Assessment of Precision and Accuracy of Brain White Matter Microstructure using Combined Diffusion MRI and Relaxometry

Santiago Coelho,Ying Liao,Filip Szczepankiewicz,Jelle Veraart,Sohae Chung,Yvonne W. Lui,Dmitry S. Novikov,Els Fieremans

2024-02-27

Abstract:Joint modeling of diffusion and relaxation has seen growing interest due to its potential to provide complementary information about tissue microstructure. For brain white matter, we designed an optimal diffusion-relaxometry MRI protocol that samples multiple b-values, B-tensor shapes, and echo times (TE). This variable-TE protocol (27 min) has as subsets a fixed-TE protocol (15 min) and a 2-shell dMRI protocol (7 min), both characterizing diffusion only. We assessed the sensitivity, specificity and reproducibility of these protocols with synthetic experiments and in six healthy volunteers. Compared with the fixed-TE protocol, the variable-TE protocol enables estimation of free water fractions while also capturing compartmental $T_2$ relaxation times. Jointly measuring diffusion and relaxation offers increased sensitivity and specificity to microstructure parameters in brain white matter with voxelwise coefficients of variation below 10%.

Medical Physics,Biological Physics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to improve the precision and accuracy of white matter microstructure imaging. Specifically, the author designed an optimized diffusion - relaxation MRI protocol, aiming to provide more abundant tissue microstructure information by combining multiple b - values, B - tensor shapes and echo times (TE). This multivariable TE protocol (27 minutes) includes a fixed - TE protocol (15 minutes) and a 2 - shell dMRI protocol (7 minutes), with the latter only characterizing diffusion properties. The study evaluated the sensitivity, specificity and reproducibility of these protocols in synthetic experiments and six healthy volunteers. Compared with the fixed - TE protocol, the multivariable TE protocol can estimate the free - water fraction and capture the T2 relaxation time of each compartment simultaneously, thereby improving the sensitivity and specificity of white matter microstructure parameters, and the voxel coefficient variation is less than 10%.

### Core problems of the paper
1. **Improve the sensitivity and specificity of microstructure parameter estimation**: Improve the sensitivity and specificity of white matter microstructure parameters by jointly measuring diffusion and relaxation.
2. **Optimize the MRI protocol**: Design a comprehensive diffusion - relaxation MRI protocol that can be completed within 30 minutes, combining multiple TEs, high b - values and multi - dimensional diffusion encoding.
3. **Evaluate protocol performance**: Use the sensitivity - specificity matrix (SSM) and coefficient of variation (COV) to evaluate the performance of different protocols.

### Key technical points
- **Multivariable TE protocol**: Combine multiple b - values, B - tensor shapes and echo times (TE) to obtain more abundant tissue microstructure information.
- **Fixed - TE protocol**: A multi - dimensional diffusion protocol using only fixed TE.
- **2 - shell dMRI protocol**: A commonly used double - shell diffusion - weighted imaging protocol.
- **Machine - learning parameter estimation**: Use machine - learning - based algorithms to estimate standard model (SM) parameters and reduce estimation errors.
- **Sensitivity - specificity matrix (SSM)**: Quantify the sensitivity and specificity of parameter estimation.
- **Coefficient of variation (COV)**: Evaluate the reproducibility and stability of the protocol.

### Main findings
- **Advantages of the multivariable TE protocol**: Compared with the fixed - TE protocol, the multivariable TE protocol can estimate the free - water fraction more accurately and capture the T2 relaxation time of each compartment, thereby significantly improving the sensitivity and specificity of parameter estimation.
- **Application of machine learning**: The machine - learning - based parameter estimation method significantly reduces the estimation error and improves the overall estimation accuracy.
- **Clinical feasibility**: A clinically feasible 8 - minute protocol is proposed. By combining the 2 - shell dMRI protocol and b0 measurements of multiple TEs, the feasibility of accurately estimating the free - water fraction in clinical applications is evaluated.

### Formula presentation
- **Signal attenuation formula**:
\[
S(B) = e^{-\sum_{ij} B_{ij} D_{ij}} = e^{-b (D_{\parallel} - D_{\perp}) (\beta (\xi^2 - \frac{1}{3}) + \frac{1}{3}) - b D_{\perp}}
\]
where \(\xi = \hat{g} \cdot \hat{n}\).

- **Kernel function**:
\[
K(b, \beta, \xi, TE) = f e^{-b D_a (\beta (\xi^2 - \frac{1}{3}) + \frac{1}{3}) - \frac{TE}{T_a^2}} + f_e e^{-b (D_{\parallel}^e - D_{\perp}^e) (\beta (\xi^2 - \frac{1}{3}) + \frac{1}{3}) - b D_{\perp}^e - \frac{TE}{T_e^2}} + f_w e^{-b D_w - \frac{TE}{T_w^2}}
\]

- **Elements of the sensitivity - specificity matrix**:
\[
S_{ij} = \left\langle \frac{\partial \hat{\theta}_j}{\partial \theta_i} \right\rangle