Augmenting gesture recognition with erlang-cox models to identify neurological disorders in premature babies

In this paper we demonstrate a Markov model based technique for recognizing gestures from accelerometers that explicitly represents duration. We do this by embedding an Erlang-Cox state transition model, which has been shown to accurately represent the first three moments of a general distribution, within a Dynamic Bayesian Network (DBN). The transition probabilities in the DBN can be learned via Expectation-Maximization or by using closed-form solutions. We test this modeling technique on 10 hours of data collected from accelerometers worn by babies pre-categorized as high-risk in the Newborn Intensive Care Unit (NICU) at UCI. We show that by treating instantaneous machine learning classification values as observations and explicitly modeling duration, we improve the recognition of Cramped Synchronized General Movements, a motion highly correlated with an eventual diagnosis of Cerebral Palsy.

[1]  Giovanni Cioni,et al.  An early marker for neurological deficits after perinatal brain lesions , 1997, The Lancet.

[2]  Bernt Schiele,et al.  A model for human interruptability: experimental evaluation and automatic estimation from wearable sensors , 2004, Eighth International Symposium on Wearable Computers.

[3]  A. Okely,et al.  Methodological considerations in using accelerometers to assess habitual physical activity in children aged 0-5 years. , 2009, Journal of science and medicine in sport.

[4]  Jeffrey M Gossett,et al.  Effect of early intervention on 8-year growth status of low-birth-weight preterm infants. , 2009, Archives of pediatrics & adolescent medicine.

[5]  Donald J. Patterson,et al.  Involuntary gesture recognition for predicting cerebral palsy in high-risk infants , 2010, International Symposium on Wearable Computers (ISWC) 2010.

[6]  Lucy E. Dunne,et al.  Garment-based body sensing using foam sensors , 2006, AUIC.

[7]  Michael A Babcock,et al.  Injury to the Preterm Brain and Cerebral Palsy: Clinical Aspects, Molecular Mechanisms, Unanswered Questions, and Future Research Directions , 2009, Journal of child neurology.

[8]  Zhen Wang,et al.  uWave: Accelerometer-based Personalized Gesture Recognition and Its Applications , 2009, PerCom.

[9]  Ingeborg Krägeloh-Mann,et al.  Cerebral palsy update , 2009, Brain and Development.

[10]  Svetha Venkatesh,et al.  Activity recognition and abnormality detection with the switching hidden semi-Markov model , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[11]  Paul Lukowicz,et al.  Using multiple sensors for mobile sign language recognition , 2003, Seventh IEEE International Symposium on Wearable Computers, 2003. Proceedings..

[12]  Ian H. Witten,et al.  The WEKA data mining software: an update , 2009, SKDD.

[13]  G Cioni,et al.  Comparison between observation of spontaneous movements and neurologic examination in preterm infants. , 1997, The Journal of pediatrics.

[14]  Desney S. Tan,et al.  Enabling always-available input with muscle-computer interfaces , 2009, UIST '09.

[15]  Yang Li,et al.  Gestures without libraries, toolkits or training: a $1 recognizer for user interface prototypes , 2007, UIST.

[16]  Thad Starner,et al.  MAGIC: a motion gesture design tool , 2010, CHI.

[17]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[18]  Pattie Maes,et al.  WUW - wear Ur world: a wearable gestural interface , 2009, CHI Extended Abstracts.

[19]  Roslyn N Boyd,et al.  A systematic review of the clinimetric properties of neuromotor assessments for preterm infants during the first year of life , 2008, Developmental medicine and child neurology.

[20]  Henry A. Kautz,et al.  Fine-grained activity recognition by aggregating abstract object usage , 2005, Ninth IEEE International Symposium on Wearable Computers (ISWC'05).

[21]  H. Prechtl,et al.  State of the art of a new functional assessment of the young nervous system. An early predictor of cerebral palsy. , 1997, Early human development.

[22]  Shunzheng Yu,et al.  Hidden semi-Markov models , 2010, Artif. Intell..

[23]  Blake Hannaford,et al.  A Hybrid Discriminative/Generative Approach for Modeling Human Activities , 2005, IJCAI.

[24]  G. Birbeck,et al.  A Systematic Review of Neuroimaging for Cerebral Palsy , 2007, Journal of child neurology.

[25]  Alan R Fleischman,et al.  Surgeon General’s Conference on the Prevention of Preterm Birth , 2009, Obstetrics and gynecology.

[26]  Pai H. Chou,et al.  Eco: ultra-wearable and expandable wireless sensor platform , 2006, International Workshop on Wearable and Implantable Body Sensor Networks (BSN'06).

[27]  Mor Harchol-Balter,et al.  Closed form solutions for mapping general distributions to quasi-minimal PH distributions , 2006, Perform. Evaluation.

[28]  Satoru Morita,et al.  Time Series Analysis of Spontaneous Upper-Extremity Movements of Premature Infants With Brain Injuries , 2008, Physical Therapy.

[29]  Daphne Koller,et al.  Expectation Maximization and Complex Duration Distributions for Continuous Time Bayesian Networks , 2005, UAI.

[30]  G Cioni,et al.  Activity patterns assessed throughout 24-hour recordings in preterm and near term infants. , 2001, Developmental psychobiology.

[31]  Ling Bao,et al.  Activity Recognition from User-Annotated Acceleration Data , 2004, Pervasive.

[32]  Thomas Schmitz-Rode,et al.  Movement analysis by accelerometry of newborns and infants for the early detection of movement disorders due to infantile cerebral palsy , 2010, Medical & Biological Engineering & Computing.

[33]  Daniel P. Siewiorek,et al.  Performance Analysis of an HMM-Based Gesture Recognition Using a Wristwatch Device , 2009, 2009 International Conference on Computational Science and Engineering.

[34]  Niels Henze,et al.  Gesture recognition with a Wii controller , 2008, TEI.

[35]  Yoav Freund,et al.  Experiments with a New Boosting Algorithm , 1996, ICML.

[36]  A. Rowlands.  Accelerometer assessment of physical activity in children: an update. , 2007, Pediatric exercise science.

[37]  Emma C. Cameron,et al.  The Effects of an Early Physical Therapy Intervention for Very Preterm, Very Low Birth Weight Infants: A Randomized Controlled Clinical Trial , 2005, Pediatric physical therapy : the official publication of the Section on Pediatrics of the American Physical Therapy Association.

[38]  N. L. Adams,et al.  Surgeon Generalʼs Conference on the Prevention of Preterm Birth , 2010 .

[39]  Per Ola Kristensson,et al.  Continuous recognition of one-handed and two-handed gestures using 3D full-body motion tracking sensors , 2012, IUI '12.

[40]  Kent Lyons,et al.  GART: The Gesture and Activity Recognition Toolkit , 2007, HCI.

[41]  Paul Lukowicz,et al.  Active Capacitive Sensing: Exploring a New Wearable Sensing Modality for Activity Recognition , 2010, Pervasive.

[42]  Paul Lukowicz,et al.  Performance metrics for activity recognition , 2011, TIST.

[43]  Mijna Hadders-Algra,et al.  The assessment of minor neurological dysfunction in infancy using the Touwen Infant Neurological Examination: strengths and limitations , 2010, Developmental medicine and child neurology.

[44]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[45]  Daqing Zhang,et al.  Gesture Recognition with a 3-D Accelerometer , 2009, UIC.

[46]  R. Reading,et al.  Cramped synchronized general movements in preterm infants as an early marker for cerebral palsy. , 2002 .

[47]  Alex Pentland,et al.  Graphical Models for Recognizing Human Interactions , 1998, NIPS.