MIT Media LAb, Cambridge, MA

September 27, 2011

Talks, discussion, and poster presentations related to the development and use of technologies for the diagnosis and treatment of autism and other behavioral disorders

Guest Speakers

Jim Rehg (video)
TITLE: Behavior Imaging and the Study of Autism
ABSTRACT:Current research efforts in Behavior Imaging, a new research field which encompasses the measurement, modeling, analysis, and visualization of social and communicative behaviors from multi-modal sensor data will be described. The goal of our research in Behavior Imaging is to develop computational methods that can support the fine-grained and large-scale measurement and analysis of social behaviors, with the potential to positively impact diagnosis and treatment. An overview of these research efforts in Behavior Imaging, with a particular emphasis on the use of computer vision techniques, specifically a new approach to video analysis based on the concept of temporal causality, which leverages a novel representation of video events as multiple point processes as well as a presention of results for retrieving and categorizing social interactions in collections of real-world video footage will be described.
BIOGRAPHYJames M. Rehg (pronounced "ray") is a Professor in the School of Interactive Computing at the Georgia Institute of Technology, where he is the Director of the Center for Behavior Imaging, co-Director of the Computational Perception Lab, and Associate Director of Research in the Center for Robotics and Intelligent Machines. He received his Ph.D. from CMU in 1995 and worked at the Cambridge Research Lab of DEC (and then Compaq) from 1995-2001, where he managed the computer vision research group. He received the National Science Foundation (NSF) CAREER award in 2001, and the Raytheon Faculty Fellowship from Georgia Tech in 2005. He and his students have received a number of best paper awards, including best student paper awards at ICML 2005 and BMVC 2010. Dr. Rehg is active in the organizing committees of the major conferences in computer vision, most-recently serving as the General co-Chair for IEEE CVPR 2009. He has served on the Editorial Board of the International Journal of Computer Vision since 2004. He has authored more than 100 peer-reviewed scientific papers and holds 23 issued US patents. Dr. Rehg is currently leading a multi-institution effort to develop the science and technology of Behavior Imaging, funded by an NSF Expedition award.

Cathy Lord (video)
Privacy regulations mandate Dr. Lord's talk only be made available to those people who originally RSVPed to attend the workshop and consequently have been provided with the required password
TITLE: New Directions in Clinical Research in Autism: Opportunities, Priorities and Challenges
ABSTRACT: Advances and current issues in screening, diagnosis and treatment of autism spectrum disorders will be briefly reviewed. Future directions, including how to integrate basic and clinical research findings with practice, will be discussed. From a clinical perspective, opportunities and challenges afforded by technology will be considered in the light of priorities for research and practice.
BIOGRAPHY: Catherine Lord, Ph.D. is the Director of the Institute for Brain Development, a new autism center at New York Presbyterian Hospital, a collaborative effort among Weill-Cornell Medical College, Columbia College of Physicians and Surgeons and the New York Center for Autism, a not-for-profit autism advocacy organization. Previously she was the director of the University of Michigan Autism and Communication Disorders Center (UMACC) and the Urie Bronfenbrenner Collegiate Professor of Psychology, Psychiatry and Pediatrics.

Takeo Kanade (video)
TITLE: First Person Vision
ABSTRACT: For behavior imaging to understand the behavior and environment of a person, the surveillance metaphor is traditional; that is, observe her and her interaction with other people and environment from a camera placed in the environment. We argue that instead the First-Person Vision that senses the environment and her activities from her point of view is more advantageous with readily available information about her head motion and gaze. This talk will present the current progresses in the First Person Vision – the ideas, devices, algorithms, and example applications.
BIOGRAPHY: Takeo Kanade is the U. A. and Helen Whitaker University Professor of Computer Science and Robotics and the director of Quality of Life Technology Engineering Research Center at Carnegie Mellon University. He received his Doctoral degree in Electrical Engineering from Kyoto University, Japan, in 1974. After holding a faculty position in the Department of Information Science, Kyoto University, he joined Carnegie Mellon University in 1980.

Panel Discussion I : Jim Rehg, Cathy Lord, and Takeo Kanade (video)

Jeff Cohn (video)
TITLE: Advances in Behavioral Science using Automated Facial and Vocal Image Analysis and Synthesis
ABSTRACT: Computational behavioral science is making significant efforts in analysis and understanding of naturally occurring behavior. Active appearance models are an especially exciting approach. They may be used both to measure naturally occurring facial behavior and to synthesize near-photorealistic real-time avatars to experimentally perturb and reveal the dynamics of social behavior.
BIOGRAPHY: Jeffrey Cohn is Professor of Psychology at the University of Pittsburgh and Adjunct Faculty at the Robotics Institute, Carnegie Mellon University. He received his PhD in psychology from the University of Massachusetts at Amherst. Dr. Cohn has led interdisciplinary and inter-institutional efforts to develop advanced methods of automatic analysis of facial expression and prosody and applied those tools to research in human emotion, interpersonal processes, social development, and psychopathology.

Jan van Santene (video)
TITLE: Applications of Natural Language Processing Technologies to Autism
ABSTRACT: Atypical features of language and speech have long been considered among the core symptoms of Autism Spectrum Disorders. However, the measurement of these features has proven to be elusive. Recent developments in Natural Language Processing (NLP; technologies ranging from speech signal processing and automatic speech recognition to stochastic parsing and automatic topic detection) are now on the verge of producing objective measures for these features, measures that can be obtained largely automatically and -- critically -- require no human judgment. This has the potential not only to aid in early detection and diagnosis, but also to help discover new behavioral phenotypes; in addition, their automation and minimal hardware requirements bode well for broader access to early detection and diagnosis. We will present a variety of results obtained by researchers at the Center for Spoken Language Understanding, including results on prosody and on statistical features of dialog.
BIOGRAPHY: Jan van Santen obtained a PhD in Mathematical Psychology at the University of Michigan in 1979, worked for 2 decades at Bell Labs Research on Natural Language Processing (text-to-speech synthesis, quantitative modeling of speech prosody), before joining and directing the Center for Spoken Language Understanding (CSLU) in 2000, then located at the Oregon Graduate Institute and now at the Oregon Health & Science University (OHSU)

Panel Discussion II : Jeff Cohn, Jan van Santen, Gregory Abowd (video)

Organized by:

Jim Rehg Georgia Institute of Technology

Matthew Goodwin Northeastern University

Gregory Abowd Georgia Institute of Technology

Stephanie Tofighi Georgia Institute of Technology