Diagnostic classification of intrinsic functional connectivity highlights somatosensory, default mode, and visual regions in autism

Autor: Christopher L. Keown, Ralph-Axel Müller, Colleen P. Chen, Mark E. Pflieger, Aarti Nair, Afrooz Jahedi, Barbara A. Bailey
Jazyk: angličtina
Rok vydání: 2015
Předmět:
Male
Autism Spectrum Disorder
Autism
Somatosensory
VIS
visual

lcsh:RC346-429
0302 clinical medicine
Feature (machine learning)
SMM
somatosensory and motor [mouth]

CEB
cerebellum

Child
Default mode network
Cerebral Cortex
0303 health sciences
VA
ventral attention

DA
dorsal attention

Default mode
Magnetic Resonance Imaging
Random forest
Neurology
Autism spectrum disorder
Connectome
lcsh:R858-859.7
Female
Psychology
Visual
Adult
UN
unknown

Adolescent
SMH
somatosensory and motor [hand]

Cognitive Neuroscience
lcsh:Computer applications to medicine. Medical informatics
Article
SUB
subcortical

Young Adult
03 medical and health sciences
SAL
salience

Machine learning
medicine
Humans
Radiology
Nuclear Medicine and imaging

lcsh:Neurology. Diseases of the nervous system
030304 developmental biology
FPTC
frontal parietal task control

Functional connectivity MRI
MR
memory retrieval

medicine.disease
Ensemble learning
Support vector machine
DMN
default mode network

Neurology (clinical)
COTC
cingulo-opercular task control

Neuroscience
030217 neurology & neurosurgery
AUD
audio
Zdroj: NeuroImage: Clinical, Vol 8, Iss C, Pp 238-245 (2015)
NeuroImage : Clinical
ISSN: 2213-1582
Popis: Despite consensus on the neurological nature of autism spectrum disorders (ASD), brain biomarkers remain unknown and diagnosis continues to be based on behavioral criteria. Growing evidence suggests that brain abnormalities in ASD occur at the level of interconnected networks; however, previous attempts using functional connectivity data for diagnostic classification have reached only moderate accuracy. We selected 252 low-motion resting-state functional MRI (rs-fMRI) scans from the Autism Brain Imaging Data Exchange (ABIDE) including typically developing (TD) and ASD participants (n = 126 each), matched for age, non-verbal IQ, and head motion. A matrix of functional connectivities between 220 functionally defined regions of interest was used for diagnostic classification, implementing several machine learning tools. While support vector machines in combination with particle swarm optimization and recursive feature elimination performed modestly (with accuracies for validation datasets
Highlights • Machine learning of resting fMRI attains high diagnostic accuracy for autism. • Peak accuracy is seen for a complex pattern of 100 connectivities. • Somatosensory regions are overall most informative. • Default mode and visual regions also contribute to diagnostic accuracy.
Databáze: OpenAIRE