Accelerated 3D qCEST of the Spine in a Porcine Model Using MR Multitasking at 3T

Area of Science:

• Biomedical Imaging
• Magnetic Resonance Imaging
• Pain Assessment

Background:

• Lower back pain is a significant health issue.
• Quantitative chemical exchange saturation transfer (qCEST) imaging offers potential for non-invasive assessment.
• Developing accurate predictive models for pain is crucial.

Purpose of the Study:

• To evaluate multitasking qCEST for assessing lower back pain in a porcine model.
• To compare exchange rate maps from multitasking qCEST with conventional qCEST.
• To develop a predictive model for pain scores using qCEST-derived biomarkers.

Main Methods:

• A porcine model with intervertebral disc injury was used.
• Conventional and multitasking steady-state (SS) qCEST imaging were performed.
• A permuted random forest (PRF) model was trained on magnetization transfer ratio (MTR) and exchange rate (k_sw) values to predict pain scores.

Main Results:

• Multitasking qCEST showed strong agreement (r=0.82) with conventional qCEST for exchange rate maps.
• Injured discs showed significantly higher k_sw values compared to healthy discs.
• The PRF model achieved 80% accuracy in predicting pain scores, outperforming Modic changes correlation (r=0.45).

Conclusions:

• 3D steady-state qCEST is a viable technique for whole-spine imaging within a clinically relevant timeframe.
• qCEST-derived biomarkers (MTR and k_sw) can accurately predict pain scores.
• This approach holds promise for objective lower back pain assessment.