Quantification of MRI T2 Interstitial Lung Disease Signal‐Intensity Volume in Idiopathic Pulmonary Fibrosis: A Pilot Study

Abstract

Background Imaging has played a pivotal role in the diagnosis of idiopathic pulmonary fibrosis (IPF). Recent reports suggest that T2‐weighted MRI could be sensitive to monitor signal‐intensity modifications of the lung parenchyma, which may relate to the disease activity in IPF. However, there is a lack of automated tools to reproducibly quantify the extent of the disease, especially using MRI. Purpose To assess the feasibility of T2 interstitial lung disease signal‐intensity volume quantification using a semiautomated method in IPF. Study Type Single center, retrospective. Population A total of 21 adult IPF patients and four control subjects without lung interstitial abnormalities. Field Strength/Sequence Both free‐breathing ultrashort echo time (TE) lung MRI using the spiral volume interpolated breath hold examination (VIBE) sequence (3D‐UTE) and T2‐BLADE at 1.5T. Assessment Semiautomated segmentation of the lung volume was done using 3D‐UTE and registered to the T2‐BLADE images. The interstitial lung disease signal‐intensity volume (ISIV) was quantified using a Gaussian mixture model clustering and then normalized to the lung volume to calculate T2‐ISIV. The composite physiological index (CPI) and forced vital capacity (FVC) were measured as known biomarkers of IPF severity. Measurements were performed independently by three readers and averaged. The reproducibility between measurements was also assessed. Statistical Tests Reproducibility was assessed using the intraclass correlation coefficient (ICC) and Bland–Altman analysis. Correlations were assessed using Spearman test. Comparison of median was assessed using the Mann–Whitney test. Results The reproducibility of T2‐ISIV was high, with ICCs = 0.99. Using Bland–Altman analysis, the mean differences were found between –0.8 to 0.1. T2‐ISIV significantly correlated with CPI and FVC (rho = 0.48 and 0.50, respectively; P < 0.05). T2‐ISIV was significantly higher in IPF than in controls (P < 0.05). Data Conclusion T2‐ISIV appears to be able to reproducibly assess the volumetric extent of abnormal interstitial lung signal‐intensity modifications in patients with IPF, and correlate with disease severity. Level of Evidence 4 Technical Efficacy Stage 1

Three-dimensional ultrashort echo time magnetic resonance imaging in assessment of idiopathic pulmonary fibrosis, in comparison with high-resolution computed tomography

“…Although the bSSFP sequence is less valuable in depicting minor fibrotic changes, the diagnostic accuracy of MRI tends to increase in more advanced ILD stages (7). The ILD signal intensity volume (ISIV) on T2-BLADE can assess the volumetric extent of abnormal interstitial lung signal intensity modifications in patients with IPF in a reproducible manner (8). Recently, the value of ultrashort echo time MRI (UTE-MRI) of the lung has been demonstrated in imaging lungs and lung diseases, providing high signal-to-noise rations (SNRs) and high-resolution images for the quantitative and qualitative assessment of changes in lung parenchyma (9)(10)(11)(12)(13)(14)(15)(16).…”

Editorial for “Implementable Deep Learning for Multi‐sequence Proton MRI Lung Segmentation: A Multi‐center, Multi‐vendor and Multi‐disease Study”

“…However, non-AI tools such as region growing may fail at segmenting the peripheral lung contours when areas of ground glass opacification of consolidation, within the lung parenchyma, may abrupt the lung periphery. 6 This is especially the case in interstitial lung diseases such as COVID-19 or lung fibrosis, where the opacifications are likely to involve the subpleural spaces. Also, there is a growing interest in 4D imaging, for which a ponderous manual segmentation of dynamic series would have to be performed.…”

“…When performed manually, the segmentation is time‐consuming, cumbersome, and may lack reproducibility. However, non‐AI tools such as region growing may fail at segmenting the peripheral lung contours when areas of ground glass opacification of consolidation, within the lung parenchyma, may abrupt the lung periphery 6 . This is especially the case in interstitial lung diseases such as COVID‐19 or lung fibrosis, where the opacifications are likely to involve the subpleural spaces.…”

Lung Magnetic Resonance Imaging for Prediction of Progression in Patients With Nonidiopathic Pulmonary Fibrosis Interstitial Lung Disease

“…Magnetic resonance imaging (MRI) has been proposed as an additional imaging method for ILD characterization, with different sequences and techniques being studied alongside LFTs and CT. [11][12][13][14][15][16][17][18] However, few studies have assessed the prognostic value of the MRI findings. [19][20][21][22] .…”