Lung CAD Research Articles

Diagnostic performance of artificial intelligence for pediatric pulmonary nodule detection on chest computed tomography: comparison of simulated lower radiation doses.

The Lung CAD shows low sensitivity at all simulated lower doses for the detection of pulmonary nodules in this pediatric population. However, radiation dose may be reduced from standard without further compromise to the Lung CAD performance. • High diagnostic performance of Lung CAD for detection of pulmonary nodules in adults. • Several imaging techniques are applied to reduce pediatric radiation dose. • Low sensitivity at all simulated lower doses for the detection of pulmonary nodules in our pediatric population. • Radiation dose may be reduced from standard without further compromise to the Lung CAD performance.

Research on lung nodule recognition algorithm based on deep feature fusion and MKL-SVM-IPSO

Lung CAD system can provide auxiliary third-party opinions for doctors, improve the accuracy of lung nodule recognition. The selection and fusion of nodule features and the advancement of recognition algorithms are crucial improving lung CAD systems. Based on the HDL model, this paper mainly focuses on the three key algorithms of feature extraction, feature fusion and nodule recognition of lung CAD system. First, CBAM is embedded into VGG16 and VGG19, and feature extraction models AE-VGG16 and AE-VGG19 are constructed, so that the network can pay more attention to the key feature information in nodule description. Then, feature dimensionality reduction based on PCA and feature fusion based on CCA are sequentially performed on the extracted depth features to obtain low-dimensional fusion features. Finally, the fusion features are input into the proposed MKL-SVM-IPSO model based on the improved Particle Swarm Optimization algorithm to speed up the training speed, get the global optimal parameter group. The public dataset LUNA16 was selected for the experiment. The results show that the accuracy of lung nodule recognition of the proposed lung CAD system can reach 99.56%, and the sensitivity and F1-score can reach 99.3% and 0.9965, respectively, which can reduce the possibility of false detection and missed detection of nodules.

Improved cost-sensitive multikernel learning support vector machine algorithm based on particle swarm optimization in pulmonary nodule recognition

In the lung computer-aided detection (Lung CAD) system, the region of interest (ROI) of lung nodules has more false positives, making the imbalance between positive and negative (true positive and false positive) samples more likely to lead to misclassification of true positive nodules, a cost-sensitive multikernel learning support vector machine (CS-MKL-SVM) algorithm is proposed. Different penalty coefficients are assigned to positive and negative samples, so that the model can better learn the features of true positive nodules and improve the classification effect. To further improve the detection rate of pulmonary nodules and overall recognition accuracy, a score function named F-new based on the harmonic mean of accuracy (ACC) and sensitivity (SEN) is proposed as a fitness function for subsequent particle swarm optimization (PSO) parameter optimization, and a feasibility analysis of this function is performed. Compared with the fitness function that considers only accuracy or sensitivity, both the detection rate and the recognition accuracy of pulmonary nodules can be improved by this new algorithm. Compared with the grid search algorithm, using PSO for parameter search can reduce the model training time by nearly 20 times and achieve rapid parameter optimization. The maximum F-new obtained on the test set is 0.9357 for the proposed algorithm. When the maximum value of F-new is achieved, the corresponding recognition ACC is 91%, and SEN is 96.3%. Compared with the radial basis function in the single kernel, the F-new of the algorithm in this paper is 2.16% higher, ACC is 1.00% higher and SEN is equal. Compared with the polynomial kernel function in the single kernel, the F-new of the algorithm is 3.64% higher, ACC is 1.00% higher and SEN is 7.41% higher. The experimental results show that the F-new, ACC and SEN of the proposed algorithm is the best among them, and the results obtained by using multikernel function combined with F-new index are better than the single kernel function. Compared with the MKL-SVM algorithm of grid search, the ACC of the algorithm in this paper is reduced by 1%, and the results are equal to those of the MKL-SVM algorithm based on PSO only. Compared with the above two algorithms, SEN is increased by 3.71% and 7.41%, respectively. Therefore, it can be seen that the cost sensitive method can effectively reduce the missed detection of nodules, and the availability of the new algorithm can be further verified.

Research on Key Algorithms of the Lung CAD System Based on Cascade Feature and Hybrid Swarm Intelligence Optimization for MKL-SVM.

Feature selection and lung nodule recognition are the core modules of the lung computer-aided detection (Lung CAD) system. To improve the performance of the Lung CAD system, algorithmic research is carried out for the above two parts, respectively. First, in view of the poor interpretability of deep features and the incomplete expression of clinically defined handcrafted features, a feature cascade method is proposed to obtain richer feature information of nodules as the final input of the classifier. Second, to better map the global characteristics of samples, the multiple kernel learning support vector machine (MKL-SVM) algorithm with a linear convex combination of polynomial kernel and sigmoid kernel is proposed. Furthermore, this paper applied the methods for speed contraction factor and roulette strategy, and a mixture of simulated annealing (SA) and particle swarm optimization (PSO) is used for global optimization, so as to solve the problem that the PSO is easy to lose particle diversity and fall into the local optimal solution as well as improve the model's training speed. Therefore, the MKL-SVM algorithm is presented in this paper, which is based on swarm intelligence optimization is proposed for lung nodule recognition. Finally, the algorithm construction experiments are conducted on the cooperative hospital dataset and compared with 8 advanced algorithms on the public dataset LUNA16. The experimental results show that the proposed algorithms can improve the accuracy of lung nodule recognition and reduce the missed detection of nodules.

Corrections to “An Automatic Approach for Learning and Tuning Gaussian Interval Type-2 Fuzzy Membership Functions Applied to Lung CAD Classification System” [Apr 12 224-234

The potential of type-2 fuzzy sets to manage high levels of uncertainty in the subjective knowledge of experts or of numerical information has focused on control and pattern classification systems in recent years. One of the main challenges in designing a type-2 fuzzy logic system (FLS) is how to estimate the parameters of the type-2 fuzzy membership function (T2MF) and the footprint of uncertainty (FOU) from imperfect and noisy datasets. This paper presents an automatic approach to learn and tune Gaussian interval type-2 membership functions (IT2MFs) with application to multidimensional pattern classification problems. T2MFs and their FOUs are tuned according to the uncertainties in the training dataset by a combination of genetic algorithm (GA) and cross-validation techniques. In our GA-based approach, the structure of the chromosome has fewer genes than other GA methods, and chromosome initialization is more precise. The proposed approach addresses the application of the interval type-2 fuzzy logic system (IT2FLS) for the problem of nodule classification in a lung computer-aided detection system. The designed IT2FLS is compared with its type-1 fuzzy logic system (T1FLS) counterpart. The results demonstrate that the IT2FLS outperforms the T1FLS by more than 30% in terms of classification accuracy.

Pulmonary Nodule Recognition Based on Multiple Kernel Learning Support Vector Machine-PSO.

Pulmonary nodule recognition is the core module of lung CAD. The Support Vector Machine (SVM) algorithm has been widely used in pulmonary nodule recognition, and the algorithm of Multiple Kernel Learning Support Vector Machine (MKL-SVM) has achieved good results therein. Based on grid search, however, the MKL-SVM algorithm needs long optimization time in course of parameter optimization; also its identification accuracy depends on the fineness of grid. In the paper, swarm intelligence is introduced and the Particle Swarm Optimization (PSO) is combined with MKL-SVM algorithm to be MKL-SVM-PSO algorithm so as to realize global optimization of parameters rapidly. In order to obtain the global optimal solution, different inertia weights such as constant inertia weight, linear inertia weight, and nonlinear inertia weight are applied to pulmonary nodules recognition. The experimental results show that the model training time of the proposed MKL-SVM-PSO algorithm is only 1/7 of the training time of the MKL-SVM grid search algorithm, achieving better recognition effect. Moreover, Euclidean norm of normalized error vector is proposed to measure the proximity between the average fitness curve and the optimal fitness curve after convergence. Through statistical analysis of the average of 20 times operation results with different inertial weights, it can be seen that the dynamic inertial weight is superior to the constant inertia weight in the MKL-SVM-PSO algorithm. In the dynamic inertial weight algorithm, the parameter optimization time of nonlinear inertia weight is shorter; the average fitness value after convergence is much closer to the optimal fitness value, which is better than the linear inertial weight. Besides, a better nonlinear inertial weight is verified.

Maximum-Intensity-Projection and Computer-Aided-Detection Algorithms as Stand-Alone Reader Devices in Lung Cancer Screening Using Different Dose Levels and Reconstruction Kernels.

The objective of our study was to evaluate lung nodule detection rates on standard and microdose chest CT with two different computer-aided detection systems (SyngoCT-CAD, VA 20, Siemens Healthcare [CAD1]; Lung CAD, IntelliSpace Portal DX Server, Philips Healthcare [CAD2]) as well as maximum-intensity-projection (MIP) images. We also assessed the impact of different reconstruction kernels. Standard and microdose CT using three reconstruction kernels (i30, i50, i70) was performed with an anthropomorphic chest phantom. We placed 133 ground-glass and 133 solid nodules (diameters of 5 mm, 8 mm, 10 mm, and 12 mm) in 55 phantoms. Four blinded readers evaluated the MIP images; one recorded the results of CAD1 and CAD2. Sensitivities for CAD and MIP nodule detection on standard dose and microdose CT were calculated for each reconstruction kernel. Dose for microdose CT was significantly less than that for standard-dose CT (0.1323 mSv vs 1.65 mSv; p < 0.0001). CAD1 delivered superior results compared with CAD2 for standard-dose and microdose CT (p < 0.0001). At microdose level, the best stand-alone sensitivity (97.6%) was comparable with CAD1 sensitivity (96.0%; p = 0.36; both with i30 reconstruction kernel). Pooled sensitivities for all nodules, doses, and reconstruction kernels on CAD1 ranged from 88.9% to 97.3% versus 49.6% to 73.9% for CAD2. The best sensitivity was achieved with standard-dose CT, i50 kernel, and CAD1 (97.3%) versus 96% with microdose CT, i30 or i50 kernel, and CAD1. MIP images and CAD1 had similar performance at both dose levels (p = 0.1313 and p = 0.48). Submillisievert CT is feasible for detecting solid and ground-glass nodules that require soft-tissue kernels for MIP and CAD systems to achieve acceptable sensitivities. MIP reconstructions remain a valuable adjunct to the interpretation of chest CT for increasing sensitivity and have the advantage of significantly lower false-positive rates.

An Automatic Approach for Learning and Tuning Gaussian Interval Type-2 Fuzzy Membership Functions Applied to Lung CAD Classification System

The potential of type-2 fuzzy sets to manage high levels of uncertainty in the subjective knowledge of experts or of numerical information has focused on control and pattern classification systems in recent years. One of the main challenges in designing a type-2 fuzzy logic system (FLS) is how to estimate the parameters of the type-2 fuzzy membership function (T2MF) and the footprint of uncertainty (FOU) from imperfect and noisy datasets. This paper presents an automatic approach to learn and tune Gaussian interval type-2 membership functions (IT2MFs) with application to multidimensional pattern classification problems. T2MFs and their FOUs are tuned according to the uncertainties in the training dataset by a combination of genetic algorithm (GA) and cross-validation techniques. In our GA-based approach, the structure of the chromosome has fewer genes than other GA methods, and chromosome initialization is more precise. The proposed approach addresses the application of the interval type-2 fuzzy logic system (IT2FLS) for the problem of nodule classification in a lung computer-aided detection system. The designed IT2FLS is compared with its type-1 fuzzy logic system (T1FLS) counterpart. The results demonstrate that the IT2FLS outperforms the T1FLS by more than 30% in terms of classification accuracy.

Automated classification of lung bronchovascular anatomy in CT using AdaBoost.

Lung CAD systems require the ability to classify a variety of pulmonary structures as part of the diagnostic process. The purpose of this work was to develop a methodology for fully automated voxel-by-voxel classification of airways, fissures, nodules, and vessels from chest CT images using a single feature set and classification method. Twenty-nine thin section CT scans were obtained from the Lung Image Database Consortium (LIDC). Multiple radiologists labeled voxels corresponding to the following structures: airways (trachea to 6th generation), major and minor lobar fissures, nodules, and vessels (hilum to peripheral), and normal lung parenchyma. The labeled data was used in conjunction with a supervised machine learning approach (AdaBoost) to train a set of ensemble classifiers. Each ensemble classifier was trained to detect voxels part of a specific structure (either airway, fissure, nodule, vessel, or parenchyma). The feature set consisted of voxel attenuation and a small number of features based on the eigenvalues of the Hessian matrix (used to differentiate structures by shape). When each ensemble classifier was composed of 20 weak classifiers, the AUC values for the airway, fissure, nodule, vessel, and parenchyma classifiers were 0.984+/-0.011, 0.949+/-0.009, 0.945+/-0.018, 0.953+/-0.016, and 0.931+/-0.015, respectively. The strong results suggest that this could be an effective input to higher-level anatomical based segmentation models with the potential to improve CAD performance.

SU‐FF‐I‐03: Evaluation of Computer Aided Diagnosis On Thin‐Slice Pulmonary CT Images

Purpose: Pulmonary stereotactic radiation therapy achieves a &gt;90% local control rate for small tumors (&lt;30mm); motivating early detection of small lung nodules. With the increasing image resolution afforded by multi‐slice CT scanners, radiologists can discern more easily these very small lung nodules. However, to overcome the challenge of image overload, assistance from lung CAD systems becomes a necessity. The purpose of this work is to evaluate the performance of our lung CAD algorithm on low‐dose, thin‐slice CT images from the Early Lung Cancer Action Program (ELCAP) database. Method and Materials: Our CAD algorithm models lung nodules as spheres (a modification of the 3D template‐matching method presented in AAPM'03) and searched for structures similar to the templates throughout the whole lung volume. This approach eliminates the need for segmentation of blood vessel‐nodule complex and is insensitive to noise and low‐contrast of some nodules. In the extraction of lung volume, a watershed transform was applied to separate left and right lungs when necessary. A 3D spherical structure element was used to smooth out the surface of both lungs in order to incorporate pleural nodules. An improved padding scheme was adapted on the templates to make the detection capacity of templates more isotropic. A test dataset was used, consisting of 55 central contiguous slices (voxel size: 0.57×0.57×1.25mm3) from a particularly challenging subject in the ELCAP database. 41 nodules were identified by an experienced radiologist from our institution. Results: 37 out of 41 nodules were detected by our CAD algorithm, a sensitivity of 90.2%. Among them, 28 out of 32 nodules with diameter &lt;3mm were detected. The smallest detected nodule had a diameter of only ∼3 pixels (1.7mm) and a volume of ∼17 voxels. Conclusion: Our CAD method is capable of detecting with high sensitivity very small lung nodules in low‐dose thin‐slice chest CT.