Optimize Flip Angle Schedules In MR Fingerprinting Using Reinforcement Learning

TL;DR

This paper introduces a reinforcement learning framework to optimize flip angle schedules in Magnetic Resonance Fingerprinting, resulting in improved fingerprint separability and potential acquisition speed-up.

Contribution

The work presents a novel RL-based method for designing flip angle schedules in MRF, demonstrating non-periodic patterns that enhance performance and reduce scan time.

Findings

RL-optimized schedules improve fingerprint separability

Non-periodic flip angle patterns are effective

Potential to reduce repetition time and accelerate scans

Abstract

Magnetic Resonance Fingerprinting (MRF) leverages transient-state signal dynamics generated by the tunable acquisition parameters, making the design of an optimal, robust sequence a complex, high-dimensional sequential decision problem, such as optimizing one of the key parameters, flip angle. Reinforcement learning (RL) offers a promising approach to automate parameter selection, to optimize pulse sequences that maximize the distinguishability of fingerprints across the parameter space. In this work, we introduce an RL framework for optimizing the flip-angle schedule in MRF and demonstrate a learned schedule exhibiting non-periodic patterns that enhances fingerprint separability. Additionally, an interesting observation is that the RL-optimized schedule may enable a reduction in the number of repetition time, potentially accelerate MRF acquisitions.