Author Correction: Atypical postural control can be detected via computer vision analysis in toddlers with autism spectrum disorder

Géraldine Dawson, Kathleen Campbell, Jordan Hashemi, Steven Lippmann, Valerie A. Smith, Kimberly L. H. Carpenter, Helen L. Egger, Steven Espinosa, Saritha Vermeer, Jeffrey P. Baker, Guillermo SapiroScientific 2020 · DOI · OpenAlex

Abstract

This article presents a formal correction to the descriptive statistics reported in the 2018 study "Atypical postural control can be detected via computer vision analysis in toddlers with autism spectrum disorder." The amendment specifically addresses the mean and standard deviation of the Autism Diagnostic Observation Schedule - Toddler Module (ADOS-T) scores used to characterize the participant cohort. Ensuring the accuracy of these baseline metrics is essential for the integrity of the study's findings regarding digital biomarkers for Autism Spectrum Disorder (ASD).

Introduction

Early detection of Autism Spectrum Disorder (ASD) remains a primary challenge in pediatric healthcare. Traditional diagnostic pathways rely heavily on clinical observations of social and communication behaviors. While effective, these methods can be subjective, time-intensive, and may not be easily quantifiable in very young children.

The original research by Dawson et al. (2018), published in Scientific Reports, addressed this need by exploring "computational phenotyping." The study proposed that atypical postural control—subtle differences in how a child maintains balance or shifts their weight—could serve as an objective, scalable digital biomarker for ASD. By utilizing computer vision to automate the detection of these motor markers, the researchers aimed to create tools that could facilitate earlier intervention. This current document serves as a formal correction to the clinical characterization of the toddlers involved in that study.

The Role of ADOS-T in Diagnostic Assessment

To validate the computer vision findings, the researchers utilized the Autism Diagnostic Observation Schedule - Toddler Module (ADOS-T). The ADOS-T is a gold-standard, semi-structured assessment used to confirm ASD diagnoses and measure the severity of symptoms in young children.

In the context of the original study, the ADOS-T scores provided the clinical baseline against which the computer vision data was compared. Accurate reporting of these scores is critical for defining the study population, ensuring that the cohort is well-defined for future replication, meta-analysis, and cross-study comparisons.

Details of the Correction

The published correction updates the "Diagnostic Assessments" portion of the methodology. The authors identified an error in the descriptive statistics originally reported for the participant group. The adjustments are as follows:

  • Original ADOS-T Mean Score: 18.81 (SD = 4.20)
  • Corrected ADOS-T Mean Score: 18.00 (SD = 4.67)

While the numerical shift is relatively small, the correction provides a more precise representation of the symptom severity within the toddler group analyzed by the computer vision system.

Significance of Data Integrity in Clinical Research

In clinical research and the development of AI-driven diagnostic tools, data integrity is paramount. Even minor adjustments in mean scores and standard deviations are significant because they define the "baseline" of the study population. Accurate reporting ensures that the correlations found between postural control and ASD severity are grounded in a correctly characterized clinical profile.

Furthermore, the issuance of a formal correction reflects a commitment to the standards of Open Access publishing and scientific transparency. By maintaining a precise and reproducible permanent record, the authors ensure that the scientific community can rely on the data for ongoing research into motor markers and neurodevelopmental disorders.

Conclusion

The original study's exploration of motor markers as digital phenotypes represents a significant step toward objective ASD screening. Postural control is a particularly compelling target because motor delays often precede the emergence of social-communication symptoms. This correction ensures that the foundational clinical data supporting these insights is accurate, preserving the validity of the study’s contributions to the field of computational phenotyping.

Summary updated

Abstract

An amendment to this paper has been published and can be accessed via a link at the top of the paper.
BibTeX
@inproceedings{dawson2020author,
  author = {Géraldine Dawson and Kathleen Campbell and Jordan Hashemi and Steven Lippmann and Valerie A. Smith and Kimberly L. H. Carpenter and Helen L. Egger and Steven Espinosa and Saritha Vermeer and Jeffrey P. Baker and Guillermo Sapiro},
  title = {{Author Correction: Atypical postural control can be detected via computer vision analysis in toddlers with autism spectrum disorder}},
  booktitle = {Scientific},
  year = {2020},
  doi = {10.1038/s41598-020-57570-1},
}

Loading related content...

Figure