Across-subject ensemble-learning alleviates the need for large samples for fMRI decoding

TL;DR

This paper introduces an ensemble-learning method for fMRI decoding that combines data from multiple subjects, significantly improving accuracy and reducing the need for large individual datasets, especially with voxel-based classifiers.

Contribution

It presents a novel across-subject ensemble-learning approach that outperforms traditional methods, particularly with limited data, and identifies effective classifier choices like Multi-layer Perceptron.

Findings

Ensemble approach improves decoding accuracy by up to 20%.

Method is especially effective with limited per-subject data.

Multi-layer Perceptron is a strong default ensemble classifier.

Abstract

Decoding cognitive states from functional magnetic resonance imaging is central to understanding the functional organization of the brain. Within-subject decoding avoids between-subject correspondence problems but requires large sample sizes to make accurate predictions; obtaining such large sample sizes is both challenging and expensive. Here, we investigate an ensemble approach to decoding that combines the classifiers trained on data from other subjects to decode cognitive states in a new subject. We compare it with the conventional decoding approach on five different datasets and cognitive tasks. We find that it outperforms the conventional approach by up to 20% in accuracy, especially for datasets with limited per-subject data. The ensemble approach is particularly advantageous when the classifier is trained in voxel space. Furthermore, a Multi-layer Perceptron turns out to be a good default choice as an ensemble method. These results show that the pre-training strategy reduces the need for large per-subject data.