Dante Capaldi, PhD, simultaneously completed both PhD and MClSc degrees in Medical Biophysics at the University of Western Ontario, Canada, in 2018. His PhD research, funded by the Natural Sciences and Engineering Research Council of Canada, focused on the development and application of novel image acquisition and analysis methods to measure pulmonary ventilation in patients with lung disease.

Dante joined the Stanford University Medical Physics Residency in 2018.

Academic Appointments

All Publications

  • Ultra-short echo-time magnetic resonance imaging lung segmentation with under-Annotations and domain shift. Medical image analysis Guo, F., Capaldi, D. P., McCormack, D. G., Fenster, A., Parraga, G. 2021; 72: 102107


    Ultra-short echo-time (UTE) magnetic resonance imaging (MRI) provides enhanced visualization of pulmonary structural and functional abnormalities and has shown promise in phenotyping lung disease. Here, we describe the development and evaluation of a lung segmentation approach to facilitate UTE MRI methods for patient-based imaging. The proposed approach employs a k-means algorithm in kernel space for pair-wise feature clustering and imposes image domain continuous regularization, coined as continuous kernel k-means (CKKM). The high-order CKKM algorithm was simplified through upper bound relaxation and solved within an iterative continuous max-flow framework. We combined the CKKM with U-net and atlas-based approaches and comprehensively evaluated the performance on 100 images from 25 patients with asthma and bronchial pulmonary dysplasia enrolled at Robarts Research Institute (Western University, London, Canada) and Centre Hospitalier Universitaire (Sainte-Justine, Montreal, Canada). For U-net, we trained the network five times on a mixture of five different images with under-annotations and applied the model to 64 images from the two centres. We also trained a U-net on five images with full and brush annotations from one centre, and tested the model on 32 images from the other centre. For an atlas-based approach, we employed three atlas images to segment 64 target images from the two centres through straightforward atlas registration and label fusion. We applied the CKKM algorithm to the baseline U-net and atlas outputs and refined the initial segmentation through multi-volume image fusion. The integration of CKKM substantially improved baseline results and yielded, with minimal computational cost, segmentation accuracy, and precision that were greater than some state-of-the-art deep learning models and similar to experienced observer manual segmentation. This suggests that deep learning and atlas-based approaches may be utilized to segment UTE MRI datasets using relatively small training datasets with under-annotations.

    View details for DOI 10.1016/

    View details for PubMedID 34153626

  • Prior-image-based CT reconstruction using attenuation-mismatched priors. Physics in medicine and biology Zhang, H., Capaldi, D., Zeng, D., Ma, J., Xing, L. 2021; 66 (6): 064007


    Prior-image-based reconstruction (PIBR) methods are powerful tools for reducing radiation doses and improving the image quality of low-dose computed tomography (CT). Apart from anatomical changes, prior and current images can also have different attenuations because they originated from different scanners or from the same scanner but with different x-ray beam qualities (e.g., kVp settings, beam filters) during data acquisition. In such scenarios, with attenuation-mismatched priors, PIBR is challenging. In this work, we investigate a specific PIBR method, called statistical image reconstruction, using normal-dose image-induced nonlocal means regularization (SIR-ndiNLM), to address PIBR with such attenuation-mismatched priors and achieve quantitative low-dose CT imaging. We propose two corrective schemes for the original SIR-ndiNLM method, (1) a global histogram-matching approach and (2) a local attenuation correction approach, to account for the attenuation differences between the prior and current images in PIBR. We validate the efficacy of the proposed schemes using images acquired from dual-energy CT scanners to simulate attenuation mismatches. Meanwhile, we utilize different CT slices to simulate anatomical mismatches or changes between the prior and the current low-dose image. We observe that the original SIR-ndiNLM introduces artifacts to the reconstruction when an attenuation-mismatched prior is used. Furthermore, we find that a larger attenuation mismatch between the prior and current images results in more severe artifacts in the SIR-ndiNLM reconstruction. Our two proposed corrective schemes enable SIR-ndiNLM to effectively handle the attenuation mismatch and anatomical changes between the two images and successfully eliminate the artifacts. We demonstrate that the proposed techniques permit SIR-ndiNLM to leverage the attenuation-mismatched prior and achieve quantitative low-dose CT reconstruction from both low-flux and sparse-view data acquisitions. This work permits robust and reliable PIBR for CT data acquired using different beam settings.

    View details for DOI 10.1088/1361-6560/abe760

    View details for PubMedID 33729997

  • Prior-image-based CT reconstruction using attenuation mismatched prior. Physics in medicine and biology Zhang, H., Capaldi, D. P., Zeng, D., Ma, J., Xing, L. 2021


    Prior-image-based reconstruction (PIBR) methods are powerful in reducing radiation dose and improving image quality for low-dose CT. Besides anatomical changes, the prior and current images can also have different attenuation due to different scanners or the same scanner but with different x-ray beam quality (e.g., kVp setting, beam filtration) during data acquisitions. PIBR is challenged in such scenarios with attenuation mismatched prior. In this work, we investigate a specific PIBR method, called statistical image reconstruction using normal dose image induced nonlocal means regularization (SIR-ndiNLM), to address PIBR with such attenuation mismatched prior and achieve quantitative low-dose CT imaging. We proposed two corrective schemes for the original SIR-ndiNLM method, 1) a global histogram matching approach and 2) a local attenuation correction approach, to account for the attenuation differences between the prior and current images in PIBR. We validated the efficacy of the proposed schemes using images acquired from dual-energy CT scanners to emulate attenuation mismatches. Meanwhile, we utilized different CT slices to emulate anatomical mismatches/changes between the prior and the current low-dose images. We observed that the original SIR-ndiNLM introduces artifacts to the reconstruction when using attenuation mismatched prior. Furthermore, we found that larger attenuation mismatch between the prior and current images results in more severe artifacts in the SIR-ndiNLM reconstruction. Our proposed two corrective schemes enabled SIR-ndiNLM to effectively handle attenuation mismatch and anatomical changes between two images and successfully eliminate the artifacts. We demonstrated that the proposed techniques permit SIR-ndiNLM to leverage the attenuation mismatched prior and achieve quantitative low-dose CT reconstruction from both low-flux and sparse-view data acquisitions. This work permits robust and reliable PIBR for CT data acquired using different beam settings.

    View details for DOI 10.1088/1361-6560/abe760

    View details for PubMedID 33596553

  • A robotically assisted 3D printed quality assurance lung phantom for Calypso. Physics in medicine and biology Capaldi, D. P., Skinner, L. B., Dubrowski, P. n., Zhang, H. n., Xing, L. n., Chuang, C. F., Loo, B. W., Bush, K. K., Fahimian, B. P., Yu, A. S. 2021


    Purpose:Radiation dose delivered to targets located near the upper-abdomen or in the thorax are significantly affected by respiratory-motion. Relatively large-margins are commonly added to compensate for this motion, limiting radiation-dose-escalation. Internal-surrogates of target motion, such as a radiofrequency (RF) tracking system, i.e. Calypso® System, are used to overcome this challenge and improve normal-tissue sparing. RF tracking systems consist of implanting transponders in the vicinity of the tumor to be tracked using radiofrequency-waves. Unfortunately, although the manufacture provides a universal quality-assurance (QA) phantom, QA-phantoms specifically for lung-applications are limited, warranting the development of alternative solutions to fulfil the tests mandated by AAPM's TG142. Accordingly, our objective was to design and develop a motion-phantom to evaluate Calypso for lung-applications that allows the Calypso® Beacons to move in different directions to better simulate true lung-motion.Methods and Materials:A Calypso lung QA-phantom was designed, and 3D-printed. The design consists of three independent arms where the transponders were attached. A pinpoint-chamber with a buildup-cap was also incorporated. A 4-axis robotic arm was programmed to drive the motion-phantom to mimic breathing. After acquiring a four-dimensional-computed-tomography (4DCT) scan of the motion-phantom, treatment-plans were generated and delivered on a Varian TrueBeam® with Calypso capabilities. Stationary and gated-treatment plans were generated and delivered to determine the dosimetric difference between gated and non-gated treatments. Portal cine-images were acquired to determine the temporal-accuracy of delivery by calculating the difference between the observed versus expected transponders locations with the known speed of the transponders' motion.Results:Dosimetric accuracy is better than TG142 tolerance of 2%. Temporal accuracy is greater than, TG142 tolerance of 100ms for beam-on, but less than 100ms for beam-hold.Conclusions:The robotic QA-phantom designed and developed in this study provides an independent phantom for performing Calypso lung-QA for commissioning and acceptance testing of Calypso for lung treatments.

    View details for DOI 10.1088/1361-6560/abebaa

    View details for PubMedID 33657537

  • Pulmonary Ventilation Maps Generated with Free-breathing Proton MRI and a Deep Convolutional Neural Network. Radiology Capaldi, D. P., Guo, F., Xing, L., Parraga, G. 2020: 202861


    Background Hyperpolarized noble gas MRI helps measure lung ventilation, but clinical translation remains limited. Free-breathing proton MRI may help quantify lung function using existing MRI systems without contrast material and may assist in providing information about ventilation not visible to the eye or easily extracted with segmentation methods. Purpose To explore the use of deep convolutional neural networks (DCNNs) to generate synthetic MRI ventilation scans from free-breathing MRI (deep learning [DL] ventilation MRI)-derived specific ventilation maps as a surrogate of noble gas MRI and to validate this approach across a wide range of lung diseases. Materials and Methods In this secondary analysis of prospective trials, 114 paired noble gas MRI and two-dimensional free-breathing MRI scans were obtained in healthy volunteers with no history of chronic or acute respiratory disease and in study participants with a range of different obstructive lung diseases, including asthma, bronchiectasis, chronic obstructive pulmonary disease, and non-small-cell lung cancer between September 2013 and April 2018 ( identifiers: NCT03169673, NCT02351141, NCT02263794, NCT02282202, NCT02279329, and NCT02002052). A U-Net-based DCNN model was trained to map free-breathing proton MRI to hyperpolarized helium 3 (3He) MRI ventilation and validated using a sixfold validation. During training, the DCNN ventilation maps were compared with noble gas MRI scans using the Pearson correlation coefficient (r) and mean absolute error. DCNN ventilation images were segmented for ventilation and ventilation defects and were compared with noble gas MRI scans using the Dice similarity coefficient (DSC). Relationships were evaluated with the Spearman correlation coefficient (rS). Results One hundred fourteen study participants (mean age, 56 years ± 15 [standard deviation]; 66 women) were evaluated. As compared with 3He MRI, DCNN model ventilation maps had a mean r value of 0.87 ± 0.08. The mean DSC for DL ventilation MRI and 3He MRI ventilation was 0.91 ± 0.07. The ventilation defect percentage for DL ventilation MRI was highly correlated with 3He MRI ventilation defect percentage (rS = 0.83, P < .001, mean bias = -2.0% ± 5). Both DL ventilation MRI (rS = -0.51, P < .001) and 3He MRI (rS = -0.61, P < .001) ventilation defect percentage were correlated with the forced expiratory volume in 1 second. The DCNN model required approximately 2 hours for training and approximately 1 second to generate a ventilation map. Conclusion In participants with diverse pulmonary pathologic findings, deep convolutional neural networks generated ventilation maps from free-breathing proton MRI trained with a hyperpolarized noble-gas MRI ventilation map data set. The maps showed correlation with noble gas MRI ventilation and pulmonary function measurements. © RSNA, 2020 See also the editorial by Vogel-Claussen in this issue.

    View details for DOI 10.1148/radiol.2020202861

    View details for PubMedID 33289613

  • Precision radiotherapy using monochromatic inverse Compton x-ray sources. Medical physics Simiele, E. A., Breitkreutz, D. Y., Capaldi, D. P., Liu, W., Bush, K. K., Skinner, L. B. 2020


    PURPOSE: The dosimetric properties of Inverse Compton (IC) x-ray sources were investigated to determine their utility for stereotactic radiation therapy.METHODS: Monte Carlo simulations were performed using the egs brachy user code of EGSnrc. Nominal IC source x-ray energies of 80 keV and 150 keV were considered in this work. Depth-dose and lateral dose-profiles in water were calculated, as was dose enhancement in bone. Further simulations were performed for brain and spine treatment sites. The impact of gold nanoparticle doping was also investigated for the brain treatment site. Analogous dose calculations were performed in a clinical treatment planning system using a clinical 6 MV photon beam model and were compared to the Monte Carlo simulations.RESULTS: Both 80 keV and 150 keV IC beams were observed to have sharp 80-20 penumbra (i.e., < 0.1 mm) with broad low-dose tails in water. For reference, the calculated penumbra for the 6 MV clinical beam was 3 mm. Maximum dose enhancement factors in bone of 3.1, 1.4, and 1.1 were observed for the 80 keV, 150keV, and clinical 6 MV beams, respectively. The plan quality for the single brain metastasis case was similar between the IC beams and the 6 MV beam without gold nanoparticles. As the concentration of gold within the target increased, the V12 Gy to the normal brain tissue and Dmax within the target volume significantly decreased and the conformity significantly improved, which resulted in superior plan quality over the clinical 6 MV beam plan. In the spine cases, the sharp penumbra and enhanced dose to bone of the IC beams produced superior plan quality (i.e., better conformity, normal tissue sparing, and spinal cord sparing) as compared to the clinical 6 MV beam plans.CONCLUSIONS: The findings from this work indicate that inverse Compton x-ray sources are well-suited for stereotactic radiotherapy treatments due to their sharp penumbra and dose enhancement around high atomic-number materials. Future work includes investigating the properties of intensity-modulated inverse Compton x-ray sources to improve the homogeneity within the target tissue.

    View details for DOI 10.1002/mp.14552

    View details for PubMedID 33107049

  • Parametric Response Mapping of Co-registered Positron Emission Tomography and Dynamic Contrast Enhanced Computed Tomography to Identify Radio-resistant Sub-volumes in Locally Advanced Cervical Cancer. International journal of radiation oncology, biology, physics Capaldi, D. P., Hristov, D. H., Kidd, E. A. 2020


    PURPOSE: To identify sub-volumes that may predict treatment response to definitive concurrent chemoradiation therapy (CCRT) using parametric-response-mapping (PRM) of co-registered positron-emission-tomography (PET) and dynamic-contrast-enhanced (DCE) computed-tomography (CT) in locally advanced cervical carcinoma.METHODS AND MATERIALS: Pre- and mid-treatment (after 23±4days of CCRT) DCE CT and PET imaging were performed on 21 cervical cancer patients who were enrolled in a pilot study to evaluate the prognostic-value of CT perfusion for primary cervical cancer (NCT01805141). Three-dimensional co-registered maps of PET/CT standardized-uptake-value (SUV) and DCE CT blood-flow (BF) were generated. PRM was performed using voxel-wise joint histogram analysis to classify voxels within the tumor as highly-metabolic and perfused (SUVhiBFhi), highly-metabolic and hypoxic (SUVhiBFlo), low-metabolically active and hypoxic (SUVloBFlo), or low-metabolically active and perfused (SUVloBFhi) tissue based on thresholds determined from population means of pre-treatment PET SUV and DCE CT BF. Relationships between baseline pre-treatment imaging metrics and relative changes in metabolic-tumor-volume (DeltaMTV), calculated from pre- and during-treatment imaging, were determined using univariable and multivariable linear regression models.RESULTS: The relative volume of three PRM sub-volumes significantly changed during treatment (SUVhiBFhi: p=.04; SUVhiBFlo: p=.0008; SUVloBFhi: p=.02), while SUVloBFlo did not (p=.9). Pre-treatment PET SUVmax (r=-.58,p=.006), PET SUVmean (rho=-.59,p=.005), DCE CT BFmean (r=-.50,p=.02), tumor-volume (rho=-.65,p=.001) and PRM SUVhiBFhi (rho=-.59,p=.004) were negatively correlated with DeltaMTV, while PRM SUVloBFlo was positively-related with DeltaMTV (r=.77,p<.0001). In a multivariable model that predicted DeltaMTV, PRM SUVloBFlo, which combines both PET/CT and DCE CT, was the only significant variable (beta=1.825,p=.03), over both imaging modalities independently.CONCLUSIONS: PRM was applied in locally advanced cervical carcinoma treated definitively with chemoradiation and radioresistant sub-volumes were identified which correlated with changes in MTV and predict treatment-response. Identification of these sub-volumes may assist in clinical decision-making to tailor therapies, such as brachytherapy, in an effort to improve patient outcomes.

    View details for DOI 10.1016/j.ijrobp.2020.03.023

    View details for PubMedID 32251757

  • Pulmonary Imaging Phenotypes of Chronic Obstructive Pulmonary Disease Using Multiparametric Response Maps. Radiology MacNeil, J. L., Capaldi, D. P., Westcott, A. R., Eddy, R. L., Barker, A. L., McCormack, D. G., Kirby, M., Parraga, G. 2020: 191735


    Background Pulmonary imaging of chronic obstructive pulmonary disease (COPD) has focused on CT or MRI measurements, but these have not been evaluated in combination. Purpose To generate multiparametric response map (mPRM) measurements in ex-smokers with or without COPD by using volume-matched CT and hyperpolarized helium 3 (3He) MRI. Materials and Methods In this prospective study (, NCT02279329), participants underwent MRI and CT and completed pulmonary function tests, questionnaires, and the 6-minute walk test between December 2010 and January 2019. Disease status was determined by using Global initiative for chronic Obstructive Lung Disease (GOLD) criteria. The mPRM voxel values were generated by using co-registered MRI and CT labels. Kruskal-Wallis and Bonferroni tests were used to determine differences across disease severity, and correlations were determined by using Spearman coefficients. Results A total of 175 ex-smokers (mean age, 69 years ± 9 [standard deviation], 108 men) with or without COPD were evaluated. Ex-smokers without COPD had a larger fraction of normal mPRM voxels (60% vs 37%, 20%, and 7% for GOLD I, II, and III/IV disease, respectively; all P ≤ .001) and a smaller fraction of abnormal voxels, including small airways disease (normal CT, not ventilated: 5% vs 6% [not significant], 11%, and 19% [P ≤ .001 for both] for GOLD I, II, and III/IV disease, respectively) and mild emphysema (normal CT, abnormal apparent diffusion coefficient [ADC]: 33% vs 54%, 56%, and 54% for GOLD I, II, and III/IV disease respectively; all P ≤ .001). Normal mPRM measurements were positively correlated with forced expiratory volume in 1 second (FEV1) (r = 0.65, P < .001), the FEV1-to-forced vital capacity ratio (r = 0.81, P < .001), and diffusing capacity (r = 0.75, P < .001) and were negatively correlated with worse quality of life (r = -0.48, P < .001). Abnormal mPRM measurements of small airways disease (normal CT, not ventilated) and mild emphysema (normal CT, abnormal ADC) were negatively correlated with FEV1 (r = -0.65 and -0.42, respectively; P < .001) and diffusing capacity (r = -0.53 and -0.60, respectively; P < .001) and were positively correlated with worse quality of life (r = 0.45 and r = 0.33, respectively; P < .001), both of which were present in ex-smokers without COPD. Conclusion Multiparametric response maps revealed two abnormal structure-function results related to emphysema and small airways disease, both of which were unexpectedly present in ex-smokers with normal spirometry and CT findings. © RSNA, 2020 Online supplemental material is available for this article.

    View details for DOI 10.1148/radiol.2020191735

    View details for PubMedID 32096708

  • Incorporating prior knowledge via volumetric deep residual network to optimize the reconstruction of sparsely sampled MRI MAGNETIC RESONANCE IMAGING Wu, Y., Ma, Y., Capaldi, D., Liu, J., Zhao, W., Du, J., Xing, L. 2020; 66: 93–103
  • Technical Note: Performance of CyberKnife® Tracking Using Low-Dose CT and kV Imaging. Medical physics Nano, T. F., Capaldi, D. P., Yeung, T. n., Chuang, C. F., Wang, L. n., Descovich, M. n. 2020


    To investigate the effects of CT protocol and in-room x-ray technique on CyberKnife® (Accuray Inc.) tracking accuracy by evaluating end-to-end tests.End-to-end (E2E) tests were performed for the different tracking methods (6D skull, fiducial, spine and lung) using an anthropomorphic head phantom (Accuray Inc.) and thorax phantom (CIRS Inc.). Bolus was added to the thorax phantom to simulate a large patient and to evaluate the performance of lung tracking in a more realistic condition. The phantoms were scanned with a Siemens Sensation Open 24 slice CT at low-dose (120kV, 70mAs, 1.5mm slice thickness) and high-dose (120kV, 700mAs, 1.5mm slice thickness) to generate low-dose and high-dose digitally reconstructed radiographs (DRRs). The difference in initial phantom alignment, Δ(Align), and in total targeting accuracy, E2E, were obtained for all tracking methods with low and high dose DRRs. Additionally, Δ(Align) was determined for different in-room x-ray imaging techniques (0.5 to 50mAs and 100 to 140kV) using a low-dose lung tracking plan.Low-dose CT scans produced images with high noise, however, for these phantoms the targets could be easily delineated on all scans. End-to-end results were less than 0.95mm for all tracking methods and all plans. The greatest difference in initial alignment Δ(Align) and E2E results between low and high dose CT protocols was 0.32mm and 0.24mm, respectively. Similar results were observed with a large thorax phantom. Tracking using di_erent in-room x-ray imaging techniques (mAs) corresponding to low exposures (resulting in high image noise) or high exposure (resulting in image saturation) had alignment accuracy Δ(Align) greater than 1mm.End-to-end targeting accuracy within tolerance (<0.95mm) was obtained for all tracking methods using low-dose CT protocols, suggesting that CT protocol should be set by target contouring needs. Additionally, high tracking accuracy was achieved for in in-room x-ray imaging techniques that produce high quality images.

    View details for DOI 10.1002/mp.14537

    View details for PubMedID 33064863

  • An integrated quality assurance phantom for frameless single-isocenter multitarget stereotactic radiosurgery. Physics in medicine and biology Capaldi, D. P., Skinner, L. B., Dubrowski, P. n., Yu, A. S. 2020


    Purpose:Brain stereotactic-radiosurgery (SRS) treatments require multiple quality-assurance (QA) procedures to ensure accurate and precise treatment delivery. As single-isocenter multitarget SRS treatments become more popular, the quantification of off-axis accuracy of the linear-accelerator is crucial. In this study, a novel brain SRS integrated phantom was developed and validated to enable SRS QA with a single phantom to facilitate implementation of a frameless single-isocenter, multitarget SRS program. This phantom combines the independent verification of each positioning system, the Winston-Lutz, off-axis accuracy evaluation (i.e. off-axis Winston-Lutz), and the dosimetric accuracy utilizing both point-dose-measurements as well as film-measurement, without moving the phantom.Methods and Materials:A novel 3D-printed phantom, coinedOneIso, was designed with a movable insert which can switch between the Winston-Lutz test target and dose measurement without moving the phantom itself. For dose verification, eight brain SRS clinical-treatment-plans with 10MV Flattening-Filter-Free (FFF) beams were delivered on a Varian TrueBeam with a high-definition-multi-leaf-collimator (HD-MLC). Radiochromic film and pinpoint ion chamber comparison measurements were made between the OneIso and solid water (SW) phantom setups. For the off-axis Winston-Lutz measurements, a row of off-axis ball-bearings (BBs) was integrated into the OneIso. To quantify the spatial accuracy versus distance from isocenter, two-dimensional displacements were calculated between the planned and delivered BB locations relative to their respective MLC defined field border.Results:OneIso and the SW phantoms agree within 1%, for both film and point-dose measurements. OneIso identified a reduction in spatial accuracy further away from isocenter. Differences increased as distance from isocenter increased exceeding recommended SRS accuracy tolerances at 3-4cm away from isocenter.Conclusions:OneIso provides a streamlined, single-setup workflow for single-isocenter multitarget frameless linac-based SRS QA. Additionally, with the ability to quantify off-axis spatial-discrepancies, we can determine limitations on the maximum distance between targets to ensure a single-isocenter multitarget SRS program meets recommended guidelines.

    View details for DOI 10.1088/1361-6560/ab8534

    View details for PubMedID 32235050

  • Technical Note: Evaluation of Audiovisual Biofeedback Smartphone Application for Respiratory Monitoring in Radiation Oncology. Medical physics Capaldi, D. P., Nano, T. F., Zhang, H. n., Skinner, L. B., Xing, L. n. 2020


    Radiation dose delivered to targets located near the upper abdomen or thorax are significantly affected by respiratory motion, necessitating large margins, limiting dose escalation. Surrogate motion management devices, such as the Real-time Position Management (RPM™) system (Varian Medical Systems, Palo Alto, CA), are commonly used to improve normal tissue sparing. Alternative to current solutions, we have developed and evaluated the feasibility of a real-time position management system that leverages the motion data from the onboard hardware of Apple iOS devices to provide patients with visual coaching with the potential to improve the reproducibility of breathing as well as improve patient compliance and reduce treatment delivery time.The iOS application, coined the Instant Respiratory Feedback (IRF) system, was developed in Swift (Apple Inc., Cupertino, CA) using the Core-Motion library and implemented on an Apple iPhone® devices. Operation requires an iPhone®, a 3D printed arm, and a radiolucent projector screen system for feedback. Direct comparison between IRF, which leverages sensor fusion data from the iPhone®, and RPM™, an optical based system, was performed on multiple respiratory motion phantoms and volunteers. The IRF system and RPM™ camera tracking marker were placed on the same location allowing for simultaneous data acquisition. The IRF surrogate measurement of displacement was compared to the signal trace acquired using RPM™ with univariate linear regressions and Bland-Altman analysis.Periodic motion shows excellent agreement between both systems, and subject motion shows good agreement during regular and irregular breathing motion. Comparison of IRF and RPM™ show very similar signal traces that were significantly related across all phantoms, including those motion with different amplitude and frequency, and subjects' waveforms (all r>0.9, p<0.0001). We demonstrate the feasibility of four-dimensional cone beam computed tomography reconstruction using IRF can acquire dynamic phantom images with similar image quality as RPM™.Feasibility of an iOS application to provide real-time respiratory motion is demonstrated. This system generated comparable signal traces to a commercially available system and offers an alternative method to monitor respiratory motion.

    View details for DOI 10.1002/mp.14484

    View details for PubMedID 32969075

  • Chronic Obstructive Pulmonary Disease: Thoracic CT Texture Analysis and Machine Learning to Predict Pulmonary Ventilation. Radiology Westcott, A., Capaldi, D. P., McCormack, D. G., Ward, A. D., Fenster, A., Parraga, G. 2019: 190450


    Background Fixed airflow limitation and ventilation heterogeneity are common in chronic obstructive pulmonary disease (COPD). Conventional noncontrast CT provides airway and parenchymal measurements but cannot be used to directly determine lung function. Purpose To develop, train, and test a CT texture analysis and machine-learning algorithm to predict lung ventilation heterogeneity in participants with COPD. Materials and Methods In this prospective study ( NCT02723474; conducted from January 2010 to February 2017), participants were randomized to optimization (n = 1), training (n = 67), and testing (n = 27) data sets. Hyperpolarized (HP) helium 3 (3He) MRI ventilation maps were co-registered with thoracic CT to provide ground truth labels, and 87 quantitative imaging features were extracted and normalized to lung averages to generate 174 features. The volume-of-interest dimension and the training data sampling method were optimized to maximize the area under the receiver operating characteristic curve (AUC). Forward feature selection was performed to reduce the number of features; logistic regression, linear support vector machine, and quadratic support vector machine classifiers were trained through fivefold cross validation. The highest-performing classification model was applied to the test data set. Pearson coefficients were used to determine the relationships between the model, MRI, and pulmonary function measurements. Results The quadratic support vector machine performed best in training and was applied to the test data set. Model-predicted ventilation maps had an accuracy of 88% (95% confidence interval [CI]: 88%, 88%) and an AUC of 0.82 (95% CI: 0.82, 0.83) when the HP 3He MRI ventilation maps were used as the reference standard. Model-predicted ventilation defect percentage (VDP) was correlated with VDP at HP 3He MRI (r = 0.90, P < .001). Both model-predicted and HP 3He MRI VDP were correlated with forced expiratory volume in 1 second (FEV1) (model: r = -0.65, P < .001; MRI: r = -0.70, P < .001), ratio of FEV1 to forced vital capacity (model: r = -0.73, P < .001; MRI: r = -0.75, P < .001), diffusing capacity (model: r = -0.69, P < .001; MRI: r = -0.65, P < .001), and quality-of-life score (model: r = 0.59, P = .001; MRI: r = 0.65, P < .001). Conclusion Model-predicted ventilation maps generated by using CT textures and machine learning were correlated with MRI ventilation maps (r = 0.90, P < .001). © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Fain in this issue.

    View details for DOI 10.1148/radiol.2019190450

    View details for PubMedID 31638491

  • Incorporating prior knowledge via volumetric deep residual network to optimize the reconstruction of sparsely sampled MRI. Magnetic resonance imaging Wu, Y., Ma, Y., Capaldi, D. P., Liu, J., Zhao, W., Du, J., Xing, L. 2019


    For sparse sampling that accelerates magnetic resonance (MR) image acquisition, non-linear reconstruction algorithms have been developed, which incorporated patient specific a prior information. More generic a prior information could be acquired via deep learning and utilized for image reconstruction. In this study, we developed a volumetric hierarchical deep residual convolutional neural network, referred to as T-Net, to provide a data-driven end-to-end mapping from sparsely sampled MR images to fully sampled MR images, where cartilage MR images were acquired using an Ultra-short TE sequence and retrospectively undersampled using pseudo-random Cartesian and radial acquisition schemes. The network had a hierarchical architecture that promoted the sparsity of feature maps and increased the receptive field, which were valuable for signal synthesis and artifact suppression. Relatively dense local connections and global shortcuts were established to facilitate residual learning and compensate for details lost in hierarchical processing. Additionally, volumetric processing was adopted to fully exploit spatial continuity in three-dimensional space. Data consistency was further enforced. The network was trained with 336 three-dimensional images (each consisting of 32 slices) and tested by 24 images. The incorporation of a priori information acquired via deep learning facilitated high acceleration factors (as high as 8) while maintaining high image fidelity (quantitatively evaluated using the structural similarity index measurement). The proposed T-Net had an improved performance as compared to several state-of-the-art networks.

    View details for PubMedID 30880112

  • CT and Functional MRI to Evaluate Airway Mucus in Severe Asthma. Chest Svenningsen, S. n., Haider, E. n., Boylan, C. n., Mukherjee, M. n., Eddy, R. L., Capaldi, D. P., Parraga, G. n., Nair, P. n. 2019; 155 (6): 1178–89


    Intraluminal contributor(s) to airflow obstruction in severe asthma are patient-specific and must be evaluated to personalize treatment. The occurrence and functional consequence of airway mucus in the presence or absence of airway eosinophils remain undetermined.The objective of this study was to understand the functional consequence of airway mucus in the presence or absence of eosinophils and to identify biomarkers of mucus-related airflow obstruction.Mucus plugs were quantified on CT scans, and their contribution to ventilation heterogeneity (using MRI ventilation defect percent [VDP]) was evaluated in 27 patients with severe asthma. Patients were dichotomized based on sputum eosinophilia such that the relationship between mucus, eosinophilia, and ventilation heterogeneity could be investigated. Fractional exhaled nitric oxide (Feno) and related cytokines in sputum were measured.Mucus plugging was present in 100% of asthma patients with sputum eosinophils and 36% of those without sputum eosinophils (P = .0006) and was correlated with MRI VDP prebronchodilator (r = 0.68; P = .0001) and postbronchodilator (r = 0.72; P < .0001). In a multivariable regression, both mucus and eosinophils contributed to the prediction of postbronchodilator MRI VDP (R2 = 0.75; P < .0001). Patients with asthma in whom the mucus score was high had raised Feno (P = .03) and IL-4 (P = .02) values. Mucus plugging correlated with Feno (r = 0.63; P = .005).Both airway eosinophils and mucus can contribute to ventilation heterogeneity in patients with severe asthma. Patients in whom mucus is the dominant cause of airway obstruction have evidence of an upregulated IL-4/IL-13 pathway that could be identified according to increased Feno level.

    View details for DOI 10.1016/j.chest.2019.02.403

    View details for PubMedID 30910637

  • Fourier decomposition free-breathing H-1 MRI perfusion maps in asthma Matheson, A. M., Capaldi, D. P., Guo, F., Eddy, R. L., McCormack, D. G., Parraga, G., Angelini, E. D., Landman, B. A. SPIE-INT SOC OPTICAL ENGINEERING. 2019

    View details for DOI 10.1117/12.2512436

    View details for Web of Science ID 000483012700036

  • Texture Analysis of Thoracic CT to Predict Hyperpolarized Gas MRI Lung Function Westcott, A., Capaldi, D. P., McCormack, D. G., Fenster, A., Parraga, G., Gimi, B., Krol, A. SPIE-INT SOC OPTICAL ENGINEERING. 2019

    View details for DOI 10.1117/12.2512851

    View details for Web of Science ID 000483014900014

  • Development and Evaluation of Pulmonary Imaging Multi-Parametric Response Maps for Deep Phenotyping of Chronic Obstructive Pulmonary Disease MacNeil, J. L., Capaldi, D. P., Eddy, R. L., Westcott, A., Matheson, A. M., Barker, A. L., Ong-Ly, C., McCormack, D. G., Parraga, G., Gimi, B., Krol, A. SPIE-INT SOC OPTICAL ENGINEERING. 2019

    View details for DOI 10.1117/12.2512849

    View details for Web of Science ID 000483014900016

  • A framework for Fourier-decomposition free-breathing pulmonary 1 H MRI ventilation measurements. Magnetic resonance in medicine Guo, F. n., Capaldi, D. P., McCormack, D. G., Fenster, A. n., Parraga, G. n. 2019; 81 (3): 2135–46


    To develop a rapid Fourier decomposition (FD) free-breathing pulmonary 1 H MRI (FDMRI) image processing and biomarker pipeline for research use.We acquired MRI in 20 asthmatic subjects using a balanced steady-state free precession (bSSFP) sequence optimized for ventilation imaging. 2D 1 H MRI series were segmented by enforcing the spatial similarity between adjacent images and the right-to-left lung volume-ratio. The segmented lung series were co-registered using a coarse-to-fine deformable registration framework that used dual optimization techniques. All pairwise registrations were implemented in parallel and FD was performed to generate 2D ventilation-weighted maps and ventilation-defect-percent (VDP). Lung segmentation and registration accuracy were evaluated by comparing algorithm and manual lung-masks, deformed manual lung-masks, and fiducials in the moving and fixed images using Dice-similarity-coefficient (DSC), mean-absolute-distance (MAD), and target-registration-error (TRE). The relationship of FD-VDP and 3 He-VDP was evaluated using the Pearson-correlation-coefficient (r) and Bland Altman analysis. Algorithm reproducibility was evaluated using the coefficient-of-variation (CoV) and intra-class-correlation-coefficient (ICC) for segmentation, registration, and FD-VDP components.For lung segmentation, there was a DSC of 95 ± 1.5% and MAD of 2.3 ± 0.5 mm, and for registration there was a DSC of 97 ± 0.8%, MAD of 1.6 ± 0.4 mm and TRE of 3.6 ± 1.2 mm. Reproducibility for segmentation DSC (CoV/ICC = 0.5%/0.92), registration TRE (CoV/ICC = 0.4%/0.98), and FD-VDP (Cov/ICC = 3.9%/0.97) was high. The pipeline required 10 min/subject. FD-VDP was correlated with 3 He-VDP (r = 0.69, P < 0.001) although there was a bias toward lower FD-VDP (bias = -4.9%).We developed and evaluated a pipeline that provides a rapid and precise method for FDMRI ventilation maps.

    View details for DOI 10.1002/mrm.27527

    View details for PubMedID 30362609

  • Hyperpolarized 3 He MRI ventilatory apparent diffusion coefficient of alpha-1 antitrypsin deficiency. Journal of magnetic resonance imaging : JMRI Westcott, A. n., Capaldi, D. P., Ouriadov, A. n., McCormack, D. G., Parraga, G. n. 2019; 49 (1): 311–13

    View details for DOI 10.1002/jmri.26202

    View details for PubMedID 30102430

  • Trainee Research Prizes from the 2017 RSNA Scientific Assembly and Annual Meeting RADIOLOGY Aboutalib, S. S., Abdelfadeel, A., Schreuder, A., Jang, J., Pandey, P., Shiradkar, R., Galimzianova, A., Qi, X., Larimer, B., McDonald, R. J., Orita, E., Dercle, L., Khosravi, M., Kim, J., Harrison, A. P., Ng, S., Chen, M., Braman, N., Tang, A., Chen, Y., Lebovic, J., Singh, V., Wang, C., Kundu, S., Armstrong, T., Ji, X., Van Wickle, J., Zhong, B., Lin, L., Lin, L., Capaldi, D., Hemachandran, N., Motkoski, J. W., Hectors, S., Adams, S. J., Jambor, I., Daye, D., Kunz, W. G., Mihal, D. C., Jahromi, A., Fritz, B., Scipione, R., Bai, H. X., Winzeck, S., Rathore, H., Tajmir, S. H., Hardy, A., Silva, S., Thompson, S. M. 2018; 287 (1): 1–4

    View details for DOI 10.1148/radiol.2018184002

    View details for Web of Science ID 000427992600001

    View details for PubMedID 29558305

  • Development of a pulmonary imaging biomarker pipeline for phenotyping of chronic lung disease. Journal of medical imaging (Bellingham, Wash.) Guo, F. n., Capaldi, D. n., Kirby, M. n., Sheikh, K. n., Svenningsen, S. n., McCormack, D. G., Fenster, A. n., Parraga, G. n. 2018; 5 (2): 026002


    We designed and generated pulmonary imaging biomarker pipelines to facilitate high-throughput research and point-of-care use in patients with chronic lung disease. Image processing modules and algorithm pipelines were embedded within a graphical user interface (based on the .NET framework) for pulmonary magnetic resonance imaging (MRI) and x-ray computed-tomography (CT) datasets. The software pipelines were generated using C++ and included: (1) inhaled He 3 / Xe 129   MRI ventilation and apparent diffusion coefficients, (2) CT-MRI coregistration for lobar and segmental ventilation and perfusion measurements, (3) ultrashort echo-time H 1   MRI proton density measurements, (4) free-breathing Fourier-decomposition H 1   MRI ventilation/perfusion and free-breathing H 1   MRI specific ventilation, (5) multivolume CT and MRI parametric response maps, and (6) MRI and CT texture analysis and radiomics. The image analysis framework was implemented on a desktop workstation/tablet to generate biomarkers of regional lung structure and function related to ventilation, perfusion, lung tissue texture, and integrity as well as multiparametric measures of gas trapping and airspace enlargement. All biomarkers were generated within 10 min with measurement reproducibility consistent with clinical and research requirements. The resultant pulmonary imaging biomarker pipeline provides real-time and automated lung imaging measurements for point-of-care and high-throughput research.

    View details for DOI 10.1117/1.JMI.5.2.026002

    View details for PubMedID 29963580

    View details for PubMedCentralID PMC6022861

  • On the Potential Role of MRI Biomarkers of COPD to Guide Bronchoscopic Lung Volume Reduction. Academic radiology Adams, C. J., Capaldi, D. P., Di Cesare, R. n., McCormack, D. G., Parraga, G. n. 2018; 25 (2): 159–68


    In patients with severe emphysema and poor quality of life, bronchoscopic lung volume reduction (BLVR) may be considered and guided based on lobar emphysema severity. In particular, x-ray computed tomography (CT) emphysema measurements are used to identify the most diseased and the second-most diseased lobes as BLVR targets. Inhaled gas magnetic resonance imaging (MRI) also provides chronic obstructive pulmonary disease (COPD) biomarkers of lobar emphysema and ventilation abnormalities. Our objective was to retrospectively evaluate CT and MRI biomarkers of lobar emphysema and ventilation in patients with COPD eligible for BLVR. We hypothesized that MRI would provide complementary biomarkers of emphysema and ventilation that help determine the most appropriate lung lobar targets for BLVR in patients with COPD.We retrospectively evaluated 22 BLVR-eligible patients from the Thoracic Imaging Network of Canada cohort (diffusing capacity of the lung for carbon monoxide = 37 ± 12%predicted, forced expiratory volume in 1 second = 34 ± 7%predicted, total lung capacity = 131 ± 17%predicted, and residual volume = 216 ± 36%predicted). Lobar CT emphysema, measured using a relative area of <-950 Hounsfield units (RA950) and MRI ventilation defect percent, was independently used to rank lung lobe disease severity.In 7 of 22 patients, there were different CT and MRI predictions of the most diseased lobe. In some patients, there were large ventilation defects in lobes not targeted by CT, indicative of a poorly ventilated lung. CT and MRI classification of the most diseased and the second-most diseased lobes showed a fair-to-moderate intermethod reliability (Cohen κ = 0.40-0.59).In this proof-of-concept retrospective analysis, quantitative MRI ventilation and CT emphysema measurements provided different BLVR targets in over 30% of the patients. The presence of large MRI ventilation defects in lobes next to CT-targeted lobes might also change the decision to proceed or to guide BLVR to a different lobar target.

    View details for DOI 10.1016/j.acra.2017.08.010

    View details for PubMedID 29051040

  • Free-breathing Pulmonary MR Imaging to Quantify Regional Ventilation. Radiology Capaldi, D. P., Eddy, R. L., Svenningsen, S. n., Guo, F. n., Baxter, J. S., McLeod, A. J., Nair, P. n., McCormack, D. G., Parraga, G. n. 2018; 287 (2): 693–704


    Purpose To measure regional specific ventilation with free-breathing hydrogen 1 (1H) magnetic resonance (MR) imaging without exogenous contrast material and to investigate correlations with hyperpolarized helium 3 (3He) MR imaging and pulmonary function test measurements in healthy volunteers and patients with asthma. Materials and Methods Subjects underwent free-breathing 1H and static breath-hold hyperpolarized 3He MR imaging as well as spirometry and plethysmography; participants were consecutively recruited between January and June 2017. Free-breathing 1H MR imaging was performed with an optimized balanced steady-state free-precession sequence; images were retrospectively grouped into tidal inspiration or tidal expiration volumes with exponentially weighted phase interpolation. MR imaging volumes were coregistered by using optical flow deformable registration to generate 1H MR imaging-derived specific ventilation maps. Hyperpolarized 3He MR imaging- and 1H MR imaging-derived specific ventilation maps were coregistered to quantify regional specific ventilation within hyperpolarized 3He MR imaging ventilation masks. Differences between groups were determined with the Mann-Whitney test and relationships were determined with Spearman (ρ) correlation coefficients. Statistical analyses were performed with software. Results Thirty subjects (median age: 50 years; interquartile range [IQR]: 30 years), including 23 with asthma and seven healthy volunteers, were evaluated. Both 1H MR imaging-derived specific ventilation and hyperpolarized 3He MR imaging-derived ventilation percentage were significantly greater in healthy volunteers than in patients with asthma (specific ventilation: 0.14 [IQR: 0.05] vs 0.08 [IQR: 0.06], respectively, P < .0001; ventilation percentage: 99% [IQR: 1%] vs 94% [IQR: 5%], P < .0001). For all subjects, 1H MR imaging-derived specific ventilation correlated with plethysmography-derived specific ventilation (ρ = 0.54, P = .002) and hyperpolarized 3He MR imaging-derived ventilation percentage (ρ = 0.67, P < .0001) as well as with forced expiratory volume in 1 second (FEV1) (ρ = 0.65, P = .0001), ratio of FEV1 to forced vital capacity (ρ = 0.75, P < .0001), ratio of residual volume to total lung capacity (ρ = -0.68, P < .0001), and airway resistance (ρ = -0.51, P = .004). 1H MR imaging-derived specific ventilation was significantly greater in the gravitational-dependent versus nondependent lung in healthy subjects (P = .02) but not in patients with asthma (P = .1). In patients with asthma, coregistered 1H MR imaging specific ventilation and hyperpolarized 3He MR imaging maps showed that specific ventilation was diminished in corresponding 3He MR imaging ventilation defects (0.05 ± 0.04) compared with well-ventilated regions (0.09 ± 0.05) (P < .0001). Conclusion 1H MR imaging-derived specific ventilation correlated with plethysmography-derived specific ventilation and ventilation defects seen by using hyperpolarized 3He MR imaging. © RSNA, 2018 Online supplemental material is available for this article.

    View details for DOI 10.1148/radiol.2018171993

    View details for PubMedID 29470939

  • Through the Looking Glass and What Was Found There: Imaging Biomarkers of Chronic Obstructive Pulmonary Disease. American journal of respiratory and critical care medicine Capaldi, D. P., Parraga, G. n. 2017; 196 (11): 1364–66

    View details for DOI 10.1164/rccm.201707-1473ED

    View details for PubMedID 28813162

  • Free-breathing Functional Pulmonary MRI: Response to Bronchodilator and Bronchoprovocation in Severe Asthma. Academic radiology Capaldi, D. P., Sheikh, K. n., Eddy, R. L., Guo, F. n., Svenningsen, S. n., Nair, P. n., McCormack, D. G., Parraga, G. n. 2017; 24 (10): 1268–76


    Ventilation heterogeneity is a hallmark feature of asthma. Our objective was to evaluate ventilation heterogeneity in patients with severe asthma, both pre- and post-salbutamol, as well as post-methacholine (MCh) challenge using the lung clearance index, free-breathing pulmonary 1H magnetic resonance imaging (FDMRI), and inhaled-gas MRI ventilation defect percent (VDP).Sixteen severe asthmatics (49 ± 10 years) provided written informed consent to an ethics board-approved protocol. Spirometry, plethysmography, and multiple breath nitrogen washout to measure the lung clearance index were performed during a single visit within 15 minutes of MRI. Inhaled-gas MRI and FDMRI were performed pre- and post-bronchodilator to generate VDP. For asthmatics with forced expiratory volume in 1 second (FEV1) >70%predicted, MRI was also performed before and after MCh challenge. Wilcoxon signed-rank tests, Spearman correlations, and a repeated-measures analysis of variance were performed.Hyperpolarized 3He (P = .02) and FDMRI (P = .02) VDP significantly improved post-salbutamol and for four asthmatics who could perform MCh (n = 4). 3He and FDMRI VDP significantly increased at the provocative concentration of MCh, resulting in a 20% decrease in FEV1 (PC20) and decreased post-bronchodilator (P = .02), with a significant difference between methods (P = .01). FDMRI VDP was moderately correlated with 3He VDP (ρ = .61, P = .01), but underestimated VDP relative to 3He VDP (-6 ± 9%). Whereas 3He MRI VDP was significantly correlated with the lung clearance index, FDMRI was not (ρ = .49, P = .06).FDMRI VDP generated in free-breathing asthmatic patients was correlated with static inspiratory breath-hold 3He MRI VDP but underestimated VDP relative to 3He MRI VDP. Although less sensitive to salbutamol and MCh, FDMRI VDP may be considered for asthma patient evaluations at centers without inhaled-gas MRI.

    View details for DOI 10.1016/j.acra.2017.04.012

    View details for PubMedID 28551402

  • Thoracic CT-MRI coregistration for regional pulmonary structure-function measurements of obstructive lung disease. Medical physics Guo, F. n., Svenningsen, S. n., Kirby, M. n., Capaldi, D. P., Sheikh, K. n., Fenster, A. n., Parraga, G. n. 2017; 44 (5): 1718–33


    Recent pulmonary imaging research has revealed that in patients with chronic obstructive pulmonary disease (COPD) and asthma, structural and functional abnormalities are spatially heterogeneous. This novel information may help optimize treatment in individual patients, monitor interventional efficacy, and develop new treatments. Moreover, by automating the measurement of regional biomarkers for the 19 different anatomical lung segments, there is an opportunity to embed imaging biomarkers into clinically acceptable clinical workflows and improve lung disease clinical care. Therefore, to exploit the regional structure-function information provided by thoracic imaging, and as a first step toward this goal, our objective was to develop a fully automated registration pipeline for thoracic x-ray computed tomography (CT) and inhaled gas functional magnetic resonance imaging (MRI) whole lung and segmental structure-function biomarkers.Thirty-five patients including 15 severe, poorly controlled asthmatics and 20 COPD patients [classified according to the global initiative for chronic obstructive lung disease (GOLD) criteria)] provided written informed consent to a study protocol approved by Health Canada and underwent pulmonary function tests, MRI, and CT during a single 2-hour visit. Using this diverse patient dataset, we developed and evaluated a joint deformable registration approach to simultaneously coregister CT with both 1 H and 3 He MRI by enforcing the similarity of the deformation fields from the two individual registrations. We derived a simpler model that was equivalent to the original challenging optimization problem through variational analysis and the simpler model gave rise to an efficient numerical solver that was parallelized on a graphics processing unit. The coregistered CT-3 He MRI and whole lung/segmental lung masks were used to generate whole lung and segmental 3 He MRI ventilation defect percent (VDP). To estimate fiducial localization reproducibility, a single observer manually identified 109 pairs of CT and 3 He MRI fiducials for 35 patient images on five separate occasions and determined the fiducial localization error (FLE). CT-3 He MRI registration accuracy was evaluated using the target registration error (TRE). Whole lung VDP generated using the algorithm was compared with VDP generated using a previously validated semiautomated approach and computational efficiency was evaluated using run time.In 35 patients including 15 with severe asthma and 20 with COPD, mean forced expiratory volume in 1 s (FEV1 ) was 63±24%pred and FEV1 /forced vital capacity (FVC) was 54 ± 17%. FLE was 0.16 mm and 0.34 mm for 3 He MRI and CT, respectively. TRE was 4.5 ± 2.0 mm, 4.0 ± 1.7 mm, 4.8 ± 2.3 mm for asthma, COPD GOLD II, and GOLD III groups, respectively, with a mean of 4.4 ± 2.0 mm for the entire dataset. TRE was significantly improved for joint CT-1 H/3 He MRI registration compared with CT-1 H MRI rigid registration (P < 0.0001). Whole lung VDP generated using the pipeline was not significantly different (P = 0.37) compared to a semiautomated method with which it was strongly correlated (r = 0.93, P < 0.0001). The fully automated pipeline required 11 ± 0.4 min to generate whole lung and segmental VDP.For a diverse group of patients with COPD and asthma, whole lung and segmental VDP was measured using an automated lung image analysis pipeline which provides a way to incorporate lung functional biomarkers into clinical research and patient care.

    View details for DOI 10.1002/mp.12160

    View details for PubMedID 28206676

  • Ultrashort echo time MRI biomarkers of asthma. Journal of magnetic resonance imaging : JMRI Sheikh, K. n., Guo, F. n., Capaldi, D. P., Ouriadov, A. n., Eddy, R. L., Svenningsen, S. n., Parraga, G. n. 2017; 45 (4): 1204–15


    To develop and assess ultrashort echo-time (UTE) magnetic resonance imaging (MRI) biomarkers of lung function in asthma patients.Thirty participants including 13 healthy volunteers and 17 asthmatics provided written informed consent to UTE and pulmonary function tests in addition to hyperpolarized-noble-gas 3T MRI and computed tomography (CT) for asthmatics only. The difference in MRI signal-intensity (SI) across four lung volumes (full-expiration, functional-residual-capacity [FRC], FRC+1L, and full-inspiration) was determined on a voxel-by-voxel basis to generate dynamic proton-density (DPD) maps. MRI ventilation-defect-percent (VDP), UTE SI, and DPD values as well as CT radiodensity were determined for whole lung and individual lobes.Mean SI at full-expiration (P < 0.01), FRC (P < 0.05), and DPD (P < 0.01) were greater in healthy volunteers compared to asthmatics. In asthmatics, UTE SI at full-expiration and DPD were correlated with FEV1 /FVC (SI r = 0.73/P = 0.002; DPD r = 0.75/P = 0.003), RV/TLC (SI r = -0.57/P = 0.02), or RV (DPD r = -0.62/P = 0.02), CT radiodensity (SI r = 0.83/P = 0.006; DPD r = 0.71/P = 0.01), and lobar VDP (SI rs  = -0.33/P = 0.02; DPD rs  = -0.47/P = 0.01).In patients with asthma, UTE SI and dynamic proton-density were related to pulmonary function measurements, whole lung and lobar VDP, as well as CT radiodensity. Thus, UTE MRI biomarkers may reflect ventilation heterogeneity and/or gas-trapping in asthmatics using conventional equipment, making this approach potentially amenable for clinical use.2 J. Magn. Reson. Imaging 2017;45:1204-1215.

    View details for DOI 10.1002/jmri.25503

    View details for PubMedID 27731948

  • A pilot randomised clinical trial of mepolizumab in COPD with eosinophilic bronchitis. The European respiratory journal Dasgupta, A. n., Kjarsgaard, M. n., Capaldi, D. n., Radford, K. n., Aleman, F. n., Boylan, C. n., Altman, L. C., Wight, T. N., Parraga, G. n., O'Byrne, P. M., Nair, P. n. 2017; 49 (3)

    View details for DOI 10.1183/13993003.02486-2016

    View details for PubMedID 28298405

  • Anatomical pulmonary magnetic resonance imaging segmentation for regional structure-function measurements of asthma. Medical physics Guo, F., Svenningsen, S., Eddy, R. L., Capaldi, D. P., Sheikh, K., Fenster, A., Parraga, G. 2016; 43 (6): 2911-2926


    Pulmonary magnetic-resonance-imaging (MRI) and x-ray computed-tomography have provided strong evidence of spatially and temporally persistent lung structure-function abnormalities in asthmatics. This has generated a shift in their understanding of lung disease and supports the use of imaging biomarkers as intermediate endpoints of asthma severity and control. In particular, pulmonary (1)H MRI can be used to provide quantitative lung structure-function measurements longitudinally and in response to treatment. However, to translate such biomarkers of asthma, robust methods are required to segment the lung from pulmonary (1)H MRI. Therefore, their objective was to develop a pulmonary (1)H MRI segmentation algorithm to provide regional measurements with the precision and speed required to support clinical studies.The authors developed a method to segment the left and right lung from (1)H MRI acquired in 20 asthmatics including five well-controlled and 15 severe poorly controlled participants who provided written informed consent to a study protocol approved by Health Canada. Same-day spirometry and plethysmography measurements of lung function and volume were acquired as well as (1)H MRI using a whole-body radiofrequency coil and fast spoiled gradient-recalled echo sequence at a fixed lung volume (functional residual capacity + 1 l). We incorporated the left-to-right lung volume proportion prior based on the Potts model and derived a volume-proportion preserved Potts model, which was approximated through convex relaxation and further represented by a dual volume-proportion preserved max-flow model. The max-flow model led to a linear problem with convex and linear equality constraints that implicitly encoded the proportion prior. To implement the algorithm, (1)H MRI was resampled into ∼3 × 3 × 3 mm(3) isotropic voxel space. Two observers placed seeds on each lung and on the background of 20 pulmonary (1)H MR images in a randomized dataset, on five occasions, five consecutive days in a row. Segmentation accuracy was evaluated using the Dice-similarity-coefficient (DSC) of the segmented thoracic cavity with comparison to five-rounds of manual segmentation by an expert observer. The authors also evaluated the root-mean-squared-error (RMSE) of the Euclidean distance between lung surfaces, the absolute, and percent volume error. Reproducibility was measured using the coefficient of variation (CoV) and intraclass correlation coefficient (ICC) for two observers who repeated segmentation measurements five-times.For five well-controlled asthmatics, forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC) was 83% ± 7% and FEV1 was 86 ± 9%pred. For 15 severe, poorly controlled asthmatics, FEV1/FV C = 66% ± 17% and FEV1 = 72 ± 27%pred. The DSC for algorithm and manual segmentation was 91% ± 3%, 92% ± 2% and 91% ± 2% for the left, right, and whole lung, respectively. RMSE was 4.0 ± 1.0 mm for each of the left, right, and whole lung. The absolute (percent) volume errors were 0.1 l (∼6%) for each of right and left lung and ∼0.2 l (∼6%) for whole lung. Intra- and inter-CoV (ICC) were <0.5% (>0.91%) for DSC and <4.5% (>0.93%) for RMSE. While segmentation required 10 s including ∼6 s for user interaction, the smallest detectable difference was 0.24 l for algorithm measurements which was similar to manual measurements.This lung segmentation approach provided the necessary and sufficient precision and accuracy required for research and clinical studies.

    View details for DOI 10.1118/1.4948999

    View details for PubMedID 27277040

  • Regional Heterogeneity of Chronic Obstructive Pulmonary Disease Phenotypes: Pulmonary (3)He Magnetic Resonance Imaging and Computed Tomography. COPD Pike, D. n., Kirby, M. n., Eddy, R. L., Guo, F. n., Capaldi, D. P., Ouriadov, A. n., McCormack, D. G., Parraga, G. n. 2016; 13 (5): 601–9


    Pulmonary ventilation may be visualized and measured using hyperpolarized (3)He magnetic resonance imaging (MRI) while emphysema and its distribution can be quantified using thoracic computed tomography (CT). Our objective was to phenotype ex-smokers with COPD based on the apical-to-basal distribution of ventilation abnormalities and emphysema to better understand how these phenotypes change regionally as COPD progresses. We evaluated 100 COPD ex-smokers who provided written informed consent and underwent spirometry, CT and (3)He MRI. (3)He MRI ventilation imaging was used to quantify the ventilation defect percent (VDP) for whole-lung and individual lung lobes. Regional VDP was used to generate the apical-lung (AL)-to-basal-lung (BL) difference (ΔVDP); a positive ΔVDP indicated AL-predominant and negative ΔVDP indicated BL-predominant ventilation defects. Emphysema was quantified using the relative-area-of-the-lung ≤-950HU (RA950) of the CT density histogram for whole-lung and individual lung lobes. The AL-to-BL RA950 difference (ΔRA950) was generated with a positive ΔRA950 indicating AL-predominant emphysema and a negative ΔRA950 indicating BL-predominant emphysema. Seventy-two ex-smokers reported BL-predominant MRI ventilation defects and 71 reported AL-predominant CT emphysema. BL-predominant ventilation defects (AL/BL: GOLD I = 18%/82%, GOLD II = 24%/76%) and AL-predominant emphysema (AL/BL: GOLD I = 84%/16%, GOLD II = 72%/28%) were the major phenotypes in mild-moderate COPD. In severe COPD there was a more uniform distribution for ventilation defects (AL/BL: GOLD III = 40%/60%, GOLD IV = 43%/57%) and emphysema (AL/BL: GOLD III = 64%/36%, GOLD IV = 43%/57%). Basal-lung ventilation defects predominated in mild-moderate GOLD grades, and a more homogeneous distribution of ventilation defects was observed in more advanced grade COPD; these differences suggest that over time, regional ventilation abnormalities become more homogenously distributed during disease progression.

    View details for DOI 10.3109/15412555.2015.1123682

    View details for PubMedID 26788765

  • Second-order Texture Measurements of (3)He Ventilation MRI: Proof-of-concept Evaluation of Asthma Bronchodilator Response. Academic radiology Zha, N. n., Pike, D. n., Svenningsen, S. n., Capaldi, D. P., McCormack, D. G., Parraga, G. n. 2016; 23 (2): 176–85


    (3)He magnetic resonance imaging (MRI) can be used to quantify functional responses to asthma therapy and provocation. Ventilation imaging offers quantitative information beyond ventilation defects that have not yet been exploited. Therefore, our objective was to evaluate hyperpolarized (3)He MRI ventilation defect percent (VDP) and compare this and pulmonary function measurements to ventilation image texture features and their changes post-bronchodilator administration in patients with asthma.Volunteers with a diagnosis of asthma provided written informed consent to an ethics board-approved protocol and underwent pulmonary function tests and MRI before and after salbutamol inhalation. MR images were analyzed using VDP, and their texture was evaluated via gray-level run-length matrices. These texture classifiers were compared to VDP in responders to bronchodilation based on VDP (VDP responders) and forced expiratory volume in 1 s (FEV1) (FEV1 responders).In total, 47 patients with asthma (18 males 39 ± 13 years, FEV1 = 79 ± 21%) reported significantly improved FEV1, FEV1/forced vital capacity (FVC), residual volume (RV)/total lung capacity (TLC) (all P = .0001) and VDP (P = .01) post-salbutamol. Post-salbutamol, VDP responders and nonresponders to salbutamol were significantly different for coarse-texture features including long-run emphasis (LRE) and long-run, low gray-level emphasis (LRLGE, both P < .05) and for FEV1 responders to salbutamol, there was significantly different long-run, high gray-level emphasis (LRHGE, P = .04). There were significant relationships for VDP with LRE (R = .50, P = .0003), LRLGE (R = .34, P = .02), and LRHGE (R = .56, P = .0001). Receiver operating characteristic curves showed VDP with the strongest performance (AUC = .92), followed by coarse-texture classifier LRHGE (AUC = .83), FEV1 (AUC = .80), LRE (AUC = .66), FVC (AUC = .58), and LRLGE (AUC = .42).In patients with asthma, differences in ventilation patchiness post-salbutamol can be quantified using coarse-texture classifiers that are significantly different in bronchodilator responders.

    View details for DOI 10.1016/j.acra.2015.10.010

    View details for PubMedID 26601971

  • Imaging how and where we breathe oxygen: another Big Short? Journal of thoracic disease Capaldi, D. P., Guo, F. n., Parraga, G. n. 2016; 8 (3): E204–7

    View details for DOI 10.21037/jtd.2016.01.83

    View details for PubMedID 27076971

    View details for PubMedCentralID PMC4805788

  • Pulmonary Imaging Biomarkers of Gas Trapping and Emphysema in COPD: (3)He MR Imaging and CT Parametric Response Maps. Radiology Capaldi, D. P., Zha, N. n., Guo, F. n., Pike, D. n., McCormack, D. G., Kirby, M. n., Parraga, G. n. 2016; 279 (2): 597–608


    To directly compare magnetic resonance (MR) imaging and computed tomography (CT) parametric response map (PRM) measurements of gas trapping and emphysema in ex-smokers both with and without chronic obstructive pulmonary disease (COPD).Participants provided written informed consent to a protocol that was approved by a local research ethics board and Health Canada and was compliant with the HIPAA (Institutional Review Board Reg. #00000940). The prospectively planned study was performed from March 2014 to December 2014 and included 58 ex-smokers (mean age, 73 years ± 9) with (n = 32; mean age, 74 years ± 7) and without (n = 26; mean age, 70 years ± 11) COPD. MR imaging (at functional residual capacity plus 1 L), CT (at full inspiration and expiration), and spirometry or plethysmography were performed during a 2-hour visit to generate ventilation defect percent (VDP), apparent diffusion coefficient (ADC), and PRM gas trapping and emphysema measurements. The relationships between pulmonary function and imaging measurements were determined with analysis of variance (ANOVA), Holm-Bonferroni corrected Pearson correlations, multivariate regression modeling, and the spatial overlap coefficient (SOC).VDP, ADC, and PRM gas trapping and emphysema (ANOVA, P < .001) measurements were significantly different in healthy ex-smokers than they were in ex-smokers with COPD. In all ex-smokers, VDP was correlated with PRM gas trapping (r = 0.58, P < .001) and with PRM emphysema (r = 0.68, P < .001). VDP was also significantly correlated with PRM in ex-smokers with COPD (gas trapping: r = 0.47 and P = .03; emphysema: r = 0.62 and P < .001) but not in healthy ex-smokers. In a multivariate model that predicted PRM gas trapping, the forced expiratory volume in 1 second normalized to the forced vital capacity (standardized coefficients [βS] = -0.69, P = .001) and airway wall area percent (βS = -0.22, P = .02) were significant predictors. PRM emphysema was predicted by the diffusing capacity for carbon monoxide (βS = -0.29, P = .03) and VDP (βS = 0.41, P = .001). Helium 3 ADC values were significantly elevated in PRM gas-trapping regions (P < .001). The spatial relationship for ventilation defects was significantly greater with PRM gas trapping than with PRM emphysema in patients with mild (for gas trapping, SOC = 36% ± 28; for emphysema, SOC = 1% ± 2; P = .001) and moderate (for gas trapping, SOC = 34% ± 28; for emphysema, SOC = 7% ± 15; P = .006) COPD. For severe COPD, the spatial relationship for ventilation defects with PRM emphysema (SOC = 64% ± 30) was significantly greater than that for PRM gas trapping (SOC = 36% ± 18; P = .01).In all ex-smokers, ADC values were significantly elevated in regions of PRM gas trapping, and VDP was quantitatively and spatially related to both PRM gas trapping and PRM emphysema. In patients with mild to moderate COPD, VDP was related to PRM gas trapping, whereas in patients with severe COPD, VDP correlated with both PRM gas trapping and PRM emphysema.

    View details for DOI 10.1148/radiol.2015151484

    View details for PubMedID 26744928

  • Magnetic resonance imaging biomarkers of chronic obstructive pulmonary disease prior to radiation therapy for non-small cell lung cancer. European journal of radiology open Sheikh, K. n., Capaldi, D. P., Hoover, D. A., Palma, D. A., Yaremko, B. P., Parraga, G. n. 2015; 2: 81–89


    In this prospectively planned interim-analysis, the prevalence of chronic obstructive lung disease (COPD) phenotypes was determined using magnetic resonance imaging (MRI) and X-ray computed tomography (CT) in non-small-cell-lung-cancer (NSCLC) patients.Stage-III-NSCLC patients provided written informed consent for pulmonary function tests, imaging and the 6-min-walk-test. Ventilation defect percent (VDP) and CT lung density (relative-of-CT-density-histogram <-950, RA950) were measured. Patients were classified into three subgroups based on qualitative and quantitative COPD and tumour-specific imaging phenotypes: (1) tumour-specific ventilation defects (TSD), (2) tumour-specific and other ventilation defects without emphysema (TSDV), and, (3) tumour-specific and other ventilation defects with emphysema (TSDVE).Seventeen stage-III NSCLC patients were evaluated (68 ± 7 years, 7 M/10 F, mean FEV1 = 77%pred) including seven current and 10 ex-smokers and eight patients with a prior lung disease diagnosis. There was a significant difference for smoking history (p = .02) and FEV1/FVC (p = .04) for subgroups classified using quantitative imaging. Patient subgroups classified using qualitative imaging findings were significantly different for emphysema (RA950, p < .001). There were significant relationships for whole-lung VDP (p < .05), but not RECIST or tumour-lobe VDP measurements with pulmonary function and exercise measurements. Preliminary analysis for non-tumour burden ventilation abnormalities using Reader-operator-characteristic (ROC) curves reflected a 94% classification rate for smoking pack-years, 93% for FEV1/FVC and 82% for RA950. ROC sensitivity/specificity/positive/negative likelihood ratios were also generated for pack-years, (0.92/0.80/4.6/0.3), FEV1/FVC (0.92/0.80/4.6/0.3), RA950 (0.92/0.80/4.6/0.3) and RECIST (0.58/0.80/2.9/1.1).In this prospectively planned interim-analysis of a larger clinical trial, NSCLC patients were classified based on COPD imaging phenotypes. A proof-of-concept evaluation showed that FEV1/FVC and smoking history identified NSCLC patients with ventilation abnormalities appropriate for functional lung avoidance radiotherapy.

    View details for DOI 10.1016/j.ejro.2015.05.003

    View details for PubMedID 26937440

    View details for PubMedCentralID PMC4750562

  • Principal Component Analysis of the CT Density Histogram to Generate Parametric Response Maps of COPD Zha, N., Capaldi, D. I., Pike, D., McCormack, D. G., Cunningham, I. A., Parraga, G., Gimi, B., Molthen, R. C. SPIE-INT SOC OPTICAL ENGINEERING. 2015

    View details for DOI 10.1117/12.2076396

    View details for Web of Science ID 000355665600038

  • Fourier-Based Linear Systems Description of Free-Breathing Pulmonary Magnetic Resonance Imaging Capaldi, D. I., Svenningsen, S., Cunningham, I. A., Parraga, G., Gimi, B., Molthen, R. C. SPIE-INT SOC OPTICAL ENGINEERING. 2015

    View details for DOI 10.1117/12.2081503

    View details for Web of Science ID 000355665600042

  • Free-breathing pulmonary 1H and Hyperpolarized 3He MRI: comparison in COPD and bronchiectasis. Academic radiology Capaldi, D. P., Sheikh, K. n., Guo, F. n., Svenningsen, S. n., Etemad-Rezai, R. n., Coxson, H. O., Leipsic, J. A., McCormack, D. G., Parraga, G. n. 2015; 22 (3): 320–29


    In this proof-of-concept demonstration, we aimed to quantitatively and qualitatively compare pulmonary ventilation abnormalities derived from Fourier decomposition of free-breathing (1)H magnetic resonance imaging (FDMRI) to hyperpolarized (3)He MRI in subjects with chronic obstructive pulmonary disease (COPD) and bronchiectasis.All subjects provided written informed consent to a protocol approved by a local research ethics board and Health, Canada, and they underwent MRI, computed tomography (CT), spirometry, and plethysmography during a single 2-hour visit. Semiautomated segmentation was used to generate ventilation defect measurements derived from FDMRI and (3)He MRI, and these were compared using analysis of variance and Pearson correlations.Twenty-six subjects were evaluated including 12 COPD subjects (67 ± 9 years) and 14 bronchiectasis subjects (70 ± 11 years). For COPD subjects, FDMRI and (3)He MRI ventilation defect percent (VDP) was 7 ± 6% and 24 ± 14%, respectively (P < .001; bias = -16 ± 9%). In COPD subjects, FDMRI was significantly correlated with (3)He MRI VDP (r = .88; P = .0001), (3)He MRI apparent diffusion coefficient (r = .71; P < .05), airways resistance (r = .60; P < .05), and RA950 (r = .80; P < .01). In subjects with bronchiectasis, FDMRI VDP (5 ± 3%) and (3)He MRI VDP (18 ± 9%) were significantly different (P < .001) and not correlated (P > .05). The Dice similarity coefficient (DSC) for FDMRI and (3)He MRI ventilation was 86 ± 7% for COPD and 86 ± 4% for bronchiectasis subjects (P > .05); the DSC for FDMRI ventilation defects and CT RA950 was 19 ± 20% in COPD and 2 ± 3% in bronchiectasis subjects (P < .01).FDMRI and (3)He MRI VDP were strongly related in COPD but not in bronchiectasis subjects. In COPD only, FDMRI ventilation defects were spatially related with (3)He ventilation defects and emphysema.

    View details for DOI 10.1016/j.acra.2014.10.003

    View details for PubMedID 25491735

  • Globally optimal co-segmentation of three-dimensional pulmonary ¹H and hyperpolarized ³He MRI with spatial consistence prior. Medical image analysis Guo, F. n., Yuan, J. n., Rajchl, M. n., Svenningsen, S. n., Capaldi, D. P., Sheikh, K. n., Fenster, A. n., Parraga, G. n. 2015; 23 (1): 43–55


    Pulmonary imaging using hyperpolarized (3)He/(129)Xe gas is emerging as a new way to understand the regional nature of pulmonary ventilation abnormalities in obstructive lung diseases. However, the quantitative information derived is completely dependent on robust methods to segment both functional and structural/anatomical data. Here, we propose an approach to jointly segment the lung cavity from (1)H and (3)He pulmonary magnetic resonance images (MRI) by constraining the spatial consistency of the two segmentation regions, which simultaneously employs the image features from both modalities. We formulated the proposed co-segmentation problem as a coupled continuous min-cut model and showed that this combinatorial optimization problem can be solved globally and exactly by means of convex relaxation. In particular, we introduced a dual coupled continuous max-flow model to study the convex relaxed coupled continuous min-cut model under a primal and dual perspective. This gave rise to an efficient duality-based convex optimization algorithm. We implemented the proposed algorithm in parallel using general-purpose programming on graphics processing unit (GPGPU), which substantially increased its computational efficiency. Our experiments explored a clinical dataset of 25 subjects with chronic obstructive pulmonary disease (COPD) across a wide range of disease severity. The results showed that the proposed co-segmentation approach yielded superior performance compared to single-channel image segmentation in terms of precision, accuracy and robustness.

    View details for DOI 10.1016/

    View details for PubMedID 25958028

  • Functional lung avoidance for individualized radiotherapy (FLAIR): study protocol for a randomized, double-blind clinical trial. BMC cancer Hoover, D. A., Capaldi, D. P., Sheikh, K., Palma, D. A., Rodrigues, G. B., Dar, A. R., Yu, E., Dingle, B., Landis, M., Kocha, W., Sanatani, M., Vincent, M., Younus, J., Kuruvilla, S., Gaede, S., Parraga, G., Yaremko, B. P. 2014; 14: 934


    Although radiotherapy is a key component of curative-intent treatment for locally advanced, unresectable non-small cell lung cancer (NSCLC), it can be associated with substantial pulmonary toxicity in some patients. Current radiotherapy planning techniques aim to minimize the radiation dose to the lungs, without accounting for regional variations in lung function. Many patients, particularly smokers, can have substantial regional differences in pulmonary ventilation patterns, and it has been hypothesized that preferential avoidance of functional lung during radiotherapy may reduce toxicity. Although several investigators have shown that functional lung can be identified using advanced imaging techniques and/or demonstrated the feasibility and theoretical advantages of avoiding functional lung during radiotherapy, to our knowledge this premise has never been tested via a prospective randomized clinical trial.Eligible patients will have Stage III NSCLC with intent to receive concurrent chemoradiotherapy (CRT). Every patient will undergo a pre-treatment functional lung imaging study using hyperpolarized 3He MRI in order to identify the spatial distribution of normally-ventilated lung. Before randomization, two clinically-approved radiotherapy plans will be devised for all patients on trial, termed standard and avoidance. The standard plan will be designed without reference to the functional state of the lung, while the avoidance plan will be optimized such that dose to functional lung is as low as reasonably achievable. Patients will then be randomized in a 1:1 ratio to receive either the standard or the avoidance plan, with both the physician and the patient blinded to the randomization results. This study aims to accrue a total of 64 patients within two years. The primary endpoint will be a pulmonary quality of life (QOL) assessment at 3 months post-treatment, measured using the functional assessment of cancer therapy-lung cancer subscale. Secondary endpoints include: pulmonary QOL at other time-points, provider-reported toxicity, overall survival, progression-free survival, and quality-adjusted survival.This randomized, double-blind trial will comprehensively assess the impact of functional lung avoidance on pulmonary toxicity and quality of life in patients receiving concurrent CRT for locally advanced identifier: NCT02002052.

    View details for DOI 10.1186/1471-2407-14-934

    View details for PubMedID 25496482

    View details for PubMedCentralID PMC4364501