Thymus chemical shift MRI

Normal/hyperplastic thymus has microscopic fat mixed with water, so it loses signal on opposed-phase GRE. Most thymic tumours (e.g. thymoma) have far less microscopic fat, so little/no drop.

Acquire

Dual-echo T1 GRE: in-phase (IP) + opposed-phase (OP) on the same slice
Use a matching ROI (copy/paste if your PACS allows)
- Include solid thymic tissue
- Avoid: vessels, cyst/necrosis, obvious macroscopic fat, edge artefact

1) Signal Intensity Index (SII) (most practical)

SII (%) = \frac{{SI}_{I P} - {SI}_{O P}}{{SI}_{I P}} \times 100

Interpretation (literature-based, not a universal law):

Higher % drop strongly favours thymic hyperplasia/rebound.
In a large adult study, an SII cutoff around 8.9% separated hyperplasia from tumours with very high accuracy.

Quick example:
${SI}_{I P} = 130, {SI}_{O P} = 77$

SII (%) = \frac{130 - 77}{130} \times 100 = 40.8 %

Interpretation: pretty convincing for thymic tissue.

2) Chemical Shift Ratio (CSR) (muscle-normalised)

CSR = \frac{(\frac{{tSI}_{O P}}{{mSI}_{O P}})}{(\frac{{tSI}_{I P}}{{mSI}_{I P}})}

t = thymus (or lesion), m = paraspinal muscle

Interpretation:

Lower CSR suggests more signal drop (more microscopic fat) and supports hyperplasia.
Caveat: CSR can be influenced by age/body habitus because muscle fat content varies, so SII is often more robust.

Warning

Do not let one number bully you. If it is nodular, growing, very FDG-avid, or has new extra-thymic disease, call it indeterminate even if there is some drop. (Some mimics like thymolipoma can also suppress on OP.)