Time-Series Prediction of Proximal Aggression Onset in Minimally-Verbal Youth with Autism Spectrum Disorder Using Physiological Biosignals

TL;DR

This study demonstrates that physiological biosignals can predict aggressive behavior in minimally-verbal youth with autism spectrum disorder within one minute before onset, using time-series analysis and machine learning models.

Contribution

It introduces a novel approach applying ridge-regularized logistic regression to physiological biosignals for early aggression prediction in MV-ASD youth.

Findings

Successful prediction of aggression onset 1 minute prior

Feasibility of real-time biosignal analysis in clinical settings

Proof-of-concept for physiological markers as early warning signs

Abstract

It has been suggested that changes in physiological arousal precede potentially dangerous aggressive behavior in youth with autism spectrum disorder (ASD) who are minimally verbal (MV-ASD). The current work tests this hypothesis through time-series analyses on biosignals acquired prior to proximal aggression onset. We implement ridge-regularized logistic regression models on physiological biosensor data wirelessly recorded from 15 MV-ASD youth over 64 independent naturalistic observations in a hospital inpatient unit. Our results demonstrate proof-of-concept, feasibility, and incipient validity predicting aggression onset 1 minute before it occurs using global, person-dependent, and hybrid classifier models.