Objectives
To develop and test a new multifeature‐based computer‐aided diagnosis (CADx) scheme of lung cancer by fusing quantitative imaging (QI) features and serum biomarkers to improve CADx ...performance in classifying between malignant and benign pulmonary nodules.
Methods
First, a dataset involving 173 patients was retrospectively assembled, which includes computed tomography (CT) images and five serum biomarkers extracted from blood samples. Second, a CADx scheme using a four‐step–based semiautomatic segmentation method was applied to segment the targeted lung nodules, and compute 78 QI features from each segmented nodule from CT images. Third, two support vector machine (SVM) classifiers were built using QI features and serum biomarkers, respectively. SVM classifiers were trained and tested using the overall dataset with a Relief feature selection method, a synthetic minority oversampling technique and a leave‐one‐case‐out validation method. Finally, to further improve CADx performance, an information‐fusion method was used to combine the prediction scores generated by two SVM classifiers.
Results
Areas under receiver operating characteristic curves (AUC) generated by QI feature and serum biomarker‐based SVMs were 0.81 ± 0.03 and 0.69 ± 0.05, respectively. Using an optimal weighted fusion method to combine prediction scores generated by two SVMs, AUC value significantly increased to 0.85 ± 0.03 (P < 0.05).
Conclusions
This study demonstrates (a) higher CADx performance by using QI features than using the serum biomarkers and (b) feasibility of further improving CADx performance by fusion of QI features and serum biomarkers, which indicates that QI features and serum biomarkers contain the complementary classification information.
•3D tensor filtering and local image features help detect lung nodule.•Image features contributes to improve the performance of CAD scheme.•False-positive nodules are reduced by using random forest ...classifier.
Computer-aided detection (CAD) technology has been developed and demonstrated its potential to assist radiologists in detecting pulmonary nodules especially at an early stage. In this paper, we present a novel scheme for automatic detection of pulmonary nodules in CT images based on a 3D tensor filtering algorithm and local image feature analysis. We first apply a series of preprocessing steps to segment the lung volume and generate the isotropic volumetric CT data. Next, a unique 3D tensor filtering approach and local image feature analysis are used to detect nodule candidates. A 3D level set segmentation method is used to correct and refine the boundaries of nodule candidates subsequently. Then, we extract the features of the detected candidates and select the optimal features by using a CFS (Correlation Feature Selection) subset evaluator attribute selection method. Finally, a random forest classifier is trained to classify the detected candidates. The performance of this CAD scheme is validated using two datasets namely, the LUNA16 (Lung Nodule Analysis 2016) database and the ANODE09 (Automatic Nodule Detection 2009) database. By applying a 10-fold cross-validation method, the CAD scheme yielded a sensitivity of 79.3% at an average of 4 false positive detections per scan (FP/Scan) for the former dataset, and a sensitivity of 84.62% and 2.8 FP/Scan for the latter dataset, respectively. Our detection results show that the use of 3D tensor filtering algorithm combined with local image feature analysis constitutes an effective approach to detect pulmonary nodules.
Radiogenomics investigates radiographic imaging phenotypes associated with gene expression patterns. This study aims to explore relationships between CT imaging radiomics features and gene expression ...data in non-small cell lung cancer (NSCLC).
Eighty-nine NSCLC patients are included in the study. Radiomics features are extracted and selected to quantify the phenotype of tumors on CT-scans. Co-expressed genes are also clustered and the first principal component of the cluster is represented, which is defined as a metagene. Then, statistical analysis was performed to assess association of CT radiomics features with metagenes. In addition, predictive models are built and metagene enrichment are conducted to further evaluate performance of NSCLC radiogenomics statistically and biologically.
There are 187 significant pairwise correlations between a CT radiomics feature and a metagene of NSCLC, where eighteen metagenes are annotated with Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) terms. Metagenes are predicted in terms of radiomics features with an accuracy of 41.89% -89.93%.
This study reveals the associations between CT imaging radiomics features and NSCLC co-expressed gene sets. The findings suggest that CT radiomics features can reflect important biological information of NSCLC patients, which may have a significant clinical impact as CT is routinely used in clinical practice, assisting in improving medical decision-support at low cost.
Automatic segmentation of pulmonary airway tree is a challenging task in many clinical applications, including developing computer-aided detection and diagnosis schemes of lung diseases.
To segment ...the pulmonary airway tree from the computed tomography (CT) chest images using a novel automatic method proposed in this study.
This method combines a two-pass region growing algorithm with gray-scale morphological reconstruction and leakage elimination. The first-pass region growing is implemented to obtain a rough airway tree. The second-pass region growing and gray-scale morphological reconstruction are used to detect the distal airways. Finally, leakage detection is performed to remove leakage and refine the airway tree.
Our methods were compared with the gold standards. Forty-five clinical CT lung image scan cases were used in the experiments. Statistics on tree division order, branch number, and airway length were adopted for evaluation. The proposed method detected up to 12 generations of bronchi. On average, 148.85 branches were extracted with a false positive rate of 0.75%.
The results show that our method is accurate for pulmonary airway tree segmentation. The strategy of separating the leakage detection from the segmenting process is feasible and promising for ensuring a high branch detected rate with a low leakage volume.
Highlights • A new CAD scheme for pulmonary nodule detection is proposed. • Dynamic self-adaptive template matching is used to detect nodules. • FLDA classifier can filter false positive detection ...nodules.
Automatic segmentation of pulmonary vascular tree in the thoracic computed tomography (CT) image is a promising but challenging task with great clinical potential values. It is difficult to segment ...the whole vascular tree in reasonable time and acceptable accuracy.
To develop a novel pulmonary vessel segmentation approach by incorporating vessel enhancement filters and the anisotropic diffusion filter with the variational region growing.
First, the airway wall from the lung lobes is eliminated from CT images by using multi-scale morphological operations. Second, a Hessian-based multi-scale vesselness filter and medialness filter are applied to detect and enhance the potential vessel. Third, an anisotropic diffusion filter is used to remove noise and enhance the tube-like structures in CT images. Last, the vascular tree is segmented by applying variational region growing algorithm.
Applying to the CT images collected from the entire dataset of VESSEL12 challenge, we achieved an average sensitivity of 92.9%, specificity of 91.6% and the area under the ROC curve of AUC = 0.972.
This study demonstrated feasibility of segmenting the pulmonary vessel effectively by incorporating vessel enhancement filters and the anisotropic diffusion filter with the variational region growing algorithm. Our method cannot only segment both large and peripheral vessels, but also distinguish the vessels from the adjacent tissues, especially the airway walls.
It is common sense that CAD has great significance in the lung nodule detection. But it is still controversial whether the CAD can also automatically differentiates between malignant and benign ...pulmonary nodules. The primary cause of this controversy is due to the subjective definition of 9 characteristics of nodules which are important basis of nodule identification. In other word, these characteristics are too dependent on the doctor scoring, and no objective standard of them has built which make these characteristics can be obtained by calculation.The main aim of this paper is to establish a quantitative method of the characteristics and refine these nine characteristics. This new method is used to find the objective replacement (a series features which can be measured through algorithms) of these subjective characteristics of the pulmonary nodule detection with Bayesian analysis.The experiment of our method proves that it is feasible to substitute the features of Pulmonary Nodule obtained by calculating for the characteristics of the nodule which only used to be gotten by the subjective judgment of doctors.
A method based on two passes of 3D region growing and morphological reconstruction for segmenting pulmonary airway tree from computed tomography (CT) chest scans is presented to solve the problem of ...leakage and under-segmentation caused by the partial volume effect and motion artifact. Firstly, the first pass of 3D region growing with optimal threshold range is used to extract the rough airway. Then, three location maps of possible distal bronchi are located by using the grayscale morphological reconstruction on axial, coronal and sagittal slices respectively. Finally, on basis of rough airway extracted in first pass of 3D region growing, the second pass of 3D region growing constrained by the three location maps is implemented to obtain the completed airway. 25 clinical CT scans with thickness between 0.75 mm and 2 mm were used to test the proposed method by recording the number of tracheal branches of each order, the total number of tracheal branches and the average number of branches. Up to 12 generations of bronchi and average 156 branches were detected in the experiment which proves that our adaptive and automated method can segment the pulmonary airway with a better performance.
In order to diagnose Parkinson disease (PD) at an early stage, it is important to develop a sensitive method for detecting structural changes in the substantia nigra (SN). Diffusion weighted imaging ...(DWI) and diffusion tensor imaging (DTI) have become important tools in supporting diagnosis of PD, with findings based on increased apparent diffusion coefficients (ADCs) in basal ganglia and decreased fractional anisotropy (FA) in SN. Based on the hypothesis that a diffusion kurtosis imaging (DKI) theory is a valuable method for PD diagnosis based on the non-Gaussian diffusion of water in biologic systems, the purpose of this study is to develop an image processing scheme (software) based on Image-J for the facilitating the application of DKI to assist PD diagnosis. Using the new DKI software enables to estimate the diffusional kurtosis and diffusion coefficients, which reflect the structural differences between regions of interest. The experimental results of applying the new software showed that diffusional kurtosis was highly sensitive to microstructural tissue changes, which were not noticeable in the diffusion coefficient values. Thus, the study results may suggest that applying the new image processing software can be useful for assessing tissue structural abnormalities, monitoring and following disease progression.
A New Error-Fitting Inversion Method for 2-D NMR Spectrum Nie, Sheng Dong; Zhou, Xiao Long
Applied Mechanics and Materials,
02/2014, Volume:
513-517, Issue:
Applied Science, Materials Science and Information Technologies in Industry
Journal Article
Peer reviewed
Two-dimensional (2-D) inversion algorithms with Nuclear Magnetic Resonance had been intensively studied all over the world. To improve the existing method, a new Error-Fitting inversion method was ...proposed. The proposal algorithm consists of two steps: firstly, error extraction and estimation is implemented; the next step is to fit the measured data to such a state that the fitting residual and the estimated noise are in a comparable level. This method is applied to simulation and experiments. The results have shown great robustness and accuracy of the algorithm. In conclusion, the proposed Error-Fitting inversion method can meet different application requirements.