4026
Correlation between CT and QSM in the brain1Department of Diagnostic Radiology and Nuclear Medicine, Graduate School of Medicine, Kyoto University, Kyoto, Japan, 2Human Brain Research Center, Graduate School of Medicine, Kyoto University, Kyoto, Japan, 3Integrated Clinical Education Center, Kyoto University Hospital, Kyoto, Japan, Kyoto, Japan
Synopsis
Quantitative susceptibility mapping (QSM) is a technique which can provide quantitative values of magnetic susceptibility, and thus is a useful tool for differentiating hemorrhage from calcification. Computed tomography (CT) is also used for evaluation of hemorrhage and calcification by using CT values. However, the correlation between CT and QSM has not been much investigated. In this study, we found positive correlations between CT values and susceptibility in the globus pallidus and hemorrhagic lesions. Negative correlations were observed in the choroid plexus and calcified lesions. Our results may help radiologists revisit the CT values based on the susceptibility values calculated from QSM.
Introduction
Differentiation between calcification and hemorrhage is important for clinical diagnosis of brain diseases. Computed tomography (CT) has been widely used as a quantitative method to diagnose hemorrhage and calcification. Also on MRI, quantitative susceptibility mapping (QSM) has been developed to evaluate susceptibility of substances quantitatively.1-5 Paramagnetic materials like hemosiderin or iron and diamagnetic materials such as calcium have opposite signals on QSM. However, the relationship between CT values and susceptibility has not been much investigated. In this study, we evaluated the correlation between CT values and susceptibility in normal structures and pathological lesions in the brain.Methods
Subjects
This study was approved by the institutional review board. We retrospectively enrolled 84 patients and 20 healthy elderly volunteers (46 male and 58 female subjects; mean age ± SD, 66.2 ± 18.2) who underwent both CT and multi-echo gradient echo (GRE) MR imaging.
Image acquisition
CT scans were performed with 64-, 80- or 320- detector CT scanners (Aquilion, Aquilion PRIME, and Aquilion ONE, respectively, Canon Medical Systems Corporation, Otawara, Japan). Parameters for patients: helical acquisition; beam pitch, 0.825; exposure, 150-300 mAs; X-ray energy, 120 kVp. For volunteers, exposure dose was reduced.
MR images were acquired on 3T scanners (MAGNETOM Skyra, Siemens Healthineer AG, Erlangen, Germany) with 32-channel head coils. Imaging parameters: TR 44 ms/TE 3.6-39.1 ms/ΔTE 5.92 ms; field of view, 240×240 mm; resolution, 0.9×0.9×1.0 mm; GRAPPA, 2; and acquisition time, 5 min 3 sec.
Post-imaging procedure
QSM calculation was conducted using STISuite version 2.2 (http://people.duke.edu/~cl160/). Mask images were generated from magnitude images by threshold and applied to phage images. For phase unwrapping and background phase removal, a sophisticated harmonic artifact reduction for phase data (SHARP) method with a variable radius of the spherical kernel at the brain boundary (V-SHARP) was conducted, and then QSM was calculated from each local tissue phase by solving an inverse problem using the improved sparse linear equations and sparse least squares (iLSQR) method.
CT images were registered to QSM images by using Syngo workstation (Siemens Healthineer AG, Germany) and SPM12 (Wellcome Trust Centre for Neuroimaging. Institute of Neurology, University College London, United Kingdom).
Image analysis
ROIs were manually placed on QSM images and applied to CT images by using ImageJ (National Institutes of Health, Bethesda, Maryland, United States) at the globus pallidus (GP), putamen (PT), caudate nucleus (CN), substantia nigra (SN), red nucleus (RN), dentate nucleus (DN), and choroid plexus (CP) (Figure 1), and hemorrhagic and calcified pathological lesions.
In addition to mean values of ROIs, maximum and 95th percentile or minimum and 5th percentile values were assessed for better understanding of the characteristics of each ROI. The correlations between CT values and susceptibility were assessed: CTmean_QSMmean, CTmax_QSMmax and CT95th_QSM95th for paramagnetic ROIs, and CTmean_QSMmean, CTmax_QSMmin and CT95th_QSM5th for diamagnetic ROIs.
Linear regression method was used. A P value < 0.05 was considered statistically significant.
Results
The scatter plot of mean values of all ROIs is shown in Figure 2. Twenty-eight hemorrhagic and 17 calcified lesions were identified. Significant positive correlations between CT values and susceptibility were observed in GP and hemorrhagic lesions, and negative correlations were found in CP and calcified lesions except for CTmean_QSMmean of calcified lesions (Table 1 and Figure 3). In GP, R2 of CTmax_QSMmax and CT95th_QSM95th were higher than that of CTmean_QSMmean. There was no significant correlation in other ROIs.Discussion
In our study, a good correlation was observed in GP. Various metals (e.g., zinc, iron, copper, manganese and aluminum) are known to deposit accompanying calcium deposition in the basal ganglia.6 It is known that physiologic calcification is the most obvious in GP on CT images,7 and GP is also the most iron-rich structure.8 Calcium and iron show high attenuation on CT, while iron has strong paramagnetic property and may overwhelm calcium’s relatively weak negative susceptibility on QSM. This might be the reason why a positive correlation was seen between CT value and susceptibility in GP. Iron also deposits in PT, CN, RN, SN and DN, but less than GP.8 We consider there is not enough iron or calcium to generate correlation between CT values and susceptibility.
The observed negative correlations in CP, which contains calcification, and calcified lesions were compatible with the fact that calcification cause negative susceptibility on QSM. Hemorrhagic lesions increase both CT values and susceptibility, and also showed a significant positive correlation.
Conclusion
Positive correlations between CT values and susceptibility were observed in the globus pallidus and hemorrhagic lesions, and negative correlations were found in the choroid plexus and calcified lesions.Acknowledgements
We are grateful to Mr. Katsutoshi Murata, Mr. Yuta Urushibata, and Mr. Hirokazu Kawaguchi, Siemens Healthcare K. K., for their kind help.References
1. Shmueli K, de Zwart JA, van Gelderen P, Li TQ, Dodd SJ, Duyn JH. Magnetic susceptibility mapping of brain tissue in vivo using MRI phase data. Magn Reson Med 2009;62(6):1510-1522.
2. Schweser F, Deistung A, Lehr BW, Reichenbach JR. Differentiation between diamagnetic and paramagnetic cerebral lesions based on magnetic susceptibility mapping. Med Phys 2010;37(10):5165-5178.
3. Wang Y, Liu T. Quantitative susceptibility mapping (QSM): Decoding MRI data for a tissue magnetic biomarker. Magn Reson Med 2015;73(1):82-101. doi: 10.1002/mrm.25358
4. Li W, Wu B, Liu C. Quantitative susceptibility mapping of human brain reflects spatial variation in tissue composition. Neuroimage 2011;55(4):1645-1656.
5. Sun H, Klahr AC, Kate M, Gioia LC, Emery DJ, Butcher KS, Wilman AH. Quantitative Susceptibility Mapping for Following Intracranial Hemorrhage. Radiology 2018;288(3):830-839.
6. Atlas SW, Grossman RI, Hackney DB, Gomori JM, Campagna N, Goldberg HI, Bilaniuk LT, Zimmerman RA. Calcified intracranial lesions: detection with gradient-echo-acquisition rapid MR imaging. AJR Am J Roentgenol 1988;150(6):1383-1389.
7. Adams AE. Basal ganglia calcification. Characteristics of CT scans and clinical findings. Neurosurg Rev. 1980;3(3):201-3.
8. Langkammer C, Schweser F, Krebs N, Deistung A, Goessler W, Scheurer E, Sommer K, Reishofer G, Yen K, Fazekas F, Ropele S, Reichenbach JR. Quantitative susceptibility mapping (QSM) as a means to measure brain iron? A post mortem validation study. Neuroimage 2012;62(3):1593-1599.
Figures
