Effect of intravenously injected gadolinium-based contrast agents on functional lung parameters derived by PREFUL MRI

Sequence comparison of spoiled gradient echo and balanced steady-state free precession for pulmonary free-breathing proton MRI in patients and healthy volunteers: Correspondence, repeatability, and validation with dynamic contrast-enhanced MRI

Phase-resolved functional lung (PREFUL) MRI is a proton-based, contrast agent-free technique derived from the Fourier decomposition approach to measure regional ventilation and perfusion dynamics during free-breathing. Besides the necessity of extensive PREFUL postprocessing, the utilized MRI sequence must fulfill specific requirements. This study investigates the impact of sequence selection on PREFUL-MRI-derived functional parameters by comparing the standard spoiled gradient echo (SPGRE) sequence with a lung-optimized balanced steady-state free precession (bSSFP) sequence, thereby facilitating PREFULs clinical application in pulmonary disease assessment. This study comprised a prospective dataset of healthy volunteers and a retrospective dataset of patients with suspected chronic thromboembolic pulmonary hypertension. Both cohorts underwent PREFUL-MRI with both sequences to assess the correspondence of PREFUL ventilation and perfusion parameters (A). Additionally, healthy subjects were scanned a second time to evaluate repeatability (B), whereas patients received dynamic contrast-enhanced (DCE)-MRI, considered the perfusion gold standard for comparison with PREFUL-MRI (C). Signal-to-noise ratio (SNR), calculated from the unprocessed images, was compared alongside median differences of PREFUL-MRI-derived parameters using a paired Wilcoxon signed rank test. Further evaluations included calculation of the Pearson correlation, intraclass-correlation coefficient for repeatability assessment, and spatial overlap (SO) for regional comparison of PREFUL-MRI and DCE-MRI. bSSFP showed a clear SNR advantage over SPGRE (median: 23 vs. 9, p < 0.001). (A) Despite significant differences, parameter values were strongly correlated (r ≥ 0.75). After thresholding, binary maps showed high healthy overlap across both cohorts (SOHealthy > 86%) and high defect overlap in the patient cohort (SODefect ≥ 48%). (B) bSSFP demonstrated slightly higher repeatability across most parameters. (C) Both sequences demonstrated comparable correspondence to DCE-MRI, with SPGRE excelling in absolute quantification and bSSFP in spatial agreement. Although bSSFP showed superior SNR results, both sequences displayed spatial defect concordance and highly correlated PREFUL parameters with deviations regarding repeatability and alignment with DCE-MRI.

Diagnostic accuracy of perfusion-weighted phase-resolved functional lung magnetic resonance imaging in patients with chronic pulmonary embolism

Purpose

This study aimed to evaluate the diagnostic performance of perfusion-weighted phase-resolved functional lung (PW-PREFUL) magnetic resonance imaging (MRI) in patients with chronic pulmonary embolism (CPE).

Materials and methods

This study included 86 patients with suspected chronic thromboembolic pulmonary hypertension (CTEPH), who underwent PREFUL MRI and ventilation/perfusion (V/Q) single-photon emission computed tomography/computed tomography (SPECT/CT). PREFUL MRI was performed at 1.5 T using a balanced steady-state free precession sequence during free breathing. Color-coded PW images and quantitative parameters were obtained by postprocessing. Meanwhile, V/Q SPECT/CT imaging was performed as a reference standard. Hypoperfused areas in the lungs were scored for each lobe and segment using V/Q SPECT/CT images and PW-PREFUL MR images, respectively. Normalized perfusion (QN) and perfusion defect percentage (QDP) were calculated for all slices. For intra- and interobserver variability, the MRI images were analyzed 2 months after the first analysis by the same radiologist and another radiologist (11 years of lung MRI experience) blinded to the results of the first reader.

Results

Of the 86 enrolled patients, 77 met the inclusion criteria (36 diagnosed with CPE using V/Q SPECT/CT and 41 diagnosed with non-CPE etiology). For the PW-PREFUL MRI, the sensitivity, specificity, accuracy, and positive and negative predictive values for the diagnosis of CPE were 97, 95, 96, 95, and 98% at the patient level; 91, 94, 93, 91, and 94% at the lobe level, and 85, 94, 92, 88, and 94% at the segment level, respectively. The detection of segmental and subsegmental hypoperfusion using PW-PREFUL MRI revealed a moderate agreement with V/Q SPECT/CT (κ = 0.65; 95% confidence interval: 0.61-0.68). The quantitative results indicated that the QN was lower in the CPE group than in the non-CPE group [median score (interquartile range, IQR) 6.3 (2.8-9.2) vs. 13.0 (8.8-16.7), p < 0.001], and the QDP was higher [median score (IQR) 33.8 (15.7-51.7) vs. 2.2 (1.4-2.9), p < 0.001].

Conclusion

PREFUL MRI could be an alternative test to detect CPE without requiring breath-hold, contrast agents, or ionizing radiation.

Artificially-generated consolidations and balanced augmentation increase performance of U-net for lung parenchyma segmentation on MR images

PURPOSE

To improve automated lung segmentation on 2D lung MR images using balanced augmentation and artificially-generated consolidations for training of a convolutional neural network (CNN).

MATERIALS AND METHODS

From 233 healthy volunteers and 100 patients, 1891 coronal MR images were acquired. Of these, 1666 images without consolidations were used to build a binary semantic CNN for lung segmentation and 225 images (187 without consolidations, 38 with consolidations) were used for testing. To increase CNN performance of segmenting lung parenchyma with consolidations, balanced augmentation was performed and artificially-generated consolidations were added to all training images. The proposed CNN (CNNBal/Cons) was compared to two other CNNs: CNNUnbal/NoCons-without balanced augmentation and artificially-generated consolidations and CNNBal/NoCons-with balanced augmentation but without artificially-generated consolidations. Segmentation results were assessed using Sørensen-Dice coefficient (SDC) and Hausdorff distance coefficient.

RESULTS

Regarding the 187 MR test images without consolidations, the mean SDC of CNNUnbal/NoCons (92.1 ± 6% (mean ± standard deviation)) was significantly lower compared to CNNBal/NoCons (94.0 ± 5.3%, P = 0.0013) and CNNBal/Cons (94.3 ± 4.1%, P = 0.0001). No significant difference was found between SDC of CNNBal/Cons and CNNBal/NoCons (P = 0.54). For the 38 MR test images with consolidations, SDC of CNNUnbal/NoCons (89.0 ± 7.1%) was not significantly different compared to CNNBal/NoCons (90.2 ± 9.4%, P = 0.53). SDC of CNNBal/Cons (94.3 ± 3.7%) was significantly higher compared to CNNBal/NoCons (P = 0.0146) and CNNUnbal/NoCons (P = 0.001).

CONCLUSIONS

Expanding training datasets via balanced augmentation and artificially-generated consolidations improved the accuracy of CNNBal/Cons, especially in datasets with parenchymal consolidations. This is an important step towards a robust automated postprocessing of lung MRI datasets in clinical routine.

Comparison of phase-resolved functional lung (PREFUL) MRI derived perfusion and ventilation parameters at 1.5T and 3T in healthy volunteers

Purpose

The purpose of this study is to evaluate the influence of different field strengths on perfusion and ventilation parameters, SNR and CNR derived by PREFUL MRI using predefined sequence parameters.

Methods

Data sets of free breathing 2d FLASH lung MRI were acquired from 15 healthy subjects at 1.5T and 3T (Magnetom Avanto and Skyra, Siemens Healthcare, Erlangen, Germany) with a maximum period of 3 days in between. The processed functional parameters regional ventilation (RVent), perfusion (Q), quantified perfusion (Q_Quant), perfusion defect percentage (QDP), ventilation defect percentage (VDP) and ventilation-perfusion match (VQM) were compared for systematic differences. Signal- and contrast-to-noise ratio (SNR and CNR) of both acquisitions were analyzed.

Results

RVent, Q, VDP, SNR and CNR presented no significant differences between 1.5T and 3T. Q_Quant (1.5T vs. 3T, P = 0.04), and QDP (1.5T vs. 3T, P≤0.01) decreased significantly at 3T. Consequently, VQM increased significantly (1.5T vs. 3T, P≤0.01). Skewness and kurtosis of the Q-values increased significantly at 3T (P≤0.01). The mean Sørensen-Dice coefficients between both series were 0.91 for QDP and 0.94 for VDP. The Bland-Altman analysis of both series showed mean differences of 4.29% for QDP, 1.23% for VDP and -5.15% for VQM. Using the above-mentioned parameters for three-day repeatability at two different scanners and field strengths, the retrospective power calculation showed, that a sample size of 15 can detect differences of 3.7% for QDP, of 2.9% for VDP and differences of 2.6% for VQM.

Conclusion

Significant differences in QDP may be related to field inhomogeneities, which is expressed by increasing skewness and kurtosis at 3T. Q_Quant reveals only poor reproducibility between 1.5T and 3T. RVent, Q, VDP, SNR and CNR were not altered significantly at the used sequence parameters. Healthy participants with minimal defects present high spatial agreement of QDP and VDP.