Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Apr 9, 2026

Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable
09:24

Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable

Published on: May 17, 2024

2.4K

SHAP-based explainable AI framework for autism severity classification using 3D motor biomarkers.

Yelda Fırat1

  • 1Department of Computer Engineering, Mudanya University, Bursa, Türkiye.

Frontiers in Psychiatry
|April 8, 2026
PubMed
Summary

Objective autism spectrum disorder (ASD) severity classification is now possible using 3D motor analysis. This study developed a Random Forest model identifying key motor biomarkers for improved ASD diagnosis and intervention.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Explainable machine learning-based identification of transcriptomic biomarkers in CD1c+ dendritic cells for non-infectious uveitis: an integrative analysis of bulk RNA-seq data.

Frontiers in bioinformatics·2026
Same author

Exploratory analysis of pupillography-based machine learning for assessing autonomic dysfunction in multiple sclerosis.

Multiple sclerosis and related disorders·2026
Same author

An explainable ensemble machine learning framework for predicting cartilage biomechanical degradation from transcriptomic profiles: toward smart orthopedic device design.

Frontiers in bioengineering and biotechnology·2026
Same author

An explainable ensemble machine learning model using baseline blood transcriptomics to predict Parkinson's disease motor progression.

Frontiers in digital health·2026

Area of Science:

  • Biomedical Engineering
  • Neuroscience
  • Developmental Psychology

Background:

  • Early diagnosis of autism spectrum disorder (ASD) is crucial for effective intervention.
  • Current diagnostic methods for ASD rely heavily on subjective clinical observations, leading to potential inconsistencies.
  • Motor abnormalities are recognized as core features of ASD, offering a potential avenue for objective assessment.

Purpose of the Study:

  • To develop objective tools for classifying ASD severity using 3D motor movement analysis.
  • To address motor abnormalities as core diagnostic features for improved ASD assessment.
  • To identify critical motor biomarkers indicative of ASD severity.

Main Methods:

  • A Random Forest (RF) model was developed to classify three ASD severity levels.
Keywords:
SHAP analysisautism spectrum disordermotor biomarkersrandom forestviolence level classification

More Related Videos

An Objective and Child-friendly Assessment of Arm Function by Using a 3-D Sensor
07:25

An Objective and Child-friendly Assessment of Arm Function by Using a 3-D Sensor

Published on: February 12, 2018

7.3K

Related Experiment Videos

Last Updated: Apr 9, 2026

Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable
09:24

Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable

Published on: May 17, 2024

2.4K
An Objective and Child-friendly Assessment of Arm Function by Using a 3-D Sensor
07:25

An Objective and Child-friendly Assessment of Arm Function by Using a 3-D Sensor

Published on: February 12, 2018

7.3K
  • The model utilized 463 motor features derived from 25 Kinect V2 joint points.
  • Data from 109 children were analyzed and validated using 5-fold cross-validation and held-out test sets, with Shapley Additive Explanations (SHAP) for biomarker identification.

    • Main Results:

    • The RF model achieved 84.6±10.9% accuracy via cross-validation and 86.4% accuracy on test sets for typical and moderate ASD.
    • A 100% classification accuracy was observed for severe ASD on synthetic data, serving as a methodological proof-of-concept.
    • SHAP analysis identified wrist movements, knee trajectories, and elbow-to-foot distances as significant motor biomarkers for severity classification.

    Conclusions:

    • The Kinect-based approach using RF and SHAP provides an effective and interpretable method for assessing ASD severity in typical and moderate cases.
    • The study demonstrates promising methodological foundations for severe ASD classification, pending validation with real-world data.
    • Objective motor analysis offers a valuable complementary tool to clinical observations in the diagnosis and severity assessment of ASD.