{rfName}
In

Indexed in

License and use

Citations

2

Altmetrics

Grant support

This publication is part of the R&D&I project with reference (IDI-20201146) , funded by The CDTI of the Spanish MCIU; part of the R&D&I project with reference PID2020-116422RB-C21, funded by MI-CIU/AEI/10.13039/501100011033; and by the Generalitat Valenciana (CIGE/2023/160) . MEM received partial Funding by the Generalitat Valenciana (GRISOLIAP/2019/137) .

Analysis of institutional authors

Altozano, AlbertoCorresponding AuthorMinissi, Maria EleonoraAuthorAlcaniz, MarianoAuthorMarin-Morales, JavierAuthor

Share

February 3, 2025
Publications
>
Article
No

Introducing 3DCNN ResNets for ASD full-body kinematic assessment: A comparison with hand-crafted features

Publicated to:Expert Systems With Applications. 270 126295-126295 - 2025-04-25 270(), DOI: 10.1016/j.eswa.2024.126295

Authors: Altozano, Alberto; Minissi, Maria Eleonora; Alcaniz, Mariano; Marin-Morales, Javier

Affiliations

Univ Politecn Valencia, Univ Res Inst Human Ctr Technol, Valencia 46021, Spain - Author

Abstract

Autism Spectrum Disorder (ASD) is characterized by challenges in social communication and restricted patterns, with motor abnormalities gaining traction for early detection. However, kinematic analysis in ASD is limited, often lacking robust validation and relying on hand-crafted features for single tasks, leading to inconsistencies across studies. End-to-end models have emerged as promising methods to overcome the need for feature engineering. Our aim is to propose a newly adapted 3DCNN ResNet from action recognition and compare it to widely used hand-crafted features for motor ASD assessment. Specifically, we developed a virtual reality environment with multiple motor tasks and trained models using both approaches. We prioritized a reliable validation framework with subject-wise nestedrepeated cross-validation. Results show the proposed model achieves a maximum accuracy of 85 +/- 3%, outperforming state-of-the-art end-to-end models with short 1-to-3 min samples. Our comparative analysis with hand-crafted features shows feature-engineered models outperformed our end-to-end model in certain tasks. However, generalized linear mixed-effects models showed that our end-to-end model achieved a statistically higher mean AUC (0.80 +/- 0.03) and Sensitivity (66 +/- 3%), while showing less variability across all VR tasks, demonstrating domain generalization and reliability. These findings show that end-to-end models enable less variable and context-independent ASD classification without requiring domain knowledge or task specificity. However, they also recognize the effectiveness of hand-crafted features in specific task scenarios.

Keywords

Autism spectrum discorder (asd)Autism spectrum disorderBehaviorChildrenConvolutional neural network (cnn3d)CoordinationDeep learningDeficitEnd-to-endFeature engineeringImmersive virtual environmentsKinematic dataMotor stereotypiesPerformance analysisRecognitionResidual network (resnetResidual network (resnet)

Quality index

Bibliometric impact. Analysis of the contribution and dissemination channel

The work has been published in the journal Expert Systems With Applications due to its progression and the good impact it has achieved in recent years, according to the agency WoS (JCR), it has become a reference in its field. In the year of publication of the work, 2025, it was in position 28/204, thus managing to position itself as a Q1 (Primer Cuartil), in the category Computer Science, Artificial Intelligence.

Independientemente del impacto esperado determinado por el canal de difusión, es importante destacar el impacto real observado de la propia aportación.

Según las diferentes agencias de indexación, el número de citas acumuladas por esta publicación hasta la fecha 2025-08-09:

  • WoS: 1

Impact and social visibility

From the perspective of influence or social adoption, and based on metrics associated with mentions and interactions provided by agencies specializing in calculating the so-called "Alternative or Social Metrics," we can highlight as of 2025-08-09:

  • The use, from an academic perspective evidenced by the Altmetric agency indicator referring to aggregations made by the personal bibliographic manager Mendeley, gives us a total of: 20.
  • The use of this contribution in bookmarks, code forks, additions to favorite lists for recurrent reading, as well as general views, indicates that someone is using the publication as a basis for their current work. This may be a notable indicator of future more formal and academic citations. This claim is supported by the result of the "Capture" indicator, which yields a total of: 20 (PlumX).

With a more dissemination-oriented intent and targeting more general audiences, we can observe other more global scores such as:

  • The Total Score from Altmetric: 52.75.
  • The number of mentions on the social network X (formerly Twitter): 1 (Altmetric).
  • The number of mentions in news outlets: 7 (Altmetric).

Leadership analysis of institutional authors

There is a significant leadership presence as some of the institution’s authors appear as the first or last signer, detailed as follows: First Author (Altozano Fernández, Alberto) and Last Author (Marín Morales, Javier).

the author responsible for correspondence tasks has been Altozano Fernández, Alberto.