Machine learning models using mobile game play accurately classify children with autism

Nicholas Deveau; Peter Washington; Emilie Leblanc; Arman Husic; Kaitlyn Dunlap; Yordan Penev; Aaron Kline; Onur Cezmi Mutlu; Dennis P Wall

doi:10.1016/j.ibmed.2022.100057

Machine learning models using mobile game play accurately classify children with autism

Intell Based Med. 2022:6:100057. doi: 10.1016/j.ibmed.2022.100057. Epub 2022 Aug 24.

Authors

Nicholas Deveau¹, Peter Washington², Emilie Leblanc³, Arman Husic³, Kaitlyn Dunlap³, Yordan Penev³, Aaron Kline³, Onur Cezmi Mutlu⁴, Dennis P Wall^{1

3}

Affiliations

¹ Biomedical Data Science, Stanford University, Stanford, 94305, California, United States.
² Bioengineering, Stanford University, Stanford, 94305, California, United States.
³ Pediatrics, Stanford University, Stanford, 94305, California, United States.
⁴ Electrical Engineering, Stanford University, Stanford, 94305, California, United States.

Abstract

Digitally-delivered healthcare is well suited to address current inequities in the delivery of care due to barriers of access to healthcare facilities. As the COVID-19 pandemic phases out, we have a unique opportunity to capitalize on the current familiarity with telemedicine approaches and continue to advocate for mainstream adoption of remote care delivery. In this paper, we specifically focus on the ability of GuessWhat? a smartphone-based charades-style gamified therapeutic intervention for autism spectrum disorder (ASD) to generate a signal that distinguishes children with ASD from neurotypical (NT) children. We demonstrate the feasibility of using "in-the-wild", naturalistic gameplay data to distinguish between ASD and NT by children by training a random forest classifier to discern the two classes (AU-ROC = 0.745, recall = 0.769). This performance demonstrates the potential for GuessWhat? to facilitate screening for ASD in historically difficult-to-reach communities. To further examine this potential, future work should expand the size of the training sample and interrogate differences in predictive ability by demographic.

Abstract

Grants and funding