BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Decoding Epileptogenesis: A Dynamical System Approach
DTSTART:20160208T120000
DTEND:20160208T130000
DTSTAMP:20260407T045504Z
UID:bee3693134813b62c8490905235853f3cbb04f8cf765a6bfe65c272f
CATEGORIES:Conferences - Seminars
DESCRIPTION:Professor Francois G Meyer\, University of Colorado at Boulder
 \nEpilepsy is the most common chronic neurological disorder\, affecting ov
 er 65 million people worldwide. This work addresses the design of a bioma
 rker of epileptogenesis. A reliable biomarker can improve the treatment o
 f the disease\, reduce health-care costs\, and help develop new drugs.\nWe
  describe here the recent results of a multidisciplinary effort to design 
 a biomarker that can actively and continuously decode the progressive cha
 nges in neuronal organization leading to epilepsy\, a process known as ep
 ileptogenesis. Using an animal model of acquired epilepsy\, we chronicall
 y record hippocampal evoked potentials elicited by an auditory stimulus. U
 sing a set of reduced coordinates\, our algorithm can identify universal 
 smooth low-dimensional configurations of the auditory evoked potentials t
 hat correspond to distinct stages of epileptogenesis. We use a hidden Mar
 kov model to learn the dynamics of the evoked potential\, as it evolves al
 ong these smooth low-dimensional subsets. We provide experimental evidenc
 e that the biomarker is able to exploit subtle changes in the evoked pote
 ntial to reliably decode the stage of epileptogenesis and predict whether
  an animal will eventually recover from the injury\, or develop spontaneou
 s seizures. This is work in collaboration with Daniel Barth\, Alexander B
 enison\, and Zachariah Smith.\nPlease note this Seminar is videotranfered 
 to EPLF at SV 3715
LOCATION:CB H8-01 144-165 AND EPFL SV 3715 http://cnp.epfl.ch/files/conten
 t/sites/cnp/files/Seminars/PlanSallesCampusBiotechH8-01.pdf
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR