Effective Autism Classification Through Grasping Kinematics / Zoha AHMAD ; Eitan SHELEF ; Bat Sheva HADAD in Autism Research, 18-6 (June 2025)  

Public ISBD [article]  inAutism Research > 18-6 (June 2025) . - p.1170-1181 Titre : Effective Autism Classification Through Grasping Kinematics Type de document : Texte imprimé et/ou numérique Auteurs : Zoha AHMAD, Auteur ; Eitan SHELEF, Auteur ; Bat Sheva HADAD, Auteur Article en page(s) : p.1170-1181 Langues : Anglais (eng) Mots-clés : autism  grasping  machine learning  motion tracking  visuomotor control Index. décimale : PER Périodiques Résumé : ABSTRACT Autism is a complex neurodevelopmental condition, where motor abnormalities play a central role alongside social and communication difficulties. These motor symptoms often manifest in early childhood, making them critical targets for early diagnosis and intervention. This study aimed to assess whether kinematic features from a naturalistic grasping task could accurately distinguish autistic participants from non-autistic ones. We analyzed grasping movements of autistic and non-autistic young adults, tracking two markers placed on the thumb and index finger. Using a subject-wise cross-validated classifiers, we achieved accuracy scores of above 84%. Receiver operating characteristic analysis revealed strong classification performance with area under the curve values of above 0.95 at the subject-wise analysis and above 0.85 at the trial-wise analysis. These findings indicate strong reliability in accurately distinguishing autistic participants from non-autistic ones. These findings suggest that subtle motor control differences can be effectively captured, offering a promising approach for developing accessible and reliable diagnostic tools for autism. En ligne : https://doi.org/10.1002/aur.70049 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=558 [article] Effective Autism Classification Through Grasping Kinematics [Texte imprimé et/ou numérique] / Zoha AHMAD, Auteur ; Eitan SHELEF, Auteur ; Bat Sheva HADAD, Auteur . - p.1170-1181. Langues : Anglais (eng) in Autism Research > 18-6 (June 2025) . - p.1170-1181 Mots-clés : autism  grasping  machine learning  motion tracking  visuomotor control Index. décimale : PER Périodiques Résumé : ABSTRACT Autism is a complex neurodevelopmental condition, where motor abnormalities play a central role alongside social and communication difficulties. These motor symptoms often manifest in early childhood, making them critical targets for early diagnosis and intervention. This study aimed to assess whether kinematic features from a naturalistic grasping task could accurately distinguish autistic participants from non-autistic ones. We analyzed grasping movements of autistic and non-autistic young adults, tracking two markers placed on the thumb and index finger. Using a subject-wise cross-validated classifiers, we achieved accuracy scores of above 84%. Receiver operating characteristic analysis revealed strong classification performance with area under the curve values of above 0.95 at the subject-wise analysis and above 0.85 at the trial-wise analysis. These findings indicate strong reliability in accurately distinguishing autistic participants from non-autistic ones. These findings suggest that subtle motor control differences can be effectively captured, offering a promising approach for developing accessible and reliable diagnostic tools for autism. En ligne : https://doi.org/10.1002/aur.70049 Permalink : https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=558

---

1 (1 - 1 / 1)