SCREENER: A general framework for task-specific experiment design in quantitative MRI

TL;DR

The paper tackles the problem that quantitative MRI protocol design is not tuned to downstream clinical tasks. It proposes SCREENER, a general framework that combines a task-specific objective module with a deep reinforcement learning optimizer to customize acquisition protocols for a given task. Using the IVIM diffusion model with $S(b) = S_0 \exp(-TE/T_2)\left( f \exp(-b D^*) + (1-f) \exp(-b D) \right)$ and task-driven simulations, the authors show that task-specific protocols yield higher classification accuracy and robustness to SNR, including zero-shot generalization to unseen SNRs. These findings suggest SCREENER can increase the clinical uptake of qMRI by efficiently tailoring scans to clinical goals.

Abstract

Quantitative magnetic resonance imaging (qMRI) is increasingly investigated for use in a variety of clinical tasks from diagnosis, through staging, to treatment monitoring. However, experiment design in qMRI, the identification of the optimal acquisition protocols, has been focused on obtaining the most precise parameter estimations, with no regard for the specific requirements of downstream tasks. Here we propose SCREENER: A general framework for task-specific experiment design in quantitative MRI. SCREENER incorporates a task-specific objective and seeks the optimal protocol with a deep-reinforcement-learning (DRL) based optimization strategy. To illustrate this framework, we employ a task of classifying the inflammation status of bone marrow using diffusion MRI data with intravoxel incoherent motion (IVIM) modelling. Results demonstrate SCREENER outperforms previous ad hoc and optimized protocols under clinical signal-to-noise ratio (SNR) conditions, achieving significant improvement, both in binary classification tasks, e.g. from 67% to 89%, and in a multi-class classification task, from 46% to 59%. Additionally, we show this improvement is robust to the SNR. Lastly, we demonstrate the advantage of DRL-based optimization strategy, enabling zero-shot discovery of near-optimal protocols for a range of SNRs not used in training. In conclusion, SCREENER has the potential to enable wider uptake of qMRI in the clinic.

Figures (2)

• Figure 1: Our proposed task-specific objective function, SCREENER versus previous CRLB function. Components unique to SCREENER are colored in dark cyan.
• Figure 2: The accuracies of the multi-class classification task with the Ad hoc method, CRLB method, and SCREENER, SCREENER (zero-shot) under different SNR.