Diagnosis of Autism Spectrum Disorder Based on Eigenvalues of Brain Networks

Mostafa, Sakib; Tang, Lingkai; Wu, Fang‐Xiang

doi:10.1109/access.2019.2940198

Cited by 71 publications

(41 citation statements)

References 74 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We further combine the DNN with the pretrained AE and then train the AE-based feature extractor and the classification network with raw features. The mixed encoder and classification network achieve an accuracy of 79.2% and the AUC of 82.4%, and it gives better results than that reported in Mostafa et al (2019). We also adapt the learnt hidden representations by AE network and DNNs to traditional machine learning algorithms such as SVM, KNN, and subspace discriminant.…”

Section: Introductionmentioning

confidence: 93%

“…In this study, we are extending the work proposed in Mostafa et al (2019) and Mostafa et al (2020) and studying the performance with different configurations and parameterizations. The same 871 subjects from the ABIDE 1 data set are used to conduct this study to accommodate the site variations and scanner configuration differences.…”

Section: Introductionmentioning

confidence: 99%

“…This type of generalizations is highly sought after by hands-on practitioners as the real-world applications are not necessarily well aligned with protocol, and it is very common for medical research and data acquisition that some data fields or items are missing. In Mostafa et al (2019), a new set of features was proposed for the diagnosis of ASD by turning to graph theories for better hidden feature extractions and compressions. In this article, the spectrum of the Laplacian matrix of brain functional networks was introduced as new high-level features and then combined with three network centralities: assortativity, clustering coefficient, and average degree.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Diagnosis of Autism Spectrum Disorder Based on Functional Brain Networks with Deep Learning

Yin

Mostafa

2021

Journal of Computational Biology

Self Cite

View full text Add to dashboard Cite

Autism spectrum disorder (ASD) is a neurological and developmental disorder. Traditional diagnosis of ASD is typically performed through the observation of behaviors and interview of a patient. However, these diagnosis methods are time-consuming and can be misleading sometimes. Integrating machine learning algorithms with neuroimages, a diagnosis method, can possibly be established to detect ASD subjects from typical control subjects. In this study, we develop deep learning methods for diagnosis of ASD from functional brain networks constructed with brain functional magnetic resonance imaging (fMRI) data. The entire Autism Brain Imaging Data Exchange 1 (ABIDE 1) data set is utilized to investigate the performance of our proposed methods. First, we construct the brain networks from brain fMRI images and define the raw features based on such brain networks. Second, we employ an autoencoder (AE) to learn the advanced features from the raw features. Third, we train a deep neural network (DNN) with the advanced features, which achieves the classification accuracy of 76.2% and the receiving operating characteristic curve (AUC) of 79.7%. As a comparison, we also apply the same advanced features to train several traditional machine learning algorithms to benchmark the classification performance. Finally, we combine the DNN with the pretrained AE and train it with the raw features, which achieves the classification accuracy of 79.2% and the AUC of 82.4%. These results show that our proposed deep learning methods outperform the state-of-the-art methods.

show abstract

Section: Introductionmentioning

confidence: 93%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Diagnosis of Autism Spectrum Disorder Based on Functional Brain Networks with Deep Learning

Yin

Mostafa

2021

Journal of Computational Biology

Self Cite

View full text Add to dashboard Cite

show abstract

“…Support Vector Machine-based Recursive Feature Elimination (SVM-RFE) was proposed by [ 38 ], where the correlation-based connectivity matrix was recursively pruned for discriminating features using SVM classifier, resulting in 90% accuracy on the dataset combined from all sites. Eigen features corresponding to 256 brain regions using the Laplacian matrix were proposed in [ 39 ] with the accuracy of 77%.…”

Section: Related Research Workmentioning

confidence: 99%

A Three-Stage Teacher, Student Neural Networks and Sequential Feed Forward Selection-Based Feature Selection Approach for the Classification of Autism Spectrum Disorder

et al. 2020

View full text Add to dashboard Cite

Autism disorder, generally known as Autism Spectrum Disorder (ASD) is a brain disorder characterized by lack of communication skills, social aloofness and repetitions in the actions in the patients, which is affecting millions of the people across the globe. Accurate identification of autistic patients is considered a challenging task in the domain of brain disorder science. To address this problem, we have proposed a three-stage feature selection approach for the classification of ASD on the preprocessed Autism Brain Imaging Data Exchange (ABIDE) rs-fMRI Dataset. In the first stage, a large neural network which we call a “Teacher ” was trained on the correlation-based connectivity matrix to learn the latent representation of the input. In the second stage an autoencoder which we call a “Student” autoencoder was given the task to learn those trained “Teacher” embeddings using the connectivity matrix input. Lastly, an SFFS-based algorithm was employed to select the subset of most discriminating features between the autistic and healthy controls. On the combined site data across 17 sites, we achieved the maximum 10-fold accuracy of 82% and for the individual site-wise data, based on 5-fold accuracy, our results outperformed other state of the art methods in 13 out of the total 17 site-wise comparisons.

show abstract

“…Although this kind of algorithms has good prediction accuracy in private data, its performance is seriously dependent on the Region Of Interest (ROI) selected by doctors and researchers, which is subjective, timeconsuming and not universal. Therefore, most researchers classify ASD based on feature analysis method [15] - [18]. Feature analysis mainly relies on computer to extract, recognize and diagnose from voxel or morphological features.…”

Section: Related Workmentioning

confidence: 99%

Classification and Biomarker Exploration of Autism Spectrum Disorders Based on Recurrent Attention Model

Yang

2020

IEEE Access

View full text Add to dashboard Cite

Diagnosis of Autism Spectrum Disorder Based on Eigenvalues of Brain Networks

Cited by 71 publications

References 74 publications

Diagnosis of Autism Spectrum Disorder Based on Functional Brain Networks with Deep Learning

Diagnosis of Autism Spectrum Disorder Based on Functional Brain Networks with Deep Learning

A Three-Stage Teacher, Student Neural Networks and Sequential Feed Forward Selection-Based Feature Selection Approach for the Classification of Autism Spectrum Disorder

Classification and Biomarker Exploration of Autism Spectrum Disorders Based on Recurrent Attention Model

Contact Info

Product

Resources

About