Classify autism and control based on deep learning and community structure on resting-state fMRI

It is unsatisfied to diagnose brain disorders based on subjective judgment. In this paper, we proposed a novel method to classify autism disorders and normal subjects objectively and automatically. The method firstly detects community structure in network of every subject. The NMI statistic matrix,...

Full description

Saved in:
Bibliographic Details
Published in2018 Tenth International Conference on Advanced Computational Intelligence (ICACI) pp. 289 - 294
Main Authors Liao, Dingan, Lu, Hu
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.03.2018
Subjects
Autism
community detection
Correlation
deep learning
fMRI
Functional magnetic resonance imaging
Machine learning
Noise reduction
Support vector machines
Symmetric matrices
Online AccessGet full text

Cover

More Information
Summary:It is unsatisfied to diagnose brain disorders based on subjective judgment. In this paper, we proposed a novel method to classify autism disorders and normal subjects objectively and automatically. The method firstly detects community structure in network of every subject. The NMI statistic matrix, which can effectively represent the features of all subject in a certain dataset, was developed and then was imported into denoising autoencoder to classify. We tested our method on three datasets. The results show that the accuracy of our method is higher than that of traditional one. Additionally, our method is more efficient than import Pearson correlation matrix into classifier. Our method is effective to help doctors diagnose autism objectively.
DOI:10.1109/ICACI.2018.8377471