Computer-aided detection of lung nodules: false positive reduction using a 3D gradient field method and 3D ellipsoid fitting.

We are developing a computer-aided detection system to aid radiologists in diagnosing lung cancer in thoracic computed tomographic (CT) images. The purpose of this study was to improve the false-positive (FP) reduction stage of our algorithm by developing and incorporating a gradient field technique. This technique extracts 3D shape information from the gray-scale values within a volume of interest. The gradient field feature values are higher for spherical objects, and lower for elongated and irregularly-shaped objects. A data set of 55 thin CT scans from 40 patients was used to evaluate the usefulness of the gradient field technique. After initial nodule candidate detection and rule-based first stage FP reduction, there were 3487 FP and 65 true positive (TP) objects in our data set. Linear discriminant classifiers with and without the gradient field feature were designed for the second stage FP reduction. The accuracy of these classifiers was evaluated using the area Az under the receiver operating characteristic (ROC) curve. The Az values were 0.93 and 0.91 with and without the gradient field feature, respectively. The improvement with the gradient field feature was statistically significant (p=0.01).

Automated Computer-Aided Detection of Lung Nodules in Metastatic Colorectal Cancer Patients for the Identification of Pulmonary Oligometastatic Disease

Diminution index: A novel 3D feature for pulmonary nodule detection

Computer-aided Lung Nodule Detection in CT: Results of Large-Scale Observer Test 1

The authors presented a novel system for automated nodule detection in lung CT exams. The approach is based on (1) a lung tissue segmentation preprocessing step, composed of histogram thresholding, seeded region growing, and mathematical morphology; (2) a filtering step, whose aim is the preliminary detection of candidate nodules (via 3D fast radial filtering) and estimation of their geometrical features (via scale space analysis); and (3) a false positive reduction (FPR) step, comprising a heuristic FPR, which applies thresholds based on geometrical features, and a supervised FPR, which is based on support vector machines classification, which in turn, is enhanced by a feature extraction algorithm based on maximum intensity projection processing and Zernike moments. The system was validated on 154 chest axial CT exams provided by the lung image database consortium public database. The authors obtained correct detection of 71% of nodules marked by all radiologists, with a false positive rate of 6.5 false positives per patient (FP/patient). A higher specificity of 2.5 FP/patient was reached with a sensitivity of 60%. An independent test on the ANODE09 competition database obtained an overall score of 0.310. The system shows a novel approach to the problem of lung nodule detection in CT scans: It relies on filtering techniques, image transforms, and descriptors rather than region growing and nodule segmentation, and the results are comparable to those of other recent systems in literature and show little dependency on the different types of nodules, which is a good sign of robustness.

Computer Aided Detection (CAD) systems provide a second opinion to radiologists in detecting lung cancer by providing automated analysis of the scans. The proposed CAD system consists of five processing steps: image acquisition, preprocessing, lung segmentation, nodule detection and false positive reduction. First, 400 CT scans are downloaded from the Lung Image Database Consortium (LIDC). Preprocessing is implemented using contrast stretching and enhancing. Lung segmentation and nodule detection stages are performed using a combination of region growing, thresholding and morphological operations. Each 3D structure is then subjected to tabular structure elimination to provide nodule candidates. In the false positive reduction stage, some of the basic nodule features are extracted from the training data to set thresholds for a simple rule-based classifier. The CAD achieved sensitivity of 77.77%, specificity of 69.5% and accuracy 70.53 % with an average 4.1 FPs/scan.

Early detection of lung cancer is crucial in reducing mortality. Magnetic resonance imaging (MRI) may be a viable imaging technique for lung cancer detection. Numerous lung nodule detection methods have been studied for computed tomography (CT) images. However, to the best of our knowledge, no detection methods have been carried out for the MR images. In this paper, a lung nodule detection method based on deep learning is proposed for thoracic MR images. With parameter optimizing, spatial three-channel input construction, and transfer learning, a faster R-convolution neural network (CNN) is designed to locate the lung nodule region. Then, a false positive (FP) reduction scheme based on anatomical characteristics is designed to reduce FPs and preserve the true nodule. The proposed method is tested on 142 T2-weighted MR scans from the First Affiliated Hospital of Guangzhou Medical University. The sensitivity of the proposed method is 85.2% with 3.47 FPs per scan. The experimental results demonstrate that the designed faster R-CNN network and the FP reduction scheme are effective in the lung nodule detection and the FP reduction for MR images.

A major challenge in the latest computer-aided detection (CADe) of polyps in CT colonography (CTC) is to improve the false positive (FP) rate while maintaining detection sensitivity. Radiologists prefer CADe system produce small number of false positive detections, otherwise they might not consider CADe system improve their workflow. Towards this end, in this study, we applied a nonlinear regression model operating on CTC image voxels directly and a nonlinear classification model with extracted image features based on support vector machines (SVMs) in order to improve the specificity of CADe of polyps. We investigated the feasibility of a support vector regression (SVR) in the massive-training framework, and we developed a massive-training SVR (MTSVR) in order to reduce the long training time associated with the massive-training artificial neural network (MTANN) for reduction of FPs in CADe of polyps in CTC. In addition, we proposed a feature selection method directly coupled with an SVM classifier to maximize the CADe system performance. We compared the proposed feature selection method with the conventional stepwise feature selection based on Wilks' lambda with a linear discriminant analysis classifier. The FP reduction system based on the proposed feature selection method was able to achieve a 96.0% by-polyp sensitivity with an FP rate of 4.1 per patient. The performance is better than that of the stepwise feature selection based on Wilks' lambda (which yielded the same sensitivity with 18.0 FPs/patient). To test the performance of the proposed MTSVR, we compared it with the original MTANN in the distinction between actual polyps and various types of FPs in terms of the training time reduction and FP reduction performance. The CTC database used in this study consisted of 240 CTC datasets obtained from 120 patients in the supine and prone positions. With MTSVR, we reduced the training time by a factor of 190, while achieving a performance (by-polyp sensitivity of 94.7% with 2.5 FPs/patient) comparable to that of the original MTANN (which has the same sensitivity with 2.6 FPs/patient).

Early detection of lung cancer is an effective way to improve the survival rate of patients. It is a critical step to have accurate detection of lung nodules in computed tomography (CT) images for the diagnosis of lung cancer. However, due to the heterogeneity of the lung nodules and the complexity of the surrounding environment, it is a challenge to develop a robust nodule detection method. In this study, we propose a two-stage convolutional neural networks (TSCNN) for lung nodule detection. The first stage based on the improved U-Net segmentation network is to establish an initial detection of lung nodules. During this stage, in order to obtain a high recall rate without introducing excessive false positive nodules, we propose a new sampling strategy for training. Simultaneously, a two-phase prediction method is also proposed in this stage. The second stage in the TSCNN architecture based on the proposed dual pooling structure is built into three 3D-CNN classification networks for false positive reduction. Since the network training requires a significant amount of training data, we designed a random mask as the data augmentation method in this study. Furthermore, we have improved the generalization ability of the false positive reduction model by means of ensemble learning. We verified the proposed architecture on the LUNA dataset in our experiments, which showed that the proposed TSCNN architecture did obtain competitive detection performance.

Initial candidate segmentation is an important task in lung nodule detection. If an accurate segmentation is used, the false positives (FPs) can be decreased in subsequent stages and the desired region of each candidate can be distinguished. In this paper, a new hybrid method for nodule candidate segmentation and FPs reduction is proposed. First, the images are transferred to the neutrosophic domain. Then, three filters, named blob-like structure enhancement (BSE), line structure enhancement (LSE), and central adaptive medialness (CAM) filters, are used for filtering the output of the last step. Afterward, the outputs of BSE, LSE, and CAM filters are used for initial candidate detection and candidate segmentation, respectively. Also, line tracking method is proposed for extending the candidate voxels, and then, it is used for candidate segmentation. After feature extraction, the sparse coding is used for learning feature vector. In the last step, the generalized linear regression model (GLRM) is used for classification. The output of classifier for sensitivity and FP/scan is 98.32% and 2.8, respectively. AUC values of different link functions in GLRM, before and after feature learning, were also calculated, and the best value of AUC was obtained with probit link function after using the sparse coding method. The experimental results demonstrate the power of the proposed algorithm in nodules detection and false positive reduction.

In this paper, we employed neural network ensemble for FPs reduction in detecting lung nodules in chest radiographs. In our scheme, the ensemble consists of four modified forward neural networks, each one of them was trained with the back propagation algorithm to distinct a different type of non-nodules from nodules. The outputs of all the individual neural networks were combined by a modified forward mixing ANN. The performance of our scheme for false positive reduction was evaluated by use of FROC. With neural network ensemble, the false positive rate of CAD scheme1 was reduced for 44% (from 2.86 to 1.6 positives per image), at an overall sensitivity of 60%. We also compared our scheme with other researches. The result demonstrates the superiority of it over other ones. We believe that the proposed method is useful in false positives reduction in the diagnosis of lung nodules in chest radiograph.

Computer-aided diagnoses programs are developed for alerting the radiologist by indicating potential sites of lesions. One of the important tasks in the development of a computational system for detecting lung nodules is to diminish the number of false positives keeping on high sensitivities. In this work we describe a system for automatic lung nodule detection. The detection is carried out in several stages. First, a knowledge-based segmentation process delimits the lung boundaries. Then, a progressive thresholding of an image in which the conspicuity of nodules has been enhanced by means of filter matching and a set of growth and circularity tests fix the areas suspicious of being nodules into region previously labelled as lungs. Finally, these suspicious regions are confirmed as nodules in a new feature (curvature) space, which gives us an important help in the task of distinguishing true and false nodules from previously extracted suspicious regions. Preliminary results are very promising, achieving high sensitivities with a little ratio of false positives.

Background and ObjectiveTo investigate the effect of the slab thickness in maximum intensity projections (MIPs) on the candidate detection performance of a deep learning-based computer-aided detection (DL-CAD) system for pulmonary nodule detection in CT scans. MethodsThe public LUNA16 dataset includes 888 CT scans with 1186 nodules annotated by four radiologists. From those scans, MIP images were reconstructed with slab thicknesses of 5 to 50 mm (at 5 mm intervals) and 3 to 13 mm (at 2 mm intervals). The architecture in the nodule candidate detection part of the DL-CAD system was trained separately using MIP images with various slab thicknesses. Based on ten-fold cross-validation, the sensitivity and the F2 score were determined to evaluate the performance of using each slab thickness at the nodule candidate detection stage. The free-response receiver operating characteristic (FROC) curve was used to assess the performance of the whole DL-CAD system that took the results combined from 16 MIP slab thickness settings. ResultsAt the nodule candidate detection stage, the combination of results from 16 MIP slab thickness settings showed a high sensitivity of 98.0% with 46 false positives (FPs) per scan. Regarding a single MIP slab thickness of 10 mm, the highest sensitivity of 90.0% with 8 FPs/scan was reached before false positive reduction. The sensitivity increased (82.8% to 90.0%) for slab thickness of 1 to 10 mm and decreased (88.7% to 76.6%) for slab thickness of 15–50 mm. The number of FPs was decreasing with increasing slab thickness, but was stable at 5 FPs/scan at a slab thickness of 30 mm or more. After false positive reduction, the DL-CAD system, utilizing 16 MIP slab thickness settings, had the sensitivity of 94.4% with 1 FP/scan. ConclusionsThe utilization of multi-MIP images could improve the performance at the nodule candidate detection stage, even for the whole DL-CAD system. For a single slab thickness of 10 mm, the highest sensitivity for pulmonary nodule detection was reached at the nodule candidate detection stage, similar to the slab thickness usually applied by radiologists.

This paper examines the effectiveness of geometric feature descriptors, common in computer vision, for false positive reduction and for classification of lung nodules in low dose CT (LDCT) scans. A data-driven lung nodule modeling approach creates templates for common nodule types, using active appearance models (AAM); which are then used to detect candidate nodules based on optimum similarity measured by the normalized cross-correlation (NCC). Geometric feature descriptors (e.g., SIFT, LBP and SURF) are applied to the output of the detection step, in order to extract features from the nodule candidates, for further enhancement of output and possible reduction of false positives. Results on the clinical ELCAP database showed that the descriptors provide 2% enhancements in the specificity of the detected nodule above the NCC results when used in a k-NN classifier. Thus quantitative measures of enhancements of the performance of CAD models based on LDCT are now possible and are entirely model-based. Most importantly, our approach is applicable for classification of nodules into categories and pathologies.

A survey and taxonomy of 2.5D approaches for lung segmentation and nodule detection in CT images