A Technical Primer on the Physical Modeling of Diffusion-Encoded Magnetic Resonance Experiments: A Random Process Perspective

TL;DR

This paper provides a detailed overview of diffusion-encoded magnetic resonance experiments from a random process perspective, offering alternative derivations that may enhance understanding for technically skilled readers.

Contribution

It introduces a novel random process perspective on diffusion-encoded MR, complementing existing physics-based descriptions and offering new insights.

Findings

Standard results derived from a random process perspective

Alternative derivations may aid understanding for technical readers

Provides a bridge between signal processing and diffusion MR

Abstract

Diffusion-encoded magnetic resonance (MR) experiments can provide important insights into the microstructural characteristics of a variety of biological tissues and other fluid- or gas-filled porous media. The physics of diffusion encoding has been studied extensively over the span of many decades, and many excellent descriptions can be found in the literature -- see, e.g., Refs. [1-5]. However, many of these descriptions spend relatively little time focusing on random process descriptions of the diffusion process, instead relying on different abstractions. In this primer, we describe diffusion-encoded MR experiments from a random process perspective. While the results we derive from this perspective are quite standard (and match the results obtained with other arguments), we expect that the alternative derivations may be insightful for some readers. This primer is intended for technical readers who have a graduate-level understanding of random processes. Readers are also expected to already have good familiarity with the basics of MR, and we anticipate that a signal processing perspective on MR [6] will be especially complementary to the random process perspectives presented herein.