Classification of autism spectrum disorders individuals and controls using phase and envelope features from resting-state fMRI data

ABSTRACT

This study uses phase and envelope features extracted from fMRI data for discriminating between Autism Spectrum Disorders (ASD) and control subjects. We exploit the transfer function perturbation (TFP) method to estimate the instantaneous phase and envelope of intrinsic resting-state brain network components from fMRI data. Then we calculated power, entropy, and coherency features. We examined three different classifiers and two different feature selection algorithms, in a way that the subsets of features that best predict classes were selected using a sequential forward feature selection (SFFS) algorithm and principal component analysis (PCA). Afterwards, three different categories of calculated features, including phase features, envelope features, and a combination of phase and envelope features, fed into non-linear Support Vector Machine (SVM), K-Nearest Neighbours (K-NN), and Deep Neural Network (DNN). Results illustrate that phase features are significantly discriminative and considerably improve ADSs and control subjects’ classification accuracy and lead to robust prediction. Moreover, 91% of classification accuracy was obtained when the dimension of phase features was reduced by PCA and fed to the non-linear SVM. Eventually, the two-sample t-test and Pearson’s correlation coefficient illustrated that phase features had a significantly lower correlation than envelope features.

KEYWORDS:

• Resting-state fMRI
• ASD
• instantaneous phase and envelope
• classification
• features
• brain connectivity

Acknowledgments

In this study, we used ABIDEII (Autism Brain Imaging Data Exchange) dataset that was established to further promote discovery science on the brain connectome in ASD. ABIDEII involves 1114 datasets from 521 individuals with ASD and 593 controls (age range: 5-64 years).

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Mahshid Naghashzadeh

Mahshid Naghashzadeh received the B.Sc. degree in electrical engineering from the Department of Electrical and Electronic Engineering, Shahid Chamran University, Ahvaz, Iran, in 2016, and the M.Sc. degree in communication systems from the Department of Electrical and Electronic Engineering, Shiraz University, Shiraz, Iran, in 2020. Her undergraduate thesis was concentrated on implementing an image noise reduction algorithm using FPGA. Her master’s thesis was focused on biomedical imaging and signal processing, as well as classifications using machine learning and deep learning methods. Her major research interests are in the fields of signal processing, biomedical image analysis, machine learning, and deep learning.

Mehran Yazdi

Mehran Yazdi received the B.Sc. degree in communication systems from the Department of Electrical and Electronic Engineering, Shiraz University, Shiraz, Iran, in 1992, and the M.Sc. and Ph.D. degrees in digital vision and image processing from the Department of Electrical Engineering, Laval University, Quebec, QC, Canada, in 1996 and 2003, respectively. He is currently Full Professor in the Department of Communications and Electronics Engineering at Shiraz University. He passed a sabbatical period in MIA Lab. of La Rochelle University in La Rochelle, France from September 2015 to May 2016. He conducted several projects in the area of hyper-spectral image compression and denoising, CT metal artifact reduction, and video compression. His major research interests are in the field of image and video processing, remote sensing, multimedia networks, multidimensional signal processing, and medical image analysis.

Alireza Zolghadrasli

Alireza Zolghadrasliwas born in Fasa (Iran) in 1955 and received the B.Sc. degree (program of five years) in electrical and electronic engineering from Shiraz University of Iran in July 1978. He continued his graduate studies at the Polytechnic Institute of Grenoble (INPG) at France. He received two M.Sc. degrees and a Ph.D. degree in electronic, signal & data processing, and image processing respectively in 1980, 1981, and 1985. In 1985, he joined the Department of Physics of University of Chambery (France) as Assistant Professor. In 1988, he joined the University of Science of Grenoble (IUT), France, as Associate Professor and researcher. Since 1990 he is a permanent academic member of EE-Dept. at Shiraz University (Iran) and actually Full Professor of communication systems at this department. He is author or co-author of more than 120 papers in international journals and conference proceedings.