Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Oct 24:9:1022981.
doi: 10.3389/fmed.2022.1022981. eCollection 2022.

Unsupervised clustering algorithms improve the reproducibility of dynamic contrast-enhanced magnetic resonance imaging pulmonary perfusion quantification in muco-obstructive lung diseases

Marilisa Konietzke  1   2   3 , Simon M F Triphan  2   3   4 , Monika Eichinger  2   3   4 , Sebastian Bossert  1 , Hartmut Heller  1 , Sabine Wege  5 , Ralf Eberhardt  3   5 , Michael U Puderbach  2   3   4   6 , Hans-Ulrich Kauczor  2   3   4 , Gudula Heußel  2   3   4 , Claus P Heußel  2   3   4 , Frank Risse  1 , Mark O Wielpütz  2   3   4
Affiliations

Unsupervised clustering algorithms improve the reproducibility of dynamic contrast-enhanced magnetic resonance imaging pulmonary perfusion quantification in muco-obstructive lung diseases

Marilisa Konietzke et al. Front Med (Lausanne). .

Abstract

Background: Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) allows the assessment of pulmonary perfusion, which may play a key role in the development of muco-obstructive lung disease. One problem with quantifying pulmonary perfusion is the high variability of metrics. Quantifying the extent of abnormalities using unsupervised clustering algorithms in residue function maps leads to intrinsic normalization and could reduce variability.

Purpose: We investigated the reproducibility of perfusion defects in percent (QDP) in clinically stable patients with cystic fibrosis (CF) and chronic obstructive pulmonary disease (COPD).

Methods: 15 CF (29.3 ± 9.3y, FEV1%predicted = 66.6 ± 15.8%) and 20 COPD (66.5 ± 8.9y, FEV1%predicted = 42.0 ± 13.3%) patients underwent DCE-MRI twice 1 month apart. QDP, pulmonary blood flow (PBF), and pulmonary blood volume (PBV) were computed from residue function maps using an in-house quantification pipeline. A previously validated MRI perfusion score was visually assessed by an expert reader.

Results: Overall, mean QDP, PBF, and PBV did not change within 1 month, except for QDP in COPD (p < 0.05). We observed smaller limits of agreement (± 1.96 SD) related to the median for QDP (CF: ± 38%, COPD: ± 37%) compared to PBF (CF: ± 89%, COPD: ± 55%) and PBV (CF: ± 55%, COPD: ± 51%). QDP correlated moderately with the MRI perfusion score in CF (r = 0.46, p < 0.05) and COPD (r = 0.66, p < 0.001). PBF and PBV correlated poorly with the MRI perfusion score in CF (r =-0.29, p = 0.132 and r =-0.35, p = 0.067, respectively) and moderately in COPD (r =-0.57 and r =-0.57, p < 0.001, respectively).

Conclusion: In patients with muco-obstructive lung diseases, QDP was more robust and showed a higher correlation with the MRI perfusion score compared to the traditionally used perfusion metrics PBF and PBV.

Keywords: chronic obstructive pulmonary disease (COPD); contrast agent lung perfusion; cystic fibrosis (CF); functional imaging; muco-obstructive lung disease.

PubMed Disclaimer

Conflict of interest statement

Authors MW, H-UK, and CH declared advisory board membership with Boehringer Ingelheim unrelated to the present study. Authors MK, SB, HH, and FR were employed by the Boehringer Ingelheim. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

FIGURE 1
FIGURE 1
Coronal residue function maps at the time point of maximum contrast-enhancement in the whole lung (Rmax maps), corresponding clustering maps for perfusion defects in percent maps (QDP maps) computation (3 clusters from 2 thresholds determined with Otsu’s method), QDP maps (defects highlighted in blue) and pulmonary blood flow maps (PBF maps) and pulmonary blood volume maps (PBV maps) from a representative cystic fibrosis and chronic obstructive pulmonary disease (COPD) patient at baseline (MRI1) and 1 month follow-up (MRI2). QDP was 21.57% at MRI1 and 20.61% at MRI2 in the cystic fibrosis patient, and 42.96% at MRI1 and 43.56% at MRI2 in the COPD patient (highlighted in blue). Perfusion scores were 8 at MRI1 and 7 at MRI2 for the cystic fibrosis patient, and 12 at MRI1 and 12 at MRI2 for the COPD patient.
FIGURE 2
FIGURE 2
Bland-Altman plots showing the midterm reproducibility of pulmonary perfusion defects in percent (QDP), pulmonary blood flow (PBF) and pulmonary blood volume (PBV) in adults with clinically stable cystic fibrosis (CF, crosses) and chronic obstructive pulmonary disease (COPD, circles). 13 CF patients and 14 COPD patients were evaluated. Mean differences, limits of agreement (LoA) and median values are given for each panel. Solid lines indicate the mean difference between MRI1 and MRI2, dashed lines the LoA (± 1.96*SD).
FIGURE 3
FIGURE 3
Correlations between perfusion defects in percent (QDP), pulmonary blood flow (PBF), and pulmonary blood volume (PBV) with the visual MRI perfusion score in 13 cystic fibrosis (CF, crosses) and 14 chronic obstructive pulmonary disease (COPD, circles) patients. MRI1 and MRI2 were combined. Spearman’s r correlation coefficients and corresponding p-values are given in each panel. Solid lines indicate the linear regression, dashed lines the minimum and maximum observed value for the parameter on the y-axis.
FIGURE 4
FIGURE 4
Correlations between visual MRI perfusion score, perfusion defect percent (QDP), pulmonary blood flow (PBF) and pulmonary blood volume (PBV) with forced expiratory volume in 1s percent predicted (FEV1%predicted) in 13 cystic fibrosis (CF, crosses) and post-bronchodilator FEV1%predicted in 14 chronic obstructive pulmonary disease (COPD, circles) patients. MRI1 and MRI2 were combined. Spearman’s r correlation coefficients and corresponding p-values are given in each panel. Solid lines indicate the linear regression lines, dashed lines the minimum and maximum observed value of the parameter on the y-axis.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Eichinger M, Optazaite DE, Kopp-Schneider A, Hintze C, Biederer J, Niemann A, et al. Morphologic and functional scoring of cystic fibrosis lung disease using MRI. Eur J Radiol. (2012) 81:1321–9. 10.1016/j.ejrad.2011.02.045 - DOI - PubMed
    1. Puderbach M, Eichinger M, Haeselbarth J, Ley S, Kopp-Schneider A, Tuengerthal S, et al. Assessment of morphological mri for pulmonary changes in cystic fibrosis (CF) patients. Invest Radiol. (2007) 42:715–24. 10.1097/rli.0b013e318074fd81 - DOI - PubMed
    1. Sileo C, Corvol H, Boelle P-Y, Blondiaux E, Clement A, Pointe HDL. HRCT and MRI of the lung in children with cystic fibrosis: comparison of different scoring systems. J Cyst Fibros. (2014) 13:198–204. 10.1016/j.jcf.2013.09.003 - DOI - PubMed
    1. Wielpütz MO, von Stackelberg O, Stahl M, Jobst BJ, Eichinger M, Puderbach MU, et al. Multicentre standardisation of chest MRI as radiation-free outcome measure of lung disease in young children with cystic fibrosis. J Cyst Fibros. (2018) 17:518–27. 10.1016/j.jcf.2018.05.003 - DOI - PubMed
    1. Wielpütz MO, Puderbach M, Kopp-Schneider A, Stahl M, Fritzsching E, Sommerburg O, et al. Magnetic resonance imaging detects changes in structure and perfusion, and response to therapy in early cystic fibrosis lung disease. Am J Respir Crit Care Med. (2014) 189:956–65. 10.1164/rccm.201309-1659oc - DOI - PubMed

LinkOut - more resources