Effects of CT section thickness and reconstruction kernel on emphysema quantification relationship to the magnitude of the CT emphysema index

Artificial Intelligence-Based Emphysema Quantification in Routine Chest Computed Tomography: Correlation With Spirometry and Visual Emphysema Grading

OBJECTIVE

The aim of the study is to assess the correlation between artificial intelligence (AI)-based low attenuation volume percentage (LAV%) with forced expiratory volume in the first second to forced vital capacity (FEV1/FVC) and visual emphysema grades in routine chest computed tomography (CT). Furthermore, optimal LAV% cutoff values for predicting a FEV1/FVC < 70% or moderate to more extensive visual emphysema grades were calculated.

METHODS

In a retrospective study of 298 consecutive patients who underwent routine chest CT and spirometry examinations, LAV% was quantified using an AI-based software with a threshold < -950 HU. The FEV1/FVC was derived from spirometry, with FEV1/FVC < 70% indicating airway obstruction. The mean time interval of CT from spirometry was 3.87 ± 4.78 days. Severity of emphysema was visually graded by an experienced chest radiologist using an established 5-grade ordinal scale (Fleischner Society classification system). Spearman correlation coefficient between LAV% and FEV1/FVC was calculated. Receiver operating characteristic determined the optimal LAV% cutoff values for predicting a FEV1/FVC < 70% or a visual emphysema grade of moderate or higher (Fleischner grade 3-5).

RESULTS

Significant correlation between LAV% and FEV1/FVC was found (ϱ = -0.477, P < 0.001). Increasing LAV% corresponded to higher visual emphysema grades. For patients with absent visual emphysema, mean LAV% was 2.98 ± 3.30, for patients with trace emphysema 3.22 ± 2.75, for patients with mild emphysema 3.90 ± 3.33, for patients with moderate emphysema 6.41 ± 3.46, for patients with confluent emphysema 9.02 ± 5.45, and for patients with destructive emphysema 16.90 ± 8.19. Optimal LAV% cutoff value for predicting a FEV1/FVC < 70 was 6.1 (area under the curve = 0.764, sensitivity = 0.773, specificity = 0.665), while for predicting a visual emphysema grade of moderate or higher, it was 4.7 (area under the curve = 0.802, sensitivity = 0.766, specificity = 0.742). Furthermore, correlation between visual emphysema grading and FEV1/FVC was found. In patients with FEV1/FVC < 70% a high proportion of subjects had emphysema grade 3 (moderate) or higher, whereas in patients with FEV1/FVC ≥ 70%, a larger proportion had emphysema grade 3 (moderate) or lower. The sensitivity for visual emphysema grading predicting a FEV1/FVC < 70% was 56.3% with an optimal cutoff point at a visual grade of 4 (confluent), demonstrating a lower sensitivity compared with LAV% (77.3%).

CONCLUSIONS

A significant correlation between AI-based LAV% and FEV1/FVC as well as visual CT emphysema grades can be found in routine chest CT suggesting that AI-based LAV% measurement might be integrated as an add-on functional parameter in the evaluation of chest CT in the future.

Measurement Variability of Same-Day CT Quantification of Interstitial Lung Disease: A Multicenter Prospective Study

Purpose To investigate quantitative CT (QCT) measurement variability in interstitial lung disease (ILD) on the basis of two same-day CT scans. Materials and Methods Participants with ILD were enrolled in this multicenter prospective study between March and October 2022. Participants underwent two same-day CT scans at an interval of a few minutes. Deep learning-based texture analysis software was used to segment ILD features. Fibrosis extent was defined as the sum of reticular opacity and honeycombing cysts. Measurement variability between scans was assessed with Bland-Altman analyses for absolute and relative differences with 95% limits of agreement (LOA). The contribution of fibrosis extent to variability was analyzed using a multivariable linear mixed-effects model while adjusting for lung volume. Eight readers assessed ILD fibrosis stability with and without QCT information for 30 randomly selected samples. Results Sixty-five participants were enrolled in this study (mean age, 68.7 years ± 10 [SD]; 47 [72%] men, 18 [28%] women). Between two same-day CT scans, the 95% LOA for the mean absolute and relative differences of quantitative fibrosis extent were -0.9% to 1.0% and -14.8% to 16.1%, respectively. However, these variabilities increased to 95% LOA of -11.3% to 3.9% and -123.1% to 18.4% between CT scans with different reconstruction parameters. Multivariable analysis showed that absolute differences were not associated with the baseline extent of fibrosis (P = .09), but the relative differences were negatively associated (β = -0.252, P < .001). The QCT results increased readers' specificity in interpreting ILD fibrosis stability (91.7% vs 94.6%, P = .02). Conclusion The absolute QCT measurement variability of fibrosis extent in ILD was 1% in same-day CT scans. Keywords: CT, CT-Quantitative, Thorax, Lung, Lung Diseases, Interstitial, Pulmonary Fibrosis, Diagnosis, Computer Assisted, Diagnostic Imaging Supplemental material is available for this article. © RSNA, 2024.

A systematic assessment and optimization of photon-counting CT for lung density quantifications

BACKGROUND

Photon-counting computed tomography (PCCT) has recently emerged into clinical use; however, its optimum imaging protocols and added benefits remains unknown in terms of providing more accurate lung density quantification compared to energy-integrating computed tomography (EICT) scanners.

PURPOSE

To systematically assess the performance of a clinical PCCT scanner for lung density quantifications and compare it against EICT.

METHODS

This cross-sectional study involved a retrospective analysis of subjects scanned (August-December 2021) using a clinical PCCT system. The influence of altering reconstruction parameters was studied (reconstruction kernel, pixel size, slice thickness). A virtual CT dataset of anthropomorphic virtual subjects was acquired to demonstrate the correspondence of findings to clinical dataset, and to perform systematic imaging experiments, not possible using human subjects. The virtual subjects were imaged using a validated, scanner-specific CT simulator of a PCCT and two EICT (defined as EICT A and B) scanners. The images were evaluated using mean absolute error (MAE) of lung and emphysema density against their corresponding ground truth.

RESULTS

Clinical and virtual PCCT datasets showed similar trends, with sharper kernels and smaller voxel sizes increasing percentage of low-attenuation areas below -950 HU (LAA-950) by up to 15.7 ± 6.9% and 11.8 ± 5.5%, respectively. Under the conditions studied, higher doses, thinner slices, smaller pixel sizes, iterative reconstructions, and quantitative kernels with medium sharpness resulted in lower lung MAE values. While using these settings for PCCT, changes in the dose level (13 to 1.3 mGy), slice thickness (0.4 to 1.5 mm), pixel size (0.49 to 0.98 mm), reconstruction technique (70 keV-VMI to wFBP), and kernel (Qr48 to Qr60) increased lung MAE by 15.3 ± 2.0, 1.4 ± 0.6, 2.2 ± 0.3, 4.2 ± 0.8, and 9.1 ± 1.6 HU, respectively. At the optimum settings identified per scanner, PCCT images exhibited lower lung and emphysema MAE than those of EICT scanners (by 2.6 ± 1.0 and 9.6 ± 3.4 HU, compared to EICT A, and by 4.8 ± 0.8 and 7.4 ± 2.3 HU, compared to EICT B). The accuracy of lung density measurements was correlated with subjects' mean lung density (p < 0.05), measured by PCCT at optimum setting under the conditions studied.

CONCLUSION

Photon-counting CT demonstrated superior performance in density quantifications, with its influences of imaging parameters in line with energy-integrating CT scanners. The technology offers improvement in lung quantifications, thus demonstrating potential toward more objective assessment of respiratory conditions.

Lung CT harmonization of paired reconstruction kernel images using generative adversarial networks

BACKGROUND

The kernel used in CT image reconstruction is an important factor that determines the texture of the CT image. Consistency of reconstruction kernel choice is important for quantitative CT-based assessment as kernel differences can lead to substantial shifts in measurements unrelated to underlying anatomical structures.

PURPOSE

In this study, we investigate kernel harmonization in a multi-vendor low-dose CT lung cancer screening cohort and evaluate our approach's validity in quantitative CT-based assessments.

METHODS

Using the National Lung Screening Trial, we identified CT scan pairs of the same sessions with one reconstructed from a soft tissue kernel and one from a hard kernel. In total, 1000 pairs of five different paired kernel types (200 each) were identified. We adopt the pix2pix architecture to train models for kernel conversion. Each model was trained on 100 pairs and evaluated on 100 withheld pairs. A total of 10 models were implemented. We evaluated the efficacy of kernel conversion based on image similarity metrics including root mean squared error (RMSE), peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM) as well as the capability of the models to reduce measurement shifts in quantitative emphysema and body composition measurements. Additionally, we study the reproducibility of standard radiomic features for all kernel pairs before and after harmonization.

RESULTS

Our approach effectively converts CT images from one kernel to another in all paired kernel types, as indicated by the reduction in RMSE (p < 0.05) and an increase in the PSNR (p < 0.05) and SSIM (p < 0.05) for both directions of conversion for all pair types. In addition, there is an increase in the agreement for percent emphysema, skeletal muscle area, and subcutaneous adipose tissue (SAT) area for both directions of conversion. Furthermore, radiomic features were reproducible when compared with the ground truth features.

CONCLUSIONS

Kernel conversion using deep learning reduces measurement variation in percent emphysema, muscle area, and SAT area.

Deep Learning-Based Kernel Adaptation Enhances Quantification of Emphysema on Low-Dose Chest CT for Predicting Long-Term Mortality

OBJECTIVES

The aim of this study was to ascertain the predictive value of quantifying emphysema using low-dose computed tomography (LDCT) post deep learning-based kernel adaptation on long-term mortality.

MATERIALS AND METHODS

This retrospective study investigated LDCTs obtained from asymptomatic individuals aged 60 years or older during health checkups between February 2009 and December 2016. These LDCTs were reconstructed using a 1- or 1.25-mm slice thickness alongside high-frequency kernels. A deep learning algorithm, capable of generating CT images that resemble standard-dose and low-frequency kernel images, was applied to these LDCTs. To quantify emphysema, the lung volume percentage with an attenuation value less than or equal to -950 Hounsfield units (LAA-950) was gauged before and after kernel adaptation. Low-dose chest CTs with LAA-950 exceeding 6% were deemed emphysema-positive according to the Fleischner Society statement. Survival data were sourced from the National Registry Database at the close of 2021. The risk of nonaccidental death, excluding causes such as injury or poisoning, was explored according to the emphysema quantification results using multivariate Cox proportional hazards models.

RESULTS

The study comprised 5178 participants (mean age ± SD, 66 ± 3 years; 3110 males). The median LAA-950 (18.2% vs 2.6%) and the proportion of LDCTs with LAA-950 exceeding 6% (96.3% vs 39.3%) saw a significant decline after kernel adaptation. There was no association between emphysema quantification before kernel adaptation and the risk of nonaccidental death. Nevertheless, after kernel adaptation, higher LAA-950 (hazards ratio for 1% increase, 1.01; P = 0.045) and LAA-950 exceeding 6% (hazards ratio, 1.36; P = 0.008) emerged as independent predictors of nonaccidental death, upon adjusting for age, sex, and smoking status.

CONCLUSIONS

The application of deep learning for kernel adaptation proves instrumental in quantifying pulmonary emphysema on LDCTs, establishing itself as a potential predictive tool for long-term nonaccidental mortality in asymptomatic individuals.

DiffusionCT: Latent Diffusion Model for CT Image Standardization

Computed tomography (CT) is one of the modalities for effective lung cancer screening, diagnosis, treatment, and prognosis. The features extracted from CT images are now used to quantify spatial and temporal variations in tumors. However, CT images obtained from various scanners with customized acquisition protocols may introduce considerable variations in texture features, even for the same patient. This presents a fundamental challenge to downstream studies that require consistent and reliable feature analysis. Existing CT image harmonization models rely on GAN-based supervised or semi-supervised learning, with limited performance. This work addresses the issue of CT image harmonization using a new diffusion-based model, named DiffusionCT, to standardize CT images acquired from different vendors and protocols. DiffusionCT operates in the latent space by mapping a latent non-standard distribution into a standard one. DiffusionCT incorporates a U-Net-based encoder-decoder, augmented by a diffusion model integrated into the bottleneck part. The model is designed in two training phases. The encoder-decoder is first trained, without embedding the diffusion model, to learn the latent representation of the input data. The latent diffusion model is then trained in the next training phase while fixing the encoder-decoder. Finally, the decoder synthesizes a standardized image with the transformed latent representation. The experimental results demonstrate a significant improvement in the performance of the standardization task using DiffusionCT.

Generative Adversarial Network-Based Image Conversion Among Different Computed Tomography Protocols and Vendors: Effects on Accuracy and Variability in Quantifying Regional Disease Patterns of Interstitial Lung Disease

OBJECTIVE

To assess whether computed tomography (CT) conversion across different scan parameters and manufacturers using a routable generative adversarial network (RouteGAN) can improve the accuracy and variability in quantifying interstitial lung disease (ILD) using a deep learning-based automated software.

MATERIALS AND METHODS

This study included patients with ILD who underwent thin-section CT. Unmatched CT images obtained using scanners from four manufacturers (vendors A-D), standard- or low-radiation doses, and sharp or medium kernels were classified into groups 1-7 according to acquisition conditions. CT images in groups 2-7 were converted into the target CT style (Group 1: vendor A, standard dose, and sharp kernel) using a RouteGAN. ILD was quantified on original and converted CT images using a deep learning-based software (Aview, Coreline Soft). The accuracy of quantification was analyzed using the dice similarity coefficient (DSC) and pixel-wise overlap accuracy metrics against manual quantification by a radiologist. Five radiologists evaluated quantification accuracy using a 10-point visual scoring system.

RESULTS

Three hundred and fifty CT slices from 150 patients (mean age: 67.6 ± 10.7 years; 56 females) were included. The overlap accuracies for quantifying total abnormalities in groups 2-7 improved after CT conversion (original vs. converted: 0.63 vs. 0.68 for DSC, 0.66 vs. 0.70 for pixel-wise recall, and 0.68 vs. 0.73 for pixel-wise precision; P < 0.002 for all). The DSCs of fibrosis score, honeycombing, and reticulation significantly increased after CT conversion (0.32 vs. 0.64, 0.19 vs. 0.47, and 0.23 vs. 0.54, P < 0.002 for all), whereas those of ground-glass opacity, consolidation, and emphysema did not change significantly or decreased slightly. The radiologists' scores were significantly higher (P < 0.001) and less variable on converted CT.

CONCLUSION

CT conversion using a RouteGAN can improve the accuracy and variability of CT images obtained using different scan parameters and manufacturers in deep learning-based quantification of ILD.

The solid component within part-solid nodules: 3-dimensional quantification, correlation with the malignant grade of nonmucinous pulmonary adenocarcinomas, and comparisons with 2-dimentional measures and semantic features in low-dose computed tomography

BACKGROUND

There is no consensus on 3-dimensional (3D) quantification method for solid component within part-solid nodules (PSNs). This study aimed to find the optimal attenuation threshold for the 3D solid component proportion in low-dose computed tomography (LDCT), namely the consolidation/tumor ratio of volume (CTRV), basing on its correlation with the malignant grade of nonmucinous pulmonary adenocarcinomas (PAs) according to the 5th edition of World Health Organization classification. Then we tested the ability of CTRV to predict high-risk nonmucinous PAs in PSNs, and compare its performance with 2-dimensional (2D) measures and semantic features.

METHODS

A total of 313 consecutive patients with 326 PSNs, who underwent LDCT within one month before surgery and were pathologically diagnosed with nonmucinous PAs, were retrospectively enrolled and were divided into training and testing cohorts according to scanners. The CTRV were automatically generated by setting a series of attenuation thresholds from - 400 to 50 HU with an interval of 50 HU. The Spearman's correlation was used to evaluate the correlation between the malignant grade of nonmucinous PAs and semantic, 2D, and 3D features in the training cohort. The semantic, 2D, and 3D models to predict high-risk nonmucinous PAs were constructed using multivariable logistic regression and validated in the testing cohort. The diagnostic performance of these models was evaluated by the area under curve (AUC) of receiver operating characteristic curve.

RESULTS

The CTRV at attenuation threshold of -250 HU (CTRV_- 250HU) showed the highest correlation coefficient among all attenuation thresholds (r = 0.655, P < 0.001), which was significantly higher than semantic, 2D, and other 3D features (all P < 0.001). The AUCs of CTRV_- 250HU to predict high-risk nonmucinous PAs were 0.890 (0.843-0.927) in the training cohort and 0.832 (0.737-0.904) in the testing cohort, which outperformed 2D and semantic models (all P < 0.05).

CONCLUSIONS

The optimal attenuation threshold was - 250 HU for solid component volumetry in LDCT, and the derived CTRV_- 250HU might be valuable for the risk stratification and management of PSNs in lung cancer screening.

Pulmonary emphysema and coronary artery calcifications at baseline LDCT and long-term mortality in smokers and former smokers of the ITALUNG screening trial

OBJECTIVES

Cardiovascular disease (CVD), lung cancer (LC), and respiratory diseases are main causes of death in smokers and former smokers undergoing low-dose computed tomography (LDCT) for LC screening. We assessed whether quantification of pulmonary emphysematous changes at baseline LDCT has a predictive value concerning long-term mortality.

METHODS

In this longitudinal study, we assessed pulmonary emphysematous changes with densitometry (volume corrected relative area below - 950 Hounsfield units) and coronary artery calcifications (CAC) with a 0-3 visual scale in baseline LDCT of 524 participants in the ITALUNG trial and analyzed their association with mortality after 13.6 years of follow-up using conventional statistics and a machine learning approach.

RESULTS

Pulmonary emphysematous changes were present in 32.3% of subjects and were mild (6% ≤ RA950 ≤ 9%) in 14.9% and moderate-severe (RA950 > 9%) in 17.4%. CAC were present in 67% of subjects (mild in 34.7%, moderate-severe in 32.2%). In the follow-up, 81 (15.4%) subjects died (20 of LC, 28 of other cancers, 15 of CVD, 4 of respiratory disease, and 14 of other conditions). After adjusting for age, sex, smoking history, and CAC, moderate-severe emphysema was significantly associated with overall (OR 2.22; 95CI 1.34-3.70) and CVD (OR 3.66; 95CI 1.21-11.04) mortality. Machine learning showed that RA950 was the best single feature predictive of overall and CVD mortality.

CONCLUSIONS

Moderate-severe pulmonary emphysematous changes are an independent predictor of long-term overall and CVD mortality in subjects participating in LC screening and should be incorporated in the post-test calculation of the individual mortality risk profile.

KEY POINTS

• Densitometry allows quantification of pulmonary emphysematous changes in low-dose CT examinations for lung cancer screening. • Emphysematous lung density changes are an independent predictor of long-term overall and cardio-vascular disease mortality in smokers and former smokers undergoing screening. • Emphysematous changes quantification should be included in the post-test calculation of the individual mortality risk profile.

Prediction of exercise-induced desaturation in COPD patients without resting hypoxemia: a retrospective study

Background

There is no universally accepted criterion for assessing exercise-induced desaturation (EID). The purpose of this study is to compare the two methods regularly used for determining EID in COPD patients, as well as to explore the risk factors and predictors related to EID.

Methods

The 6MWT was performed with continuous SpO₂ monitoring on patients with stable COPD. Using two methods (method A: “SpO_2rest–SpO_2min ≥ 4% and/or SpO_2min < 90%”, method B: “SpO_2rest–SpO_2end ≥ 4% and/or SpO_2end < 90%”) as EID determination criteria to assess the incidence of EID. The differences and consistency of the two methods are compared. Moreover, we collected data through the pulmonary function test, mMRC dyspnea score, COPD assessment test, BODE index and CT-defined emphysema. Univariate and multivariate logistic regression analyses were used to identify factors affecting the EID. For the parameters that predict EID in 6MWT, a receiver operating characteristic (ROC) curve analysis was employed.

Results

The analysis included 124 patients. The overall incidence of EID was 62.1% by using method A as the criterion and 51.6% by method B. All of the EID patients found by method B were included in the EID patients identified by method A, as well as 13 new-EID patients. The difference in diagnostic outcomes between the two approaches was not statistically significant (P > 0.05), but they were in excellent agreement (Kappa = 0.807, P = 0.001). Logistic regression analyses found that D_LCO SB% pred, D_LCO/VA% pred, CAT score, mean density, PD15, emphysema volume and %LAA were significant determinants of the EID. For predicting EID, the ROC analysis produced AUC and cutoffs of 0.689 and 50.45% (D_LCO SB% pred), 0.707 and 75.0% (D_LCO/VA% pred), 0.727 and 15 points (CAT score), 0.691 and − 955.00HU (PD15), 0.671 and − 856.46HU (mean density), 0.668 and 338.14 ml (emphysema volume) and 0.656 and 7.63% (%LAA), respectively.

Conclusions

Two methods evaluating EID in this research are in a good agreement, method A can find more EID patients by focusing on SpO_2min. When conditions are constrained, it is also sufficient to assess EID in COPD patients by method B. In terms of the predictors of EID, D_LCO SB% pred, D_LCO/VA% pred, CAT score and CT-defined emphysema are all statistically significant test variables to determine EID.

When Treatment of Pulmonary Emphysema with Endobronchial Valves Did Not Work: Evaluation of Quantitative CT Analysis and Pulmonary Function Tests Before and After Valve Explantation

Purpose

To investigate changes in quantitative CT analysis (QCT) and pulmonary function tests (PFT) in pulmonary emphysema patients who required premature removal of endobronchial valves (EBV).

Patients and Methods

Our hospital’s medical records listed 274 patients with high-grade COPD (GOLD stages 3 and 4) and pulmonary emphysema who were treated with EBV to reduce lung volume. Prior to intervention, a complete evaluation was performed that included quantitative computed tomography analysis (QCT) of scans acquired at full inspiration and full expiration, pulmonary function tests (PFT), and paraclinical findings (6-minute walking distance test (6MWDT) and quality of life questionnaires). In 41 of these 274 patients, EBV treatment was unsuccessful and the valves had to be removed for various reasons. A total of 10 of these 41 patients ventured a second attempt at EBV therapy and underwent complete reevaluation. In our retrospective study, results from three time points were compared: Before EBV implantation (BL), after EBV implantation (TP2), and after EBV explantation (TP3). QCT parameters included lung volume, total emphysema score (TES, ie, the emphysema index) and the 15th percentile of lung attenuation (P15) for the whole lung and each lobe separately. Differences in these parameters between inspiration and expiration were calculated (Vol. Diff (%), TES Diff (%), P15 Diff (%)). The results of PFT and further clinical tests were taken from the patient’s records.

Results

We found persistent therapy effect in the target lobe even after valve explantation together with a compensatory hyperinflation of the rest of the lung. As a result of these two divergent effects, the volume of the total lung remained rather constant. Furthermore, there was a slight deterioration of the emphysema score for the whole lung, whereas the TES of the target lobe persistently improved.

Conclusion

Interestingly, we found evidence that, contrary to our expectations, unsuccessful EBV therapy can have a persistent positive effect on target lobe QCT scores.

Deep Learning-Based Automatic Assessment of Lung Impairment in COVID-19 Pneumonia: Predicting Markers of Hypoxia With Computer Vision

Background

Hypoxia is a potentially life-threatening condition that can be seen in pneumonia patients.

Objective

We aimed to develop and test an automatic assessment of lung impairment in COVID-19 associated pneumonia with machine learning regression models that predict markers of respiratory and cardiovascular functioning from radiograms and lung CT.

Materials and Methods

We enrolled a total of 605 COVID-19 cases admitted to Al Ain Hospital from 24 February to 1 July 2020 into the study. The inclusion criteria were as follows: age ≥ 18 years; inpatient admission; PCR positive for SARS-CoV-2; lung CT available at PACS. We designed a CNN-based regression model to predict systemic oxygenation markers from lung CT and 2D diagnostic images of the chest. The 2D images generated by averaging CT scans were analogous to the frontal and lateral view radiograms. The functional (heart and breath rate, blood pressure) and biochemical findings (SpO₂, HCO3-, K⁺, Na⁺, anion gap, C-reactive protein) served as ground truth.

Results

Radiologic findings in the lungs of COVID-19 patients provide reliable assessments of functional status with clinical utility. If fed to ML models, the sagittal view radiograms reflect dyspnea more accurately than the coronal view radiograms due to the smaller size and the lower model complexity. Mean absolute error of the models trained on single-projection radiograms was approximately 11÷12% and it dropped by 0.5÷1% if both projections were used (11.97 ± 9.23 vs. 11.43 ± 7.51%; p = 0.70). Thus, the ML regression models based on 2D images acquired in multiple planes had slightly better performance. The data blending approach was as efficient as the voting regression technique: 10.90 ± 6.72 vs. 11.96 ± 8.30%, p = 0.94. The models trained on 3D images were more accurate than those on 2D: 8.27 ± 4.13 and 11.75 ± 8.26%, p = 0.14 before lung extraction; 10.66 ± 5.83 and 7.94 ± 4.13%, p = 0.18 after the extraction. The lung extraction boosts 3D model performance unsubstantially (from 8.27 ± 4.13 to 7.94 ± 4.13%; p = 0.82). However, none of the differences between 3D and 2D were statistically significant.

Conclusion

The constructed ML algorithms can serve as models of structure-function association and pathophysiologic changes in COVID-19. The algorithms can improve risk evaluation and disease management especially after oxygen therapy that changes functional findings. Thus, the structural assessment of acute lung injury speaks of disease severity.

Single material beam hardening correction via an analytical energy response model for diagnostic CT

BACKGROUND

Various clinical studies show the potential for a wider quantitative role of diagnostic X-ray computed tomography (CT) beyond size measurements. Currently, the clinical use of attenuation values is however limited due to their lack of robustness. This issue can be observed even on the same scanner across patient size and positioning. There are different causes for the lack of robustness in the attenuation values; one possible source of error is beam hardening of the X-ray source spectrum. The conventional and well-established approach to address this issue is a calibration-based single material beam hardening correction (BHC) using a water cylinder.

PURPOSE

We investigate an alternative approach for single material BHC with the aim of producing a more robust result for the attenuation values. The underlying hypothesis of this investigation is that calibration based BHC automatically corrects for scattered radiation in a manner that is sub-optimal in terms of bias as soon as the scanned object strongly deviates from the water cylinder used for calibration.

METHODS

The approach we propose performs BHC via an analytical energy response model that is embedded into a correction pipeline that efficiently estimates and subtracts scattered radiation in a patient-specific manner prior to BHC. The estimation of scattered radiation is based on minimizing, in average, the squared difference between our corrected data and the vendor-calibrated data. The used energy response model is considering the spectral effects of the detector response and of the pre-filtration of the source spectrum including a beam-shaping bowtie filter. The performance of the correction pipeline is first characterized with computer simulated data. Afterwards, it is tested using real 3-D CT data sets of two different phantoms, with various kV settings and phantom positions, assuming a circular data acquisition. The results are compared in the image domain to those from the scanner.

RESULTS

For experiments with a water cylinder, the proposed correction pipeline leads to similar results as the vendor. For reconstructions of a QRM liver phantom with extension ring, the proposed correction pipeline achieved a more uniform and stable outcome in the attenuation values of homogeneous materials within the phantom. For example, the root mean squared deviation between centered and off-centered phantom positioning was reduced from 6.6 HU to 1.8 HU in one profile.

CONCLUSIONS

We have introduced a patient-specific approach for single material BHC in diagnostic CT via the use of an analytical energy response model. This approach shows promising improvements in terms of robustness of attenuation values for large patient sizes. Our results contribute towards improving CT images so as to make CT attenuation values more reliable for use in clinical practice. This article is protected by copyright. All rights reserved.

Data harmonisation for information fusion in digital healthcare: A state-of-the-art systematic review, meta-analysis and future research directions

Removing the bias and variance of multicentre data has always been a challenge in large scale digital healthcare studies, which requires the ability to integrate clinical features extracted from data acquired by different scanners and protocols to improve stability and robustness. Previous studies have described various computational approaches to fuse single modality multicentre datasets. However, these surveys rarely focused on evaluation metrics and lacked a checklist for computational data harmonisation studies. In this systematic review, we summarise the computational data harmonisation approaches for multi-modality data in the digital healthcare field, including harmonisation strategies and evaluation metrics based on different theories. In addition, a comprehensive checklist that summarises common practices for data harmonisation studies is proposed to guide researchers to report their research findings more effectively. Last but not least, flowcharts presenting possible ways for methodology and metric selection are proposed and the limitations of different methods have been surveyed for future research.

Utility of Noncancerous Chest CT Features for Predicting Overall Survival and Noncancer Death in Patients With Stage I Lung Cancer Treated With Stereotactic Body Radiotherapy

Background: Noncancerous imaging markers can be readily derived from pretreatment diagnostic and radiotherapy planning chest CT examinations. Objective: To explore the ability of noncancerous features on chest CT to predict overall survival (OS) and noncancer-related death in patients with stage I lung cancer treated with stereotactic body radiation therapy (SBRT). Methods: This retrospective study included 282 patients (168 female, 114 male; median age, 75 years) with stage I lung cancer treated with SBRT between January 2009 and June 2017. Pretreatment chest CT was used to quantify coronary artery calcium (CAC) score, pulmonary artery (PA)-to-aorta ratio, emphysema, and body composition in terms of the cross-sectional area and attenuation of skeletal muscle and subcutaneous adipose tissue at the T5, T8, and T10 vertebral levels. Associations of clinical and imaging features with OS were quantified using a multivariable Cox proportional hazards (PH) model. Penalized multivariable Cox PH models to predict OS were constructed using clinical features only and using both clinical and imaging features. Models' discriminatory ability was assessed by constructing time-varying ROC curves and computing AUC at prespecified times. Results: After a median OS of 60.8 months (95% CI 55.8-68.9), 148 (52.5%) patients died, including 83 (56.1%) with noncancer deaths. Higher CAC score (11-399: hazard ratio [HR] 1.83 [95% CI 1.15-2.91], P=.01; ≥400: HR 1.63 [95% CI 1.01-2.63], P=.04), higher PA-to-aorta ratio (HR 1.33 [95% CI 1.16-1.52], P<.001, per 0.1-unit increase), and lower thoracic skeletal muscle index (HR 0.88 [95% CI 0.79-0.98], P=.02, per 10 cm²/m² increase) were independently associated with shorter OS. Discriminatory ability for 5-year OS was greater for the model including clinical and imaging features than for the model including clinical features only (AUC, 0.75 [95% CI 0.68-0.83] versus 0.61 [95% CI 0.53-0.70], p < .01). The model's most important clinical or imaging feature based on mean standardized regression coefficients was the PA-to-aorta ratio. Conclusions: In patients undergoing SBRT for stage I lung cancer, higher CAC score, higher PA-to-aorta ratio, and lower thoracic skeletal muscle index independently predicted worse OS. Clinical Impact: Noncancerous imaging features on chest CT performed before SBRT improve survival prediction compared with clinical features alone.

Impact of pixel value truncation on image quality of low dose chest CT

PURPOSE

In some noisy low-dose CT lung cancer screening images, we noticed that the CT density values of air were increased and the visibility of emphysema was distinctly decreased. By examining histograms of these images, we found that the CT density values were truncated at -1,024 HU. The purpose of this study was to investigate the effect of pixel value truncation on the visibility of emphysema using mathematical models.

METHODS AND MATERIALS

Assuming CT noise follows a normal distribution, we derived the relationship between the mean CT density value and the standard deviation (SD) when the pixel values below -1,024 HU are truncated and replaced by -1,024 HU. To validate our mathematical model, 20 untruncated phantom CT images were truncated by simulation, and the mean CT density values and SD of air in the images were measured and compared with the theoretical values. In addition, the mean CT density values and SD of air were measured in 100 cases of real clinical images obtained by GE, Siemens, and Philips scanners, respectively, and the agreement with the theoretical values was examined. Next, the contrast-to-noise ratio (CNR) between air (-1,000 HU) and lung parenchyma (-850 HU) was derived from the mathematical model in the presence and absence of truncation as a measure of the visibility of emphysema. In addition, the radiation dose ratios required to obtain the same CNR in the case with and without truncation were also calculated.

RESULTS

The mathematical model revealed that when the pixel values are truncated, the mean CT density values are proportional to the noise magnitude when the magnitude exceeds a certain level. The mean CT density values and SD measured in the images with pixel values truncated by simulation and in the real clinical images acquired by GE and Philips scanners agreed well with the theoretical values from our mathematical model. In the Siemens images, the measured and theoretical values agreed well when a portion of the truncated values were replaced by random values instead of simply replacing by -1,024 HU. The CNR of air and lung parenchyma was lowered by truncating CT density values compared to that of no truncation. Furthermore, it was found that higher radiation dose was required to obtain the same CNR with truncation as without. As an example, when the noise SD was 60 HU, the radiation dose required for the GE and Philips truncation method was about 1.2 times higher than that without truncation, and that for the Siemens truncation method was about 1.4 times higher.

CONCLUSIONS

It was demonstrated mathematically that pixel value truncation causes a brightening of the mean CT density value and decreases the CNR of emphysema. Our results indicate that it is advisable to turn off truncation at -1,024 HU, especially when scanning at low and ultra-low radiation doses in the thorax. This article is protected by copyright. All rights reserved.

Inter-Observer Agreement between Low-Dose and Standard-Dose CT with Soft and Sharp Convolution Kernels in COVID-19 Pneumonia

Computed tomography (CT) has been an essential diagnostic tool during the COVID-19 pandemic. The study aimed to develop an optimal CT protocol in terms of safety and reliability. For this, we assessed the inter-observer agreement between CT and low-dose CT (LDCT) with soft and sharp kernels using a semi-quantitative severity scale in a prospective study (Moscow, Russia). Two consecutive scans with CT and LDCT were performed in a single visit. Reading was performed by ten radiologists with 3–25 years’ experience. The study included 230 patients, and statistical analysis showed LDCT with a sharp kernel as the most reliable protocol (percentage agreement 74.35 ± 43.77%), but its advantage was marginal. There was no significant correlation between radiologists’ experience and average percentage agreement for all four evaluated protocols. Regarding the radiation exposure, CTDI_vol was 3.6 ± 0.64 times lower for LDCT. In conclusion, CT and LDCT with soft and sharp reconstructions are equally reliable for COVID-19 reporting using the “CT 0-4” scale. The LDCT protocol allows for a significant decrease in radiation exposure but may be restricted by body mass index.

Kernel Conversion for Robust Quantitative Measurements of Archived Chest Computed Tomography Using Deep Learning-Based Image-to-Image Translation

Chest computed tomography (CT) is used to screen for lung cancer and evaluate pulmonary and extra-pulmonary abnormalities such as emphysema and coronary artery calcification, particularly in smokers. In real-world practice, lung abnormalities are visually assessed using high-contrast thin-slice images which are generated from raw scan data using sharp reconstruction kernels with the sacrifice of increased image noise. In contrast, accurate CT quantification requires low-contrast thin-slice images with low noise, which are generated using soft reconstruction kernels. However, only sharp-kernel thin-slice images are archived in many medical facilities due to limited data storage space. This study aimed to establish deep neural network (DNN) models to convert sharp-kernel images to soft-kernel-like images with a final goal to reuse historical chest CT images for robust quantitative measurements, particularly in completed previous longitudinal studies. By using pairs of sharp-kernel (input) and soft-kernel (ground-truth) images from 30 patients with chronic obstructive pulmonary disease (COPD), DNN models were trained. Then, the accuracy of kernel conversion based on the established DNN models was evaluated using CT from independent 30 smokers with and without COPD. Consequently, differences in CT values between new images converted from sharp-kernel images using the established DNN models and ground-truth soft-kernel images were comparable with the inter-scans variability derived from repeated phantom scans (6 times), showing that the conversion error was the same level as the measurement error of the CT device. Moreover, the Dice coefficients to quantify the similarity between low attenuation voxels on given images and the ground-truth soft-kernel images were significantly higher on the DNN-converted images than the Gaussian-filtered, median-filtered, and sharp-kernel images (p < 0.001). There were good agreements in quantitative measurements of emphysema, intramuscular adipose tissue, and coronary artery calcification between the converted and the ground-truth soft-kernel images. These findings demonstrate the validity of the new DNN model for kernel conversion and the clinical applicability of soft-kernel-like images converted from archived sharp-kernel images in previous clinical studies. The presented method to evaluate the validity of the established DNN model using repeated scans of phantom could be applied to various deep learning-based image conversions for robust quantitative evaluation.

Quantitative measurements of emphysema in ultra-high resolution computed tomography using model-based iterative reconstruction in comparison to that using hybrid iterative reconstruction

The percentage of low attenuation volume ratio (LAVR), which is measured using computed tomography (CT), is an index of the severity of emphysema. For LAVR evaluation, ultra-high-resolution (U-HR) CT images are useful. To improve the image quality of U-HRCT, iterative reconstruction is used. There are two types of iterative reconstruction: hybrid iterative reconstruction (HIR) and model-based iterative reconstruction (MBIR). In this study, we physically and clinically evaluated U-HR images reconstructed with HIR and MBIR, and demonstrated the usefulness of U-HR images with MBIR for quantitative measurements of emphysema. Both images were reconstructed with a slice thickness of 0.25 mm and an image matrix size of 1024 × 1024 pixels. For physical evaluation, the modulation transfer function (MTF) and noise power spectrum (NPS) of HIR and MBIR were compared. For clinical evaluation, LAVR calculated from HIR and MBIR were compared using the Wilcoxon matched-pairs signed-rank test. In addition, the correlation between LAVR and forced expiratory volume in one second (FEV₁%) was evaluated using the Spearman rank correlation test. The MTFs of HIR and MBIR were comparable. The NPS of MBIR was lower than that of HIR. The mean LAVR values calculated from HIR and MBIR were 19.5 ± 12.6% and 20.4 ± 11.7%, respectively (p = 0.84). The correlation coefficients between LAVR and FEV₁% that were taken from HIR and MBIR were 0.64 and 0.74, respectively (p < 0.01). MBIR is more useful than HIR for the quantitative measurements of emphysema with U-HR images.

CT Findings From Interstitial Lung Diseases in Patients With Metastatic Breast Cancer Treated With Fam-Trastuzumab Deruxtecan: A Single Institutional Experience

This single-institute, retrospective cohort study enrolled patients with human epidermal growth factor receptor 2-positive metastatic breast cancer treated with trastuzumab deruxtecan between August 2017 and January 2021 from four previous studies. Of 31 patients, 4 (12.9%) had interstitial lung disease (ILD). The dominant pattern observed on computed tomography was organizing pneumonia (100%), comprising subpleural consolidations in the lung periphery. However, no dominant distribution was observed in radiological lesions of the lungs. Of all the tested patients, lower lobe predominance was noted in 2 (50.0%) patients, upper lobe predominance in 1 (25.0%) patient, and diffused lobe distribution in 1 (25.0%) patient. All events were confined to the Common Terminology Criteria for Adverse Events grade 1 or 2 (100%). None of the patients died. Despite the small number of cases investigated, the incidence of trastuzumab deruxtecan-induced ILD in the Korean population was comparable to that previously reported.

X-ray dark-field chest imaging for detection and quantification of emphysema in patients with chronic obstructive pulmonary disease: a diagnostic accuracy study

BACKGROUND

Although advanced medical imaging technologies give detailed diagnostic information, a low-dose, fast, and inexpensive option for early detection of respiratory diseases and follow-ups is still lacking. The novel method of x-ray dark-field chest imaging might fill this gap but has not yet been studied in living humans. Enabling the assessment of microstructural changes in lung parenchyma, this technique presents a more sensitive alternative to conventional chest x-rays, and yet requires only a fraction of the dose applied in CT. We studied the application of this technique to assess pulmonary emphysema in patients with chronic obstructive pulmonary disease (COPD).

METHODS

In this diagnostic accuracy study, we designed and built a novel dark-field chest x-ray system (Technical University of Munich, Munich, Germany)-which is also capable of simultaneously acquiring a conventional thorax radiograph (7 s, 0·035 mSv effective dose). Patients who had undergone a medically indicated chest CT were recruited from the department of Radiology and Pneumology of our site (Klinikum rechts der Isar, Technical University of Munich, Munich, Germany). Patients with pulmonary pathologies, or conditions other than COPD, that might influence lung parenchyma were excluded. For patients with different disease stages of pulmonary emphysema, x-ray dark-field images and CT images were acquired and visually assessed by five readers. Pulmonary function tests (spirometry and body plethysmography) were performed for every patient and for a subgroup of patients the measurement of diffusion capacity was performed. Individual patient datasets were statistically evaluated using correlation testing, rank-based analysis of variance, and pair-wise post-hoc comparison.

FINDINGS

Between October, 2018 and December, 2019 we enrolled 77 patients. Compared with CT-based parameters (quantitative emphysema ρ=-0·27, p=0·089 and visual emphysema ρ=-0·45, p=0·0028), the dark-field signal (ρ=0·62, p<0·0001) yields a stronger correlation with lung diffusion capacity in the evaluated cohort. Emphysema assessment based on dark-field chest x-ray features yields consistent conclusions with findings from visual CT image interpretation and shows improved diagnostic performance than conventional clinical tests characterising emphysema. Pair-wise comparison of corresponding test parameters between adjacent visual emphysema severity groups (CT-based, reference standard) showed higher effect sizes. The mean effect size over the group comparisons (absent-trace, trace-mild, mild-moderate, and moderate-confluent or advanced destructive visual emphysema grades) for the COPD assessment test score is 0·21, for forced expiratory volume in 1 s (FEV₁)/functional vital capacity is 0·25, for FEV₁% of predicted is 0·23, for residual volume % of predicted is 0·24, for CT emphysema index is 0·35, for dark-field signal homogeneity within lungs is 0·38, for dark-field signal texture within lungs is 0·38, and for dark-field-based emphysema severity is 0·42.

INTERPRETATION

X-ray dark-field chest imaging allows the diagnosis of pulmonary emphysema in patients with COPD because this technique provides relevant information representing the structural condition of lung parenchyma. This technique might offer a low radiation dose alternative to CT in COPD and potentially other lung disorders.

FUNDING

European Research Council, Deutsche Forschungsgemeinschaft, Royal Philips, and Karlsruhe Nano Micro Facility.

Influence of Radiation Dose and Reconstruction Kernel on Fat Fraction Analysis in Dual-energy CT: A Phantom Study

BACKGROUND/AIM

The quantitative evaluation of fat tissue, mainly for the determination of liver steatosis, is possible by using dual-energy computed tomography. Different photon energy acquisitions allow for estimation of attenuation coefficients. The effect of variation in radiation doses and reconstruction kernels on fat fraction estimation was investigated.

MATERIALS AND METHODS

A six-probe-phantom with fat concentrations of 0%, 20%, 40%, 60%, 80%, and 100% were scanned in Sn140/100 kV with radiation doses ranging between 20 and 200 mAs before and after calibration. Images were reconstructed using iterative kernels (I26,Q30,I70).

RESULTS

Fat fractions measured in dual-energy computed tomography (DECT) were consistent with the 20%-stepwise varying actual concentrations. Variation in radiation dose resulted in 3.1% variation of fat fraction. Softer reconstruction kernel (I26) underestimated the fat fraction (-9.1%), while quantitative (Q30) and sharper kernel (I70) overestimated fat fraction (10,8% and 13,1, respectively).

CONCLUSION

The fat fraction in DECT approaches the actual fat concentration when calibrated to the reconstruction kerneö. Variation of radiation dose caused an acceptable 3% variation.

Deep learning for lung disease segmentation on CT: Which reconstruction kernel should be used?

PURPOSE

The purpose of this study was to determine whether a single reconstruction kernel or both high and low frequency kernels should be used for training deep learning models for the segmentation of diffuse lung disease on chest computed tomography (CT).

MATERIALS AND METHODS

Two annotated datasets of COVID-19 pneumonia (323,960 slices) and interstitial lung disease (ILD) (4,284 slices) were used. Annotated CT images were used to train a U-Net architecture to segment disease. All CT slices were reconstructed using both a lung kernel (LK) and a mediastinal kernel (MK). Three different trainings, resulting in three different models were compared for each disease: training on LK only, MK only or LK+MK images. Dice similarity scores (DSC) were compared using the Wilcoxon signed-rank test.

RESULTS

Models only trained on LK images performed better on LK images than on MK images (median DSC = 0.62 [interquartile range (IQR): 0.54, 0.69] vs. 0.60 [IQR: 0.50, 0.70], P < 0.001 for COVID-19 and median DSC = 0.62 [IQR: 0.56, 0.69] vs. 0.50 [IQR 0.43, 0.57], P < 0.001 for ILD). Similarly, models only trained on MK images performed better on MK images (median DSC = 0.62 [IQR: 0.53, 0.68] vs. 0.54 [IQR: 0.47, 0.63], P < 0.001 for COVID-19 and 0.69 [IQR: 0.61, 0.73] vs. 0.63 [IQR: 0.53, 0.70], P < 0.001 for ILD). Models trained on both kernels performed better or similarly than those trained on only one kernel. For COVID-19, median DSC was 0.67 (IQR: =0.59, 0.73) when applied on LK images and 0.67 (IQR: 0.60, 0.74) when applied on MK images (P < 0.001 for both). For ILD, median DSC was 0.69 (IQR: 0.63, 0.73) when applied on LK images (P = 0.006) and 0.68 (IQR: 0.62, 0.72) when applied on MK images (P > 0.99).

CONCLUSION

Reconstruction kernels impact the performance of deep learning-based models for lung disease segmentation. Training on both LK and MK images improves the performance.

Fractal dimension in CT low attenuation areas is predictive of long-term oxygen therapy initiation in COPD patients: Results from two observational cohort studies

BACKGROUND

Some chronic obstructive pulmonary disease (COPD) patients develop hypoxemia with disease progression, with some even requiring long-term oxygen therapy (LTOT). Lung function, especially diffusing capacity, and the annual decline in PaO₂, are reported to be predictive factors of chronic respiratory failure. However, the association between lung morphometry evaluated using computed tomography (CT) images and LTOT initiation is unknown.

METHODS

We retrospectively evaluated the relationship between clinical indices, including pulmonary function, body mass index (BMI), and CT parameters, at baseline and LTOT initiation in two prospective COPD cohorts. In the Nara Medical University cohort (n = 76), the low attenuation area (LAA) and its fractal dimension (fractal D) were adapted as the indices for parenchymal destruction in CT images. The association between these CT measurements and LTOT initiation was replicated in the Kyoto University cohort (n = 130).

RESULTS

In the Nara Medical University cohort, lower BMI (hazard ratio [HR]:0.70, p = 0.006), lower % diffusing capacity (%DLCO) (HR: 0.92, p = 0.006), lower %DLCO/VA (HR, 0.90, p = 0.008), higher RV/TLC (HR, 1.26, p = 0.012), higher LAA% (HR: 1.18, p = 0.001), and lower fractal D (HR: 3.27 × 10^-8, p < 0.001) were associated with LTOT initiation. Multivariate analysis in the Kyoto University cohort confirmed that lower %DLCO and lower fractal D were independently associated with LTOT initiation, whereas LAA% was not.

CONCLUSION

Fractal D, which is the index for morphometric complexity of LAA in CT analysis, is predictive of LTOT initiation in COPD patients.

Association of quantitative computed tomography ındices with lung function and extent of pulmonary fibrosis in patients with systemic sclerosis

OBJECTIVES

The aim was to investigate the discriminative value of a wide range of quantitative computed tomography (qCT) parameters in systemic sclerosis (SSc) patients with and without pulmonary fibrosis (PF) and their association with pulmonary function tests (PFTs) and visual fibrosis scores (VFS).

METHOD

Thoracic high-resolution computed tomography (HRCT) images of SSc patients with and without PF were analyzed with Vitrea® Advanced Visualization software. The mean lung attenuation (MLA), skewness, kurtosis, and threshold-based volumes [low-density volume (LDV), medium-density volume (MDV), and high-density volume (HDV)] derived from the attenuation histograms of the right and left lungs were evaluated separately. Visual scores were measured semi-quantitatively and the overall extent of pulmonary parenchymal abnormality was calculated.

RESULTS

Forty-one SSc patients with PF (85.4% female; mean age 50.4 ± 15.6 years) were compared with 94 without PF (88.3% female; mean age 50 ± 11.5 years). All qCT parameters were significantly different between those with and without PF (p < 0.05). Amongst the qCT measurements, R-MLA, L-MLA, R-MDV, L-MDV, and left total lung volume (L-TLV) correlated with all three of forced vital capacity, carbon monoxide diffusion capacity, and VFS, even after adjustment for sex and age (|r|> 0.300 and p < 0.05). R-MLA, L-MLA, R-HDV/TLV, and L-HDV/TLV exhibited diagnostic accuracy in discriminating patients with PF (AUC value > 0.7).

CONCLUSION

QCT parameters differentiated SSc patients with PF from the ones without and showed a good correlation with VFS. With the application of user-friendly and less operator-dependent software, qCT analysis may become an objective tool for analysis of PF in SSc, complementary to PFTs and VFS. Key Points • Quantitative computed tomography parameters can accurately and objectively differentiate between SSc patients with and without PF. • Furthermore, in SSc patients with fibrosis, a moderate to a high correlation was identified between many of the qCT parameters, PFT results, and VFS.

Measuring pulmonary function in COPD using quantitative chest computed tomography analysis

COPD is diagnosed and evaluated by pulmonary function testing (PFT). Chest computed tomography (CT) primarily serves a descriptive role for diagnosis and severity evaluation. CT densitometry-based emphysema quantification and lobar fissure integrity assessment are most commonly used, mainly for lung volume reduction purposes and scientific efforts.

A shift towards a more quantitative role for CT to assess pulmonary function is a logical next step, since more, currently underutilised, information is present in CT images. For instance, lung volumes such as residual volume and total lung capacity can be extracted from CT; these are strongly correlated to lung volumes measured by PFT.

This review assesses the current evidence for use of quantitative CT as a proxy for PFT in COPD and discusses challenges in the movement towards CT as a more quantitative modality in COPD diagnosis and evaluation. To better understand the relevance of the traditional PFT measurements and the role CT might play in the replacement of these parameters, COPD pathology and traditional PFT measurements are discussed.

CT may be used as a proxy for lung function in COPD diagnosis and evaluation, particularly for the hyperinflation markershttps://bit.ly/2RrGAZf

A homological approach to a mathematical definition of pulmonary fibrosis and emphysema on computed tomography

Three-dimensional imaging is essential to evaluate local abnormalities and understand structure-function relationships in an organ. However, quantifiable and interpretable methods to localize abnormalities remain unestablished. Visual assessments are prone to bias, machine learning methods depend on training images, and the underlying decision principle is usually difficult to interpret. Here, we developed a homological approach to mathematically define emphysema and fibrosis in the lungs on computed tomography (CT). With the use of persistent homology, the density of homological features, including connected components, tunnels, and voids, was extracted from the volumetric CT scans of lung diseases. A pair of CT values at which each homological feature appeared (birth) and disappeared (death) was computed by sweeping the threshold levels from higher to lower CT values. Consequently, fibrosis and emphysema were defined as voxels with dense voids having a longer lifetime (birth-death difference) and voxels with dense connected components having a lower birth, respectively. In an independent dataset including subjects with idiopathic pulmonary fibrosis (IPF), chronic obstructive pulmonary disease (COPD), and combined pulmonary fibrosis and emphysema (CPFE), the proposed definition enabled accurate segmentation with comparable quality to deep learning in terms of Dice coefficients. Persistent homology-defined fibrosis was closely associated with physiological abnormalities such as impaired diffusion capacity and long-term mortality in subjects with IPF and CPFE, and persistent homology-defined emphysema was associated with impaired diffusion capacity in subjects with COPD. The present persistent homology-based evaluation of structural abnormalities could help explore the clinical and physiological impacts of structural changes and morphological mechanisms of disease progression.NEW & NOTEWORTHY This study proposes a homological approach to mathematically define a three-dimensional texture feature of emphysema and fibrosis on chest computed tomography using persistent homology. The proposed definition enabled accurate segmentation with comparable quality to deep learning while offering higher interpretability than deep learning-based methods.

Unraveling the interplay of image formation, data representation and learning in CT-based COPD phenotyping automation: The need for a meta-strategy

PURPOSE

In the literature on automated phenotyping of chronic obstructive pulmonary disease (COPD), there is a multitude of isolated classical machine learning and deep learning techniques, mostly investigating individual phenotypes, with small study cohorts and heterogeneous meta-parameters, e.g., different scan protocols or segmented regions. The objective is to compare the impact of different experimental setups, i.e., varying meta-parameters related to image formation and data representation, with the impact of the learning technique for subtyping automation for a variety of phenotypes. The identified associations of these parameters with automation performance and their interactions might be a first step towards a determination of optimal meta-parameters, i.e., a meta-strategy.

METHODS

A clinical cohort of 981 patients (53.8 ± 15.1 years, 554 male) was examined. The inspiratory CT images were analyzed to automate the diagnosis of 13 COPD phenotypes given by two radiologists. A benchmark feature set that integrates many quantitative criteria was extracted from the lung and trained a variety of learning algorithms on the first 654 patients (two thirds) and the respective algorithm retrospectively assessed the remaining 327 patients (one third). The automation performance was evaluated by the area under the receiver operating characteristic curve (AUC). 1717 experiments were conducted with varying meta-parameters such as reconstruction kernel, segmented regions and input dimensionality, i.e., number of extracted features. The association of the meta-parameters with the automation performance was analyzed by multivariable general linear model decomposition of the automation performance in the contributions of meta-parameters and the learning technique.

RESULTS

The automation performance varied strongly for varying meta-parameters. For emphysema-predominant phenotypes, an AUC of 93%-95% could be achieved for the best meta-configuration. The airways-predominant phenotypes led to a lower performance of 65%-85%, while smooth kernel configurations on average were unexpectedly superior to those with sharp kernels. The performance impact of meta-parameters, even that of often neglected ones like the missing-data imputation, was in general larger than that of the learning technique. Advanced learning techniques like 3D deep learning or automated machine learning yielded inferior automation performance for non-optimal meta-configurations in comparison to simple techniques with suitable meta-configurations. The best automation performance was achieved by a combination of modern learning techniques and a suitable meta-configuration.

CONCLUSIONS

Our results indicate that for COPD phenotype automation, study design parameters such as reconstruction kernel and the model input dimensionality should be adapted to the learning technique and may be more important than the technique itself. To achieve optimal automation and prediction results, the interaction between input those meta-parameters and the learning technique should be considered. This might be particularly relevant for the development of specific scan protocols for novel learning algorithms, and towards an understanding of good study design for automated phenotyping.

Automatic pulmonary vessel segmentation on noncontrast chest CT: deep learning algorithm developed using spatiotemporally matched virtual noncontrast images and low-keV contrast-enhanced vessel maps

OBJECTIVES

To develop a deep learning-based pulmonary vessel segmentation algorithm (DLVS) from noncontrast chest CT and to investigate its clinical implications in assessing vascular remodeling of chronic obstructive lung disease (COPD) patients.

METHODS

For development, 104 pulmonary CT angiography scans (49,054 slices) using a dual-source CT were collected, and spatiotemporally matched virtual noncontrast and 50-keV images were generated. Vessel maps were extracted from the 50-keV images. The 3-dimensional U-Net-based DLVS was trained to segment pulmonary vessels (with a vessel map as the output) from virtual noncontrast images (as the input). For external validation, vendor-independent noncontrast CT images (n = 14) and the VESSEL 12 challenge open dataset (n = 3) were used. For each case, 200 points were selected including 20 intra-lesional points, and the probability value for each point was extracted. For clinical validation, we included 281 COPD patients with low-dose noncontrast CTs. The DLVS-calculated volume of vessels with a cross-sectional area < 5 mm² (PVV5) and the PVV5 divided by total vessel volume (%PVV5) were measured.

RESULTS

DLVS correctly segmented 99.1% of the intravascular points (1,387/1,400) and 93.1% of the extravascular points (1,309/1,400). The areas-under-the receiver-operating characteristic curve (AUROCs) were 0.977 and 0.969 for the two external validation datasets. For the COPD patients, both PPV5 and %PPV5 successfully differentiated severe patients whose FEV1 < 50 (AUROCs; 0.715 and 0.804) and were significantly correlated with the emphysema index (Ps < .05).

CONCLUSIONS

DLVS successfully segmented pulmonary vessels on noncontrast chest CT by utilizing spatiotemporally matched 50-keV images from a dual-source CT scanner and showed promising clinical applicability in COPD.

KEY POINTS

• We developed a deep learning pulmonary vessel segmentation algorithm using virtual noncontrast images and 50-keV enhanced images produced by a dual-source CT scanner. • Our algorithm successfully segmented vessels on diseased lungs. • Our algorithm showed promising results in assessing the loss of small vessel density in COPD patients.

The combined use of 2D scout and 3D axial CT images to accurately determine the catheter tips for high-dose-rate brachytherapy plans

Purpose

To develop a method combining CT scout images with axial images to improve the localization accuracy of catheter tips in high‐dose‐rate (HDR) brachytherapy treatments.

Materials and Methods

CT scout images were utilized along with conventionally reconstructed axial images to aid the localization of catheter tips used during HDR treatment planning. A method was developed to take advantage of the finer image resolution of the scout images to more precisely identify the tip coordinates. The accuracies of this method were compared with the conventional method based on the axial CT images alone, for various slice thicknesses, in a computed tomography dose index (CTDI) head phantom. A clinical case which involved multiple interstitial catheters was also selected for the evaluation of this method. Locations of the catheter tips were reconstructed with the conventional CT‐based method and this newly developed method, respectively. Location coordinates obtained via both methods were quantitatively compared.

Results

Combination of the scout and axial CT images improved the accuracy of identification and reconstruction of catheter tips along the longitudinal direction (i.e., head‐to‐foot direction, more or less parallel to the catheter tracks), compared to relying on the axial CT images alone. The degree of improvement was dependent on CT slice thickness. For the clinical patient case, the coordinate differences of the reconstructed catheter tips were 2.6 mm ± 0.9 mm in the head‐to‐foot direction, 0.4 mm ± 0.2 mm in the left‐to‐right direction, and 0.6 mm ± 0.2 mm in the anterior‐to‐posterior direction, respectively.

Conclusion

Combining CT scout and axial images demonstrates the ability to provide a more accurate identification and reconstruction of the interstitial catheter tips for HDR brachytherapy treatment, especially in the longitudinal direction. The method developed in this work has the potential to be implemented clinically together with automatic segmentation in modern brachytherapy treatment planning systems, in order to improve the reconstruction accuracy of HDR catheters.

The Emerging Role of Quantification of Imaging for Assessing the Severity and Disease Activity of Emphysema, Airway Disease, and Interstitial Lung Disease

There has been an explosion of use for quantitative image analysis in the setting of lung disease due to advances in acquisition protocols and postprocessing technology, including machine and deep learning. Despite the plethora of published papers, it is important to understand which approach has clinical validation and can be used in clinical practice. This paper provides an introduction to quantitative image analysis techniques being used in the investigation of lung disease and focusses on the techniques that have a reasonable clinical validation for being used in clinical trials and patient care.

STAN-CT: Standardizing CT Image using Generative Adversarial Networks

Computed Tomography (CT) plays an important role in lung malignancy diagnostics, therapy assessment, and facilitating precision medicine delivery. However, the use of personalized imaging protocols poses a challenge in large-scale cross-center CT image radiomic studies. We present an end-to-end solution called STAN-CT for CT image standardization and normalization, which effectively reduces discrepancies in image features caused by using different imaging protocols or using different CT scanners with the same imaging protocol. STAN-CT consists oftwo components: 1)a Generative Adversarial Networks (GAN) model where a latent-feature-based loss function is adopted to learn the data distribution of standard images within a few rounds of generator training, and 2) an automatic DICOM reconstruction pipeline with systematic image quality control that ensures the generation ofhigh-quality standard DICOM images. Experimental results indicate that the training efficiency and model performance of STAN-CT have been significantly improved compared to the state-of-the-art CT image standardization and normalization algorithms.

Endobronchial Valves for the Treatment of Advanced Emphysema

Bronchoscopic lung volume reduction with one-way endobronchial valves is a guideline treatment option for patients with advanced emphysema that is supported by extensive scientific data. Patients limited by severe hyperinflation, with a suitable emphysema treatment target lobe and with absence of collateral ventilation, are the responders to this treatment. Detailed patient selection, a professional treatment performance, and dedicated follow up of the valve treatment, including management of complications, are key ingredients to success. This treatment does not stand alone; it especially requires extensive knowledge of COPD for which the most appropriate treatment is discussed in a multidisciplinary approach. We discuss the endobronchial valve treatment for emphysema and provide a guideline for patient selection, treatment guidance, and practice tools, based on our own experience and literature.

Quantification of Perinodular Emphysema in High-risk Patients Offers No Benefit in Lung Nodule Risk-Stratification of Malignancy Potential

PURPOSE

Pulmonary nodules, found either incidentally or on lung cancer screening, are common. Evaluating the benign or malignant nature of these nodules is costly in terms of patient risk and expense. The presence of both global and regional emphysema has been linked to increased lung cancer risk. We sought to determine whether the measurement of emphysema directly adjacent to a lung nodule could inform the likelihood of a nodule being malignant.

MATERIALS AND METHODS

Within a population of Veterans at high risk for lung cancer, 58 subjects with malignant nodules found on computerized tomographic chest scans were matched by lobe and nodule size to 58 controls. Lung densitometry was measured via determination of the low attenuation area percentage at -950 Hounsfield units (LAA950) and the Hounsfield unit (HU) value at which 15% of lung voxels have a lower lung density (Perc15), at predefined lung volumes that encompassed the nodule to evaluate both perinodular and regional lung fields. The association between measured lung density and malignancy was investigated using conditional logistic regression models, with densitometry measurements used as the primary predictor, adjusting for age alone, or age and computerized tomographic scan characteristics.

RESULTS

No significant differences in emphysema measurements between malignant and benign nodules were identified at lung volumes encompassing both perinodular and regional emphysema. Furthermore, emphysema quantification remained stable across lung volumes within individuals.

CONCLUSIONS

In this study, quantifying the degree of perinodular or regional emphysema did not offer any benefit in the risk stratification of lung nodules.

[Quantitative Analysis of Emphysema in Ultra-high-resolution CT by Using Deep Learning Reconstruction: Comparison with Hybrid Iterative Reconstruction]

PURPOSE

The noise generated in ultra-high-resolution computed tomography (U-HRCT) images affects the quantitative analysis of emphysema. In this study, we compared the physical properties of reconstructed images for hybrid iterative reconstruction (HIR) and deep learning reconstruction (DLR), which are reconstruction methods for reducing image noise. Using clinical evaluation, we evaluated the correlation between low attenuation volume (LAV) % obtained by CT and forced expiratory volume in 1 s per forced vital capacity (FEV₁/FVC) obtained by respiratory function tests.

MATERIALS AND METHODS

CT data obtained by HIR and DLR were used for analysis (matrix size: 1024´1024, slice thickness: 0.25 mm). The physical characteristics were evaluated for the modulation transfer function (MTF) and noise power spectrum (NPS). Display-field of view (D-FOV) was analyzed by varying between 300 mm and 400 mm. The clinical data evaluated the relationship between LAV% and FEV₁/FVC by Spearman's correlation coefficient.

RESULT

The 10% MTFs were 1.3 cycles/mm (HIR) and 1.3 cycles/mm (DLR) at D-FOV 300 mm, and 1.2 cycles/mm (HIR) and 1.1 cycles/mm (DLR) at D-FOV 400 mm. The NPS had less noise in DLR than HIR in all frequency ranges. The correlation coefficients between LAV% and FEV₁/FVC were 0.64 and 0.71, respectively, in HIR and DLR.

CONCLUSION

There was no difference in the resolution characteristics of HIR and DLR. DLR had better noise characteristics than HIR. The correlation between LAV% measured by HIR and DLR and FEV₁/FVC is equivalent. The noise characteristics of the DLR enable the reduction of exposure to emphysema quantitative analysis by CT.

Memory-efficient GAN-based domain translation of high resolution 3D medical images

Generative adversarial networks (GANs) are currently rarely applied on 3D medical images of large size, due to their immense computational demand. The present work proposes a multi-scale patch-based GAN approach for establishing unpaired domain translation by generating 3D medical image volumes of high resolution in a memory-efficient way. The key idea to enable memory-efficient image generation is to first generate a low-resolution version of the image followed by the generation of patches of constant sizes but successively growing resolutions. To avoid patch artifacts and incorporate global information, the patch generation is conditioned on patches from previous resolution scales. Those multi-scale GANs are trained to generate realistically looking images from image sketches in order to perform an unpaired domain translation. This allows to preserve the topology of the test data and generate the appearance of the training domain data. The evaluation of the domain translation scenarios is performed on brain MRIs of size 155 × 240 × 240 and thorax CTs of size up to 512³. Compared to common patch-based approaches, the multi-resolution scheme enables better image quality and prevents patch artifacts. Also, it ensures constant GPU memory demand independent from the image size, allowing for the generation of arbitrarily large images.

CT kernel conversions using convolutional neural net for super-resolution with simplified squeeze-and-excitation blocks and progressive learning among smooth and sharp kernels

PURPOSE

Computed tomography (CT) volume sets reconstructed with different kernels are helping to increase diagnostic accuracy. However, several CT volumes reconstructed with different kernels are difficult to sustain, due to limited storage and maintenance issues. A CT kernel conversion method is proposed using convolutional neural networks (CNN).

METHODS

A total of 3289 CT images from ten patients (five men and five women; mean age, 63.0 ± 8.6 years) were obtained in May 2016 (Somatom Sensation 16, Siemens Medical Systems, Forchheim, Germany). These CT images were reconstructed with various kernels, including B10f (very smooth), B30f (medium smooth), B50f (medium sharp), and B70f (very sharp) kernels. Smooth kernel images were converted into sharp kernel images using super-resolution (SR) network with Squeeze-and-Excitation (SE) blocks and auxiliary losses, and vice versa. In this study, the single-conversion model and multi-conversion model were presented. In case of the single-conversion model, for the one corresponding output image (e.g., B10f to B70), SE-Residual blocks were stacked. For the multi-conversion model, to convert an image into several output images (e.g., B10f to B30f, B50f, and B70f, and vice versa), progressive learning (PL) was employed by calculating auxiliary losses in every four SE-Residual blocks. Through auxiliary losses, the model could learn mutual relationships between different kernel types. The conversion quality was evaluated by the root-mean-square-error (RMSE), structural similarity (SSIM) index and mutual information (MI) between original and converted images.

RESULTS

The RMSE (SSIM index , MI) of the multi-conversion model was 4.541 ± 0.688 (0.998 ± 0.001 , 2.587 ± 0.137), 27.555 ± 5.876 (0.944 ± 0.021 , 1.735 ± 0.137), 72.327 ± 17.387 (0.815 ± 0.053 , 1.176 ± 0.096), 8.748 ± 1.798 (0.996 ± 0.002 , 2.464 ± 0.121), 9.470 ± 1.772 (0.994 ± 0.003 , 2.336 ± 0.133), and 9.184 ± 1.605 (0.994 ± 0.002 , 2.342 ± 0.138) in conversion between B10f-B30f, B10f-B50f, B10f-B70f, B70f-B50f, B70f-B30f, and B70f-B10f, respectively, which showed significantly better image quality than the conventional model.

CONCLUSIONS

We proposed deep learning-based CT kernel conversion using SR network. By introducing simplified SE blocks and PL, the model performance was significantly improved.

Deep Learning-based Approach for Automated Assessment of Interstitial Lung Disease in Systemic Sclerosis on CT Images

Purpose

To develop a deep learning algorithm for the automatic assessment of the extent of systemic sclerosis (SSc)-related interstitial lung disease (ILD) on chest CT images.

Materials and Methods

This retrospective study included 208 patients with SSc (median age, 57 years; 167 women) evaluated between January 2009 and October 2017. A multicomponent deep neural network (AtlasNet) was trained on 6888 fully annotated CT images (80% for training and 20% for validation) from 17 patients with no, mild, or severe lung disease. The model was tested on a dataset of 400 images from another 20 patients, independently partially annotated by three radiologist readers. The ILD contours from the three readers and the deep learning neural network were compared by using the Dice similarity coefficient (DSC). The correlation between disease extent obtained from the deep learning algorithm and that obtained by using pulmonary function tests (PFTs) was then evaluated in the remaining 171 patients and in an external validation dataset of 31 patients based on the analysis of all slices of the chest CT scan. The Spearman rank correlation coefficient (ρ) was calculated to evaluate the correlation between disease extent and PFT results.

Results

The median DSCs between the readers and the deep learning ILD contours ranged from 0.74 to 0.75, whereas the median DSCs between contours from radiologists ranged from 0.68 to 0.71. The disease extent obtained from the algorithm, by analyzing the whole CT scan, correlated with the diffusion lung capacity for carbon monoxide, total lung capacity, and forced vital capacity (ρ = -0.76, -0.70, and -0.62, respectively; P < .001 for all) in the dataset for the correlation with PFT results. The disease extents correlated with diffusion lung capacity for carbon monoxide, total lung capacity, and forced vital capacity were ρ = -0.65, -0.70, and -0.57, respectively, in the external validation dataset (P < .001 for all).

Conclusion

The developed algorithm performed similarly to radiologists for disease-extent contouring, which correlated with pulmonary function to assess CT images from patients with SSc-related ILD.Supplemental material is available for this article.© RSNA, 2020.

Emphysema quantification using low-dose computed tomography with deep learning-based kernel conversion comparison

OBJECTIVE

This study determined the effect of dose reduction and kernel selection on quantifying emphysema using low-dose computed tomography (LDCT) and evaluated the efficiency of a deep learning-based kernel conversion technique in normalizing kernels for emphysema quantification.

METHODS

A sample of 131 participants underwent LDCT and standard-dose computed tomography (SDCT) at 1- to 2-year intervals. LDCT images were reconstructed with B31f and B50f kernels, and SDCT images were reconstructed with B30f kernels. A deep learning model was used to convert the LDCT image from a B50f kernel to a B31f kernel. Emphysema indices (EIs), lung attenuation at 15th percentile (perc15), and mean lung density (MLD) were calculated. Comparisons among the different kernel types for both LDCT and SDCT were performed using Friedman's test and Bland-Altman plots.

RESULTS

All values of LDCT B50f were significantly different compared with the values of LDCT B31f and SDCT B30f (p < 0.05). Although there was a statistical difference, the variation of the values of LDCT B50f significantly decreased after kernel normalization. The 95% limits of agreement between the SDCT and LDCT kernels (B31f and converted B50f) ranged from - 2.9 to 4.3% and from - 3.2 to 4.4%, respectively. However, there were no significant differences in EIs and perc15 between SDCT and LDCT converted B50f in the non-chronic obstructive pulmonary disease (COPD) participants (p > 0.05).

CONCLUSION

The deep learning-based CT kernel conversion of sharp kernel in LDCT significantly reduced variation in emphysema quantification, and could be used for emphysema quantification.

KEY POINTS

• Low-dose computed tomography with smooth kernel showed adequate performance in quantifying emphysema compared with standard-dose CT. • Emphysema quantification is affected by kernel selection and the application of a sharp kernel resulted in a significant overestimation of emphysema. • Deep learning-based kernel normalization of sharp kernel significantly reduced variation in emphysema quantification.

Inter- and intra-software reproducibility of computed tomography lung density measurements

PURPOSE

Multiple commercial, open-source, and academic software tools exist for objective quantification of lung density in computed tomography (CT) images. The purpose of this study was to evaluate the intersoftware reproducibility of CT lung density measurements.

METHODS

Computed tomography images from 50 participants from the COPDGene^TM cohort study were randomly selected for analysis; n = 10 participants across each global initiative for chronic obstructive lung disease (GOLD) grade (GOLD 0-IV). Academic-based groups (n = 4) and commercial vendors (n = 4) participated anonymously to generate CT lung density measurements using their software tools. Computed tomography total lung volume (TLV), percentage of the low attenuation areas in the lung with Hounsfield unit (HU) values below -950HU (LAA₉₅₀ ), and the HU value corresponding to the 15th percentile on the parenchymal density histogram (Perc15) were included in the analysis. The intersoftware bias and reproducibility coefficient (RDC) was generated with and without quality assurance (QA) for manual correction of the lung segmentation; intrasoftware bias and RDC was also generated by repeated measurements on the same images.

RESULTS

Intersoftware mean bias was within ±0.22 mL, ±0.46%, and ±0.97 HU for TLV, LAA₉₅₀ and Perc15, respectively. The RDC was 0.35 L, 1.2% and 1.8 HU for TLV, LAA₉₅₀ and Perc15, respectively. Intersoftware RDC remained unchanged following QA: 0.35 L, 1.2% and 1.8 HU for TLV, LAA₉₅₀ and Perc15, respectively. All software investigated had an intrasoftware RDC of 0. The RDC was comparable for TLV, LAA₉₅₀ and Perc15 measurements, respectively, for academic-based groups/commercial vendor-based software tools: 0.39 L/0.32 L, 1.2%/1.2%, and 1.7 HU/1.6 HU. Multivariable regression analysis showed that academic-based software tools had greater within-subject standard deviation of TLV than commercial vendors, but no significant differences between academic and commercial groups were found for LAA₉₅₀ or Perc15 measurements.

CONCLUSIONS

Computed tomography total lung volume and lung density measurement bias and reproducibility was reported across eight different software tools. Bias was negligible across vendors, reproducibility was comparable for software tools generated by academic-based groups and commercial vendors, and segmentation QA had negligible impact on measurement variability between software tools. In summary, results from this study report the amount of additional measurement variability that should be accounted for when using different software tools to measure lung density longitudinally with well-standardized image acquisition protocols. However, intrasoftware reproducibility was deterministic for all cases so use of the same software tool to reduce variability for serial studies is highly recommended.

Comparison of Artificial Intelligence-Based Fully Automatic Chest CT Emphysema Quantification to Pulmonary Function Testing

OBJECTIVE. The purpose of this study was to evaluate an artificial intelligence (AI)-based prototype algorithm for fully automated quantification of emphysema on chest CT compared with pulmonary function testing (spirometry). MATERIALS AND METHODS. A total of 141 patients (72 women, mean age ± SD of 66.46 ± 9.7 years [range, 23-86 years]; 69 men, mean age of 66.72 ± 11.4 years [range, 27-91 years]) who underwent both chest CT acquisition and spirometry within 6 months were retrospectively included. The spirometry-based Tiffeneau index (TI; calculated as the ratio of forced expiratory volume in the first second to forced vital capacity) was used to measure emphysema severity; a value less than 0.7 was considered to indicate airway obstruction. Segmentation of the lung based on two different reconstruction methods was carried out by using a deep convolution image-to-image network. This multilayer convolutional neural network was combined with multilevel feature chaining and depth monitoring. To discriminate the output of the network from ground truth, an adversarial network was used during training. Emphysema was quantified using spatial filtering and attenuation-based thresholds. Emphysema quantification and TI were compared using the Spearman correlation coefficient. RESULTS. The mean TI for all patients was 0.57 ± 0.13. The mean percentages of emphysema using reconstruction methods 1 and 2 were 9.96% ± 11.87% and 8.04% ± 10.32%, respectively. AI-based emphysema quantification showed very strong correlation with TI (reconstruction method 1, ρ = -0.86; reconstruction method 2, ρ = -0.85; both p < 0.0001), indicating that AI-based emphysema quantification meaningfully reflects clinical pulmonary physiology. CONCLUSION. AI-based, fully automated emphysema quantification shows good correlation with TI, potentially contributing to an image-based diagnosis and quantification of emphysema severity.

Per cent low attenuation volume and fractal dimension of low attenuation clusters on CT predict different long-term outcomes in COPD

BACKGROUND

Fractal dimension (D) characterises the size distribution of low attenuation clusters on CT and assesses the spatial heterogeneity of emphysema that per cent low attenuation volume (%LAV) cannot detect. This study tested the hypothesis that %LAV and D have different roles in predicting decline in FEV₁, exacerbation and mortality in patients with COPD.

METHODS

Chest inspiratory CT scans in the baseline and longitudinal follow-up records for FEV₁, exacerbation and mortality prospectively collected over 10 years in the Hokkaido COPD Cohort Study were examined (n=96). The associations between CT measures and long-term outcomes were replicated in the Kyoto University cohort (n=130).

RESULTS

In the Hokkaido COPD cohort, higher %LAV, but not D, was associated with a greater decline in FEV₁ and 10-year mortality, whereas lower D, but not %LAV, was associated with shorter time to first exacerbation. Multivariable analysis for the Kyoto University cohort confirmed that lower D at baseline was independently associated with shorter time to first exacerbation and that higher LAV% was independently associated with increased mortality after adjusting for age, height, weight, FEV₁ and smoking status.

CONCLUSION

These well-established cohorts clarify the different prognostic roles of %LAV and D, whereby lower D is associated with a higher risk of exacerbation and higher %LAV is associated with a rapid decline in lung function and long-term mortality. Combination of %LAV and fractal D may identify COPD subgroups at high risk of a poor clinical outcome more sensitively.

Quantitative CT Analysis of Diffuse Lung Disease

Quantitative analysis of thin-section CT of the chest has a growing role in the clinical evaluation and management of diffuse lung diseases. This heterogeneous group includes diseases with markedly different prognoses and treatment options. Quantitative tools can assist in both accurate diagnosis and longitudinal management by improving characterization and quantification of disease and increasing the reproducibility of disease severity assessment. Furthermore, a quantitative index of disease severity may serve as a useful tool or surrogate endpoint in evaluating treatment efficacy. The authors explore the role of quantitative imaging tools in the evaluation and management of diffuse lung diseases. Lung parenchymal features can be classified with threshold, histogram, morphologic, and texture-analysis-based methods. Quantitative CT analysis has been applied in obstructive, infiltrative, and restrictive pulmonary diseases including emphysema, cystic fibrosis, asthma, idiopathic pulmonary fibrosis, hypersensitivity pneumonitis, connective tissue-related interstitial lung disease, and combined pulmonary fibrosis and emphysema. Some challenges limiting the development and practical application of current quantitative analysis tools include the quality of training data, lack of standard criteria to validate the accuracy of the results, and lack of real-world assessments of the impact on outcomes. Artifacts such as patient motion or metallic beam hardening, variation in inspiratory effort, differences in image acquisition and reconstruction techniques, or inaccurate preprocessing steps such as segmentation of anatomic structures may lead to inaccurate classification. Despite these challenges, as new techniques emerge, quantitative analysis is developing into a viable tool to supplement the traditional visual assessment of diffuse lung diseases and to provide decision support regarding diagnosis, prognosis, and longitudinal evaluation of disease. ^©RSNA, 2019.

Potential for dose reduction in CT emphysema densitometry with post-scan noise reduction: a phantom study

OBJECTIVE

The aim of this phantom study was to investigate the effect of scan parameters and noise suppression techniques on the minimum radiation dose for acceptable image quality for CT emphysema densitometry.

METHODS

The COPDGene phantom was scanned on a third generation dual-source CT system with 16 scan setups (CTDI_vol 0.035-10.680 mGy). Images were reconstructed at 1.0/0.7 mm slice thickness/increment, with three kernels (one soft, two hard), filtered backprojection and three grades of third-generation iterative reconstruction (IR). Additionally, deep learning-based noise suppression software was applied. Main outcomes: overlap in area of the normalized histograms of CT density for the emphysema insert and lung material, and the radiation dose required for a maximum of 4.3% overlap (defined as acceptable image quality).

RESULTS

In total, 384 scan reconstructions were analyzed. Decreasing radiation dose resulted in an exponential increase of the overlap in normalized histograms of CT density. The overlap was 11-91% for the lowest dose setting (CTDI_vol 0.035mGy). The soft kernel reconstruction showed less histogram overlap than hard filter kernels. IR and noise suppression also reduced overlap. Using intermediate grade IR plus noise suppression software allowed for 85% radiation dose reduction while maintaining acceptable image quality.

CONCLUSION

CT density histogram overlap can quantify the degree of discernibility of emphysema and healthy lung tissue. Noise suppression software, IR, and soft reconstruction kernels substantially decrease the dose required for acceptable image quality.

ADVANCES IN KNOWLEDGE

Noise suppression software, IR, and soft reconstruction kernels allow radiation dose reduction by 85% while still allowing differentiation between emphysema and normal lung tissue.

Quantitative Emphysema Measurement On Ultra-High-Resolution CT Scans

Purpose

To evaluate the advantages of ultra-high-resolution computed tomography (U-HRCT) scans for the quantitative measurement of emphysematous lesions over conventional HRCT scans.

Materials and methods

This study included 32 smokers under routine clinical care who underwent chest CT performed by a U-HRCT scanner. Chronic obstructive pulmonary disease (COPD) was diagnosed in 13 of the 32 participants. Scan data were reconstructed by 2 different protocols: i) U-HRCT mode with a 1024×1024 matrix and 0.25-mm slice thickness and ii) conventional HRCT mode with a 512×512 matrix and 0.5-mm slice thickness. On both types of scans, lesions of emphysema were quantitatively assessed as percentage of low attenuation volume (LAV%, <-950 Hounsfield units). LAV% values determined for scan data from the U-HRCT and conventional HRCT modes were compared by the Wilcoxon matched-pairs signed rank test. The association between LAV% and forced expiratory volume in 1 s per forced vital capacity (FEV₁/FVC) was assessed by the Spearman rank correlation test.

Results

Mean values for LAV% determined for the U-HRCT and conventional HRCT modes were 8.9 ± 8.8% and 7.3 ± 8.4%, respectively (P<0.0001). The correlation coefficients for LAV% and FEV₁/FVC on the U-HRCT and conventional HRCT modes were 0.50 and 0.49, respectively (both P<0.01).

Conclusion

Compared with conventional HRCT scans, U-HRCT scans reveal emphysematous lesions in greater detail, and provide slightly increased correlation with airflow limitation.

Quantitative CT analysis in patients with pulmonary emphysema: is lung function influenced by concomitant unspecific pulmonary fibrosis?

Purpose

Quantitative analysis of CT scans has proven to be a reproducible technique, which might help to understand the pathophysiology of chronic obstructive pulmonary disease (COPD) and combined pulmonary fibrosis and emphysema. The aim of this retrospective study was to find out if the lung function of patients with COPD with Global Initiative for Chronic Obstructive Lung Disease (GOLD) stages III or IV and pulmonary emphysema is measurably influenced by high attenuation areas as a correlate of concomitant unspecific fibrotic changes of lung parenchyma.

Patients and methods

Eighty-eight patients with COPD GOLD stage III or IV underwent CT and pulmonary function tests. Quantitative CT analysis was performed to determine low attenuation volume (LAV) and high attenuation volume (HAV), which are considered to be equivalents of fibrotic (HAV) and emphysematous (LAV) changes of lung parenchyma. Both parameters were determined for the whole lung, as well as peripheral and central lung areas only. Multivariate regression analysis was used to correlate HAV with different parameters of lung function.

Results

Unlike LAV, HAV did not show significant correlation with parameters of lung function. Even in patients with a relatively high HAV of more than 10%, in contrast to HAV (p=0.786) only LAV showed a significantly negative correlation with forced expiratory volume in 1 second (r=−0.309, R²=0.096, p=0.003). A severe decrease of DLCO% was associated with both larger HAV (p=0.045) and larger LAV (p=0.001). Residual volume and FVC were not influenced by LAV or HAV.

Conclusion

In patients with COPD GOLD stage III-IV, emphysematous changes of lung parenchyma seem to have such a strong influence on lung function, which is a possible effect of concomitant unspecific fibrosis is overwhelmed.

Harmonization of chest CT scans for different doses and reconstruction methods

PURPOSE

To develop and validate a computed tomography (CT) harmonization technique by combining noise-stabilization and autocalibration methodologies to provide reliable densitometry measurements in heterogeneous acquisition protocols.

METHODS

We propose to reduce the effects of spatially variant noise such as nonuniform patterns of noise and biases. The method combines the statistical characterization of the signal-to-noise relationship in the CT image intensities, which allows us to estimate both the signal and spatially variant variance of noise, with an autocalibration technique that reduces the nonuniform biases caused by noise and reconstruction techniques. The method is firstly validated with anthropomorphic synthetic images that simulate CT acquisitions with variable scanning parameters: different dosage, nonhomogeneous variance of noise, and various reconstruction methods. We finally evaluate these effects and the ability of our method to provide consistent densitometric measurements in a cohort of clinical chest CT scans from two vendors (Siemens, n = 54 subjects; and GE, n = 50 subjects) acquired with several reconstruction algorithms (filtered back-projection and iterative reconstructions) with high-dose and low-dose protocols.

RESULTS

The harmonization reduces the effect of nonhomogeneous noise without compromising the resolution of the images (25% RMSE reduction in both clinical datasets). An analysis through hierarchical linear models showed that the average biases induced by differences in dosage and reconstruction methods are also reduced up to 74.20%, enabling comparable results between high-dose and low-dose reconstructions. We also assessed the statistical similarity between acquisitions obtaining increases of up to 30% points and showing that the low-dose vs high-dose comparisons of harmonized data obtain similar and even higher similarity than the observed for high-dose vs high-dose comparisons of nonharmonized data.

CONCLUSION

The proposed harmonization technique allows to compare measures of low-dose with high-dose acquisitions without using a specific reconstruction as a reference. Since the harmonization does not require a precalibration with a phantom, it can be applied to retrospective studies. This approach might be suitable for multicenter trials for which a reference reconstruction is not feasible or hard to define due to differences in vendors, models, and reconstruction techniques.

Technical Note: Design and implementation of a high-throughput pipeline for reconstruction and quantitative analysis of CT image data

PURPOSE

With recent substantial improvements in modern computing, interest in quantitative imaging with CT has seen a dramatic increase. As a result, the need to both create and analyze large, high-quality datasets of clinical studies has increased as well. At present, no efficient, widely available method exists to accomplish this. The purpose of this technical note is to describe an open-source high-throughput computational pipeline framework for the reconstruction and analysis of diagnostic CT imaging data to conduct large-scale quantitative imaging studies and to accelerate and improve quantitative imaging research.

METHODS

The pipeline consists of two, primary "blocks": reconstruction and analysis. Reconstruction is carried out via a graphics processing unit (GPU) queuing framework developed specifically for the pipeline that allows a dataset to be reconstructed using a variety of different parameter configurations such as slice thickness, reconstruction kernel, and simulated acquisition dose. The analysis portion then automatically analyzes the output of the reconstruction using "modules" that can be combined in various ways to conduct different experiments. Acceleration of analysis is achieved using cluster processing. Efficiency and performance of the pipeline are demonstrated using an example 142 subject lung screening cohort reconstructed 36 different ways and analyzed using quantitative emphysema scoring techniques.

RESULTS

The pipeline reconstructed and analyzed the 5112 reconstructed datasets in approximately 10 days, a roughly 72× speedup over previous efforts using the scanner for reconstructions. Tightly coupled pipeline quality assurance software ensured proper performance of analysis modules with regard to segmentation and emphysema scoring.

CONCLUSIONS

The pipeline greatly reduced the time from experiment conception to quantitative results. The modular design of the pipeline allows the high-throughput framework to be utilized for other future experiments into different quantitative imaging techniques. Future applications of the pipeline being explored are robustness testing of quantitative imaging metrics, data generation for deep learning, and use as a test platform for image-processing techniques to improve clinical quantitative imaging.