New approach to phase correction in multi-echo T₂ relaxometry.

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Biomedical Engineering
• Medical Physics

Background:

• Transverse relaxation time (T2) estimation from multi-echo spin-echo MRI typically uses magnitude data with a least squares (LS) approach.
• Rice-distributed noise in magnitude images introduces bias in LS-based T2 estimates.
• Estimating real-valued, Gaussian-distributed data from complex MRI data can mitigate this bias.

Purpose of the Study:

• To propose and evaluate two novel phase correction algorithms for generating real-valued MRI data.
• To improve the accuracy of T2 estimation by reducing bias associated with magnitude-based methods.
• To demonstrate the practical feasibility of these algorithms in MRI data analysis.

Main Methods:

• Development of the Weighted Linear Phase Estimation (WELPE) algorithm for phase correction, supporting multi-coil input.
• Implementation of a maximum likelihood estimator for true decaying signal magnitude, efficient for linear phase variations.
• Validation through Monte Carlo simulations comparing estimation accuracy and bias reduction.

Main Results:

• Both proposed algorithms successfully generate real-valued data suitable for LS-based T2 estimation.
• Phase-corrected data significantly reduces estimation bias, especially in multi-component T2 relaxometry.
• The WELPE algorithm demonstrated practical feasibility on an in vivo human brain MRI dataset.

Conclusions:

• The developed phase correction algorithms offer a significant improvement over traditional magnitude-based T2 estimation methods.
• These methods effectively reduce bias in T2 estimates, leading to more accurate relaxometry.
• WELPE is a robust and feasible tool for enhancing T2 quantification in clinical MRI applications.