Background: Early detection of pulmonary morbidity following haematopoietic stem cell transplantation (HSCT) remains an important challenge for intervention, primarily due to the insensitivity of spirometry to early change, and in paediatrics, patient compliance provides additional challenges. Regional lung ventilation abnormalities in paediatric HSCT patients were quantified using hyperpolarised xenon-129 (129Xe) magnetic resonance imaging (MRI) and compared to spirometry.Methods: Medically stable, paediatric allogeneic HSCT patients (n=23, ages 6–16 years) underwent an outpatient MRI scan where regional ventilation was quantified with a breath-hold of hyperpolarised129Xe gas. Ventilation deficits, regions of the lung that ventilate poorly due to obstruction, were quantified as a ventilation defect percentage (VDP) and compared to forced expiratory volume in 1 s (FEV1), FEV1/forced vital capacity (FVC) ratio, and forced expiratory flow at 25–75% of FVC (FEF25–75%) from spirometry using linear regression.Results: The mean±sd129Xe VDP was 10.5±9.4% (range 2.6–41.4%).129Xe VDP correlated with FEV1, FEV1/FVC ratio and FEF25–75%(p≤0.02 for all comparisons). Ventilation deficits were detected in patients with normal spirometry (i.e.FEV1>80%), supporting the sensitivity of129Xe MRI to early obstruction reported in other pulmonary conditions. Seven (30%) patients could not perform spirometry, yet ventilation deficits were observed in five of these patients, detecting abnormalities that otherwise may have gone undetected and untreated until advanced.Conclusion: Lung ventilation deficits were detected using hyperpolarised129Xe gas MRI in asymptomatic paediatric HSCT patients and in a subgroup who were unable to perform reliable spirometry.129Xe MRI provides a reliable imaging-based assessment of pulmonary involvement in this potentially difficult to diagnose paediatric population.
Background: Chest computed tomography (CT) remains the imaging standard for demonstrating cystic fibrosis (CF) airway structural diseasein vivo. However, visual scoring systems as an outcome measure are time consuming, require training and lack high reproducibility. Our objective was to validate a fully automated artificial intelligence (AI)-driven scoring system of CF lung disease severity.Methods: Data were retrospectively collected in three CF reference centres, between 2008 and 2020, in 184 patients aged 4–54 years. An algorithm using three 2D convolutional neural networks was trained with 78 patients’ CT scans (23 530 CT slices) for the semantic labelling of bronchiectasis, peribronchial thickening, bronchial mucus, bronchiolar mucus and collapse/consolidation. 36 patients’ CT scans (11 435 CT slices) were used for testingversusground-truth labels. The method's clinical validity was assessed in an independent group of 70 patients with or without lumacaftor/ivacaftor treatment (n=10 and n=60, respectively) with repeat examinations. Similarity and reproducibility were assessed using the Dice coefficient, correlations using the Spearman test, and paired comparisons using the Wilcoxon rank test.Results: The overall pixelwise similarity of AI-drivenversusground-truth labels was good (Dice 0.71). All AI-driven volumetric quantifications had moderate to very good correlations to a visual imaging scoring (p<0.001) and fair to good correlations to forced expiratory volume in 1 s % predicted at pulmonary function tests (p0.99).Conclusion: AI allows fully automated volumetric quantification of CF-related modifications over an entire lung. The novel scoring system could provide a robust disease outcome in the era of effective CF transmembrane conductance regulator modulator therapy.