A Unified Maximum Likelihood Framework for Simultaneous Motion and $T_{1}$ Estimation in Quantitative MR $T_{1}$...

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Quantitative Imaging
• Biomedical Engineering

Background:

• Quantitative MR T1 mapping estimates tissue spin-lattice relaxation time (T1) from T1-weighted images.
• Accurate spatial alignment of these images is critical for precise T1 estimation.
• Conventional two-step approaches, registering images before T1 estimation, introduce bias.

Purpose of the Study:

• To develop a unified framework for simultaneous estimation of motion parameters and T1 maps.
• To overcome the bias introduced by conventional two-step registration and T1 estimation methods.
• To improve the accuracy of T1 mapping in the presence of motion.

Main Methods:

• A unified Maximum Likelihood (ML) estimation framework was proposed.
• Motion parameters and the T1 map are estimated simultaneously.
• The framework jointly incorporates relaxation models, motion models, and data statistics.

Main Results:

• The unified ML framework demonstrated superior accuracy in motion and T1 parameter estimation compared to conventional and state-of-the-art methods.
• Reduced mean-square error in T1 map estimation was observed.
• The method was validated through Monte Carlo simulations and real in vivo human brain data.

Conclusions:

• The proposed unified ML approach provides more accurate T1 maps by jointly estimating motion and T1 parameters.
• This method offers a significant improvement over traditional two-step approaches in quantitative MR T1 mapping.
• The framework is applicable to real-life scenarios, including in vivo human brain imaging.