Spin echoes: full numerical solution and breakdown of approximative solutions.

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Biophysics
• Computational Fluid Dynamics

Background:

• Understanding spin echo signal behavior in microvasculature is crucial for MRI applications.
• Existing approximations to the Bloch-Torrey equation simplify complex fluid dynamics and diffusion processes.
• Krogh's capillary and random distribution models represent distinct microvascular architectures.

Purpose of the Study:

• To numerically solve the Bloch-Torrey equation for spin echo signals in Krogh's capillary and random distribution vessel models.
• To compare the exact numerical solutions with established approximations: Gaussian local phase, Gaussian phase, and strong-collision.
• To elucidate the differences between approximation methods and identify regimes where they diverge from exact solutions.

Main Methods:

• Numerical solution of the Bloch-Torrey equation.
• Modeling of spin echo signals in Krogh's capillary and random distribution vessel models.
• Comparative analysis against Gaussian local phase, Gaussian phase, and strong-collision approximations.

Main Results:

• The full numerical solution exhibits oscillations in the intermediate diffusion regime, not present in approximate solutions.
• Approximations become exact in the limit of large diffusion coefficients, yielding a linear-exponential decay.
• An analytically solvable discrete two-level model explains the exact numerical solution's features.
• A correspondence exists between diffusion regimes and damped harmonic oscillator cases.

Conclusions:

• Numerical solutions provide a more accurate representation of spin echo signals, especially in intermediate diffusion regimes.
• The study clarifies the limitations of common approximations for Bloch-Torrey equation analysis.
• Findings contribute to a deeper understanding of diffusion and flow effects in MRI signal generation within microvascular networks.