Home Extending Semiautomatic Ventilation Defect Analysis for Hyperpolarized129 Xe Ventilation MRI
Post
Cancel

Extending Semiautomatic Ventilation Defect Analysis for Hyperpolarized129 Xe Ventilation MRI

Posted Nov 30, 2014 Updated Apr 10, 2023
By Mu He MS , S. Sivaram Kaushik MS , Scott H. Robertson MS , Matthew S. Freeman MS , Rohan S. Virgincar MS , H. Page McAdams MD , Bastiaan Driehuys PhD
9 min read

Rationale and Objectives

Clinical deployment of hyperpolarized 129 Xe magnetic resonance imaging requires accurate quantification and visualization of the ventilation defect percentage (VDP). Here, we improve the robustness of our previous semiautomated analysis method to reduce operator dependence, correct for B 1 inhomogeneity and vascular structures, and extend the analysis to display multiple intensity clusters.

Materials and Methods

Two segmentation methods were compared—a seeded region-growing method, previously validated by expert reader scoring, and a new linear-binning method that corrects the effects of bias field and vascular structures. The new method removes nearly all operator interventions by rescaling the 129 Xe magnetic resonance images to the 99th percentile of the cumulative distribution and applying fixed thresholds to classify 129 Xe voxels into four clusters: defect, low, medium, and high intensity. The methods were applied to 24 subjects including patients with chronic obstructive pulmonary disease ( n = 8), age-matched controls ( n = 8), and healthy normal subjects ( n = 8).

Results

Linear-binning enabled a faster and more reproducible workflow and permitted analysis of an additional 0.25 ± 0.18 L of lung volume by accounting for vasculature. Like region-growing, linear-binning VDP correlated strongly with reader scoring ( R 2 = 0.93, P < .0001), but with less systematic bias. Moreover, linear-binning maps clearly depict regions of low and high intensity that may prove useful for phenotyping subjects with chronic obstructive pulmonary disease.

Conclusions

Corrected linear-binning provides a robust means to quantify 129 Xe ventilation images yielding VDP values that are indistinguishable from expert reader scores, while exploiting the entire dynamic range to depict multiple image clusters.

The introduction of hyperpolarized (HP) 129 Xe magnetic resonance imaging (MRI) into clinical research has accelerated in recent years, with demonstrations of high-resolution imaging of pulmonary ventilation , alveolar microstructure , and gas exchange . Like 3 He MRI, which emerged over the past decade but suffers from supply limitations , 129 Xe MRI offers a means to enable radiation-free longitudinal imaging of pulmonary function. Although recent interest in 129 Xe has been centered on exploiting its intriguing properties such as solubility and chemical-shift to probe diffusive gas exchange , important questions still remain to be addressed about how best to acquire and analyze even basic 129 Xe spin-density (ventilation) images. HP 129 Xe spin-density imaging, like 3 He MRI, readily depicts ventilation impairment in a variety of pulmonary disorders such as asthma, chronic obstructive pulmonary disease (COPD), cystic fibrosis, and radiation-induced lung injury .

Although much progress has been made using reader-based scoring of HP gas MRI, it is also important to quantify these images in a way that is robust and automated. Such quantification not only enables high-throughput imaging, but also provides a means to extract the full richness of functional lung imaging. The simplest and longest-standing quantification metric is the ventilation defect percentage (VDP) introduced by Woodhouse et al. . Methods to calculate VDP were subsequently extended by Kirby et al. who exploited a breath-hold 1 H anatomical image to confine the analysis to the thoracic cavity. We have previously used this basic framework in conjunction with supervised region-growing methods to quantify VDP of 129 Xe MRI . This approach showed a good interoperator agreement and correlated well with expert reader scores. However, the study also highlighted the need to further remove operator dependence and to extend the range of analysis beyond simply calculating defect percentages. Moreover, the automated analysis can also be confounded by two additional factors. Firstly, the flexible vest coil may produce an inhomogeneous B 1 field, which causes a bias field variation across the lung . Secondly, the presence of pulmonary vasculature within the thoracic cavity can cause 129 Xe signal intensity voids that can be incorrectly classified as ventilation defects.

Get Radiology Tree app to read full this article<

Materials and methods

Get Radiology Tree app to read full this article<

129 Xe Polarization and Delivery

Get Radiology Tree app to read full this article<

Image Acquisition

Get Radiology Tree app to read full this article<

Semiautomated Image Analysis Methods

Creating the registered thoracic cavity mask

Get Radiology Tree app to read full this article<

Method 1: seeded region-growing segmentation to determine VDP

Get Radiology Tree app to read full this article<

Figure 1, Seeded region-growing segmentation workflow: the 129 Xe ventilation image (a) was segmented to obtain a ventilation mask (b) . The 1 H image was registered to the 129 Xe image and segmented to obtain a thoracic cavity mask (c) . These masks were multiplied to remove major airways from the ventilation mask, thereby obtaining the ventilated volume within the thoracic cavity, enabling calculation of the binary ventilation defect map (d) and VDP. VDP, ventilation defect percentage.

Get Radiology Tree app to read full this article<

Method 2: scaled linear-binning maps with bias field and vesselness corrections

Get Radiology Tree app to read full this article<

Figure 2, Corrected linear-binning map workflow. The bias field corrected ventilation images (a) were used to register the 1 H thoracic cavity images (b) . The proton images were then segmented and detected vasculature was removed to define the thoracic cavity mask that constrains the analysis (c) . Pixels from the bias field corrected ventilation image lying within the thoracic cavity volume were rescaled by their top percentile to range from 0 to 1 and classified into four intensity clusters to create the linear-binning map (d) .

Get Radiology Tree app to read full this article<

Linear-binning histogram scaling

Get Radiology Tree app to read full this article<

Vesselness correction of the thoracic cavity mask

Get Radiology Tree app to read full this article<

Figure 3, Tuning of the thoracic cavity mask. The thoracic cavity image (a) was segmented and morphologically filled to generate the initial thoracic cavity mask (b) . The thoracic cavity image then underwent vesselness filtering to detect the pulmonary vasculature (c) . This was then removed from the original thoracic cavity image to generate the final thoracic cavity mask (d) .

Get Radiology Tree app to read full this article<

Retrospective bias field correction

Get Radiology Tree app to read full this article<

Statistical Methods

Get Radiology Tree app to read full this article<

Results

Linear-Binning Histogram Scaling

Get Radiology Tree app to read full this article<

Figure 4, 129 Xe magnetic resonance imaging histogram rescaling: the native 129 Xe magnetic resonance image (a) has a histogram with a high-intensity “tail” (d) that must be removed for effective rescaling. If this is done by simply dividing all intensities by their top 5%, the resulting binning maps (b) overestimate the low-intensity bins because the tail has only been partially removed (e) . By instead scaling all 129 Xe intensity pixels by the 99th percentile of the 129 Xe intensity cumulative distribution, the associated binning maps are more reproducible and more consistent with reader perception (c) because the histogram tail has been effectively removed (f) .

Get Radiology Tree app to read full this article<

Vesselness Filter

Get Radiology Tree app to read full this article<

Figure 5, Impact of the “vesselness” filter. The thoracic cavity image (a) and 129 Xe ventilation image (b) in a healthy volunteer. The thoracic cavity mask after initial segmentation (c) overestimates the large vessels as shown in the intersection with 129 Xe ventilation image (d) . The thoracic cavity mask after morphologic closing (e) slightly reduces the overestimation of the larger vessels, but completely eliminates the exclusion of smaller vasculature from the mask (f) . The mask after application of the vesselness filter (g) now shows a thoracic cavity mask that best excludes the vasculature, while preserving a maximum 129 Xe image volume for quantitative analysis (h) .

Get Radiology Tree app to read full this article<

Bias Field Correction

Get Radiology Tree app to read full this article<

Figure 6, Bias field correction: column 1 shows ventilation images from two subjects (coronal and axial views) before the application of the retrospective bias correction algorithm. Column 2 depicts the estimated bias fields in both orientations (shown as a maximum intensity projection). Column 3 shows the 129 Xe images after bias field correction. Bias fields appear most intense near the coil elements.

Get Radiology Tree app to read full this article<

Seeded Region-Growing and Linear-Binning Clustering Measurements

Get Radiology Tree app to read full this article<

Figure 7, Ventilation images (before bias field correction) with associated reader-based scores, along with ventilation maps generated by the seeded region-growing method and corrected linear-binning method. Shown here are representative cases of a healthy volunteer, an age-matched control, and a COPD subject. COPD, chronic obstructive pulmonary disease; FEV 1 , forced expiratory volume in 1 second; VDP, ventilation defect percentage; VDS%, ventilation defect score percentage.

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Figure 8, Ventilation defect measurements for healthy volunteers, age-matched controls, and chronic obstructive pulmonary disease subjects obtained by expert reader scoring, seeded region-growing, and linear-binning.

Table 1

Mean Values (±Standard Deviation) of Ventilation Defect Score Percentage (VDS%) and 129 Xe Cluster Percentage Measurements Using Region-Growing and Linear-Binning

Parameter Mean Values ± Standard Deviation_P_ Values HV ( n = 8) AMC ( n = 8) COPD ( n = 8) HV, AMC AMC, COPD HV, COPD VDS% (%) 3.65 ± 6.07 13.02 ± 10.31 47.14 ± 11.93 .068 <.0001 ∗ <.0001 ∗ Seeded region-growing method VDP (%) 2.41 ± 2.55 7.77 ± 5.27 24.68 ± 7.59 .066 <.0001 ∗ <.0001 ∗ Linear-binning method VDP (%) 4.24 ± 2.91 10.29 ± 6.96 40.45 ± 15.40 .135 <.0001 ∗ <.0001 ∗ Low-intensity (%) 16.16 ± 3.76 15.94 ± 8.32 22.95 ± 7.07 .947 .055 .048 ∗ Medium-intensity (%) 70.87 ± 4.73 65.87 ± 11.76 32.69 ± 14.21 .373 <.0001 ∗ <.0001 ∗ High-intensity (%) 8.73 ± 1.28 7.90 ± 3.36 3.92 ± 1.37 .463 .0017 ∗ .0003 ∗

AMC, age-matched control; COPD, chronic obstructive pulmonary disease; HV, healthy volunteer, VDP, ventilation defect percentage.

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Correlations and Bland-Altman Analysis

Get Radiology Tree app to read full this article<

Figure 9, Correlation of the expert reader–based VDS% with VDP calculated using the seeded region-growing method and linear-binning method. Note that although both methods correlate well with reader scores, Bland-Altman plots show a systematic bias in seeded region-growing that is significantly reduced for linear binning. COPD, chronic obstructive pulmonary disease; VDP, ventilation defect percentage; VDS%, ventilation defect score percentage.

Get Radiology Tree app to read full this article<

Discussion

Advantages of the Corrected Linear-Binning Method

Get Radiology Tree app to read full this article<

Comparison of Reader-Based Scoring, Seeded Region-Growing, and Linear-Binning

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Figure 10, Examples of slight discordance between reader scoring and binning. Subject 1 depicts a healthy volunteer appearing to have no clear defects, but when viewed in the context of the registered thoracic cavity, shows ventilation defects and low intensity primarily in the apex of the right lung. The second row shows images of an age-matched control with a somewhat tortuous thoracic cavity, which may have caused readers to assign a higher VDS% of 16.67%. However, binning analysis, which incorporates the tortuous thoracic cavity shows mostly medium intensity. VPD, ventilation defect percentage; VDS%, ventilation defect score percentage.

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Study Limitations

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Conclusions

Get Radiology Tree app to read full this article<

Acknowledgments

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

References

  • 1. Svenningsen S., Kirby M., Starr D., et. al.: Hyperpolarized He and Xe MRI: differences in asthma before bronchodilation. J Magn Reson Imaging 2013; 38: pp. 1521-1530.

  • 2. Kaushik S.S., Freeman M.S., Cleveland Z.I., et. al.: Probing the regional distribution of pulmonary gas-exchange through single-breath, gas- and dissolved-phase 129Xe MR imaging. J Appl Physiol 2013; 115: pp. 850-860.

  • 3. Qing K., Ruppert K., Jiang Y., et. al.: Regional mapping of gas uptake by blood and tissue in the human lung using hyperpolarized xenon-129 MRI. J Magn Reson Imaging 2014; 39: pp. 346-359.

  • 4. Cleveland Z.I., Cofer G.P., Metz G., et. al.: Hyperpolarized Xe MR imaging of alveolar gas uptake in humans. PLoS One 2010; 5: pp. e12192.

  • 5. Kramer D.: DOE begins rationing helium-3. Phys Today 2010; 63: pp. 6.

  • 6. Virgincar R.S., Cleveland Z.I., Sivaram Kaushik S., et. al.: Quantitative analysis of hyperpolarized (129) Xe ventilation imaging in healthy volunteers and subjects with chronic obstructive pulmonary disease. NMR Biomed 2012; 26: pp. 424-435.

  • 7. Shukla Y., Wheatley A., Kirby M., et. al.: Hyperpolarized 129Xe magnetic resonance imaging: tolerability in healthy volunteers and subjects with pulmonary disease. Acad Radiol 2012; 19: pp. 941-951.

  • 8. Kirby M., Svenningsen S., Owrangi A., et. al.: Hyperpolarized 3He and 129Xe MR imaging in healthy volunteers and patients with chronic obstructive pulmonary disease. Radiology 2012; 265: pp. 600-610.

  • 9. Kirby M., Heydarian M., Svenningsen S., et. al.: Hyperpolarized 3He magnetic resonance functional imaging semiautomated segmentation. Acad Radiol 2012; 19: pp. 141-152.

  • 10. Woodhouse N., Wild J.M., Paley M.N., et. al.: Combined helium-3/proton magnetic resonance imaging measurement of ventilated lung volumes in smokers compared to never-smokers. J Magn Reson Imaging 2005; 21: pp. 365-369.

  • 11. Tustison N.J., Avants B.B., Cook P.A., et. al.: N4ITK: improved N3 bias correction. IEEE Trans Med Imaging 2010; 29: pp. 1310-1320.

  • 12. Frangi A.F., Niessen W.J., Hoogeveen R.M., et. al.: Model-based quantitation of 3-D magnetic resonance angiographic images. IEEE Trans Med Imaging 1999; 18: pp. 946-956.

  • 13. Rabe K.F., Hurd S., Anzueto A., et. al.: Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease—GOLD executive summary. Am J Respir Crit Care Med 2007; 176: pp. 532-555.

  • 14. Driehuys B., Cates G.D., Miron E., et. al.: High-volume production of laser-polarized Xe-129. Appl Phys Lett 1996; 69: pp. 1668-1670.

  • 15. Studholme C., Hill D.L.G., Hawkes D.J.: An overlap invariant entropy measure of 3D medical image alignment. Pattern Recogn 1999; 32: pp. 71-86.

  • 16. Rueckert D., Sonoda L.I., Hayes C., et. al.: Nonrigid registration using free-form deformations: application to breast MR images. IEEE Trans Med Imaging 1999; 18: pp. 712-721.

  • 17. Mistry N.N., Thomas A., Kaushik S.S., et. al.: Quantitative analysis of hyperpolarized 3He ventilation changes in mice challenged with methacholine. Magn Reson Med 2010; 63: pp. 658-666.

  • 18. Tustison N.J., Avants B.B., Flors L., et. al.: Ventilation-based segmentation of the lungs using hyperpolarized (3)He MRI. J Magn Reson Imaging 2011; 34: pp. 831-841.

  • 19. Belaroussi B., Milles J., Carme S., et. al.: Intensity non-uniformity correction in MRI: existing methods and their validation. Med Image Anal 2006; 10: pp. 234-246.

  • 20. Arnold J.B., Liow J.S., Schaper K.A., et. al.: Qualitative and quantitative evaluation of six algorithms for correcting intensity nonuniformity effects. Neuroimage 2001; 13: pp. 931-943.

  • 21. Wild J.M., Ajraoui S., Deppe M.H., et. al.: Synchronous acquisition of hyperpolarised 3He and 1H MR images of the lungs—maximising mutual anatomical and functional information. NMR Biomed 2011; 24: pp. 130-134.

  • 22. Tustison N.J., Altes T.A., Song G., et. al.: Feature analysis of hyperpolarized helium-3 pulmonary MRI: a study of asthmatics versus nonasthmatics. Magn Reson Med 2010; 63: pp. 1448-1455.

Academic Radiology, Volume 21, Issue 12
Journals General Radiology
This post is licensed under CC BY 4.0 by the author.
Recently Updated
Contents