A theoretical framework for retrospective correction to the arterial input function in quantitative myocardial...

Area of Science:

• Medical Imaging
• Biophysics

Background:

• Quantitative cardiac perfusion MRI relies on accurate arterial input function (AIF) estimation.
• Existing methods for AIF correction may have limitations in accuracy and flexibility.

Purpose of the Study:

• To develop and evaluate a flexible, Bloch-equation based framework for retrospective T2* correction to the arterial input function (AIF) in cardiac perfusion MRI.
• To improve the accuracy of gadolinium concentration ([Gd]) and myocardial blood flow (MBF) quantification.

Main Methods:

• The framework calculates initial [Gd] from T1 measurements, estimates T2* using literature values for native T2 and static magnetic field variation (ΔB0), and then refines [Gd] after T2* correction.
• The method was applied to phantom and in vivo dual-imaging perfusion data from three patient groups, using different pulse sequences and parameters.
• Sensitivity analysis was performed for T2* correction based on variations in native T2 and ΔB0.

Main Results:

• Phantom studies showed a significant reduction in normalized root-means-square-error (NRMSE) for [Gd] with the proposed correction (1.3%) compared to uncorrected data (5.1%).
• In vivo data demonstrated significant improvements in peak AIF and resting MBF values across all patient groups after T2* correction (P < .001).
• The correction method showed less than 5% variation with ±20% changes in native T2 and ΔB0.

Conclusions:

• The developed theoretical framework enables effective retrospective T2* correction for AIF estimation in dual-imaging cardiac perfusion MRI.
• This approach enhances the reliability of quantitative perfusion parameters, such as MBF.
• The method is flexible and robust across different imaging parameters and patient data.