Subtyping cognitive profiles in Autism Spectrum Disorder using a Functional Random Forest algorithm

E. Feczko, N. M. Balba, Oscar Miranda Dominguez, M. Cordova, Sarah Karalunas, L. Irwin, D. V. Demeter, A. P. Hill, B. H. Langhorst, J. Grieser Painter, Jan Van Santen, Eric Fombonne, Joel Nigg, Damien Fair

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

DSM-5 Autism Spectrum Disorder (ASD) comprises a set of neurodevelopmental disorders characterized by deficits in social communication and interaction and repetitive behaviors or restricted interests, and may both affect and be affected by multiple cognitive mechanisms. This study attempts to identify and characterize cognitive subtypes within the ASD population using our Functional Random Forest (FRF) machine learning classification model. This model trained a traditional random forest model on measures from seven tasks that reflect multiple levels of information processing. 47 ASD diagnosed and 58 typically developing (TD) children between the ages of 9 and 13 participated in this study. Our RF model was 72.7% accurate, with 80.7% specificity and 63.1% sensitivity. Using the random forest model, the FRF then measures the proximity of each subject to every other subject, generating a distance matrix between participants. This matrix is then used in a community detection algorithm to identify subgroups within the ASD and TD groups, and revealed 3 ASD and 4 TD putative subgroups with unique behavioral profiles. We then examined differences in functional brain systems between diagnostic groups and putative subgroups using resting-state functional connectivity magnetic resonance imaging (rsfcMRI). Chi-square tests revealed a significantly greater number of between group differences (p <.05) within the cingulo-opercular, visual, and default systems as well as differences in inter-system connections in the somato-motor, dorsal attention, and subcortical systems. Many of these differences were primarily driven by specific subgroups suggesting that our method could potentially parse the variation in brain mechanisms affected by ASD.

Original languageEnglish (US)
Pages (from-to)674-688
Number of pages15
JournalNeuroImage
Volume172
DOIs
StatePublished - May 15 2018

Fingerprint

Brain
Chi-Square Distribution
Interpersonal Relations
Automatic Data Processing
Autism Spectrum Disorder
Communication
Magnetic Resonance Imaging
Sensitivity and Specificity
Population
Forests
Neurodevelopmental Disorders
Machine Learning

Keywords

  • Autism
  • Functional connectivity
  • MRI
  • Random forests
  • Supervised learning

ASJC Scopus subject areas

  • Neurology
  • Cognitive Neuroscience

Cite this

Subtyping cognitive profiles in Autism Spectrum Disorder using a Functional Random Forest algorithm. / Feczko, E.; Balba, N. M.; Miranda Dominguez, Oscar; Cordova, M.; Karalunas, Sarah; Irwin, L.; Demeter, D. V.; Hill, A. P.; Langhorst, B. H.; Grieser Painter, J.; Van Santen, Jan; Fombonne, Eric; Nigg, Joel; Fair, Damien.

In: NeuroImage, Vol. 172, 15.05.2018, p. 674-688.

Research output: Contribution to journalArticle

Feczko, E. ; Balba, N. M. ; Miranda Dominguez, Oscar ; Cordova, M. ; Karalunas, Sarah ; Irwin, L. ; Demeter, D. V. ; Hill, A. P. ; Langhorst, B. H. ; Grieser Painter, J. ; Van Santen, Jan ; Fombonne, Eric ; Nigg, Joel ; Fair, Damien. / Subtyping cognitive profiles in Autism Spectrum Disorder using a Functional Random Forest algorithm. In: NeuroImage. 2018 ; Vol. 172. pp. 674-688.
@article{c227bdac599845578490b7eab06b71da,
title = "Subtyping cognitive profiles in Autism Spectrum Disorder using a Functional Random Forest algorithm",
abstract = "DSM-5 Autism Spectrum Disorder (ASD) comprises a set of neurodevelopmental disorders characterized by deficits in social communication and interaction and repetitive behaviors or restricted interests, and may both affect and be affected by multiple cognitive mechanisms. This study attempts to identify and characterize cognitive subtypes within the ASD population using our Functional Random Forest (FRF) machine learning classification model. This model trained a traditional random forest model on measures from seven tasks that reflect multiple levels of information processing. 47 ASD diagnosed and 58 typically developing (TD) children between the ages of 9 and 13 participated in this study. Our RF model was 72.7{\%} accurate, with 80.7{\%} specificity and 63.1{\%} sensitivity. Using the random forest model, the FRF then measures the proximity of each subject to every other subject, generating a distance matrix between participants. This matrix is then used in a community detection algorithm to identify subgroups within the ASD and TD groups, and revealed 3 ASD and 4 TD putative subgroups with unique behavioral profiles. We then examined differences in functional brain systems between diagnostic groups and putative subgroups using resting-state functional connectivity magnetic resonance imaging (rsfcMRI). Chi-square tests revealed a significantly greater number of between group differences (p <.05) within the cingulo-opercular, visual, and default systems as well as differences in inter-system connections in the somato-motor, dorsal attention, and subcortical systems. Many of these differences were primarily driven by specific subgroups suggesting that our method could potentially parse the variation in brain mechanisms affected by ASD.",
keywords = "Autism, Functional connectivity, MRI, Random forests, Supervised learning",
author = "E. Feczko and Balba, {N. M.} and {Miranda Dominguez}, Oscar and M. Cordova and Sarah Karalunas and L. Irwin and Demeter, {D. V.} and Hill, {A. P.} and Langhorst, {B. H.} and {Grieser Painter}, J. and {Van Santen}, Jan and Eric Fombonne and Joel Nigg and Damien Fair",
year = "2018",
month = "5",
day = "15",
doi = "10.1016/j.neuroimage.2017.12.044",
language = "English (US)",
volume = "172",
pages = "674--688",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Subtyping cognitive profiles in Autism Spectrum Disorder using a Functional Random Forest algorithm

AU - Feczko, E.

AU - Balba, N. M.

AU - Miranda Dominguez, Oscar

AU - Cordova, M.

AU - Karalunas, Sarah

AU - Irwin, L.

AU - Demeter, D. V.

AU - Hill, A. P.

AU - Langhorst, B. H.

AU - Grieser Painter, J.

AU - Van Santen, Jan

AU - Fombonne, Eric

AU - Nigg, Joel

AU - Fair, Damien

PY - 2018/5/15

Y1 - 2018/5/15

N2 - DSM-5 Autism Spectrum Disorder (ASD) comprises a set of neurodevelopmental disorders characterized by deficits in social communication and interaction and repetitive behaviors or restricted interests, and may both affect and be affected by multiple cognitive mechanisms. This study attempts to identify and characterize cognitive subtypes within the ASD population using our Functional Random Forest (FRF) machine learning classification model. This model trained a traditional random forest model on measures from seven tasks that reflect multiple levels of information processing. 47 ASD diagnosed and 58 typically developing (TD) children between the ages of 9 and 13 participated in this study. Our RF model was 72.7% accurate, with 80.7% specificity and 63.1% sensitivity. Using the random forest model, the FRF then measures the proximity of each subject to every other subject, generating a distance matrix between participants. This matrix is then used in a community detection algorithm to identify subgroups within the ASD and TD groups, and revealed 3 ASD and 4 TD putative subgroups with unique behavioral profiles. We then examined differences in functional brain systems between diagnostic groups and putative subgroups using resting-state functional connectivity magnetic resonance imaging (rsfcMRI). Chi-square tests revealed a significantly greater number of between group differences (p <.05) within the cingulo-opercular, visual, and default systems as well as differences in inter-system connections in the somato-motor, dorsal attention, and subcortical systems. Many of these differences were primarily driven by specific subgroups suggesting that our method could potentially parse the variation in brain mechanisms affected by ASD.

AB - DSM-5 Autism Spectrum Disorder (ASD) comprises a set of neurodevelopmental disorders characterized by deficits in social communication and interaction and repetitive behaviors or restricted interests, and may both affect and be affected by multiple cognitive mechanisms. This study attempts to identify and characterize cognitive subtypes within the ASD population using our Functional Random Forest (FRF) machine learning classification model. This model trained a traditional random forest model on measures from seven tasks that reflect multiple levels of information processing. 47 ASD diagnosed and 58 typically developing (TD) children between the ages of 9 and 13 participated in this study. Our RF model was 72.7% accurate, with 80.7% specificity and 63.1% sensitivity. Using the random forest model, the FRF then measures the proximity of each subject to every other subject, generating a distance matrix between participants. This matrix is then used in a community detection algorithm to identify subgroups within the ASD and TD groups, and revealed 3 ASD and 4 TD putative subgroups with unique behavioral profiles. We then examined differences in functional brain systems between diagnostic groups and putative subgroups using resting-state functional connectivity magnetic resonance imaging (rsfcMRI). Chi-square tests revealed a significantly greater number of between group differences (p <.05) within the cingulo-opercular, visual, and default systems as well as differences in inter-system connections in the somato-motor, dorsal attention, and subcortical systems. Many of these differences were primarily driven by specific subgroups suggesting that our method could potentially parse the variation in brain mechanisms affected by ASD.

KW - Autism

KW - Functional connectivity

KW - MRI

KW - Random forests

KW - Supervised learning

UR - http://www.scopus.com/inward/record.url?scp=85042292622&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85042292622&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2017.12.044

DO - 10.1016/j.neuroimage.2017.12.044

M3 - Article

VL - 172

SP - 674

EP - 688

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

ER -