Orso Pusterla

Purpose: To introduce a widely applicable workflow for pulmonary lobe segmentation of MR images using a recurrent neural network (RNN) trained with chest computed tomography (CT) datasets. The feasibility is demonstrated for 2D coronal ultra-fast balanced steady-state free precession (ufSSFP) MRI. Methods: Lung lobes of 250 publicly accessible CT datasets of adults were segmented with an open-source CT-specific algorithm. To match 2D ufSSFP MRI data of pediatric patients, both CT data and segmentations were translated into pseudo-MR images, masked to suppress anatomy outside the lung. Network-1 was trained with pseudo-MR images and lobe segmentations, and applied to 1000 masked ufSSFP images to predict lobe segmentations. These outputs were directly used as targets to train Network-2 and Network-3 with non-masked ufSSFP data as inputs, and an additional whole-lung mask as input for Network-2. Network predictions were compared to reference manual lobe segmentations of ufSSFP data in twenty pediatric cystic fibrosis patients. Manual lobe segmentations were performed by splitting available whole-lung segmentations into lobes. Results: Network-1 was able to segment the lobes of ufSSFP images, and Network-2 and Network-3 further increased segmentation accuracy and robustness. The average all-lobe Dice similarity coefficients were 95.0$\pm$2.8 (mean$\pm$pooled SD [%]), 96.4$\pm$2.0, 93.0$\pm$2.0, and the average median Hausdorff distances were 6.1$\pm$0.9 (mean$\pm$SD [mm]), 5.3$\pm$1.1, 7.1$\pm$1.3, for Network-1, Network-2, and Network-3, respectively. Conclusions: RNN lung lobe segmentation of 2D ufSSFP imaging is feasible, in good agreement with manual segmentations. The proposed workflow might provide access to automated lobe segmentations for various lung MRI examinations and quantitative analyses.

MRI showed a regional decrease of lung perfusion after exposure to tobacco smoke and a local increase of lung perfusion after electronic nicotine delivery systems use. Background Evidence regarding short-term effects of electronic nicotine delivery systems (ENDS) and tobacco smoke on lung ventilation and perfusion is limited. Purpose To examine the immediate effe...

Purpose: To introduce a widely applicable workflow for pulmonary lobe segmentation of MR images using a recurrent neural network (RNN) trained with chest CT datasets. The feasibility is demonstrated for 2D coronal ultrafast balanced SSFP (ufSSFP) MRI.Methods: Lung lobes of 250 publicly accessible CT datasets of adults were segmented with an open‐source CT‐specific algorithm. To match 2D ufSSFP MRI data of pediatric patients, both CT data and segmentations were translated into pseudo‐MR images that were masked to suppress anatomy outside the lung. Network‐1 was trained with pseudo‐MR images and lobe segmentations and then applied to 1000 masked ufSSFP images to predict lobe segmentations. These outputs were directly used as targets to train Network‐2 and Network‐3 with non‐masked ufSSFP data as inputs, as well as an additional whole‐lung mask as input for Network‐2. Network predictions were compared to reference manual lobe segmentations of ufSSFP data in 20 pediatric cystic fibrosis patients. Manual lobe segmentations were performed by splitting available whole‐lung segmentations into lobes.Results: Network‐1 was able to segment the lobes of ufSSFP images, and Network‐2 and Network‐3 further increased segmentation accuracy and robustness. The average all‐lobe Dice similarity coefficients were 95.0 ± 2.8 (mean ± pooled SD [%]) and 96.4 ± 2.5, 93.0 ± 2.0; and the average median Hausdorff distances were 6.1 ± 0.9 (mean ± SD [mm]), 5.3 ± 1.1, 7.1 ± 1.3 for Network‐1, Network‐2, and Network‐3, respectively.Conclusion: Recurrent neural network lung lobe segmentation of 2D ufSSFP imaging is feasible, in good agreement with manual segmentations. The proposed workflow might provide access to automated lobe segmentations for various lung MRI examinations and quantitative analyses.

Purpose: Lung impairment from functional MRI is frequently assessed as defect percentage. The defect distribution, however, is currently not quantified. The purpose of this work was to develop a novel measure that quantifies how clustered or scattered defects in functional lung MRI appear, and to evaluate it in pediatric cystic fibrosis.Theory: The defect distribution index (DDI) calculates a score for each lung voxel categorized as defected. The index increases according to how densely and how far an expanding circle around a defect voxel contains more than 50% defect voxels.Methods: Fractional ventilation and perfusion maps of 53 children with cystic fibrosis were previously acquired with matrix pencil decomposition MRI. In this work, the DDI is compared to a visual score of 3 raters who evaluated how clustered the lung defects appear. Further, spearman correlations between DDI and lung function parameters were determined.Results: The DDI strongly correlates with the visual scoring (r = 0.90 for ventilation; r = 0.88 for perfusion; P < .0001). Although correlations between DDI and defect percentage are moderate to strong (r = 0.61 for ventilation; r = 0.75 for perfusion; P < .0001), the DDI distinguishes between patients with comparable defect percentage.Conclusion: The DDI is a novel measure for functional lung MRI. It provides complementary information to the defect percentage because the DDI assesses defect distribution rather than defect size. The DDI is applicable to matrix pencil MRI data of cystic fibrosis patients and shows very good agreement with human perception of defect distributions.

Purpose: To investigate the repeatability and reproducibility of lung segmentation and their impact on the quantitative outcomes from functional pulmonary MRI. Additionally, to validate an artificial neural network (ANN) to accelerate whole‐lung quantification.Method: Ten healthy children and 25 children with cystic fibrosis underwent matrix pencil decomposition MRI (MP‐MRI). Impaired relative fractional ventilation (R_FV) and relative perfusion (R_Q) from MP‐MRI were compared using whole‐lung segmentation performed by a physician at two time‐points (A_t1 and A_t2), by an MRI technician (B), and by an ANN (C). Repeatability and reproducibility were assess with Dice similarity coefficient (DSC), paired t‐test and Intraclass‐correlation coefficient (ICC).Results: The repeatability within an observer (A_t1 vs A_t2) resulted in a DSC of 0.94 ± 0.01 (mean ± SD) and an unsystematic difference of −0.01% for R_FV (P = .92) and +0.1% for R_Q (P = .21). The reproducibility between human observers (A_t1 vs B) resulted in a DSC of 0.88 ± 0.02, and a systematic absolute difference of −0.81% (P < .001) for R_FV and −0.38% (P = .037) for R_Q. The reproducibility between human and the ANN (A_t1 vs C) resulted in a DSC of 0.89 ± 0.03 and a systematic absolute difference of −0.36% for R_FV (P = .017) and −0.35% for R_Q (P = .002). The ICC was >0.98 for all variables and comparisons.Conclusions: Despite high overall agreement, there were systematic differences in lung segmentation between observers. This needs to be considered for longitudinal studies and could be overcome by using an ANN, which performs as good as human observers and fully automatizes MP‐MRI post‐processing.

Purpose: Standard relaxation time quantification using phase‐cycled balanced steady‐state free precession (bSSFP), eg, motion‐insensitive rapid configuration relaxometry (MIRACLE), is subject to a considerable underestimation of tissue T₁ and T₂ due to asymmetric intra‐voxel frequency distributions. In this work, an artificial neural network (ANN) fitting approach is proposed to simultaneously extract accurate reference relaxation times (T₁, T₂) and robust field map estimates ( , ΔB₀) from the bSSFP profile.Methods: Whole‐brain bSSFP data acquired at 3T were used for the training of a feedforward ANN with N = 12, 6, and 4 phase‐cycles. The magnitude and phase of the Fourier transformed complex bSSFP frequency response served as input and the multi‐parametric reference set [T₁, T₂, , ∆B₀] as target. The ANN predicted relaxation times were validated against the target and MIRACLE.Results: The ANN prediction of T₁ and T₂ for trained and untrained data agreed well with the reference, even for only four acquired phase‐cycles. In contrast, relaxometry based on 4‐point MIRACLE was prone to severe off‐resonance‐related artifacts. ANN predicted and ∆B₀ maps showed the expected spatial inhomogeneity patterns in high agreement with the reference measurements for 12‐point, 6‐point, and 4‐point bSSFP phase‐cycling schemes.Conclusion: ANNs show promise to provide accurate brain tissue T₁ and T₂ values as well as reliable field map estimates. Moreover, the bSSFP acquisition can be accelerated by reducing the number of phase‐cycles while still delivering robust T₁, T₂, , and ∆B₀ estimates.