Automated CT image evaluation of the lung: a morphology-based concept

Association Between Lung Parenchymal Attenuation in Computed Tomography and Airflow Limitation in Adults with Cystic Fibrosis

Objectives: To determine the association between airflow limitation and the quantification of lung attenuation in computed tomography (CT) in adult patients with cystic fibrosis (CF). Methods: A cross-sectional study in a single center between January 2013 and December 2018 in adult patients with stable CF. We collected clinical data and the results of spirometry and plethysmography. A chest CT at inspiration and expiration, using a specific software that automatically measured the lung attenuation, was performed. Results: In total, 73 patients (63% males) were included. The mean age was 31.6 ± 12.3 years and the FEV1 was 67.8 ± 25.9% pred. An airflow limitation was found in 63%, the mean residual volume was 159.9% pred, and air trapping was observed in 50 (87.7%) of the patients. The patients with airflow limitations showed a higher bulla index and a percentage of lung voxels in the range of emphysema. The FEV1 and the FEV1/FVC correlated with the percentage of the lungs at a high attenuation value (HAV), the range of emphysema, and the bulla index at inspiration, as well as the mean lung density at expiration and the inspiratory-expiratory variation of the mean lung density (MLDi-e). Finally, in the multivariate model, the MLDi-e and the HAV at inspiration were associated with airflow limitations. Conclusions: The measurements obtained from the automated quantification of lung parenchymal attenuation predicts airflow limitation in CF.

Pulmonary volumes and signs of chronic airflow limitation in quantitative computed tomography

BACKGROUND

Computed tomography (CT) offers pulmonary volumetric quantification but is not commonly used in healthy individuals due to radiation concerns. Chronic airflow limitation (CAL) is one of the diagnostic criteria for chronic obstructive pulmonary disease (COPD), where early diagnosis is important. Our aim was to present reference values for chest CT volumetric and radiodensity measurements and explore their potential in detecting early signs of CAL.

METHODS

From the population-based Swedish CArdioPulmonarybioImage Study (SCAPIS), 294 participants aged 50-64, were categorized into non-CAL (n = 258) and CAL (n = 36) groups based on spirometry. From inspiratory and expiratory CT images we compared lung volumes, mean lung density (MLD), percentage of low attenuation volume (LAV%) and LAV cluster volume between groups, and against reference values from static pulmonary function test (PFT).

RESULTS

The CAL group exhibited larger lung volumes, higher LAV%, increased LAV cluster volume and lower MLD compared to the non-CAL group. Lung volumes significantly deviated from PFT values. Expiratory measurements yielded more reliable results for identifying CAL compared to inspiratory. Using a cut-off value of 0.6 for expiratory LAV%, we achieved sensitivity, specificity and positive/negative predictive values of 72%, 85% and 40%/96%, respectively.

CONCLUSION

We present volumetric reference values from inspiratory and expiratory chest CT images for a middle-aged healthy cohort. These results are not directly comparable to those from PFTs. Measures of MLD and LAV can be valuable in the evaluation of suspected CAL. Further validation and refinement are necessary to demonstrate its potential as a decision support tool for early detection of COPD.

Artificial intelligence in radiology

The interest in artificial intelligence (AI) has ballooned within radiology in the past few years primarily due to notable successes of deep learning. With the advances brought by deep learning, AI has the potential to recognize and localize complex patterns from different radiological imaging modalities, many of which even achieve comparable performance to human decision-making in recent applications. In this chapter, we review several AI applications in radiology for different anatomies: chest, abdomen, pelvis, as well as general lesion detection/identification that is not limited to specific anatomies. For each anatomy site, we focus on introducing the tasks of detection, segmentation, and classification with an emphasis on describing the technology development pathway with the aim of providing the reader with an understanding of what AI can do in radiology and what still needs to be done for AI to better fit in radiology. Combining with our own research experience of AI in medicine, we elaborate how AI can enrich knowledge discovery, understanding, and decision-making in radiology, rather than replacing the radiologist.

Categorization & Recognition of Lung Tumor Using Machine Learning Representations

BACKGROUND

Lung Cancer is the disease spreading around the world nowadays. Early recognition of lung disease is a difficult task as the cells which cause tumor will grow quickly and the majority of these cells are enclosed with each other. From the beginning of the treatment, tumor detection handling systems which are generally utilized for the diagnosis of lung cancer, recognizable proof of hereditary and ecological elements is imperative in creating a novel technique for lung tumor detection. In different cancers, for example, lung cancer, the time calculated is imperative to find the anomaly issue in target images.

METHODS

In this proposed framework, GLCM (Gray Level Co-event Matrix) is utilized for preprocessing of images and to feature extraction procedures to check the condition of the patient whether it is ordinary or irregular. Surface-based elements, for example, GLCM (Gray Level Co-event Matrix) features assume a vital part of remedial image examination which is utilized for the identification of Lung cancer. In the event that lung cancer is effectively distinguished and anticipated in its initial stages, it lessens numerous treatment choices and furthermore, decreases the danger of intrusive surgery and increment survival rate.

RESULTS & CONCLUSION

The proposed method will efficiently identify the position of the tumor in lungs using the probability framework. This will offer a promising outcome for recognition and diagnosis of lung cancer. In this manuscript, GLCM features are used for the prediction of lung tumor and tests are performed for performance analysis in comparison with the histogram and GLCM features, in which GLCM features are accurate in predicting lung tumor even if it takes more time than histogram features. In this manner, early discovery and probability of lung cancer should assume a crucial task in finding a procedure and furthermore, an increment in the survival rate of the patient. This exploration investigates machine learning systems which consider quality articulation, to perceive cancer or to identify lung cancer.

Predictive Modelling of Lung Function using Emphysematous Density Distribution

Target lung tissue selection remains a challenging task to perform for treating severe emphysema with lung volume reduction (LVR). In order to target the treatment candidate, the percentage of low attenuation volume (LAV%) representing the proportion of emphysema volume to whole lung volume is measured using computed tomography (CT) images. Although LAV% have shown to have a correlation with lung function in patients with chronic obstructive pulmonary disease (COPD), similar measurements of LAV% in whole lung or lobes may have large variations in lung function due to emphysema heterogeneity. The functional information of regional emphysema destruction is required for supporting the choice of optimal target. The purpose of this study is to develop an emphysema heterogeneity descriptor for the three-dimensional emphysematous bullae according to the size variations of emphysematous density (ED) and their spatial distribution. The second purpose is to derive a predictive model of airflow limitation based on the regional emphysema heterogeneity. Deriving the bullous representation and grouping them into four scales in the upper and lower lobes, a predictive model is computed using the linear model fitting to estimate the severity of lung function. A total of 99 subjects, 87 patients with mild to very severe COPD (Global Initiative for Chronic Obstructive Lung Disease (GOLD) stage I~IV) and 12 control participants with normal lung functions (forced expiratory volume in one second (FEV₁)/forced vital capacity (FVC) > 0.7) were evaluated. The final model was trained with stratified cross-validation on randomly selected 75% of the dataset (n = 76) and tested on the remaining dataset (n = 23). The dispersed cases of LAV% inconsistent with their lung function outcome were evaluated, and the correlation study suggests that comparing to LAV of larger bullae, the widely spread smaller bullae with equivalent LAV has a larger impact on lung function. The testing dataset has the correlation of r = -0.76 (p < 0.01) between the whole lung LAV% and FEV₁/FVC, whereas using two ED % of scales and location-dependent variables to predict the emphysema-associated FEV₁/FVC, the results shows their correlation of 0.82 (p < 0.001) with clinical FEV₁/FVC.

Quantitative Evaluation of Lobar Pulmonary Function of Emphysema Patients with Endobronchial Coils

BACKGROUND

Recent advances in bronchoscopic lung volume reduction offer new therapies for patients with emphysema and hyperinflation. Pulmonary lobe segmentation with quantification of lobar volumes and emphysema severity plays a pivotal role in treatment planning and post-interventional assessment. Computed tomography (CT)-derived lobar volumes could reflect more accurate regional changes in pulmonary function.

OBJECTIVES

The aim of our study is to validate the reliability of an in-house CT Lung Segmentation software (LungSeg; the Hamlyn Centre, Imperial College London, UK) for lung lobar volume and emphysema quantification for chronic obstructive pulmonary disease (COPD) patients.

METHODS

A total of 108 CT scans from subjects who participated in an endobronchial coil treatment trial were included. Lobar volume and emphysema quantification were performed using the LungSeg and Syngo CT Pulmo 3D package (Siemens Healthcare GmbH, Germany). The inter-user reliability of the LungSeg program was investigated. Correlation coefficients and Bland-Altman analyses were used to quantify the inter-software variability. The agreement between CT volume analysis and plethysmography analysis was also examined.

RESULTS

The high intraclass correlation coefficients (mean ICC = 0.98) of the lobar volumes and emphysema indices measured by LungSeg suggest its excellent reproducibility. The LungSeg and Syngo program have good correlation (rho ≥0.94) and agreement for both lobar volume (median difference = 94 mL and LOAnp = 214.6 mL) and emphysema index (median difference ≤1.5% and LOAnp ≤2.03%) calculations. CT analysis provides a higher estimation of total lung capacity (TLCCT) than body plethysmography (TLCpleth), while there is a fair agreement on residual volume (RVCT) by LungSeg as compared with body plethysmography (RVpleth).

CONCLUSIONS

CT-derived lobar volume and emphysema quantification using the LungSeg program is efficient and reliable in allowing lobar volume assessment. LungSeg has low inter-user variability and agrees better with plethysmography for COPD assessment in our study.

Predicting all-cause and lung cancer mortality using emphysema score progression rate between baseline and follow-up chest CT images: A comparison of risk model performances

Purpose

Normalized emphysema score is a protocol-robust CT biomarker of mortality. We aimed to improve mortality prediction by including the emphysema score progression rate–its change over time–into the models.

Method and materials

CT scans from 6000 National Lung Screening Trial CT arm participants were included. Of these, 1810 died (445 lung cancer-specific). The remaining 4190 survivors were sampled with replacement up to 24432 to approximate the full cohort. Three overlapping subcohorts were formed which required participants to have images from specific screening rounds. Emphysema scores were obtained after resampling, normalization, and bullae cluster analysis of the original images. Base models contained solely the latest emphysema score. Progression models included emphysema score progression rate. Models were adjusted by including baseline age, sex, BMI, smoking status, smoking intensity, smoking duration, and previous COPD diagnosis. Cox proportional hazard models predicting all-cause and lung cancer mortality were compared by calculating the area under the curve per year follow-up.

Results

In the subcohort of participants with baseline and first annual follow-up scans, the analysis was performed on 4940 participants (23227 after resampling). Area under the curve for all-cause mortality predictions of the base and progression models 6 years after baseline were 0.564 (0.564 to 0.565) and 0.569 (0.568 to 0.569) when unadjusted, and 0.704 (0.703 to 0.704) to 0.705 (0.704 to 0.705) when adjusted. The respective performances predicting lung cancer mortality were 0.638 (0.637 to 0.639) and 0.643 (0.642 to 0.644) when unadjusted, and 0.724 (0.723 to 0.725) and 0.725 (0.725 to 0.726) when adjusted.

Conclusion

Including emphysema score progression rate into risk models shows no clinically relevant improvement in mortality risk prediction. This is because scan normalization does not adjust for an overall change in lung density. Adjusting for changes in smoking behavior is likely required to make this a clinically useful measure of emphysema progression.

Comparison of Filtered Back Projection, Hybrid Iterative Reconstruction, Model-Based Iterative Reconstruction, and Virtual Monoenergetic Reconstruction Images at Both Low- and Standard-Dose Settings in Measurement of Emphysema Volume and Airway Wall Thickness: A CT Phantom Study

Objective

To evaluate the accuracy of emphysema volume (EV) and airway measurements (AMs) produced by various iterative reconstruction (IR) algorithms and virtual monoenergetic images (VME) at both low- and standard-dose settings.

Materials and Methods

Computed tomography (CT) images were obtained on phantom at both low- (30 mAs at 120 kVp) and standard-doses (100 mAs at 120 kVp). Each CT scan was reconstructed using filtered back projection, hybrid IR (iDose⁴; Philips Healthcare), model-based IR (IMR-R1, IMR-ST1, IMR-SP1; Philips Healthcare), and VME at 70 keV (VME70). The EV of each air column and wall area percentage (WA%) of each airway tube were measured in all algorithms. Absolute percentage measurement errors of EV (APE_vol) and AM (APE_WA%) were then calculated.

Results

Emphysema volume was most accurately measured in IMR-R1 (APE_vol in low-dose, 0.053 ± 0.002; APE_vol in standard-dose, 0.047 ± 0.003; all p < 0.001) and AM was the most accurate in IMR-SP1 on both low- and standard-doses CT (APE_WA% in low-dose, 0.067 ± 0.002; APE_WA% in standard-dose, 0.06 ± 0.003; all p < 0.001). There were no significant differences in the APE_vol of IMR-R1 between low- and standard-doses (all p > 0.05). VME70 showed a significantly higher APE_vol than iDose⁴, IMR-R1, and IMR-ST1 (all p < 0.004). VME70 also showed a significantly higher APE_WA% compared with the other algorithms (all p < 0.001).

Conclusion

IMR was the most accurate technique for measurement of both EV and airway wall thickness. However, VME70 did not show a significantly better accuracy compared with other algorithms.

Normalized emphysema scores on low dose CT: Validation as an imaging biomarker for mortality

The purpose of this study is to develop a computed tomography (CT) biomarker of emphysema that is robust across reconstruction settings, and evaluate its ability to predict mortality in patients at high risk for lung cancer. Data included baseline CT scans acquired between August 2002 and April 2004 from 1737 deceased subjects and 5740 surviving controls taken from the National Lung Screening Trial. Emphysema scores were computed in the original scans (origES) and after applying resampling, normalization and bullae analysis (normES). We compared the prognostic value of normES versus origES for lung cancer and all-cause mortality by computing the area under the receiver operator characteristic curve (AUC) and the net reclassification improvement (NRI) for follow-up times of 1–7 years. normES was a better predictor of mortality than origES. The 95% confidence intervals for the differences in AUC values indicated a significant difference for all-cause mortality for 2 through 6 years of follow-up, and for lung cancer mortality for 1 through 7 years of follow-up. 95% confidence intervals in NRI values showed a statistically significant improvement in classification for all-cause mortality for 2 through 7 years of follow-up, and for lung cancer mortality for 3 through 7 years of follow-up. Contrary to conventional emphysema score, our normalized emphysema score is a good predictor of all-cause and lung cancer mortality in settings where multiple CT scanners and protocols are used.

Peroxisome proliferator-activated receptor-γ in induced sputum is correlated with MMP-9/TIMP-1 imbalance and formation of emphysema in COPD patients

Background

The development of chronic obstructive pulmonary disease (COPD) is modulated by the symmetry of matrix metalloproteinases (MMPs) and the counter-acting tissue inhibitors of metalloproteinases (TIMPs). We investigated the interaction between peroxisome proliferator-activated receptor gamma (PPARγ) expression and the imbalance of MMP-9/TIMP-1 in the induced sputum of stable COPD patients.

Methods

Sixty-six stable COPD patients were enrolled and the induced sputum samples were gathered. The correlation between PPARγ and other index, including MMP-9, TIMP-1, pulmonary function and the index of emphysema-the percentage of low attenuation area (LAA%), was analyzed.

Results

PPARγ and TIMP-1 concentrations were decreased and the concentration of MMP-9 and the ratio of MMP9/TIMP1 were enhanced in the induced sputum of COPD patients, compared to the healthy controls. Among COPD patients, those with worse lung function or patients with emphysema exhibited increased MMP-9 expression with decreased TIMP-1 and PPARγ expression. Besides, the concentration of PPARγ of the induced sputum was correlated with the forced expiratory volume in one second percentage (FEV1%) positively and the expression of TIMP-1; while it was negatively correlated with the residual volume (RV), RV/total lung capacity (TLC), LAA%, and MMP-9 expression.

Conclusions

Our findings reveal the protective role of PPARγ in the maintenance of the dynamic balance of MMP-9/TIMP-1 in COPD, thus providing evidence on which to base the potential COPD treatment.

Results of quantitative chest-CT in chronic pulmonary graft-vs.-host disease (cGvHD) 3 years after allogeneic stem cell transplantation

BACKGROUND

To quantify lung parenchymal changes in symptomatic patients with chronic pulmonary graft-versus-host disease 3 years after allogeneic stem cell transplantation (allo-SCT) by means of CT-densitometry (CTD) and to compare results with those of established pulmonary function tests (PFT).

METHODS

The study group consisted of 26 patients with pulmonary cGvHD (19 males, 7 females; mean age, 49.29±15.89; range, 19-72 years). The diagnosis was based on clinical symptoms, PFT and chest-CT findings. CTD and PFT were performed both in the pre- and post-transplantation setting and results compared with each other. CT scans were obtained during suspended deep inspiration including the whole lungs. The mean lung attenuation (MLD), low attenuation values (LAV) and distribution of focal parenchymal abnormalities compatible with emphysema (HU <-950) were quantitatively calculated with histograms and graphics. On PFT, total lung capacity (TLC), residual volume (RV), vital capacity (VC), forced expiratory volume in 1 s (FEV1s) and diffusion capacity for carbon monoxide (DLCOSB) were registered.

RESULTS

Changes in end-inspiratory lung volume and density (MLD and LAV) in symptomatic cGvHD patients in mean three years after allo-SCT proved all not significant, but there was a clear trend towards an increase in lung volume and a decrease in lung attenuation. These results were similar throughout all classes of bronchiolitis obliterans (BO) by cGvHD. PFT showed a significant decrease in VC, FEV1s but only a minimal decrease in DLCOSB. Changes in FVC after stem cell transplantation correlated with changes in LAV (r=0.649, P=0.031). Predicted VC correlated with changes in LAV (r=0.771, P=0.005). There was a correlation between the absolute difference of FEV1 and DLCOSB (r=0.64, P=0.14) before and after stem cell transplantation.

CONCLUSIONS

End-inspiratory phase CT lung parenchyma quantification in symptomatic patients with pulmonary cGvHD 3 years after allo-SCT shows discrete changes over the pre-transplantation setting representing airway obstruction, mirroring airflow limitation on PFT. Its use enables exclusion of relevant parenchymal destruction (emphysema-equivalent lung density) at this time.

Lung densitometry: why, how and when

Lung densitometry assesses with computed tomography (CT) the X-ray attenuation of the pulmonary tissue which reflects both the degree of inflation and the structural lung abnormalities implying decreased attenuation, as in emphysema and cystic diseases, or increased attenuation, as in fibrosis. Five reasons justify replacement with lung densitometry of semi-quantitative visual scales used to measure extent and severity of diffuse lung diseases: (I) improved reproducibility; (II) complete vs. discrete assessment of the lung tissue; (III) shorter computation times; (IV) better correlation with pathology quantification of pulmonary emphysema; (V) better or equal correlation with pulmonary function tests (PFT). Commercially and open platform software are available for lung densitometry. It requires attention to technical and methodological issues including CT scanner calibration, radiation dose, and selection of thickness and filter to be applied to sections reconstructed from whole-lung CT acquisition. Critical is also the lung volume reached by the subject at scanning that can be measured in post-processing and represent valuable information per se. The measurements of lung density include mean and standard deviation, relative area (RA) at -970, -960 or -950 Hounsfield units (HU) and 1st and 15th percentile for emphysema in inspiratory scans, and RA at -856 HU for air trapping in expiratory scans. Kurtosis and skewness are used for evaluating pulmonary fibrosis in inspiratory scans. The main indication for lung densitometry is assessment of emphysema component in the single patient with chronic obstructive pulmonary diseases (COPD). Additional emerging applications include the evaluation of air trapping in COPD patients and in subjects at risk of emphysema and the staging in patients with lymphangioleiomyomatosis (LAM) and with pulmonary fibrosis. It has also been applied to assess prevalence of smoking-related emphysema and to monitor progression of smoking-related emphysema, alpha1 antitrypsin deficiency emphysema, and pulmonary fibrosis. Finally, it is recommended as end-point in pharmacological trials of emphysema and lung fibrosis.

Segmental bronchi collapsibility: computed tomography-based quantification in patients with chronic obstructive pulmonary disease and correlation with emphysema phenotype, corresponding lung volume changes and clinical parameters

BACKGROUND

Global pulmonary function tests lack region specific differentiation that might influence therapy in severe chronic obstructive pulmonary disease (COPD) patients. Therefore, the aim of this work was to assess the degree of expiratory 3^rd generation bronchial lumen collapsibility in patients with severe COPD using chest-computed tomography (CT), to evaluate emphysema-phenotype, lobar volumes and correlate results with pulmonary function tests.

METHODS

Thin-slice chest-CTs acquired at end-inspiration & end-expiration in 42 COPD GOLD IV patients (19 females, median-age: 65.9 y) from November 2011 to July 2014 were re-evaluated. The cross-sectional area of all segmental bronchi was measured 5 mm below the bronchial origin in both examinations. Lung lobes were semi-automatically segmented, volumes calculated at end-inspiratory and end-expiratory phase and visually defined emphysema-phenotypes defined. Results of CT densitometry were compared with lung functional tests including forced expiratory volume at 1 s (FEV₁), total lung capacity (TLC), vital capacity (VC), residual volume (RV), diffusion capacity parameters and the maximal expiratory flow rates (MEFs).

RESULTS

Mean expiratory bronchial collapse was 31%, stronger in lobes with homogenous (38.5%) vs. heterogeneous emphysema-phenotype (27.8%, P=0.014). The mean lobar expiratory volume reduction was comparable in both emphysema-phenotypes (volume reduction 18.6%±8.3% in homogenous vs. 17.6%±16.5% in heterogeneous phenotype). The degree of bronchial lumen collapsibility, did not correlate with expiratory volume reduction. MEF₂₅ correlated weakly with 3^rd generation airway collapsibility (r=0.339, P=0.03). All patients showed a concentric expiratory reduction of bronchial cross-sectional area.

CONCLUSIONS

Changes in collapsibility of 3^rd generation bronchi in COPD grade IV patients is significantly lower than that in the trachea and the main bronchi. Collapsibility did not correlate with the reduction in lung volume but was significantly higher in lobes with homogeneous vs. heterogeneous emphysema phenotype. Changes in the 3^rd generation bronchial calibres between inspiration and expiration are not predictive for the degree of small airway collapsibility and related airflow limitation.

Development of digital phantoms based on a finite element model to simulate low-attenuation areas in CT imaging for pulmonary emphysema quantification

PURPOSE

To develop an innovative finite element (FE) model of lung parenchyma which simulates pulmonary emphysema on CT imaging. The model is aimed to generate a set of digital phantoms of low-attenuation areas (LAA) images with different grades of emphysema severity.

METHODS

Four individual parameter configurations simulating different grades of emphysema severity were utilized to generate 40 FE models using ten randomizations for each setting. We compared two measures of emphysema severity (relative area (RA) and the exponent D of the cumulative distribution function of LAA clusters size) between the simulated LAA images and those computed directly on the models output (considered as reference).

RESULTS

The LAA images obtained from our model output can simulate CT-LAA images in subjects with different grades of emphysema severity. Both RA and D computed on simulated LAA images were underestimated as compared to those calculated on the models output, suggesting that measurements in CT imaging may not be accurate in the assessment of real emphysema extent.

CONCLUSIONS

Our model is able to mimic the cluster size distribution of LAA on CT imaging of subjects with pulmonary emphysema. The model could be useful to generate standard test images and to design physical phantoms of LAA images for the assessment of the accuracy of indexes for the radiologic quantitation of emphysema.

Adaptive quantification and longitudinal analysis of pulmonary emphysema with a hidden Markov measure field model

The extent of pulmonary emphysema is commonly estimated from CT scans by computing the proportional area of voxels below a predefined attenuation threshold. However, the reliability of this approach is limited by several factors that affect the CT intensity distributions in the lung. This work presents a novel method for emphysema quantification, based on parametric modeling of intensity distributions and a hidden Markov measure field model to segment emphysematous regions. The framework adapts to the characteristics of an image to ensure a robust quantification of emphysema under varying CT imaging protocols, and differences in parenchymal intensity distributions due to factors such as inspiration level. Compared to standard approaches, the presented model involves a larger number of parameters, most of which can be estimated from data, to handle the variability encountered in lung CT scans. The method was applied on a longitudinal data set with 87 subjects and a total of 365 scans acquired with varying imaging protocols. The resulting emphysema estimates had very high intra-subject correlation values. By reducing sensitivity to changes in imaging protocol, the method provides a more robust estimate than standard approaches. The generated emphysema delineations promise advantages for regional analysis of emphysema extent and progression.

Increased mean lung density: another independent predictor of lung cancer?

OBJECTIVES

To investigate the relationship between emphysema phenotype, mean lung density (MLD), lung function and lung cancer by using an automated multiple feature analysis tool on thin-section computed tomography (CT) data.

METHODS

Both emphysema phenotype and MLD evaluated by automated quantitative CT analysis were compared between outpatients and screening participants with lung cancer (n=119) and controls (n=989). Emphysema phenotype was defined by assessing features such as extent, distribution on core/peel of the lung and hole size. Adjusted multiple logistic regression models were used to evaluate independent associations of CT densitometric measurements and pulmonary function test (PFT) with lung cancer risk.

RESULTS

No emphysema feature was associated with lung cancer. Lung cancer risk increased with decreasing values of forced expiratory volume in 1s (FEV1) independently of MLD (OR 5.37, 95% CI: 2.63-10.97 for FEV1<60% vs. FEV1≥90%), and with increasing MLD independently of FEV1 (OR 3.00, 95% CI: 1.60-5.63 for MLD>-823 vs. MLD<-857 Hounsfield units).

CONCLUSION

Emphysema per se was not associated with lung cancer whereas decreased FEV1 was confirmed as being a strong and independent risk factor. The cross-sectional association between increased MLD and lung cancer requires future validations.

Emphysema quantification by low-dose CT: potential impact of adaptive iterative dose reduction using 3D processing

OBJECTIVE

The purpose of this study is to investigate the effect of a novel reconstruction algorithm, adaptive iterative dose reduction using 3D processing, on emphysema quantification by low-dose CT.

MATERIALS AND METHODS

Twenty-six patients who had undergone standard-dose (150 mAs) and low-dose (25 mAs) CT scans were included in this retrospective study. Emphysema was quantified by several quantitative measures, including emphysema index given by the percentage of lung region with low attenuation (lower than -950 HU), the 15th percentile of lung density, and size distribution of low-attenuation lung regions, on standard-dose CT images reconstructed without adaptive iterative dose reduction using 3D processing and on low-dose CT images reconstructed both without and with adaptive iterative dose reduction using 3D processing. The Bland-Altman analysis was used to assess whether the agreement between emphysema quantifications on low-dose CT and on standard-dose CT was improved by the use of adaptive iterative dose reduction using 3D processing.

RESULTS

For the emphysema index, the mean differences between measurements on low-dose CT and on standard-dose CT were 1.98% without and -0.946% with the use of adaptive iterative dose reduction using 3D processing. For 15th percentile of lung density, the mean differences without and with adaptive iterative dose reduction using 3D processing were -6.67 and 1.28 HU, respectively. For the size distribution of low-attenuation lung regions, the ranges of the mean relative differences without and with adaptive iterative dose reduction using 3D processing were 21.4-85.5% and -14.1% to 11.2%, respectively. For 15th percentile of lung density and the size distribution of low-attenuation lung regions, the agreement was thus improved by the use of adaptive iterative dose reduction using 3D processing.

CONCLUSION

The use of adaptive iterative dose reduction using 3D processing resulted in greater consistency of emphysema quantification by low-dose CT, with quantification by standard-dose CT.

Automated texture-based quantification of centrilobular nodularity and centrilobular emphysema in chest CT images

RATIONALE AND OBJECTIVES

Characterization of smoking-related lung disease typically consists of visual assessment of chest computed tomographic (CT) images for the presence and extent of emphysema and centrilobular nodularity (CN). Quantitative analysis of emphysema and CN may improve the accuracy, reproducibility, and efficiency of chest CT scoring. The purpose of this study was to develop a fully automated texture-based system for the detection and quantification of centrilobular emphysema (CLE) and CN in chest CT images.

MATERIALS AND METHODS

A novel approach was used to prepare regions of interest (ROIs) within the lung parenchyma for representation by texture features associated with the gray-level run-length and gray-level gap-length methods. These texture features were used to train a multiple logistic regression classifier to discriminate between normal lung tissue, CN or "smoker's lung," and CLE. This classifier was trained and evaluated on 24 and 71 chest CT scans, respectively.

RESULTS

During training, the classifier correctly classified 89% of ROIs depicting normal lung tissue, 74% of ROIs depicting CN, and 95% of ROIs manifesting CLE. When the performance of the classifier in quantifying extent of CN and CLE was evaluated on 71 chest CT scans, 65% of ROIs in smokers without CLE were classified as CN, compared to 31% in nonsmokers (P < .001) and 28% in smokers with CLE (P < .001).

CONCLUSIONS

The texture-based framework described herein facilitates successful discrimination among normal lung tissue, CN, and CLE and can be used for the automated quantification of smoking-related lung disease.

What can imaging tell us about physiology? Lung growth and regional mechanical strain

The interplay of mechanical forces transduces diverse physico-biochemical processes to influence lung morphogenesis, growth, maturation, remodeling and repair. Because tissue stress is difficult to measure in vivo, mechano-sensitive responses are commonly inferred from global changes in lung volume, shape, or compliance and correlated with structural changes in tissue blocks sampled from postmortem-fixed lungs. Recent advances in noninvasive volumetric imaging technology, nonrigid image registration, and deformation analysis provide valuable tools for the quantitative analysis of in vivo regional anatomy and air and tissue-blood distributions and when combined with transpulmonary pressure measurements, allow characterization of regional mechanical function, e.g., displacement, strain, shear, within and among intact lobes, as well as between the lung and the components of its container-rib cage, diaphragm, and mediastinum-thereby yielding new insights into the inter-related metrics of mechanical stress-strain and growth/remodeling. Here, we review the state-of-the-art imaging applications for mapping asymmetric heterogeneous physical interactions within the thorax and how these interactions permit as well as constrain lung growth, remodeling, and compensation during development and following pneumonectomy to illustrate how advanced imaging could facilitate the understanding of physiology and pathophysiology. Functional imaging promises to facilitate the formulation of realistic computational models of lung growth that integrate mechano-sensitive events over multiple spatial and temporal scales to accurately describe in vivo physiology and pathophysiology. Improved computational models in turn could enhance our ability to predict regional as well as global responses to experimental and therapeutic interventions.

Quantitative computed tomography in COPD: possibilities and limitations

Chronic obstructive pulmonary disease (COPD) is a heterogeneous disease that is characterized by chronic airflow limitation. Unraveling of this heterogeneity is challenging but important, because it might enable more accurate diagnosis and treatment. Because spirometry cannot distinguish between the different contributing pathways of airflow limitation, and visual scoring is time-consuming and prone to observer variability, other techniques are sought to start this phenotyping process. Quantitative computed tomography (CT) is a promising technique, because current CT technology is able to quantify emphysema, air trapping, and large airway wall dimensions. This review focuses on CT quantification techniques of COPD disease components and their current status and role in phenotyping COPD.

CT quantification of emphysema in young subjects with no recognizable chest disease

OBJECTIVE

The purpose of this prospective study was to evaluate volumetric CT emphysema quantification (CT densitovolumetry) in a young population with no recognizable lung disease.

SUBJECTS AND METHODS

A cohort of 30 nonsmoking patients with no recognizable lung disease (16 men, 14 women; age range, 19-41 years) underwent inspiratory and expiratory CT, after which the data were postprocessed for volumetric quantification of emphysema (threshold, -950 HU). Correlation was tested for age, weight, height, sex, body surface area (BSA), and physical activity. Normal limits were established by mean +/- 1.96 SD.

RESULTS

No correlation was found between the measured volumes and age or physical activity. Correlation was found between BSA and normal lung volume in inspiration (r = 0.69, p = 0.000), shrink volume (i.e., difference in total lung volume in inspiration and in expiration) (r = 0.66, p = 0.000), and percentage of shrink volume (r = 0.35, p = 0.05). For an alpha error of 5%, the limits of normality based on this sample are percentage of emphysema in inspiration, 0.35%; percentage of emphysema in expiration, 0.12%; and maximum lung volume in expiration, 3.6 L. The maximum predicted percentage of shrink volume can be calculated as %SV = 29.43% + 16.97% x BSA (+/- 1.96 x 7.61%).

CONCLUSION

Young healthy nonsmokers with no recognizable lung disease can also show a small proportion of emphysematous-like changes on CT densitovolumetry when a threshold of -950 HU is used. Reference values should be considered when applying the technique for early detection or grading of emphysema and when studying aging lungs.

Automatic segmentation of the lungs using robust level sets

This paper presents a method for the automatic segmentation of the lungs in X-ray computed tomography (CT) images. The proposed technique is based on the use of a robust geometric active contour that is initialized around the lungs, automatically splits in two, and performs outlier rejection during the curve evolution. The technique starts by grey-level thresholding of the images followed by edge detection. Then the edge connected points are organized into strokes and classified as valid or invalid. A confidence degree (weight) is assigned to each stroke and updated during the evolution process with the valid strokes receiving a high confidence degree and the confidence degrees of the outlier strokes tending to zero. These weights depend on the distance between the stroke points and the curve and also on the stroke size. Initialization of the curve is fully automatic. Experimental results show the effectiveness of the proposed technique.

Diffuse parenchymal lung diseases: 3D automated detection in MDCT

Characterization and quantification of diffuse parenchymal lung disease (DPLD) severity using MDCT, mainly in interstitial lung diseases and emphysema, is an important issue in clinical research for the evaluation of new therapies. This paper develops a 3D automated approach for detection and diagnosis of DPLDs (emphysema, fibrosis, honeycombing, ground glass). The proposed methodology combines multi-resolution image decomposition based on 3D morphological filtering, and graph-based classification for a full characterization of the parenchymal tissue. The very promising results obtained on a small patient database are good premises for a near implementation and validation of the proposed approach in clinical routine.

Assessment of the relationship between lung parenchymal destruction and impaired pulmonary perfusion on a lobar level in patients with emphysema

PURPOSE

To assess the relationship between lung parenchymal destruction and impaired pulmonary perfusion on a lobar level using CT and MRI in patients with emphysema.

MATERIAL AND METHODS

Forty-five patients with severe emphysema (GOLD III and IV) underwent inspiratory 3D-HRCT and contrast-enhanced MR-perfusion (1.5T; 3.5mmx1.9mmx4mm). 3D-HRCT data was analyzed using a software for detection and visualization of emphysema. Emphysema was categorized in four clusters with different volumes and presented as overlay on the CT. CT and lung perfusion were visually analyzed for three lobes on each side using a four-point-score to grade the abnormalities on CT (1: predominantly small emphysema-clusters to 4: >75% large emphysema-clusters) and MRI (1: normal perfusion to 4: no perfusion).

RESULTS

A total of 270 lobes were evaluated. At CT, the score was 1 for 9 lobes, 2 for 43, 3 for 77, and 4 for 141 lobes. At MRI, the score was 1 for 13 lobes, 2 for 45, 3 for 92, and 4 for 120 lobes. Matching of lung parenchymal destruction and reduced perfusion was found in 213 lobes (weighted kappa=0.8). The score was higher on CT in 44, and higher on MRI in 13 lobes.

CONCLUSION

3D-HRCT and 3D MR-perfusion show a high lobar agreement between parenchymal destruction and reduction of perfusion in patients with severe emphysema.

Quantitative analysis of emphysema in 3D using MDCT: influence of different reconstruction algorithms

PURPOSE

The aim of the study was to compare the influence of different reconstruction algorithms on quantitative emphysema analysis in patients with severe emphysema.

MATERIAL AND METHODS

Twenty-five patients suffering from severe emphysema were included in the study. All patients underwent inspiratory MDCT (Aquilion-16, slice thickness 1/0.8mm). The raw data were reconstructed using six different algorithms: bone kernel with beam hardening correction (BHC), soft tissue kernel with BHC; standard soft tissue kernel, smooth soft tissue kernel (internal reference standard), standard lung kernel, and high-convolution kernel. The only difference between image data sets was the algorithm employed to reconstruct the raw data, no additional radiation was required. CT data were analysed using self-written emphysema detection and quantification software providing lung volume, emphysema volume (EV), emphysema index (EI) and mean lung density (MLD).

RESULTS

The use of kernels with BHC led to a significant decrease in MLD (5%) and EI (61-79%) in comparison with kernels without BHC. The absolute difference (from smooth soft tissue kernel) in MLD ranged from -0.6 to -6.1 HU and were significant different for all kernels. The EV showed absolute differences between -0.05 and -0.4 L and was significantly different for all kernels. The EI showed absolute differences between -0.8 and -5.1 and was significantly different for all kernels.

CONCLUSION

The use of kernels with BHC led to a significant decrease in MLD and EI. The absolute differences between different kernels without BHC were small but they were larger than the known interscan variation in patients. Thus, for follow-up examinations the same reconstruction algorithm has to be used and use of BHC has to be avoided.

Multi-detector CT of the chest: influence of dose onto quantitative evaluation of severe emphysema: a simulation study

PURPOSE

Quantitative evaluation of the lung parenchyma might be impaired or unreliable by use of reduced-dose CT protocols. Aim of the study was to define the threshold where reduced dose has significant impact on quantitative emphysema parameters.

MATERIALS AND METHODS

Thirty patients with severe centrilobular emphysema underwent multidetector computed tomography (120 kV, 150 mAs). Original CT raw data were simulated using 10 mAs settings (10-100 SIMmAs). Quantitative analysis provided lung volume, emphysema volume, emphysema index, mean lung density, and 4 emphysema volume classes. Simulated low-dose results were compared with original acquisition.

RESULTS

Emphysema index showed no clinical relevant variation down to 30 SIMmAs. The large emphysema volume class was significantly different below 50 SIMmAs. The intermediate and small classes showed an overproportional variation below 50 SIMmAs.

CONCLUSIONS

Dose reduction down to 30 SIMmAs is possible for clinical routine. Settings below 50 SIMmAs significantly alter the in-detailed 3-dimensional emphysema quantification.

Local noise weighted filtering for emphysema scoring of low-dose CT images

Computed tomography (CT) has become the new reference standard for quantification of emphysema. The most popular measure of emphysema derived from CT is the pixel index (PI), which expresses the fraction of the lung volume with abnormally low intensity values. As PI is calculated from a single, fixed threshold on intensity, this measure is strongly influenced by noise. This effect shows up clearly when comparing the PI score of a high-dose scan to the PI score of a low-dose (i.e., noisy) scan of the same subject. In this paper, the noise variance (NOVA) filter is presented: a general framework for (iterative) nonlinear filtering, which uses an estimate of the spatially dependent noise variance in an image. The NOVA filter iteratively estimates the local image noise and filters the image. For the specific purpose of emphysema quantification of low-dose CT images, a dedicated, noniterative NOVA filter is constructed by using prior knowledge of the data to obtain a good estimate of the spatially dependent noise in an image. The performance of the NOVA filter is assessed by comparing characteristics of pairs of high-dose and low-dose scans. The compared characteristics are the PI scores for different thresholds and the size distributions of emphysema bullae. After filtering, the PI scores of high-dose and low-dose images agree to within 2%-3% points. The reproducibility of the high-dose bullae size distribution is also strongly improved. NOVA filtering of a CT image of typically 400 x 512 x 512 voxels takes only a couple of minutes which makes it suitable for routine use in clinical practice.

Computer analysis of computed tomography scans of the lung: a survey

Current computed tomography (CT) technology allows for near isotropic, submillimeter resolution acquisition of the complete chest in a single breath hold. These thin-slice chest scans have become indispensable in thoracic radiology, but have also substantially increased the data load for radiologists. Automating the analysis of such data is, therefore, a necessity and this has created a rapidly developing research area in medical imaging. This paper presents a review of the literature on computer analysis of the lungs in CT scans and addresses segmentation of various pulmonary structures, registration of chest scans, and applications aimed at detection, classification and quantification of chest abnormalities. In addition, research trends and challenges are identified and directions for future research are discussed.

MDCT-based 3-D texture classification of emphysema and early smoking related lung pathologies

Our goal is to enhance the ability to differentiate normal lung from subtle pathologies via multidetector row CT (MDCT) by extending a two-dimensional (2-D) texturebased tissue classification [adaptive multiple feature method (AMFM)] to use three-dimensional (3-D) texture features. We performed MDCT on 34 humans and classified volumes of interest (VOIs) in the MDCT images into five categories: EC, emphysema in severe chronic obstructive pulmonary disease (COPD); MC, mild emphysema in mild COPD; NC, normal appearing lung in mild COPD; NN, normal appearing lung in normal nonsmokers; and NS, normal appearing lung in normal smokers. COPD severity was based upon pulmonary function tests (PFTs). Airways and vessels were excluded from VOIs; 24 3-D texture features were calculated; and a Bayesian classifier was used for discrimination. A leave-one-out method was employed for validation. Sensitivity of the four-class classification in the form of 3-D/2-D was: EC: 85%/71%, MC: 90%/82%; NC: 88%/50%; NN: 100%/60%. Sensitivity and specificity for NN using a two-class classification of NN and NS in the form of 3-D/2-D were: 99%/72% and 100%/75%, respectively. We conclude that 3-D AMFM analysis of lung parenchyma improves discrimination compared to 2-D AMFM of the same VOIs. Furthermore, our results suggest that the 3-D AMFM may provide a means of discriminating subtle differences between smokers and nonsmokers both with normal PFTs.

Automatic segmentation of the lungs using multiple active contours and outlier model

This paper presents a method for the automatic segmentation of the lungs in X-ray computed tomography (CT) images. The proposed technique is based on the use of multiple active contour models (ACMs) for the simultaneous segmentation of both lungs and outlier detection. The technique starts by grey-level thresholding of the images followed by edge detection. Then the edge points are organized in strokes and a set of weights summing to one is assigned to each stroke. These weights represent the soft assignment of the stroke to each of the ACMs and depend on the distance between the stroke points and the ACM units, on gradient direction information and also on the stroke size. Both the weights and the ACMs energy minimization are computed using the generalized expectation-maximization (EM) algorithm. Initialization of the ACM's is fully automatic. Experimental results show the effectiveness of the proposed technique.

Paired Inspiratory/Expiratory Volumetric Thin-Slice CT Scan for Emphysema Analysis

PURPOSES

The aim of the study was to use three-dimensional high-resolution CT scan data sets in inspiration and expiration for the quantitative evaluation of emphysema. Using an advanced dedicated semiautomatic analysis tool, the functional inspiratory/expiratory shifts of emphysema volume and clusters were quantified. The pulmonary function test (PFT) served as the clinical "gold standard."

MATERIALS AND METHODS

Thirty-one patients (9 women and 22 men; mean [+/- SD] age, 60 +/- 8 years) who had severe emphysema due to COPD (Global Initiative for Chronic Obstructive Lung Disease [GOLD] class III and IV) were included in the study. All patients underwent paired inspiratory/expiratory multidetector CT scans (slice thickness, 1/0.8 mm) and pulmonary function tests (PFTs). CT scan data were analyzed with self-written emphysema detection solftware. It provides lung volume (LV), emphysema volume (EV), emphysema index (EI), and four clusters of emphysema with different volumes (from 2, 8, 65, and 120 mm(3)). These results were correlated with total lung capacity (TLC), intrathoracic gas volume (ITGV), and residual volume (RV) derived from PFT results.

RESULTS

Inspiratory LV correlated with TLC (r = 0.9), expiratory LV with ITGV (r = 0.87), and RV (r = 0.83). Expiratory EV correlated better with ITGV (r = 0.88) and RV (r = 0.93) than with inspiratory EV (r = 0.83 and 0.88, respectively). The mean inspiratory EI was 54 +/- 13%, and it decreased to 43 +/- 15% in expiration. However, the individuals showed a broad spectrum of changes of EI (mean, 11%; range, 1 to 28%), and no differences in inspiratory/expiratory EI and changes in EI or LV were found between GOLD III and GOLD IV patients. In expiration, there was a change from the large emphysema cluster (-37%) to the intermediate cluster (+15%) and small cluster (+13% and +11%, respectively). The change of volume of the large emphysema cluster after expiration correlated well with the changes in LV (r = 0.9), EV (r = 0.99), EI (r = 0.85), and MLD (r = 0.76).

CONCLUSION

Emphysema volumes measured from expiratory MDCT scans better reflect PFT abnormalities in patients with severe emphysema than those from inspiratory scans. Volumetric cluster analysis provided deeper insights into the local hyperinflation and expiratory obstruction of large emphysematous clusters.

Toward automated segmentation of the pathological lung in CT

Conventional methods of lung segmentation rely on a large gray value contrast between lung fields and surrounding tissues. These methods fail on scans with lungs that contain dense pathologies, and such scans occur frequently in clinical practice. We propose a segmentation-by-registration scheme in which a scan with normal lungs is elastically registered to a scan containing pathology. When the resulting transformation is applied to a mask of the normal lungs, a segmentation is found for the pathological lungs. As a mask of the normal lungs, a probabilistic segmentation built up out of the segmentations of 15 registered normal scans is used. To refine the segmentation, voxel classification is applied to a certain volume around the borders of the transformed probabilistic mask. Performance of this scheme is compared to that of three other algorithms: a conventional, a user-interactive and a voxel classification method. The algorithms are tested on 10 three-dimensional thin-slice computed tomography volumes containing high-density pathology. The resulting segmentations are evaluated by comparing them to manual segmentations in terms of volumetric overlap and border positioning measures. The conventional and user-interactive methods that start off with thresholding techniques fail to segment the pathologies and are outperformed by both voxel classification and the refined segmentation-by-registration. The refined registration scheme enjoys the additional benefit that it does not require pathological (hand-segmented) training data.

Computer-aided diagnosis in high resolution CT of the lungs

A computer-aided diagnosis (CAD) system is presented to automatically distinguish normal from abnormal tissue in high-resolution CT chest scans acquired during daily clinical practice. From high-resolution computed tomography scans of 116 patients, 657 regions of interest are extracted that are to be classified as displaying either normal or abnormal lung tissue. A principled texture analysis approach is used, extracting features to describe local image structure by means of a multi-scale filter bank. The use of various classifiers and feature subsets is compared and results are evaluated with ROC analysis. Performance of the system is shown to approach that of two expert radiologists in diagnosing the local regions of interest, with an area under the ROC curve of 0.862 for the CAD scheme versus 0.877 and 0.893 for the radiologists.