Attention-Based Node-Edge Graph Convolutional Networks for Identification of Autism Spectrum Disorder Using Multi-Modal MRI Data

Chen, Yuzhong; Yan, Jiadong; Jiang, Mingxin; Zhao, Zhongbo; Zhao, Weihua; Zhang, Rong; Kendrick, Keith M.; Jiang, Xi

doi:10.1007/978-3-030-88010-1_31

Cited by 4 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There has been a recent movement toward integrating diagnostic modalities such as fMRI, DWI, and sMRI. It may improve prognosis accuracy by using complementary information in the multimodal data ( Eill et al, 2019 ; Chen et al, 2021 ). Only one study combined non-imaging data (such as demographic data and reports) with imaging data to enhance classifier predictability and interpretability ( Dekhil et al, 2021 ).…”

Section: Discussion and Limitationsmentioning

confidence: 99%

“…Graph convolutional networks (GCN) enable graph embedding by representing graph nodes, edges, and subgraphs as low-dimensional vectors. GCN can also learn graph topological structure information, which is essential for studying population brain networks ( Chen et al, 2021 ). Research using GCN to classify autistic individuals under two distinct graph definition categories has been conducted ( Parisot et al, 2018 ; Chen et al, 2021 ).…”

Section: Highlighted Researchmentioning

confidence: 99%

“…GCN can also learn graph topological structure information, which is essential for studying population brain networks ( Chen et al, 2021 ). Research using GCN to classify autistic individuals under two distinct graph definition categories has been conducted ( Parisot et al, 2018 ; Chen et al, 2021 ). The first establishes edges between subjects through phenotype information, including age, gender, and acquisition locations, together with the imaging-based node features ( Parisot et al, 2018 ).…”

Section: Highlighted Researchmentioning

confidence: 99%

“…The first establishes edges between subjects through phenotype information, including age, gender, and acquisition locations, together with the imaging-based node features ( Parisot et al, 2018 ). The second type considers each subject as a graph ( Chen et al, 2021 ). However, these two studies relied on single-modal MRI.…”

Section: Highlighted Researchmentioning

confidence: 99%

“…However, these two studies relied on single-modal MRI. To bridge the gap, Chen et al (2021) developed an Attention-based Node-Edge GCN method that integrates sMRI and rs-fMRI data while concurrently modeling nodes and edges in graphs. Also, a gradient-based model interpretation technique was utilized to detect putative ASD biomarkers.…”

Section: Highlighted Researchmentioning

confidence: 99%

See 4 more Smart Citations

Machine learning for autism spectrum disorder diagnosis using structural magnetic resonance imaging: Promising but challenging

Bahathiq

Banjar

Bamaga

et al. 2022

Front. Neuroinform.

View full text Add to dashboard Cite

Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder that affects approximately 1% of the population and causes significant burdens. ASD’s pathogenesis remains elusive; hence, diagnosis is based on a constellation of behaviors. Structural magnetic resonance imaging (sMRI) studies have shown several abnormalities in volumetric and geometric features of the autistic brain. However, inconsistent findings prevented most contributions from being translated into clinical practice. Establishing reliable biomarkers for ASD using sMRI is crucial for the correct diagnosis and treatment. In recent years, machine learning (ML) and specifically deep learning (DL) have quickly extended to almost every sector, notably in disease diagnosis. Thus, this has led to a shift and improvement in ASD diagnostic methods, fulfilling most clinical diagnostic requirements. However, ASD discovery remains difficult. This review examines the ML-based ASD diagnosis literature over the past 5 years. A literature-based taxonomy of the research landscape has been mapped, and the major aspects of this topic have been covered. First, we provide an overview of ML’s general classification pipeline and the features of sMRI. Next, representative studies are highlighted and discussed in detail with respect to methods, and biomarkers. Finally, we highlight many common challenges and make recommendations for future directions. In short, the limited sample size was the main obstacle; Thus, comprehensive data sets and rigorous methods are necessary to check the generalizability of the results. ML technologies are expected to advance significantly in the coming years, contributing to the diagnosis of ASD and helping clinicians soon.

show abstract

Section: Discussion and Limitationsmentioning

confidence: 99%

Section: Highlighted Researchmentioning

confidence: 99%

Section: Highlighted Researchmentioning

confidence: 99%

Section: Highlighted Researchmentioning

confidence: 99%

Section: Highlighted Researchmentioning

confidence: 99%

See 3 more Smart Citations

Machine learning for autism spectrum disorder diagnosis using structural magnetic resonance imaging: Promising but challenging

Bahathiq

Banjar

Bamaga

et al. 2022

Front. Neuroinform.

View full text Add to dashboard Cite

show abstract

A scoping review of interpretability and explainability concerning artificial intelligence methods in medical imaging

Champendal,

Müller,

Prior

et al. 2023

European Journal of Radiology

View full text Add to dashboard Cite

Automated diagnosis of autism: State of the art

Valizadeh

Moassefi

Nakhostin-Ansari

et al. 2021

Preprint

View full text Add to dashboard Cite

Objective: To determine the diagnostic accuracy of the applied machine learning algorithms for the diagnosis of autism spectrum disorder (ASD) based on structural magnetic resonance imaging (sMRI), resting-state functional MRI (rs-fMRI), and electroencephalography (EEG). Methods: We will include cross-sectional studies (both single-gates and two-gates) that have evaluated the diagnostic accuracy of machine learning algorithms on the sMRI data of ASD patients regardless of age, sex, and ethnicity. On the 22nd of May 2021, we searched Embase, MEDLINE, APA PsycINFO, IEEE Xplore, Scopus, and Web of Science for eligible studies. We also searched grey literature within various sources. We will use an adapted version of the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool to assess the risk of bias and applicability. Data will be synthesized using the relatively new Split Component Synthesis (SCS) method. We plan to assess heterogeneity using the I2 statistics and assess publication bias using trim and fill tests combined with ln DOR. Certainty of evidence will be assessed using the GRADE approach for diagnostic studies. Funding: These studies are funded by Sports Medicine Research Center, Tehran, Iran. Registration: PROSPERO submission IDs: 262575, 262825, and 262831.

show abstract

Attention-Based Node-Edge Graph Convolutional Networks for Identification of Autism Spectrum Disorder Using Multi-Modal MRI Data

Cited by 4 publications

References 18 publications

Machine learning for autism spectrum disorder diagnosis using structural magnetic resonance imaging: Promising but challenging

Machine learning for autism spectrum disorder diagnosis using structural magnetic resonance imaging: Promising but challenging

A scoping review of interpretability and explainability concerning artificial intelligence methods in medical imaging

Automated diagnosis of autism: State of the art

Contact Info

Product

Resources

About