Quantitative computed tomography analysis of the airways in patients with cystic fibrosis using automated software: correlation with spirometry in the evaluation of severity

Correlation among clinical, functional and morphological indexes of the respiratory system in non-cystic fibrosis bronchiectasis patients

BACKGROUND

Non-cystic fibrosis bronchiectasis (NCFB) is a heterogeneous disease, which assessment and severity can't be defined by one particular instrument but using a multidimensional score. Thus, in additional to traditional methods, alternative tools have been developed to assist these patients' evaluation.

OBJECTIVE

To correlate functional and morphological indexes with severity and dyspnea in NCFB patients, focusing on the correlation between the impulse oscillometry system (IOS) and the quantitative analysis of computed tomography (CT).

METHODS

Clinically stable NCFB patients, between 18 and 80 years old were submitted to clinical, functional and morphological evaluations assessed by Bronchiectasis Severity Index (BSI) and Medical Research Council (MRC) scale; spirometry and IOS; and subjective and quantitative Chest CT scans analysis, respectively.

RESULTS

This study included 38 patients. The best correlations obtained between functional and morphological airway indexes were: resistance at 5 Hz-R5 and the normalized thickness of bronchial walls-Pi10 (r = 0.57), and the mean forced expiratory flow (FEF25-75%) and CT score (r = -0.39). BSI as well as MRC showed higher correlations with the quantitative automated analysis of CT (BSI and Pi10: r = 0.41; MRC and Pi10: r = 0.35) than with subjective CT score (BSI and CT score: r = 0.41; MRC and CT score: r = 0.15); and moderate and weak correlations were obtained on both functional airway indexes (BSI and peripheral airways resistance - R5-R20: r = 0.53; BSI and forced expiratory volume at the first second-FEV1: R = -0,64; MRC and R5-R20: r = 0.42; and MRC and VEF1: r = -0.45).

CONCLUSION

In NCFB patients, compartmentalized methods for assessing the respiratory system (IOS and the automated quantitative CT analysis) have a good correlation with severity and dyspnea.

Radiology of Bronchiectasis

Bronchiectasis is a radiological diagnosis made using computed tomographic (CT) imaging. Although visual CT assessment is necessary for the diagnosis of bronchiectasis, visual assessment of disease severity and progression is challenging. Computer tools offer the potential to improve the characterization of lung damage in patients with bronchiectasis. Newer imaging techniques such as MRI with hyperpolarized gas inhalation have the potential to identify early forms of disease and are without the constraints of requiring ionizing radiation exposure.

Computed tomography in cystic fibrosis lung disease: a focus on radiation exposure

Thoracic computed tomography (CT) is the imaging reference method in the diagnosis, assessment and management of lung disease. In the setting of cystic fibrosis (CF), CT demonstrates increased sensitivity compared with pulmonary function tests and chest radiography, and findings correlate with clinical outcomes. Better understanding of the aetiology of CF lung disease indicates that even asymptomatic infants with CF can have irreversible pulmonary pathology. Surveillance and early diagnosis of lung disease in CF are important to preserve lung parenchyma and to optimise long-term outcomes. CF is associated with increased cumulative radiation exposure due to the requirement for repeated imaging from a young age. Radiation dose optimisation, important for the safe use of CT in children with CF, is best achieved in a team environment where paediatric radiologists work closely with paediatric respiratory physicians, physicists and radiography technicians to achieve the best patient outcomes. Despite the radiation doses incurred, CT remains a vital imaging tool in children with CF. Radiologists with special interests in CT dose optimisation and respiratory disease are key to the appropriate use of CT in paediatric imaging. Paediatric radiologists strive to minimise radiation dose to children whilst providing the best possible assessment of lung disease.

Automatic Quantitative Computed Tomography Evaluation of the Lungs in Patients With Systemic Sclerosis Treated With Autologous Stem Cell Transplantation

BACKGROUND/OBJECTIVE

Interstitial lung disease stands among the leading causes of death in systemic sclerosis (SSc) patients. Autologous hematopoietic stem cell transplantation (AHSCT) has been proven superior to conventional immunosuppressive therapy in severe and progressive SSc. Here, pulmonary quantitative measurements were obtained in high-resolution computed tomography (HRCT) scans of patients with SSc before and after AHSCT.

METHODS

The medical records of thirthy-three patients who underwent AHSCT between 2011 and 2017 were evaluated for clinical and tomographic features at baseline (pre-AHCST) and 18 months after the procedure. Quantitative analysis of HRCT images by a fully automated program calculated lung volumes, densities, attenuation percentiles, and vascular volume. Patients were divided into 2 groups, according to changes in forced vital capacity (FVC). The "best response" group included patients that had an increased FVC of 10% or greater, and the "stable response" group included those who had a decreased or an increased FVC of less than 10%.

RESULTS

In the best response group (15 patients), there was reduction (p < 0.05) of mean lung density and density percentile values after AHSCT. In the stable response group (18 patients), there were no significant changes in lung volumes and pulmonary densities after AHSCT. Pulmonary HRCT densities showed moderate/strong correlation with function.

CONCLUSIONS

Quantitative HRCT analysis identified significant reduction in pulmonary densities in patients with improved pulmonary function after AHSCT. Lung density, as evaluated by the quantitative HRCT analysis tool, has potential to become a biomarker in the evaluation of interstitial lung disease treatment in patients with SSc.

Artificial intelligence, machine learning, computer-aided diagnosis, and radiomics: advances in imaging towards to precision medicine

The discipline of radiology and diagnostic imaging has evolved greatly in recent years. We have observed an exponential increase in the number of exams performed, subspecialization of medical fields, and increases in accuracy of the various imaging methods, making it a challenge for the radiologist to "know everything about all exams and regions". In addition, imaging exams are no longer only qualitative and diagnostic, providing now quantitative information on disease severity, as well as identifying biomarkers of prognosis and treatment response. In view of this, computer-aided diagnosis systems have been developed with the objective of complementing diagnostic imaging and helping the therapeutic decision-making process. With the advent of artificial intelligence, "big data", and machine learning, we are moving toward the rapid expansion of the use of these tools in daily life of physicians, making each patient unique, as well as leading radiology toward the concept of multidisciplinary approach and precision medicine. In this article, we will present the main aspects of the computational tools currently available for analysis of images and the principles of such analysis, together with the main terms and concepts involved, as well as examining the impact that the development of artificial intelligence has had on radiology and diagnostic imaging.

Idiopathic interstitial pneumonias: review of the latest American Thoracic Society/European Respiratory Society classification

The diagnosis of idiopathic interstitial pneumonias (IIPs) involves a multidisciplinary scenario in which the radiologist assumes a key role. The latest (2013) update of the IIP classification by the American Thoracic Society/European Respiratory Society proposed some important changes to the original classification of 2002. The novelties include the addition of a new disease (idiopathic pleuroparenchymal fibroelastosis) and the subdivision of the IIPs into four main groups: chronic fibrosing IIPs (idiopathic pulmonary fibrosis and nonspecific interstitial pneumonia); smoking-related IIPs (desquamative interstitial pneumonia and respiratory bronchiolitis-associated interstitial lung disease); acute or subacute IIPs (cryptogenic organizing pneumonia and acute interstitial pneumonia); rare IIPs (lymphoid interstitial pneumonia and idiopathic pleuroparenchymal fibroelastosis); and the so-called "unclassifiable" IIPs. In this study, we review the main clinical, tomographic, and pathological characteristics of each IIP.

Congruence Between Pulmonary Function and Computed Tomography Imaging Assessment of Cystic Fibrosis Severity

In cystic fibrosis, pulmonary function tests (PFTs) and computed tomography are used to assess lung function and structure, respectively. Although both techniques of assessment are congruent there are lingering doubts about which PFTs variables show the best congruence with computed tomography scoring. In this study we addressed the issue by reinvestigating the association between PFTs variables and the score of changes seen in computed tomography scans in patients with cystic fibrosis with and without pulmonary exacerbation. This retrospective study comprised 40 patients in whom PFTs and computed tomography were performed no longer than 3 weeks apart. Images (inspiratory: 0.625 mm slice thickness, 0.625 mm interval; expiratory: 1.250 mm slice thickness, 10 mm interval) were evaluated with the Bhalla scoring system. The most frequent structural abnormality found in scans were bronchiectases and peribronchial thickening. The strongest relationship was found between the Bhalla sore and forced expiratory volume in 1 s (FEV1). The Bhalla sore also was related to forced vital capacity (FVC), FEV1/FVC ratio, residual volume (RV), and RV/total lung capacity (TLC) ratio. We conclude that lung structural data obtained from the computed tomography examination are highly congruent to lung function data. Thus, computed tomography imaging may supersede functional assessment in cases of poor compliance with spirometry procedures in the lederly or children. Computed tomography also seems more sensitive than PFTs in the assessment of cystic fibrosis progression. Moreover, in early phases of cystic fibrosis, computed tomography, due to its excellent resolution, may be irreplaceable in monitoring pulmonary damage.

Utility of the inspiratory phase in high-resolution computed tomography evaluations of pediatric patients with bronchiolitis obliterans after allogeneic bone marrow transplant: reducing patient radiation exposure

OBJECTIVE

To evaluate the utility of the inspiratory phase in high-resolution computed tomography (HRCT) of the chest for the diagnosis of post-bone marrow transplantation bronchiolitis obliterans.

MATERIALS AND METHODS

This was a retrospective, observational, cross-sectional study. We selected patients of either gender who underwent bone marrow transplantation and chest HRCT between March 1, 2002 and December 12, 2014. Ages ranged from 3 months to 20.7 years. We included all examinations in which the HRCT was performed appropriately. The examinations were read by two radiologists, one with extensive experience in pediatric radiology and another in the third year of residency, who determined the presence or absence of the following imaging features: air trapping, bronchiectasis, alveolar opacities, nodules, and atelectasis.

RESULTS

A total of 222 examinations were evaluated (mean, 5.4 ± 4.5 examinations per patient). The expiratory phase findings were comparable to those obtained in the inspiratory phase, except in one patient, in whom a small uncharacteristic nodule was identified only in the inspiratory phase. Air trapping was identified in a larger number of scans in the expiratory phase than in the inspiratory phase, as was atelectasis, although the difference was statistically significant only for air trapping.

CONCLUSION

In children being evaluated for post-bone marrow transplantation bronchiolitis obliterans, the inspiratory phase can be excluded from the chest HRCT protocol, thus reducing by half the radiation exposure in this population.

Radiologist agreement on the quantification of bronchiectasis by high-resolution computed tomography

Objective

To evaluate radiologist agreement on the quantification of bronchiectasis by high-resolution computed tomography (HRCT).

Materials and Methods

The HRCT scans of 43 patients with bronchiectasis were analyzed by two radiologists, who used a scoring system to grade the findings. Kappa (κ) values and overall agreement were calculated.

Results

For the measurement and appearance of bronchiectasis, the interobserver agreement was moderate (κ = 0.45 and κ = 0.43, respectively), as was the intraobserver agreement (κ = 0.54 and κ = 0.47, respectively). Agreement on the presence of mucous plugging was fair, for central distribution (overall interobserver agreement of 68.3% and κ = 0.39 for intraobserver agreement) and for peripheral distribution (κ = 0.34 and κ = 0.35 for interobserver and intraobserver agreement, respectively). The agreement was also fair for peribronchial thickening (κ = 0.21 and κ = 0.30 for interobserver and intraobserver agreement, respectively). There was fair interobserver and intraobserver agreement on the detection of opacities (κ = 0.39 and 71.9%, respectively), ground-glass attenuation (64.3% and κ = 0.24, respectively), and cysts/bullae (κ = 0.47 and κ = 0.44, respectively). Qualitative analysis of the HRCT findings of bronchiectasis and the resulting individual patient scores showed that there was an excellent correlation between the observers (intraclass correlation coefficient of 0.85 and 0.81 for interobserver and intraobserver agreement, respectively).

Conclusion

In the interpretation of HRCT findings of bronchiectasis, radiologist agreement appears to be fair. In our final analysis of the findings using the proposed score, we observed excellent interobserver and intraobserver agreement.