Rapid classification of autism for public health surveillance

This presentation given August 3, 2017 describes work toward applying machine learning methods to CDC’s autism surveillance program. CDC’s population-based autism surveillance is labor-intensive and costly, as it requires clinicians to manually review children’s medical and educational records for descriptions of autism symptoms. Using the words in these records, our team is building algorithms to predict which children will meet the surveillance case definition for autism. This talk describes our early results, recent progress, and perspectives gained from working with textual data.

Presenter

Matthew Maenner, PhD, Epidemiologist, National Center on Birth Defects and Developmental Disabilities, US Centers for Disease Control and Prevention