Electromagnetic Properties: Susceptibility: QSM vs. SWI Pros & Cons
Xu Li1,2
1F.M. Kirby Research Center, Kennedy Krieger Institute, Baltimore, MD, United States, 2Radiology, Johns Hopkins University School of Medicine, Baltimore, MD, United States

Synopsis

Keywords: Contrast mechanisms: Electromagnetic tissue properties, Contrast mechanisms: Susceptibility

Tissue magnetic susceptibility is an important source of imaging contrast used in susceptibility-weighted imaging (SWI) and quantitative susceptibility mapping (QSM). In this presentation, I will explain and compare the basic technical aspects of SWI and QSM. While SWI focuses on enhancing contrast related to tissue magnetic susceptibility, QSM aims at quantifying the magnetic susceptibility distribution. QSM overcomes blooming artifacts in SWI and improves differentiation between paramagnetic and diamagnetic materials, such as iron and myelin. However, QSM requires a more complicated reconstruction pipeline.

What is magnetic susceptibility and how it is useful in MRI

Magnetic susceptibility is an intrinsic property of materials that describe their degree of magnetization when placed in an external magnetic field. Paramagnetic materials enhance their internal field, while diamagnetic materials reduce it (1).
In MRI, the magnetic susceptibility effect is a well-known phenomenon that results from differences in magnetic susceptibility between different tissues or materials. These differences can cause imaging artifacts such as signal loss and geometric distortions, especially in areas around air-tissue interfaces or near metallic implants. However, variations in tissue magnetic susceptibility can also provide valuable pathophysiological information and imaging contrast related to tissue composition such as iron deposition, myelin content, blood oxygenation, and tissue microstructures (2-4). Magnetic susceptibility has long been studied as an imaging contrast mechanism as in susceptibility weighted imaging (SWI) (5, 6) and more recently in quantitative susceptibility mapping (QSM) (7-10).

Susceptibility weighted imaging (SWI)

SWI is a specialized MRI technique that combines a specific T2*-weighted gradient echo (GRE) sequence with a unique processing design that utilizes a phase mask to enhance tissue contrast based on the underlying tissue magnetic susceptibility difference (5, 11). Initially, SWI was developed for the better visualization of venous structures and the detection of microhemorrhages (6, 12). It can also be used to visualize large tissue magnetic susceptibility sources, such as paramagnetic substances like iron and diamagnetic substances like calcium.
The fundamental principle underlying susceptibility-weighted imaging (SWI) is the tissue susceptibility-induced non-local magnetic field shift and variation. Such field shift results in not only T2* signal decay but also changes in the MR phase measurements, which are particularly sensitive to this magnetic susceptibility effect. To generate a SWI image, a high-pass filter is typically applied to the phase data to eliminate phase aliasing and large background field caused by air-tissue interfaces. The filtered phase data is then used to create a phase mask, which is multiplied with the magnitude images several times to achieve the SWI contrast. Recent advancements in SWI using multi-echo acquisition have enabled the resolution of signal dropout and wrap-like artifacts around large $$$B_0$$$ inhomogeneities, such as those seen around the frontal sinus in conventional single-echo-based SWI images (13, 14).
It is important to acknowledge that the susceptibility-induced magnetic field shift is non-local, meaning that it affects regions beyond the source of susceptibility, and is highly dependent on the orientation of the susceptibility sources relative to the magnetic field. This phenomenon is responsible for the appearance of blooming artifacts in SWI, where the contrast extends beyond the original source of susceptibility. As a result, lesions with high susceptibility values, such as micro-hemorrhages, may appear larger in SWI images than their actual size, and image contrast can vary across scans that use different head orientations and scan parameters.
When acquired with a 3D high-resolution GRE sequence, SWI contrast has demonstrated better depiction of cerebral microbleed and vascular diseases than the T2* weighted imaging (15, 16). SWI or SWI-related imaging techniques, e.g. high-pass filtered phase, have also found applications in multiple sclerosis, including the detection of the central vein sign and peri-lesion rim (17, 18), as well as in Parkinson’s disease (19) and other neurodegenerative diseases (11).

Quantitative susceptibility mapping (QSM)

QSM, as a successor of SWI, aims to quantitatively reconstruct the magnetic susceptibility distribution from the measured MR phase image. This is mainly achieved by performing a field-to-susceptibility dipole deconvolution with the known dipole kernel (20, 21) to compute the underlying tissue magnetic susceptibility from the magnetic field measurement derived from phase image (22-24). QSM uses similar T2* weighted GRE sequence as SWI, preferably with 3D and multi-echo acquisition. QSM reconstruction involves multiple processing steps from the acquired GRE dataset, including phase unwrapping and echo combination, creation of mask, background field removal and dipole inversion (10).
The phase unwrapping and echo combination step resolves the phase aliasing and calculates a total field map. Different phase unwrapping methods have been developed and applied for MR phase images (25), while multi-echo combination can be achieved either by fitting the magnetic field shift over TEs (26) or using a weighed echo averaging approach to boost SNR (27). A phase quality map or spatial noise can also be estimated from this step to help mask out unreliable phase measurement.
The creation of a mask is another important step for QSM, particularly for background field removal. The mask defines the region of interest (ROI) within which the tissue field map and QSM will be estimated. The background field removal aims at removing the field shift contributions generated from sources outside of the defined ROI. For the brain, this is normally caused by the air-tissue interface with susceptibility differences orders of magnitude larger than that between normal brain gray matter and white matter. Background field removal can be done based on the harmonic property of the background field or the approximate orthogonality between the background and tissue field (28-32).
Finally, the field-to-susceptibility inversion step performs the dipole deconvolution to calculate the susceptibility distribution map from the tissue field. Note that this inversion is ill-conditioned due to the singularity of the dipole kernel on a cone surface and ill-posed due to signal deviations from the diploe pattern, e.g. due to limited proton sensor or rapid signal decay (7, 33). As a result, most QSM inversion algorithm relies on regularization to suppress the streaking and shadowing artifacts in the calculated susceptibility maps. Sparsity-type regularization, such as L1-norm or total variation regularization applied in image space, typically performs better than generic k-space based regularizer, as demonstrated in previous QSM challenges (34, 35). Deep learning based QSM methods have emerged in recently years, but their generalization capability may need further improvements (36-38).
QSM has been applied in many clinical research areas including neurodevelopment, aging, and many neurodegenerative diseases (4, 39). QSM measures and changes are commonly linked to its potential contributing tissue sources such as iron, myelin, and oxygenation (40). QSM can also be used to visualize venous structure and micro-hemorrhages, as in SWI (41).

QSM vs. SWI Pros and Cons

Compared to SWI, QSM provides more accurate quantitative measures of tissue magnetic susceptibility values, making it a superior imaging biomarker for assessing tissue compositions and monitoring disease related changes. Unlike SWI, QSM is not affected by blooming artifacts, which can cause lesions with high susceptibility values to appear larger than their actual size and lead to varying image contrast. QSM solves this issue by solving the dipole deconvolution problem. Furthermore, SWI has limited capability to differentiate paramagnetic and diamagnetic materials as both leads to extra T2* signal decay and image hypo-intensity in SWI. Even though the filtered phase images provided in modern SWI sequences may help distinguish the paramagnetic (e.g. blood products) versus diamagnetic (e.g. calcifications) materials, such phase information may not be applicable to lesions with more complex geometries, and its interpretation may depend on the vendor (11). QSM overcomes this limitation, making it a superior option for this task.
However, QSM suffers from a more complicated image reconstruction process, requiring higher computational cost and more reconstruction parameters. While community consensus on QSM is ongoing, it is mainly limited in research studies. In contrast, SWI is a faster and more straightforward technique that has been implemented on scanners by most vendors and it is widely used in clinic routine.

Acknowledgements

No acknowledgement found.

References

1. J. F. Schenck, The role of magnetic susceptibility in magnetic resonance imaging: MRI magnetic compatibility of the first and second kinds. Med Phys 23, 815-850 (1996).

2. J. H. Duyn, J. Schenck, Contributions to magnetic susceptibility of brain tissue. NMR Biomed 10.1002/nbm.3546 (2016).

3. S. Ropele, C. Langkammer, Iron quantification with susceptibility. NMR Biomed 30 (2017).

4. Y. Wang et al., Clinical quantitative susceptibility mapping (QSM): Biometal imaging and its emerging roles in patient care. J Magn Reson Imaging 46, 951-971 (2017).

5. E. M. Haacke, Y. Xu, Y. C. Cheng, J. R. Reichenbach, Susceptibility weighted imaging (SWI). Magn Reson Med 52, 612-618 (2004).

6. E. M. Haacke, S. Mittal, Z. Wu, J. Neelavalli, Y. C. Cheng, Susceptibility-weighted imaging: technical aspects and clinical applications, part 1. AJNR Am J Neuroradiol 30, 19-30 (2009).

7. Y. Wang, T. Liu, Quantitative susceptibility mapping (QSM): Decoding MRI data for a tissue magnetic biomarker. Magn Reson Med 73, 82-101 (2015).

8. E. M. Haacke et al., Quantitative susceptibility mapping: current status and future directions. Magn Reson Imaging 33, 1-25 (2015).

9. C. Liu, W. Li, K. A. Tong, K. W. Yeom, S. Kuzminski, Susceptibility-weighted imaging and quantitative susceptibility mapping in the brain. J Magn Reson Imaging 42, 23-41 (2015).

10. A. Deistung, F. Schweser, J. R. Reichenbach, Overview of quantitative susceptibility mapping. NMR Biomed 30 (2017).

11. S. Haller, E. M. Haacke, M. M. Thurnher, F. Barkhof, Susceptibility-weighted Imaging: Technical Essentials and Clinical Neurologic Applications. Radiology 299, 3-26 (2021).

12. M. Ayaz, A. S. Boikov, E. M. Haacke, D. K. Kido, W. M. Kirsch, Imaging Cerebral Microbleeds Using Susceptibility Weighted Imaging: One Step Toward Detecting Vascular Dementia. Journal of Magnetic Resonance Imaging 31, 142-148 (2010).

13. S. S. Oh et al., Improved susceptibility weighted imaging method using multi-echo acquisition. Magn Reson Med 72, 452-458 (2014).

14. K. Eckstein et al., Improved susceptibility weighted imaging at ultra-high field using bipolar multi-echo acquisition and optimized image processing: CLEAR-SWI. Neuroimage 237, 118175 (2021).

15. W. Liu et al., Imaging Cerebral Microhemorrhages in Military Service Members with Chronic Traumatic Brain Injury. Radiology 278, 536-545 (2016).

16. S. Haller et al., Cerebral Microbleeds: Imaging and Clinical Significance. Radiology 287, 11-28 (2018).

17. G. Sparacia, F. Agnello, A. Gambino, M. Sciortino, M. Midiri, Multiple sclerosis: High prevalence of the 'central vein' sign in white matter lesions on susceptibility-weighted images. Neuroradiol J 31, 356-361 (2018).

18. M. Absinta et al., Association of Chronic Active Multiple Sclerosis Lesions With Disability In Vivo. JAMA Neurol 76, 1474-1483 (2019).

19. M. S. Perez Akly et al., Accuracy of nigrosome-1 detection to discriminate patients with Parkinson's disease and essential tremor. Neuroradiol J 32, 395-400 (2019).

20. J. P. Marques, R. Bowtell, Application of a fourier-based method for rapid calculation of field inhomogeneity due to spatial variation of magnetic susceptibility. Concepts in Magnetic Resonance Part B-Magnetic Resonance Engineering 25b, 65-78 (2005).

21. R. Salomir, B. D. De Senneville, C. T. W. Moonen, A fast calculation method for magnetic field inhomogeneity due to an arbitrary distribution of bulk susceptibility. Concepts in Magnetic Resonance Part B-Magnetic Resonance Engineering 19b, 26-34 (2003).

22. K. Shmueli et al., Magnetic susceptibility mapping of brain tissue in vivo using MRI phase data. Magn Reson Med 62, 1510-1522 (2009).

23. L. de Rochefort, R. Brown, M. R. Prince, Y. Wang, Quantitative MR susceptibility mapping using piece-wise constant regularized inversion of the magnetic field. Magn Reson Med 60, 1003-1009 (2008).

24. L. de Rochefort et al., Quantitative susceptibility map reconstruction from MR phase data using bayesian regularization: validation and application to brain imaging. Magn Reson Med 63, 194-206 (2010).

25. S. D. Robinson et al., An illustrated comparison of processing methods for MR phase imaging and QSM: combining array coil signals and phase unwrapping. NMR Biomed 30 (2017).

26. T. Liu et al., Nonlinear formulation of the magnetic field to source relationship for robust quantitative susceptibility mapping. Magn Reson Med 69, 467-476 (2013).

27. B. Wu, W. Li, A. V. Avram, S. M. Gho, C. Liu, Fast and tissue-optimized mapping of magnetic susceptibility and T2* with multi-echo and multi-shot spirals. Neuroimage 59, 297-305 (2012).

28. F. Schweser, S. D. Robinson, L. de Rochefort, W. Li, K. Bredies, An illustrated comparison of processing methods for phase MRI and QSM: removal of background field contributions from sources outside the region of interest. NMR Biomed 30 (2017).

29. F. Schweser, A. Deistung, B. W. Lehr, J. R. Reichenbach, Quantitative imaging of intrinsic magnetic tissue properties using MRI signal phase: an approach to in vivo brain iron metabolism? Neuroimage 54, 2789-2807 (2011).

30. T. Liu et al., A novel background field removal method for MRI using projection onto dipole fields (PDF). NMR Biomed 24, 1129-1136 (2011).

31. B. Wu, W. Li, A. Guidon, C. Liu, Whole brain susceptibility mapping using compressed sensing. Magn Reson Med 67, 137-147 (2012).

32. D. Zhou, T. Liu, P. Spincemaille, Y. Wang, Background field removal by solving the Laplacian boundary value problem. NMR Biomed 27, 312-319 (2014).

33. Y. Kee et al., Quantitative Susceptibility Mapping (QSM) Algorithms: Mathematical Rationale and Computational Implementations. IEEE Trans Biomed Eng 64, 2531-2545 (2017).

34. C. Langkammer et al., Quantitative susceptibility mapping: Report from the 2016 reconstruction challenge. Magn Reson Med 79, 1661-1673 (2018).

35. Q. S. M. C. O. Committee et al., QSM reconstruction challenge 2.0: Design and report of results. Magn Reson Med 86, 1241-1255 (2021).

36. J. Yoon et al., Quantitative susceptibility mapping using deep neural network: QSMnet. Neuroimage 179, 199-206 (2018).

37. S. Bollmann et al., DeepQSM - using deep learning to solve the dipole inversion for quantitative susceptibility mapping. Neuroimage 195, 373-383 (2019).

38. W. Jung, S. Bollmann, J. Lee, Overview of quantitative susceptibility mapping using deep learning: Current status, challenges and opportunities. Nmr in Biomedicine 35 (2022).

39. P. Ravanfar et al., Systematic Review: Quantitative Susceptibility Mapping (QSM) of Brain Iron Profile in Neurodegenerative Diseases. Front Neurosci 15, 618435 (2021).

40. S. Vinayagamani et al., Quantitative Susceptibility Mapping: Technical Considerations and Clinical Applications in Neuroimaging. J Magn Reson Imaging 53, 23-37 (2021).

41. S. Liu et al., Improved MR venography using quantitative susceptibility-weighted imaging. J Magn Reson Imaging 40, 698-708 (2014).

Proc. Intl. Soc. Mag. Reson. Med. 31 (2023)