Magnetic Resonance: Contrast Weighting, Safety, and Image Trade-offs
Weighting is a matter of timing
Tissue contrast in spin echo is set by TR and TE. Short TR + short TE gives T1 weighting (fat bright, fluid dark); long TR + long TE gives T2 weighting (fluid bright); long TR + short TE gives proton density. Flip angle plays the TR-like role in gradient echo. A frequent trap is forgetting that GRE sequences have no 180° refocusing pulse, so they are sensitive to susceptibility and field inhomogeneity (T2* rather than true T2).
Suppression techniques
Fat suppression exploits the short T1 of fat: STIR uses an inversion time (~150 ms at 1.5 T) to null fat signal independent of field homogeneity, whereas chemical-shift fat saturation depends on a well-shimmed field. FLAIR uses a long TI (~2000 ms) to null CSF. Candidates often confuse which one nulls what.
Safety and SAR
Gadolinium is paramagnetic, shortens T1 (and T2), and brightens on T1-weighted images; it is avoided or used with caution in severe renal impairment because of nephrogenic systemic fibrosis. SAR rises with the square of field strength, with flip angle, and with duty cycle, so high-field fast sequences are the usual culprits when scanning limits are hit. Zones I–IV, the missile effect, implant conditionality, and quench procedure (evacuate, do not re-enter for oxygen displacement) are high-yield.
Encoding and trade-offs
Slice selection = RF bandwidth with a gradient; phase encoding = the slow direction; frequency encoding = readout. k-space centre governs contrast, periphery governs resolution. SNR scales with voxel volume, with √NEX, and inversely with √bandwidth - so thinner slices, larger matrices, higher bandwidth, and smaller FOV all cost SNR.
Rule of thumb
Double NEX: SNR ×1.41, scan time ×2. Halve phase-encoding steps with parallel imaging: time halves, SNR falls by roughly √2.