Explainable artificial intelligence based analysis for interpreting infant fNIRS data in developmental cognitive neuroscience

Javier Andreu-Perez,Hani Hagras,Mehrin Kiani,Maria Laura Filippetti,Lauren L Emberson,Silvia Rigato

doi:10.1038/s42003-021-02534-y

Javier Andreu-Perez, Hani Hagras + Show 4 more

Open Access

https://doi.org/10.1038/s42003-021-02534-y

Copy DOI

Journal: Communications Biology	Publication Date: Sep 15, 2021
Citations: 14	License type: open-access

Affiliation: University of Essex, Princeton University

Abstract

In the last decades, non-invasive and portable neuroimaging techniques, such as functional near infrared spectroscopy (fNIRS), have allowed researchers to study the mechanisms underlying the functional cognitive development of the human brain, thus furthering the potential of Developmental Cognitive Neuroscience (DCN). However, the traditional paradigms used for the analysis of infant fNIRS data are still quite limited. Here, we introduce a multivariate pattern analysis for fNIRS data, xMVPA, that is powered by eXplainable Artificial Intelligence (XAI). The proposed approach is exemplified in a study that investigates visual and auditory processing in six-month-old infants. xMVPA not only identified patterns of cortical interactions, which confirmed the existent literature; in the form of conceptual linguistic representations, it also provided evidence for brain networks engaged in the processing of visual and auditory stimuli that were previously overlooked by other methods, while demonstrating similar statistical performance.

Highlights

In the last decades, non-invasive and portable neuroimaging techniques, such as functional near infrared spectroscopy, have allowed researchers to study the mechanisms underlying the functional cognitive development of the human brain, furthering the potential of Developmental Cognitive Neuroscience (DCN)
We provide a novel, explainable method for analysing and interpreting infant functional near infrared spectroscopy (fNIRS) data
The proposed xMVPA is an multivariate pattern analysis (MVPA) based on XAI that provides functional patterns characterised by conceptual labels delineating contributions between brain regions for information processing

Summary

Introduction

Non-invasive and portable neuroimaging techniques, such as functional near infrared spectroscopy (fNIRS), have allowed researchers to study the mechanisms underlying the functional cognitive development of the human brain, furthering the potential of Developmental Cognitive Neuroscience (DCN). In this article, we introduce the use of an eXplainable Artificial Intelligence (XAI) inference mechanism for infant fNIRS data that delineates the interaction patterns between brain regions activated in response to perceptual stimuli.

Results

Conclusion