Crowdsourcing airway annotations in chest computed tomography images.

Measuring airways in chest computed tomography (CT) scans is important for characterizing diseases such as cystic fibrosis, yet very time-consuming to perform manually. Machine learning algorithms offer an alternative, but need large sets of annotated scans for good performance. We investigate whether crowdsourcing can be used to gather airway annotations. We generate image slices at known locations of airways in 24 subjects and request the crowd workers to outline the airway lumen and airway wall. After combining multiple crowd workers, we compare the measurements to those made by the experts in the original scans. Similar to our preliminary study, a large portion of the annotations were excluded, possibly due to workers misunderstanding the instructions. After excluding such annotations, moderate to strong correlations with the expert can be observed, although these correlations are slightly lower than inter-expert correlations. Furthermore, the results across subjects in this study are quite variable. Although the crowd has potential in annotating airways, further development is needed for it to be robust enough for gathering annotations in practice. For reproducibility, data and code are available online: http://github.com/adriapr/crowdairway.git.

Airway tapering: an objective image biomarker for bronchiectasis.

PURPOSE: To estimate airway tapering in control subjects and to assess the usability of tapering as a bronchiectasis biomarker in paediatric populations. METHODS: Airway tapering values were semi-automatically quantified in 156 children with control CTs collected in the Normal Chest CT Study Group. Airway tapering as a biomarker for bronchiectasis was assessed on spirometer-guided inspiratory CTs from 12 patients with bronchiectasis and 12 age- and sex-matched controls. Semi-automatic image analysis software was used to quantify intra-branch tapering (reduction in airway diameter along the branch), inter-branch tapering (reduction in airway diameter before and after bifurcation) and airway-artery ratios on chest CTs. Biomarkers were further stratified in small, medium and large airways based on three equal groups of the accompanying vessel size. RESULTS: Control subjects showed intra-branch tapering of 1% and inter-branch tapering of 24-39%. Subjects with bronchiectasis showed significantly reduced intra-branch of 0.8% and inter-branch tapering of 19-32% and increased airway-artery ratios compared with controls (p < 0.01). Tapering measurements were significantly different between diseased and controls across all airway sizes. Difference in airway-artery ratio was only significant in small airways. CONCLUSION: Paediatric normal values for airway tapering were established in control subjects. Tapering showed to be a promising biomarker for bronchiectasis as subjects with bronchiectasis show significantly less airway tapering across all airway sizes compared with controls. Detecting less tapering in larger airways could potentially lead to earlier diagnosis of bronchiectasis. Additionally, compared with the conventional airway-artery ratio, this novel biomarker has the advantage that it does not require pairing with pulmonary arteries. KEY POINTS: • Tapering is a promising objective image biomarker for bronchiectasis that can be extracted semi-automatically and has good correlation with validated visual scoring methods. • Less airway tapering was observed in patients with bronchiectasis and can be observed sensitively throughout the bronchial tree, even in the more central airways. • Tapering values seemed to be less influenced by variety in scanning protocols and lung volume making it a more robust biomarker for bronchiectasis detection.

Technical challenges of quantitative chest MRI data analysis in a large cohort pediatric study.

OBJECTIVES: This study was conducted in order to evaluate the effect of geometric distortion (GD) on MRI lung volume quantification and evaluate available manual, semi-automated, and fully automated methods for lung segmentation. METHODS: A phantom was scanned with MRI and CT. GD was quantified as the difference in phantom's volume between MRI and CT, with CT as gold standard. Dice scores were used to measure overlap in shapes. Furthermore, 11 subjects from a prospective population-based cohort study each underwent four chest MRI acquisitions. The resulting 44 MRI scans with 2D and 3D Gradwarp were used to test five segmentation methods. Intraclass correlation coefficient, Bland-Altman plots, Wilcoxon, Mann-Whitney U, and paired t tests were used for statistics. RESULTS: Using phantoms, volume differences between CT and MRI varied according to MRI positions and 2D and 3D Gradwarp correction. With the phantom located at the isocenter, MRI overestimated the volume relative to CT by 5.56 ± 1.16 to 6.99 ± 0.22% with body and torso coils, respectively. Higher Dice scores and smaller intraobject differences were found for 3D Gradwarp MR images. In subjects, semi-automated and fully automated segmentation tools showed high agreement with manual segmentations (ICC = 0.971-0.993 for end-inspiratory scans; ICC = 0.992-0.995 for end-expiratory scans). Manual segmentation time per scan was approximately 3-4 h and 2-3 min for fully automated methods. CONCLUSIONS: Volume overestimation of MRI due to GD can be quantified. Semi-automated and fully automated segmentation methods allow accurate, reproducible, and fast lung volume quantification. Chest MRI can be a valid radiation-free imaging modality for lung segmentation and volume quantification in large cohort studies. KEY POINTS: • Geometric distortion varies according to MRI setting and patient positioning. • Automated segmentation methods allow fast and accurate lung volume quantification. • MRI is a valid radiation-free alternative to CT for quantitative data analysis.

Diagnosis of bronchiectasis and airway wall thickening in children with cystic fibrosis: Objective airway-artery quantification.

OBJECTIVES: To quantify airway and artery (AA)-dimensions in cystic fibrosis (CF) and control patients for objective CT diagnosis of bronchiectasis and airway wall thickness (AWT). METHODS: Spirometer-guided inspiratory and expiratory CTs of 11 CF and 12 control patients were collected retrospectively. Airway pathways were annotated semi-automatically to reconstruct three-dimensional bronchial trees. All visible AA-pairs were measured perpendicular to the airway axis. Inner, outer and AWT (outer-inner) diameter were divided by the adjacent artery diameter to compute AinA-, AoutA- and AWTA-ratios. AA-ratios were predicted using mixed-effects models including disease status, lung volume, gender, height and age as covariates. RESULTS: Demographics did not differ significantly between cohorts. Mean AA-pairs CF: 299 inspiratory; 82 expiratory. CONTROLS: 131 inspiratory; 58 expiratory. All ratios were significantly larger in inspiratory compared to expiratory CTs for both groups (p<0.001). AoutA- and AWTA-ratios were larger in CF than in controls, independent of lung volume (p<0.01). Difference of AoutA- and AWTA-ratios between patients with CF and controls increased significantly for every following airway generation (p<0.001). CONCLUSION: Diagnosis of bronchiectasis is highly dependent on lung volume and more reliably diagnosed using outer airway diameter. Difference in bronchiectasis and AWT severity between the two cohorts increased with each airway generation. KEY POINTS: • More peripheral airways are visible in CF patients compared to controls. • Structural lung changes in CF patients are greater with each airway generation. • Number of airways visualized on CT could quantify CF lung disease. • For objective airway disease quantification on CT, lung volume standardization is required.

Objective airway artery dimensions compared to CT scoring methods assessing structural cystic fibrosis lung disease.

Background: CF-CT and PRAGMA-CF are commonly used scoring methods to quantify the severity of bronchiectasis (BE) and airway wall thickening (AWT) on chest CTs of children with cystic fibrosis (CF). We aimed to validate CF-CT and PRAGMA-CF sub-scores for BE and AWT against quantitative airway­artery (AA) dimensions. Methods: This is a retrospective study with 23 spirometer guided inspiratory chest CTs (11 CF, 12 controls; age range 6 to 16 years old) included. AA-, and AWTA-ratios of all visible AA pairs were computed by dividing diameters of the outer airway and wall (outer-inner airway) by the accompanying artery diameter, respectively. BE, AWT and total airway disease (TAD) were scored using CF-CT (% max score) and PRAGMA-CF (% extent). Correlations were computed using Spearman rank. Akaike information criterion (AIC) from the mixed-effects models were used to investigate whether CF-CT or PRAGMA-CF was a better predictor for AA-, and AWTA-ratios (lower AIC equals a better fitted model). Results: 4861 AA pairs were measured in total. Correlations between CF-CT and PRAGMA-CF: BE (r = 0.93, P < 0.001); AWT (r = 0.62, P < 0.001); TAD (r = 0.88, P < 0.001). PRAGMA-CF TAD sub-score had lowest AIC in the mixed-model predicting AA-ratio. CF-CT AWT and PRAGMA-CF TAD sub-score had equal low AIC in the mixed-model predicting AWTA-ratio. Conclusion: PRAGMA-CF TAD sub-score was more precise predicting BE. CF-CT AWT and PRAGMA-CF TAD sub-scores predicted AWT equally well. CF-CT and PRAGMA-CF were both sensitive methods to score BE and AWT in children with CF lung disease, with PRAGMA-CT TAD sub-score being most accurate in predicting AA dimensions.

Automatic airway-artery analysis on lung CT to quantify airway wall thickening and bronchiectasis.

PURPOSE: Bronchiectasis and airway wall thickening are commonly assessed in computed tomography (CT) by comparing the airway size with the size of the accompanying artery. Thus, in order to automate the quantification of bronchiectasis and wall thickening following a similar principle, there is a need for methods that automatically segment the airway and vascular trees, measure their size, and pair each airway branch with its accompanying artery. METHODS: This paper combines and extends existing techniques to present a fully automated pipeline that, given a thoracic chest CT, segments, measures, and pairs airway branches with the accompanying artery, then quantifies airway wall thickening and bronchiectasis by measuring the wall-artery ratio (WAR) and lumen and outer wall airway-artery ratio (AAR). Measurements that do not use the artery size for normalization are also extracted, including wall area percentage (WAP), wall thickness ratio (WTR), and airway diameters. RESULTS: The method was thoroughly evaluated using 8000 manual annotations of airway-artery pairs from 24 full-inspiration pediatric CT scans (12 diseased and 12 controls). Limits of agreement between the automatically and manually measured diameters were comparable to interobserver limits of agreement. Differences in automatically obtained WAR, AAR, WAP, and WTR between bronchiectatic subjects and controls were similar as when manual annotations were used: WAR and outer AAR were significantly higher in the bronchiectatic subjects (p < 0.05), but lumen AAR, WAP, and WTR were not. Only measurements that use artery size for normalization led to significant differences between groups, highlighting the importance of airway-artery pairing. CONCLUSIONS: The fully automatic method presented in this paper could replace time-consuming manual annotations and visual scoring methods to quantify abnormal widening and thickening of airways.

Quantification of Diaphragm Mechanics in Pompe Disease Using Dynamic 3D MRI.

BACKGROUND: Diaphragm weakness is the main reason for respiratory dysfunction in patients with Pompe disease, a progressive metabolic myopathy affecting respiratory and limb-girdle muscles. Since respiratory failure is the major cause of death among adult patients, early identification of respiratory muscle involvement is necessary to initiate treatment in time and possibly prevent irreversible damage. In this paper we investigate the suitability of dynamic MR imaging in combination with state-of-the-art image analysis methods to assess respiratory muscle weakness. METHODS: The proposed methodology relies on image registration and lung surface extraction to quantify lung kinematics during breathing. This allows for the extraction of geometry and motion features of the lung that characterize the independent contribution of the diaphragm and the thoracic muscles to the respiratory cycle. RESULTS: Results in 16 3D+t MRI scans (10 Pompe patients and 6 controls) of a slow expiratory maneuver show that kinematic analysis from dynamic 3D images reveals important additional information about diaphragm mechanics and respiratory muscle involvement when compared to conventional pulmonary function tests. Pompe patients with severely reduced pulmonary function showed severe diaphragm weakness presented by minimal motion of the diaphragm. In patients with moderately reduced pulmonary function, cranial displacement of posterior diaphragm parts was reduced and the diaphragm dome was oriented more horizontally at full inspiration compared to healthy controls. CONCLUSION: Dynamic 3D MRI provides data for analyzing the contribution of both diaphragm and thoracic muscles independently. The proposed image analysis method has the potential to detect less severe diaphragm weakness and could thus be used to determine the optimal start of treatment in adult patients with Pompe disease in prospect of increased treatment response.

Multicentre chest computed tomography standardisation in children and adolescents with cystic fibrosis: the way forward.

Progressive cystic fibrosis (CF) lung disease is the main cause of mortality in CF patients. CF lung disease starts in early childhood. With current standards of care, respiratory function remains largely normal in children and more sensitive outcome measures are needed to monitor early CF lung disease. Chest CT is currently the most sensitive imaging modality to monitor pulmonary structural changes in children and adolescents with CF. To quantify structural lung disease reliably among multiple centres, standardisation of chest CT protocols is needed. SCIFI CF (Standardised Chest Imaging Framework for Interventions and Personalised Medicine in CF) was founded to characterise chest CT image quality and radiation doses among 16 participating European CF centres in 10 different countries. We aimed to optimise CT protocols in children and adolescents among several CF centres. A large variety was found in CT protocols, image quality and radiation dose usage among the centres. However, the performance of all CT scanners was found to be very similar, when taking spatial resolution and radiation dose into account. We conclude that multicentre standardisation of chest CT in children and adolescents with CF can be achieved for future clinical trials.

Lung MRI and impairment of diaphragmatic function in Pompe disease.

BACKGROUND: Pompe disease is a progressive metabolic myopathy. Involvement of respiratory muscles leads to progressive pulmonary dysfunction, particularly in supine position. Diaphragmatic weakness is considered to be the most important component. Standard spirometry is to some extent indicative but provides too little insight into diaphragmatic dynamics. We used lung MRI to study diaphragmatic and chest-wall movements in Pompe disease. METHODS: In ten adult Pompe patients and six volunteers, we acquired two static spirometer-controlled MRI scans during maximum inspiration and expiration. Images were manually segmented. After normalization for lung size, changes in lung dimensions between inspiration and expiration were used for analysis; normalization was based on the cranial-caudal length ratio (representing vertical diaphragmatic displacement), and the anterior-posterior and left-right length ratios (representing chest-wall movements due to thoracic muscles). RESULTS: We observed striking dysfunction of the diaphragm in Pompe patients; in some patients the diaphragm did not show any displacement. Patients had smaller cranial-caudal length ratios than volunteers (p < 0.001), indicating diaphragmatic weakness. This variable strongly correlated with forced vital capacity in supine position (r = 0.88) and postural drop (r = 0.89). While anterior-posterior length ratios also differed between patients and volunteers (p = 0.04), left-right length ratios did not (p = 0.1). CONCLUSIONS: MRI is an innovative tool to visualize diaphragmatic dynamics in Pompe patients and to study chest-walland diaphragmatic movements in more detail. Our data indicate that diaphragmatic displacement may be severely disturbed in patients with Pompe disease.

RERBEE: robust efficient registration via bifurcations and elongated elements applied to retinal fluorescein angiogram sequences.

We present RERBEE (robust efficient registration via bifurcations and elongated elements), a novel feature-based registration algorithm able to correct local deformations in high-resolution ultra-wide field-of-view (UWFV) fluorescein angiogram (FA) sequences of the retina. The algorithm is able to cope with peripheral blurring, severe occlusions, presence of retinal pathologies and the change of image content due to the perfusion of the fluorescein dye in time. We have used the computational power of a graphics processor to increase the performance of the most computationally expensive parts of the algorithm by a factor of over × 1300, enabling the algorithm to register a pair of 3900 × 3072 UWFV FA images in 5-10 min instead of the 5-7 h required using only the CPU. We demonstrate accurate results on real data with 267 image pairs from a total of 277 (96.4%) graded as correctly registered by a clinician and 10 (3.6%) graded as correctly registered with minor errors but usable for clinical purposes. Quantitative comparison with state-of-the-art intensity-based and feature-based registration methods using synthetic data is also reported. We also show some potential usage of a correctly aligned sequence for vein/artery discrimination and automatic lesion detection.

Deformable registration of retinal fluorescein angiogram sequences using vasculature structures.

State-of-the-art deformable registration algorithms do not perform as well with FA sequences because they are designed to deal with changes of content appearance (e.g., due to different sensors imaging the same organs) but not with content changes, which occur throughout a FA sequence as different portions or the vascular structure are visible (perfused) in different frames. This paper presents a frame-to-frame registration algorithm for ultra-wide-field-of-view (UWFV) fluorescein angiograms (FA) of the retina, based on deformable alignment of the retinal vasculature structure. Comparative experiments on an initial set of UWFV FAs indicate that, thanks to its specialization, our technique outperforms one of the best state-of-the-art methods for multimodal image registration when dealing with the demanding characteristics of the UWFV FA sequences.

Robust optic disc location via combination of weak detectors.

This paper presents a new robust approach for the automatic location of the optic disc. We detect several candidates independently for optic disc, macula and the main blood vessels (arcades). Candidates are sorted by reliability. The space of all possible triplets disc-macula-arcades is searched using a-priori anatomical knowledge, selecting the triplet formed by the most reliable candidates satisfying anatomical constraints best. This triplet includes the best optic disc location. The approach was tested using the well-known STARE data set (81 images: 31 healthy retinas and 50 containing pathological lesions of various types and severity) and 20 wide-field-of-view scanning laser ophthalmoscope (SLO) images acquired with an OPTOS instrument. We achieved 91.4% detection rate in the STARE data set and 100% in the OPTOS data set.