| Names |
CHESS, SPAIR/SPIR |
STIR |
Dixon, mDIXON, IDEAL |
| Principle |
Frequency-selective RF pulse hits fat peak to dephase fat (CHESS); or null the fat (SPAIR/SPIR) |
Non-selective inversion pulse → wait until fat signal is null (TI time) then RF → No transverse fat signal |
Phase separation of water and fat using multiple echo times then recombine mathematically |
| Pros |
High SNR, Preserves T1, T2 contrast |
Most robust to Off-resonance, can be used in large FOV and near metal |
Uniform FS over big FOV, multiple sequences from one acquisition, good SNR |
| Cons |
Very sensitive to off-resonance; Patchy fat suppression due to unavoidable B0 inhomogeneity in large FOV |
lower SNR → typically poorer spatial resolution due to thicker slices and longer TE needed to compensate low SNR; Kills all short T1 signals including Gd enhancement (cannot be used in post-contrast imaging). |
Sensitive to off-resonance → water-fat swaps, ugly artefact near strong inhomogeneity |
| Applications |
Workhorse of joint imaging (better cartilage signal than STIR); Good for pre/post contrast study (unlike STIR) |
Robust imaging in off-resonance zones (hand, foot, spine, whole body, metal) |
Pretty and fast large FOV imaging when B0 is ok (whole body, spine); if fail → step back to STIR |
