Susceptibility Weighted Imaging (SWI): What you really need to know

Susceptibility-weighted imaging (SWI) is (mainly) a neuroimaging technique, which uses tissues magnetic susceptibility differences to generate a unique contrast, different from that of proton density (PD), T1, T2, T2* and DWI (fig. 1).

Fig. 1 Susceptibility-weighted Imaging (SWI). Multiple views of SWI minIP with 0.5x0.5x2.5mm voxel size within 2min of scan time. Images courtesy of Bac Nguyen.

Fig. 1 Susceptibility-weighted Imaging (SWI). Multiple views of SWI minIP with 0.5x0.5x2.5mm voxel size within 2min of scan time. Images courtesy of Bac Nguyen.


SWI is a fully flow/velocity compensated, RF spoiled, high-resolution, three-dimensional (3D) gradient recalled echo (GRE) sequence. Parallel imaging is also employed to reduce imaging time. Both the magnitude and phase images are saved. The phase image is high pass (HP) filtered to remove unwanted artifacts (background field inhomogeneities caused by air-tissue interfaces and main magnetic field effects). The magnitude image is then combined with the HP filtered phase image to create an enhanced contrast magnitude image referred to as the susceptibility weighted image (SWI). It is also common to create minimum intensity projections (minIP) over 8 to 10 mm to better visualize vein connectivity. In this way four sets of images are generated (fig. 2):

  • magnitude image

  • HP filtered phase image

  • susceptibility weighted image (SWI)

  • minIP susceptibility weighted image (SWI minIP)

Fig. 2 shows a patient with brain tumour and illustrates the processing steps of SWI sequence. Images courtesy of Bac Nguyen.

Fig. 2 shows a patient with brain tumour and illustrates the processing steps of SWI sequence. Images courtesy of Bac Nguyen.


Typical imaging parameters include TR = 25-50 msec, TE = 20-40 msec, and flip angles = 15-20°. Shorter times and smaller flip angles are used as field strength increases.

SWI can be used better at higher field strengths. First of all, magnetic susceptibility increases accordingly to the square of the magnetic field strength. Moreover, the high signal-to-noise (SNR) ratio available at higher magnetic fields allows higher resolution scans. Finally, SWI can be used better at scanners with higher performance gradients because these systems allow shorter echo times (TE) without a loss of contrast. Shorter TE can reduce the scan time and motion related artifacts.

Different vendors of MRI equipment use different names for SWI technique, and table 1 provides a cross-vendor comparison.


Table 1 provides a cross-vendor comparison for SWI technique.

Table 1 provides a cross-vendor comparison for SWI technique.


Clinical Applications

SWI can be very useful in the following clinical indications:

  • Improved detection of hemorrhage (fig. 3), microbleeding (diffuse axonal injury) and hemorrhagic transformation (stroke).

  • Tumour characterization. Ability to detect tumour vasculature and micro-hemorrhages (fig. 4).

  • Detection of occult vascular disease (cavernomas, angiomas, telangiectasias).

  • Identification and differentiation of iron and other mineral deposition.

  • Helpful in MR diagnosis of neurodegenerative diseases (Alzheimer’s, multiple sclerosis, Parkinson disease etc.).

  • Outside of the CNS, SWI can be also very useful as it can help in hemorrhage detection, and in specific MSK pathologies (i.e. myositis ossificans, GCTTS, calcified tendinitis) (fig. 5).

Fig. 3 provides a direct comparison between the conventional “hemo” techniques and 3D SWI. SWI is 3 to 6 times more sensitive than conventional 2D T2*-w GRE and 2D T2*-w SS-EPI imaging for detection of hemorrhage and microbleeding. Images courtesy of Bac Nguyen.

Fig. 3 provides a direct comparison between the conventional “hemo” techniques and 3D SWI. SWI is 3 to 6 times more sensitive than conventional 2D T2*-w GRE and 2D T2*-w SS-EPI imaging for detection of hemorrhage and microbleeding. Images courtesy of Bac Nguyen.