Machine learning algorithms for the early detection of behavioral health disorders in children

Abstract: For children who have behavioral health disorders, starting therapies early is key for their future quality of life. Early and accurate screening tools are vital.

We present a clinically validated machine learning driven App to diagnose autism. We give an overview of the development and optimization of multiple machine learning algorithms that use different media to identify autism, as well as the combination of their results, and the clinical validation of the algorithms’ performance. We also show initial research into using an interactive story telling App to screen for speech and language problems. Finally, we present a technique for simultaneously screening for multiple conditions and show the potential for identifying ADHD alongside autism.

Bio: Ford Garberson develops machine learning algorithms to screen for childhood cognitive conditions such as autism at Cognoa. This development work ranges from analyses to identify and compensate for defects in data that will be used in machine learning to development and optimization of the algorithms to analyzing clinical study results to validate the performance of the algorithms in real life.

Prior to joining Cognoa, he spent two years developing machine learning algorithms for energy and consumer comfort optimization in smart thermostats at a startup called EcoFactor. Prior to that he spent nine years running statistical analyses on petabytes of data collected at particle colliders, including the Large Hadron Collider in Switzerland.