Cargando…

Adaptive noise depression for functional brain network estimation

Autism spectrum disorder (ASD) is one common psychiatric illness that manifests in neurological and developmental disorders, which can last throughout a person's life and cause challenges in social interaction, communication, and behavior. Since the standard ASD diagnosis is highly based on the...

Descripción completa

Detalles Bibliográficos
Autores principales: Ma, Di, Peng, Liling, Gao, Xin
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9871598/
https://www.ncbi.nlm.nih.gov/pubmed/36704736
http://dx.doi.org/10.3389/fpsyt.2022.1100266
_version_ 1784877212082634752
author Ma, Di
Peng, Liling
Gao, Xin
author_facet Ma, Di
Peng, Liling
Gao, Xin
author_sort Ma, Di
collection PubMed
description Autism spectrum disorder (ASD) is one common psychiatric illness that manifests in neurological and developmental disorders, which can last throughout a person's life and cause challenges in social interaction, communication, and behavior. Since the standard ASD diagnosis is highly based on the symptoms of the disease, it is difficult to make an early diagnosis to take the best cure opportunity. Compared to the standard methods, functional brain network (FBN) could reveal the statistical dependence among neural architectures in brains and provide potential biomarkers for the early neuro-disease diagnosis and treatment of some neurological disorders. However, there are few FBN estimation methods that take into account the noise during the data acquiring process, resulting in poor quality of FBN and thus poor diagnosis results. To address such issues, we provide a brand-new approach for estimating FBNs under a noise modeling framework. In particular, we introduce a noise term to model the representation errors and impose a regularizer to incorporate noise prior into FBNs estimation. More importantly, the proposed method can be formulated as conducting traditional FBN estimation based on transformed fMRI data, which means the traditional methods can be elegantly modified to support noise modeling. That is, we provide a plug-and-play noise module capable of being embedded into different methods and adjusted according to different noise priors. In the end, we conduct abundant experiments to identify ASD from normal controls (NCs) based on the constructed FBNs to illustrate the effectiveness and flexibility of the proposed method. Consequently, we achieved up to 13.04% classification accuracy improvement compared with the baseline methods.
format Online
Article
Text
id pubmed-9871598
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-98715982023-01-25 Adaptive noise depression for functional brain network estimation Ma, Di Peng, Liling Gao, Xin Front Psychiatry Psychiatry Autism spectrum disorder (ASD) is one common psychiatric illness that manifests in neurological and developmental disorders, which can last throughout a person's life and cause challenges in social interaction, communication, and behavior. Since the standard ASD diagnosis is highly based on the symptoms of the disease, it is difficult to make an early diagnosis to take the best cure opportunity. Compared to the standard methods, functional brain network (FBN) could reveal the statistical dependence among neural architectures in brains and provide potential biomarkers for the early neuro-disease diagnosis and treatment of some neurological disorders. However, there are few FBN estimation methods that take into account the noise during the data acquiring process, resulting in poor quality of FBN and thus poor diagnosis results. To address such issues, we provide a brand-new approach for estimating FBNs under a noise modeling framework. In particular, we introduce a noise term to model the representation errors and impose a regularizer to incorporate noise prior into FBNs estimation. More importantly, the proposed method can be formulated as conducting traditional FBN estimation based on transformed fMRI data, which means the traditional methods can be elegantly modified to support noise modeling. That is, we provide a plug-and-play noise module capable of being embedded into different methods and adjusted according to different noise priors. In the end, we conduct abundant experiments to identify ASD from normal controls (NCs) based on the constructed FBNs to illustrate the effectiveness and flexibility of the proposed method. Consequently, we achieved up to 13.04% classification accuracy improvement compared with the baseline methods. Frontiers Media S.A. 2023-01-10 /pmc/articles/PMC9871598/ /pubmed/36704736 http://dx.doi.org/10.3389/fpsyt.2022.1100266 Text en Copyright © 2023 Ma, Peng and Gao. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Psychiatry
Ma, Di
Peng, Liling
Gao, Xin
Adaptive noise depression for functional brain network estimation
title Adaptive noise depression for functional brain network estimation
title_full Adaptive noise depression for functional brain network estimation
title_fullStr Adaptive noise depression for functional brain network estimation
title_full_unstemmed Adaptive noise depression for functional brain network estimation
title_short Adaptive noise depression for functional brain network estimation
title_sort adaptive noise depression for functional brain network estimation
topic Psychiatry
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9871598/
https://www.ncbi.nlm.nih.gov/pubmed/36704736
http://dx.doi.org/10.3389/fpsyt.2022.1100266
work_keys_str_mv AT madi adaptivenoisedepressionforfunctionalbrainnetworkestimation
AT pengliling adaptivenoisedepressionforfunctionalbrainnetworkestimation
AT gaoxin adaptivenoisedepressionforfunctionalbrainnetworkestimation