Recognizing Mimicked Autistic Self-Stimulatory Behaviors Using HMMs

Westeyn, Tracy L.; Vadas, Kristin; Bian, Xuehai; Starner﻿﻿, Thad; Abowd, Gregory D.

doi:10.1109/iswc.2005.45

Cited by 68 publications

(47 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…E.g., Westeyn et al [30] used activity recognition based on acceleration signals for the identi cation of seven stereotype patterns in autistic children such as drumming or rocking. In a similar fashion, Goodwin et al [31] performed recognition of prede ned, periodic movements in autistic children.…”

Section: Discussionmentioning

confidence: 99%

Kinematic assessment of stereotypy in spontaneous movements in infants

et al. 2012

View full text Add to dashboard Cite

Movement variation constitutes a crucial feature of infant motor development. Reduced variation of spontaneous infant movements, i.e. stereotyped movements, may indicate severe neurological de cit at an early stage. Hitherto evaluation of movement variation has been mainly restricted to subjective assessment based on observation. is article introduces a method for quantitative assessment yielding an objective de nition of stereotyped movements which may be used for the prognosis of neurological de cits such as cerebral palsy ( ).Movements of 3-months-old infants were recorded with an electromagnetic tracking system facilitating the analysis of joint angles of the upper and lower limb. A stereotypy score based on dynamic time warping has been developed describing movements which are self-similar in multiple degrees of freedom. For clinical evaluation, this measure was calculated in a group of infants at risk for neurological disorders (n=54) and a control group of typically developing children (n=21) on the basis of spontaneous movements at the age of three months. e stereotypy score was related to outcome at the age of 24 months in terms of (n=10) or no-(n=53). Using the stereotypy score of upper limb movements cases could be identi ed with a sensitivity of 90% and a speci city of 96%. e corresponding score of the leg movements did not allow for valid discrimination of the groups. e presented stereotypy feature is a promising candidate for a marker that may be used as a simple and noninvasive quantitative measure in the prediction of . e method can be adopted for the assessment of infant movement variation in research and clinical applications.

show abstract

Section: Discussionmentioning

confidence: 99%

Kinematic assessment of stereotypy in spontaneous movements in infants

et al. 2012

View full text Add to dashboard Cite

show abstract

“…The tracked body part locations are then used to analyse for repetitive self-stimulatory behaviours. Westyn et al [7] used wrist band worn sensors to track the hand motion of the child and constructed a Hidden Markov Model using 3-axis readings obtained from the sensors to classify the behaviours. Ploetz et al [8] used sensors attached to the limbs to obtain acceleration data.…”

Section: Related Workmentioning

confidence: 99%

Detecting self-stimulatory behaviours for autism diagnosis

Rajagopalan

Goecke

2014

2014 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Autism Spectrum Disorders (ASD), often referred to as autism, are neurological disorders characterised by deficits in cognitive skills, social and communicative behaviours. A common way of diagnosing ASD is by studying behavioural cues expressed by the children. An algorithm for detecting three types of self-stimulatory behaviours from publicly available unconstrained videos is proposed here. The child's body is tracked in the video by a careful selection of poselet bounding box predictions using a nearest neighbour algorithm. A global motion descriptor -Histogram of Dominant Motions (HDM) -is computed using the dominant motion flow in the detected body regions. The motion model built using this descriptor is used for detecting the self-stimulatory behaviours. Experiments conducted on the recently released unconstrained SSBD video dataset show significant improvement in detection accuracy over the baseline approach. The robustness of the method is validated using benchmark action recognition datasets. The proposed poselet bounding box selection algorithm is validated against the ground truth annotation data provided with the UCF101 dataset.

show abstract

“…Westeyn et al used pattern recognition algorithms to analyze accelerometer readings for detecting stereotypical behaviors [10]. 69% of continuous recognition tests were accurately detected in [10] using Hidden Markov Models but the data were acquired from individuals mimicking the actual behaviors rather than collected from children with ASD.…”

Section: Related Workmentioning

confidence: 99%

Smartphone Based Autism Social Alert System

Chuah

DiBlasio

2012

2012 8th International Conference on Mobile Ad-Hoc and Sensor Networks (MSN)

View full text Add to dashboard Cite

Abstract-Autism currently affects 1 in every 88 American children impairing their social interactions, communications and daily living. Often, parents, educators, and researchers need to purchase expensive equipment to help autistic children cope with challenges in their daily living. In this paper, we present the Smartphone-Based Autism Social Alert (SASA) system which we design to help such children. The SASA system uses the inexpensive sensors embedded within smartphones to facilitate the study of the autistic children's behaviors by recording and analyzing data collected from such embedded sensors in smartphones carried by autistic children. Our system can automatically detect their stereotypical behaviors such that early interventions can be taken by caregivers or teachers. In addition, the system can correlate environmental sensor data streams, e.g. audio background, with the occurrence of stereotypical behaviors so as to identify potential environmental factors that may trigger such behaviors. We also include some preliminary classification results on the sensor data which we have collected from Androidbased phones using the WEKA J.48 classifier. Our preliminary results show that simple features extracted from accelerometer readings are sufficient to give high accuracy rates when training is performed on a per user per device basis. Our audio classifier which uses 12 MFCC coefficients, average zero crossing rate, and energy can give an accuracy of 78.6% when evaluated using audio traces collected for seven audio categories. Additional extensive experiments will be carried out in the near future at a nearby secondary school for autistic children.

show abstract

Recognizing Mimicked Autistic Self-Stimulatory Behaviors Using HMMs

Cited by 68 publications

References 4 publications

Kinematic assessment of stereotypy in spontaneous movements in infants

Kinematic assessment of stereotypy in spontaneous movements in infants

Detecting self-stimulatory behaviours for autism diagnosis

Smartphone Based Autism Social Alert System

Contact Info

Product

Resources

About