The Wall Lab at Stanford University

Menu

Machine Learning Diagnostic Classifiers

Overview

We developed eight classifiers using decision tree learning that performed optimally for classification of a wide range of individuals both on and off the autism spectrum. These classifiers are substantially shorter than the gold standard diagnostic instruments in the field today and both classifiers pinpoint several behavioral patterns that could guide future methods for expeditious observation-based screening and diagnosis in and out of clinical settings.

Specific Aims

  • To apply machine learning to medical records generated through the administration of standard diagnostic test procedures to find minimal sets of features that can be used in models to achieve highly accurate outcomes
  • To find and build efficient (mobile) ways to capture data needed by these models
  • To scale these approaches to reach a greater percentage of the population in need than what is possible today

Papers

  • KM Paskov, DP Wall (2018). A Low Rank Model for Phenotype Imputation in Autism Spectrum Disorder, AMIA Summits on Translational Science Proceedings 2017, 178 Full Text

  • H Abbas, F Garberson, E Glover, DP Wall (2017). Machine learning approach for early detection of autism by combining questionnaire and home video screening, Journal of the American Medical Informatics Association Full Text

  • DP Wall, Dally R, Luyster R, Jung JY, DeLuca TF. Use of Artificial Intelligence to Shorten the Behavioral Diagnosis of Autism. PLoS ONE. 2012 Aug 27. Full Text

  • DP Wall, Kosmicki J, DeLuca TF, Harstad E, Fusaro VA. Use of machine learning to shorten observation-based screening and diagnosis of autism. Transl Psychiatry. 2012 Apr 10. Full Text

We have developed machine learning classifiers to distinguish ASD children from typically-developing children, using feature extraction and sparsity-enforcing classifiers in order to find feature sets from ADOS (modules 2 and 3).

  • S Levy, M Duda, N Haber, DP Wall (2017). Sparsifying machine learning models identify stable subsets of predictive features for behavioral detection of autism, Molecular autism 8 (1), 65 Full Text

  • H Abbas, F Garberson, E Glover, DP Wall (2017). Machine learning for early detection of autism (and other conditions) using a parental questionnaire and home video screening, Big Data (Big Data), 2017 IEEE International Conference on, 3558-3561 Full Text

  • J.A. Kosmicki, V. Sochat, M. Duda, and D.P. Wall. Searching for a minimal set of behaviors for autism detection through feature selection-based machine learning. Translational Psychiatry. February 2015; doi:10.1038/tp.2015.7. Full Text

Using archival data, we used feature selection and classification techniques to choose a subset of features on the Social Responsiveness Scale (SRS) from which we can distinguish ASD and ADHD. In work in submission, this method has been validated on data gathered through a mobile survey system.

  • M Duda, N Haber, J Daniels, DP Wall (2017). Crowdsourced validation of a machine-learning classification system for autism and ADHD, Translational psychiatry 7 (5), e1133 Full Text

  • M Duda, R Ma, N Haber, DP Wall (2017). Use of machine learning for behavioral distinction of autism and ADHD, Translational psychiatry 6 (2), e732 Full Text

  • Duda, M., Ma, R., Haber, N., & Wall, D.P. (2016). Use of machine learning for behavioral distinction of autism and ADHD. Translational Psychiatry, 6(732), doi: 10.1038/tp2015.221. Published online 9 February 2016. Full Text