Brain imaging-based machine learning in autism spectrum disorder: methods and applications

Xu, Ming; Calhoun, Vince D.; Jiang, Rongtao; Yan, Weizheng; Sui, Jing

doi:10.1016/j.jneumeth.2021.109271

Cited by 38 publications

(28 citation statements)

References 177 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A majority of past and present studies on machine learning for clinical neuroimaging rely on features designed by domain experts; cf. recent reviews on machine learning in epilepsy (Sone & Beheshti, 2021), autism (Xu et al, 2021), stroke (Sirsat et al, 2020), and mild cognitive impairment (Ansart et al, 2021). The UK Biobank directly provides widely used feature representations (IDPs) for structural, functional, and diffusion tensor imaging.…”

Section: Methodsmentioning

confidence: 99%

Performance reserves in brain-imaging-based phenotype prediction

Schulz

Bzdok

Haufe

et al. 2022

Preprint

View full text Add to dashboard Cite

Machine learning studies have shown that various phenotypes can be predicted from structural and functional brain images. However, in most such studies, prediction performance ranged from moderate to disappointing. It is unclear whether prediction performance will substantially improve with larger sample sizes or whether insufficient predictive information in brain images impedes further progress. Here, we systematically assess the effect of sample size on prediction performance using sample sizes far beyond what is possible in common neuroimaging studies. We project 3-9 fold improvements in prediction performance for behavioral and mental health phenotypes when moving from one thousand to one million samples. Moreover, we find that moving from single imaging modalities to multimodal input data can lead to further improvements in prediction performance, often on par with doubling the sample size. Our analyses reveal considerable performance reserves for neuroimaging-based phenotype prediction. Machine learning models may benefit much more from extremely large neuroimaging datasets than currently believed.

show abstract

Section: Methodsmentioning

confidence: 99%

Performance reserves in brain-imaging-based phenotype prediction

Schulz

Bzdok

Haufe

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Recently, investigators have tried to use different methods to distinguish ASD from TD in addition to traditional statistical methods, such as the logistic regression model ( 13 ). Machine learning techniques, including the support vector machine (SVM), have been applied to identify biomarkers for ASD ( 39 – 41 ). As a subset of artificial intelligence in the field of computer science, machine learning is a procedure that trains the computer algorithm to analyze a set of observed data and statistically learn the latent patterns without being explicitly programmed.…”

Section: Introductionmentioning

confidence: 99%

Different Eye Tracking Patterns in Autism Spectrum Disorder in Toddler and Preschool Children

et al. 2022

View full text Add to dashboard Cite

BackgroundChildren with autism spectrum disorder (ASD) have been observed to be associated with fixation abnormality as measured eye tracking, but the dynamics behind fixation patterns across age remain unclear.Materials and MethodsIn this study, we investigated gaze patterns between toddlers and preschoolers with and without ASD while they viewed video clips and still images (i.e., mouth-moving face, biological motion, mouthing face vs. moving object, still face picture vs. objects, and moving toys).ResultsWe found that the fixation time percentage of children with ASD showed significant decrease compared with that of TD children in almost all areas of interest (AOI) except for moving toy (helicopter). We also observed a diagnostic group (ASD vs. TD) and chronological age (Toddlers vs. preschooler) interaction for the eye AOI during the mouth-moving video clip. Support vector machine analysis showed that the classifier could discriminate ASD from TD in toddlers with an accuracy of 80% and could discriminate ASD from TD in preschoolers with an accuracy of 71%.ConclusionOur results suggest that toddlers and preschoolers may be associated with both common and distinct fixation patterns. A combination of eye tracking and machine learning methods has the potential to shed light on the development of new early screening/diagnosis methods for ASD.

show abstract

“…A feature is any measurable property extracted from the source dataset regarding the class. Through features engineering, neuroimaging data is transformed into trustworthy and biologically relevant features that greatly influence data separation (Xu et al, 2021). The "dimensionality curse" problem 10.3389/fninf.2022.949926 FIGURE 4 Differences between (A) ML-based studies workflow and (B) DL-based studies workflow.…”

Section: Feature Extraction Selection/reductionmentioning

confidence: 99%

“…The hyperparameters that determine the model's architecture (e.g., number of neurons, activation function, batch size, etc.) are then optimized for optimum performance, model generalization, and loss function reduction (Kim and Na, 2018; 10.3389/fninf.2022.949926 Xu et al, 2021). Hyperparameter tuning/optimization is the process of determining the optimal combination of hyperparameter values to get maximum data performance in an acceptable amount of time (Rojas-Domínguez et al, 2017).…”

Section: Model Trainingmentioning

confidence: 99%

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

Brain imaging-based machine learning in autism spectrum disorder: methods and applications

Cited by 38 publications

References 177 publications

Performance reserves in brain-imaging-based phenotype prediction

Performance reserves in brain-imaging-based phenotype prediction

Different Eye Tracking Patterns in Autism Spectrum Disorder in Toddler and Preschool Children

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

Contact Info

Product

Resources

About