BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Affective computing for assessing treatments in psychiatry - Justi
 n Dauwels 
DTSTART:20170901T153000Z
DTEND:20170901T163000Z
UID:TALK78901@talks.cam.ac.uk
CONTACT:Fumiya Iida
DESCRIPTION:Abstract:\nIn order to evaluate therapies in psychiatry\, it i
 s crucial to able to assess the negative symptoms of mentally ill people. 
 In an ongoing study\, we are designing data-driven methods\, inspired by a
 ffective computing\, to assess the negative symptoms of mentally ill peopl
 e. \n\nThe underlying idea is that the severity of negative symptoms and i
 mpairments in neurocognition and social cognition in mentally ill people c
 an be evaluated objectively and conveniently by speech and motor character
 istics. Impoverished speech patterns (e.g.\, taking prolonged time to resp
 ond to a question\, decreased percentage of speaking in a conversation\, e
 tc.) and motor patterns are often associated with higher severity of negat
 ive symptoms\, and with poorer cognitive and functioning performance.\n\nS
 pecifically\, we conducted an observational study involving 56 patients wi
 th schizophrenia and 26 healthy participants without any mental illness. A
 ll participants underwent a battery comprising of clinical\, neuro- and so
 cial cognitive and functioning assessments in a single visit. Audio and vi
 deo recording were conducted during a social responsivity task. We designe
 d algorithms to extract the prosodic and conversational speech features an
 d motor features from the audio and video clips\, respectively\, and appli
 ed pattern recognition methods to identify objective sociological metrics 
 for assessing the severity of negative symptoms\, cognitive performance an
 d functioning levels of patients.\n\nIn the future\, after additional opti
 mization and testing\, this technology might be used as a convenient\, eff
 icient and objective evaluation tool in routine clinical practice\, potent
 ially integrated in mobile apps or social robots.\n\nIn this talk\, we wil
 l present our results so far\, and will elaborate on the challenges and pl
 ans for future research.\n\n\n\n\n\nBio:\nDr. Justin Dauwels is an Associa
 te Professor of the School of Electrical and Electronic Engineering at the
  Nanyang Technological University (NTU) in Singapore. He also serves as De
 puty Director of the ST Engineering – NTU corporate lab\, which comprise
 s 100+ PhD students\, research staff and engineers\, developing novel auto
 nomous systems for airport operations and transportation. \n\nHis research
  interests are in data analytics with applications to intelligent transpor
 tation systems\, autonomous systems\, and analysis of human behaviour and 
 physiology. He obtained his PhD degree in electrical engineering at the Sw
 iss Polytechnical Institute of Technology (ETH) in Zurich in December 2005
 . Moreover\, he was a postdoctoral fellow at the RIKEN Brain Science Insti
 tute (2006-2007) and a research scientist at the Massachusetts Institute o
 f Technology (2008-2010). He has been a JSPS postdoctoral fellow (2007)\, 
 a BAEF fellow (2008)\, a Henri-Benedictus Fellow of the King Baudouin Foun
 dation (2008)\, and a JSPS invited fellow (2010\, 2011). \n\nHis research 
 on intelligent transportation systems has been featured by the BBC\, Strai
 ts Times\, Lianhe Zaobao\, Channel 5\, and numerous technology websites. H
 is research team has won several best paper awards at international confer
 ences. Besides his academic efforts\, the team of Dr. Justin Dauwels also 
 collaborates intensely with local start-ups\, SMEs\, and agencies\, in add
 ition to MNCs\, in the field of data-driven transportation\, logistics\, a
 nd human behavior analysis.\n
LOCATION:Bio-Inspired Robotics Lab\,  Engineering\, Department of
END:VEVENT
END:VCALENDAR