Detecting and Classifying Self-injurious Behavior in Autism Spectrum Disorder Using Machine Learning Techniques
- PDF / 723,779 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 6 Downloads / 214 Views
ORIGINAL PAPER
Detecting and Classifying Self‑injurious Behavior in Autism Spectrum Disorder Using Machine Learning Techniques Kristine D. Cantin‑Garside1 · Zhenyu Kong1 · Susan W. White2,3 · Ligia Antezana3 · Sunwook Kim1 · Maury A. Nussbaum1,4
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Traditional self-injurious behavior (SIB) management can place compliance demands on the caregiver and have low ecological validity and accuracy. To support an SIB monitoring system for autism spectrum disorder (ASD), we evaluated machine learning methods for detecting and distinguishing diverse SIB types. SIB episodes were captured with body-worn accelerometers from children with ASD and SIB. The highest detection accuracy was found with k-nearest neighbors and support vector machines (up to 99.1% for individuals and 94.6% for grouped participants), and classification efficiency was quite high (offline processing at ~ 0.1 ms/observation). Our results provide an initial step toward creating a continuous and objective smart SIB monitoring system, which could in turn facilitate the future care of a pervasive concern in ASD. Keywords Activity recognition · Autism · Machine learning · Wearable sensors Abbreviations ASD Autism spectrum disorder SIB Self-injurious behavior SMM Stereotypical motor movement SVM Support vector machine DA Discriminant analysis DT Decision tree nB Naïve Bayes kNN k-nearest neighbor NN Neural networks SRC Sparse representation classification
* Maury A. Nussbaum [email protected] 1
Department of Industrial and Systems Engineering, Virginia Tech, Blacksburg, VA 24060, USA
2
Department of Psychology, The University of Alabama, Tuscaloosa, AB, USA
3
Department of Psychology, Virginia Tech, Blacksburg, VA 24060, USA
4
Department of Industrial and System Engineering, Virginia Tech, 250 Durham Hall (0118), Blacksburg, VA 24061, USA
Introduction Self-injurious behavior (SIB) is a leading cause of hospitalization for children with autism spectrum disorder (ASD) (Kalb et al. 2016). SIB may be repetitive and rhythmic, and can include behaviors such as head banging and self-hitting (Minshawi et al. 2014). SIB can result in physical damage (Rooker et al. 2018), including abrasions, lacerations, and contusions, especially as SIB commonly continues beyond the initial age of onset (Minshawi et al. 2014; Taylor et al. 2011; Richards et al. 2016). However, early interventions can help prevent severe consequences and alleviate the long-term persistence of SIB (Kurtz et al. 2003). Children who exhibit severe SIB often lack the verbal and cognitive abilities to report their SIB occurrence and discuss their motivation. Applied behavioral analysis suggests that functional assessments (FAs) should thus be completed before an intervention to determine potential triggers of SIB (Williams et al. 2005; Pelios et al. 1999; Iwata et al. 1994). In typical applications, FA requires clinicians or trained caregivers to track behavior times and the events surrounding the behavior
Data Loading...
