Quantitative Susceptibility Mapping Using a Multispectral Autoregressive Moving Average Model to Assess Hepatic Iron...

Area of Science:

• Medical Imaging
• Quantitative MRI
• Biomarkers

Background:

• R2*-MRI is used for hepatic iron content (HIC) assessment but is affected by coexisting pathologies like fibrosis and steatosis.
• These pathologies can confound HIC estimations, potentially impacting iron chelation therapy guidance.

Purpose of the Study:

• To evaluate the efficacy of a multispectral autoregressive moving average (ARMA) model combined with quantitative susceptibility mapping (QSM).
• To determine if ARMA-QSM can provide a confounder-free measurement of magnetic susceptibility for HIC prediction.

Main Methods:

• Utilized a phantom study and an in vivo cohort of 48 patients.
• Performed ARMA-QSM modeling on 3D multi-echo gradient echo (GRE) signals to estimate R2*, fat fraction (FF), and susceptibility.
• Calculated R2*-based dry HIC using 2D GRE acquisition and a published calibration curve.

Main Results:

• ARMA R2* and susceptibility values strongly correlated with iron concentrations in phantoms (R² ≥ 0.9).
• In patients, ARMA R2* highly correlated with clinical HIC (R² = 0.97), and susceptibility showed good correlation (R² = 0.82).
• ARMA-QSM simultaneously estimated susceptibility-based wet HIC, R2*-based dry HIC, and FFs from a single acquisition.

Conclusions:

• ARMA-QSM demonstrates feasibility for simultaneously estimating multiple parameters from a single multi-echo GRE scan.
• Both R2*- and susceptibility-based HIC values from ARMA-QSM correlate well with clinical dry HIC.
• ARMA-QSM offers a promising approach for confounder-free assessment of hepatic iron overload.