cavities, which could be used for better surgical planning of treating lung cancer. The major challenge for developing a surgical planning system is the automatic segmentation of lung lobes by identifying the lobar fissures. This paper presents a lobe segmentation algorithm that uses a two-stage approach: 1) adaptive fissure sweeping to find fissure regions and 2) wavelet transform to identify the fissure locations and curvatures within these regions. Tested on isotropic CT image stacks from nine anonymous patients with pathological lungs, the algorithm yielded an accuracy of 76.7%–94.8% with strict evaluation criteria. In comparison, surgeons obtain an accuracy of 80% for localizing the fissure regions in clinical CT images with a thickness of 2.5–7.0 mm. As well, this paper describes a procedure for visualizing lung lobes in three dimensions using software—amira—and the segmentation algorithm. The procedure, including the segmentation, needed about 5 min for each patient. These results provide promising potential for developing an automatic algorithm to segment lung lobes for surgical planning of treating lung cancer.