Cargando…

Functional connectome fingerprinting using shallow feedforward neural networks

Although individual subjects can be identified with high accuracy using correlation matrices computed from resting-state functional MRI (rsfMRI) data, the performance significantly degrades as the scan duration is decreased. Recurrent neural networks can achieve high accuracy with short-duration (72...

Descripción completa

Detalles Bibliográficos
Autores principales: Sarar, Gokce, Rao, Bhaskar, Liu, Thomas
Formato: Online Artículo Texto
Lenguaje:English
Publicado: National Academy of Sciences 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8053937/
https://www.ncbi.nlm.nih.gov/pubmed/33827923
http://dx.doi.org/10.1073/pnas.2021852118
_version_ 1783680218700972032
author Sarar, Gokce
Rao, Bhaskar
Liu, Thomas
author_facet Sarar, Gokce
Rao, Bhaskar
Liu, Thomas
author_sort Sarar, Gokce
collection PubMed
description Although individual subjects can be identified with high accuracy using correlation matrices computed from resting-state functional MRI (rsfMRI) data, the performance significantly degrades as the scan duration is decreased. Recurrent neural networks can achieve high accuracy with short-duration (72 s) data segments but are designed to use temporal features not present in the correlation matrices. Here we show that shallow feedforward neural networks that rely solely on the information in rsfMRI correlation matrices can achieve state-of-the-art identification accuracies ([Formula: see text]) with data segments as short as 20 s and across a range of input data size combinations when the total number of data points (number of regions [Formula: see text] number of time points) is on the order of [Formula: see text].
format Online
Article
Text
id pubmed-8053937
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher National Academy of Sciences
record_format MEDLINE/PubMed
spelling pubmed-80539372021-05-04 Functional connectome fingerprinting using shallow feedforward neural networks Sarar, Gokce Rao, Bhaskar Liu, Thomas Proc Natl Acad Sci U S A Biological Sciences Although individual subjects can be identified with high accuracy using correlation matrices computed from resting-state functional MRI (rsfMRI) data, the performance significantly degrades as the scan duration is decreased. Recurrent neural networks can achieve high accuracy with short-duration (72 s) data segments but are designed to use temporal features not present in the correlation matrices. Here we show that shallow feedforward neural networks that rely solely on the information in rsfMRI correlation matrices can achieve state-of-the-art identification accuracies ([Formula: see text]) with data segments as short as 20 s and across a range of input data size combinations when the total number of data points (number of regions [Formula: see text] number of time points) is on the order of [Formula: see text]. National Academy of Sciences 2021-04-13 2021-04-07 /pmc/articles/PMC8053937/ /pubmed/33827923 http://dx.doi.org/10.1073/pnas.2021852118 Text en Copyright © 2021 the Author(s). Published by PNAS. https://creativecommons.org/licenses/by-nc-nd/4.0/This open access article is distributed under Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND) (https://creativecommons.org/licenses/by-nc-nd/4.0/) .
spellingShingle Biological Sciences
Sarar, Gokce
Rao, Bhaskar
Liu, Thomas
Functional connectome fingerprinting using shallow feedforward neural networks
title Functional connectome fingerprinting using shallow feedforward neural networks
title_full Functional connectome fingerprinting using shallow feedforward neural networks
title_fullStr Functional connectome fingerprinting using shallow feedforward neural networks
title_full_unstemmed Functional connectome fingerprinting using shallow feedforward neural networks
title_short Functional connectome fingerprinting using shallow feedforward neural networks
title_sort functional connectome fingerprinting using shallow feedforward neural networks
topic Biological Sciences
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8053937/
https://www.ncbi.nlm.nih.gov/pubmed/33827923
http://dx.doi.org/10.1073/pnas.2021852118
work_keys_str_mv AT sarargokce functionalconnectomefingerprintingusingshallowfeedforwardneuralnetworks
AT raobhaskar functionalconnectomefingerprintingusingshallowfeedforwardneuralnetworks
AT liuthomas functionalconnectomefingerprintingusingshallowfeedforwardneuralnetworks