What does this research mean for the field?

Current machine learning and deep learning models for Autism Spectrum Disorder prediction exhibit significant methodological limitations, such as limited external validation and insufficient ablation analyses, which undermine their reported performance and clinical generalizability. Novelty: ClaimNovelty.SYNTHESIS. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To evaluate machine learning and deep learning approaches for autism spectrum disorder prediction, focusing on data augmentation and feature selection techniques.

September 8, 2026Health care scienceOpen Access

Machine Learning for Autism Spectrum Disorder Prediction: A Review of Data Augmentation and Feature Selection Techniques

View Full Paper

Ask AI

Bookmark

Authors

SASahar Ahmed M Alkhaibari FDFeng Dong

Discussion

Member takes

Overview

Systematic review reveals methodological gaps in machine learning models for autism spectrum disorder prediction, highlighting the need for robust validation and biological interpretability.

Key Points

To evaluate machine learning and deep learning approaches for autism spectrum disorder prediction, focusing on data augmentation and feature selection techniques.
Conducted a structured literature search across IEEE Xplore, PubMed, Scopus, and Google Scholar for studies published between 2021 and 2024.
Evaluated 26 peer-reviewed studies using the Prediction Model Risk of Bias Assessment Tool to examine methodological quality and pipeline design.
Categorized data augmentation into conventional transformations and advanced generative adversarial network methods, noting that few studies utilized ablation analyses to isolate individual performance contributions.
Identified filter, wrapper, and embedded feature selection approaches—such as recursive feature elimination and elastic net regularization—frequently applied without biological interpretability or empirical justification.
Demonstrated widespread methodological limitations across the reviewed literature, including inadequate validation frameworks and limited external testing that hinder clinical applicability.

Cite This Study

Alkhaibari et al. (2026) studied this question.

synapsesocial.com/papers/6a9fd70a58e84d0ff5b4599e https://doi.org/10.1002/hcs2.70098

View Full Paper

Ask AI

Bookmark

September 8, 2026Health care scienceOpen Access

Machine Learning for Autism Spectrum Disorder Prediction: A Review of Data Augmentation and Feature Selection Techniques

View Full Paper

Ask AI

Bookmark

Authors

SASahar Ahmed M Alkhaibari FDFeng Dong

Discussion

Member takes

Overview

Systematic review reveals methodological gaps in machine learning models for autism spectrum disorder prediction, highlighting the need for robust validation and biological interpretability.

Key Points

To evaluate machine learning and deep learning approaches for autism spectrum disorder prediction, focusing on data augmentation and feature selection techniques.
Conducted a structured literature search across IEEE Xplore, PubMed, Scopus, and Google Scholar for studies published between 2021 and 2024.
Evaluated 26 peer-reviewed studies using the Prediction Model Risk of Bias Assessment Tool to examine methodological quality and pipeline design.
Categorized data augmentation into conventional transformations and advanced generative adversarial network methods, noting that few studies utilized ablation analyses to isolate individual performance contributions.
Identified filter, wrapper, and embedded feature selection approaches—such as recursive feature elimination and elastic net regularization—frequently applied without biological interpretability or empirical justification.
Demonstrated widespread methodological limitations across the reviewed literature, including inadequate validation frameworks and limited external testing that hinder clinical applicability.

Cite This Study

Alkhaibari et al. (2026) studied this question.

synapsesocial.com/papers/6a9fd70a58e84d0ff5b4599e https://doi.org/10.1002/hcs2.70098

View Full Paper

Ask AI

Bookmark