Few-shot Decoding of Brain Activation Maps

TL;DR

This paper explores the application of few-shot learning techniques to neuroimaging data, demonstrating their ability to decode brain signals efficiently with limited examples, which could benefit clinical and cognitive neuroscience.

Contribution

It introduces a neuroimaging benchmark dataset for few-shot learning and compares multiple paradigms, including meta-learning, for decoding brain activation maps.

Findings

Few-shot methods effectively decode brain signals with limited data

Meta-learning approaches show promising results in neuroimaging tasks

Potential applications include biomarker identification and understanding brain representation generalization

Abstract

Few-shot learning addresses problems for which a limited number of training examples are available. So far, the field has been mostly driven by applications in computer vision. Here, we are interested in adapting recently introduced few-shot methods to solve problems dealing with neuroimaging data, a promising application field. To this end, we create a neuroimaging benchmark dataset for few-shot learning and compare multiple learning paradigms, including meta-learning, as well as various backbone networks. Our experiments show that few-shot methods are able to efficiently decode brain signals using few examples, which paves the way for a number of applications in clinical and cognitive neuroscience, such as identifying biomarkers from brain scans or understanding the generalization of brain representations across a wide range of cognitive tasks.