The Dependence of Parallel Imaging with Linear Predictability on the Undersampling Direction

TL;DR

This paper investigates how the direction of undersampling affects the quality of parallel imaging reconstruction using linear predictability, providing theoretical conditions and a new metric for optimal undersampling directions.

Contribution

It introduces a sufficient condition linking undersampling direction and coil arrangement for successful reconstruction, supported by theoretical justification and real data examples.

Findings

Undersampling direction significantly impacts reconstruction quality.

A new metric predicts optimal undersampling directions.

Theoretical conditions guide better sampling strategies.

Abstract

Parallel imaging with linear predictability takes advantage of information present in multiple receive coils to accurately reconstruct the image with fewer samples. Commonly used algorithms based on linear predictability include GRAPPA and SPIRiT. We present a sufficient condition for reconstruction based on the direction of undersampling and the arrangement of the sensing coils. This condition is justified theoretically and examples are shown using real data. We also propose a metric based on the fully-sampled auto-calibration region which can show which direction(s) of undersampling will allow for a good quality image reconstruction.