Stopping criteria for boosting automatic experimental design using real-time fMRI with Bayesian optimization

TL;DR

This paper introduces and empirically evaluates two stopping criteria for Bayesian optimization in real-time fMRI experiments, aiming to improve efficiency and reliability in brain pattern elicitation.

Contribution

It proposes novel stopping criteria tailored for real-time fMRI Bayesian optimization, addressing a critical gap in experimental design.

Findings

The proposed criteria effectively determine when to stop the experiment.

They reduce scanning time without compromising results.

The criteria improve the reliability of brain pattern targeting.

Abstract

Bayesian optimization has been proposed as a practical and efficient tool through which to tune parameters in many difficult settings. Recently, such techniques have been combined with real-time fMRI to propose a novel framework which turns on its head the conventional functional neuroimaging approach. This closed-loop method automatically designs the optimal experiment to evoke a desired target brain pattern. One of the challenges associated with extending such methods to real-time brain imaging is the need for adequate stopping criteria, an aspect of Bayesian optimization which has received limited attention. In light of high scanning costs and limited attentional capacities of subjects an accurate and reliable stopping criteria is essential. In order to address this issue we propose and empirically study the performance of two stopping criteria.