From Mechanisms to Engineered Interventions for Epilepsy  





The International Workshop on Seizure Prediction (IWSP) is an ongoing series of international meetings on seizure prediction, seizure dynamics and seizure control founded in 2002.  IWSP has been bringing together researchers from a wide range of backgrounds including epileptology, neurosurgery, neurosciences, physics, mathematics, computer science, and engineering.  The workshops provide a regular, structured forum for the exchange of ideas, dissemination of findings, and evaluation of metrics to gauge progress.  The workshops also form the basis for development of future collaborative projects.

Background and Significance

Epilepsy, one of the most prevalent neurological disorders, affects approximately 1% (>60 million) of the world’s population. In an estimated 20 million of the patients, seizures are not controlled even by multiple anti-seizure drugs. Uncontrolled seizures have significant negative impacts for quality of life, and have been shown to correlate with reduced life expectancy.

In the past 25 years, an interdisciplinary effort involving epileptologists, engineers, physicists, mathematicians, neurosurgeons and neuroscientists, the International Seizure Prediction Group (ISPG), has attempted to approach the problem of seizure prediction. Seizure prediction in its most basic form is the identification of a preseizure state, and serves to remove one of the most distressing aspects of epilepsy for patients: the unpredictability of seizures. The ISPG has also addressed the closely related topics of seizure localization and seizure evolution.

At a more fundamental level, the goal of seizure prediction is to identify the underlying mechanisms of seizure generation and to engineer systems that will detect those dynamics and provide for intervention, i.e. seizure control. As the field has matured, it has become clear that before we can truly make progress, we need to shift our focus to answer some crucial, basic, unresolved questions that speak to these basic mechanisms (basic mechanism CBUQs). These include:

  • What constitutes a human focal-onset seizure at the level of neural firing?
  • How do seizures evolve spatially and terminate at all scales from the cell to organism?
  • What information is - and is not - reflected in the EEG, our primary clinical diagnostic tool? What extensions to EEG – scales of electrophysiological measures - should be used?
  • What other dynamics and cycles such as sleep and stress couple into the mechanisms of seizure generation?
  • What other non-EEG measures of brain and behavior can be utilized to reveal the underlying mechanisms of seizure generation?

While these questions seem simple, it is rather astounding that so little is known about them, even though 50 years have passed since the neural signature of a seizure (the paroxysmal depolarizing shift) was first described. In our view, this lack of insight into seizure generation mechanisms and how it couples into the daily life of patients is a major barrier to many potentially valuable therapeutic epilepsy treatments. A major goal of the meeting is to focus attention on new directions for investigation that have the potential to bring real progress to the goals of seizure localization and prediction, and that require the combined expertise of clinical neurophysiology, bioengineering, and epilepsy neuroscience.

Traditionally, epilepsy neurophysiology research has been divided into two camps: a clinical approach that begins with EEG recordings and attempts to divine information on the location and behavior of neural activity, and a “bench” approach that uses a simplified model of patterns thought to be important during clinical seizures, and investigates interactions between sites or other physiological structures. Very few investigators have been able to link these two views. By bringing together investigators from both sides of the divide and focusing attention on the interaction between EEG and neural activity, we hope to facilitate new collaborations and research directions that will advance the field in meaningful ways.

A similar divide exists between the modeling community and experimentalists, clinicians, and engineers. What and how we measure limit what we know about a system. Although EEG is the diagnostic linchpin for surgical epilepsy treatment, there is an attitude in the neuroscience community that discounts EEG as a mostly meaningless and dirty metric of brain activity. Unfortunately, because epilepsy is a human disease, the reality is that EEG is currently the most accessible and widespread view into the dynamics of human brain. The overall goal of the Seizure Prediction Group is to develop treatments for epilepsy especially based on closed loop prediction and interventions. While this goal has not changed, the group has recognized that the focus of attention needs to shift away from a biological mechanism-free algorithmic approach to EEG and related electrophysiology recordings, and toward a better understanding of underlying physiological mechanisms as they specifically relate to clinical recordings, and a better insight into how to OBSERVE these mechanisms and their associated dynamics. It is the specific aim and design of the proposed meeting to effect this shift. We will push to close these divides and promote efforts that link theoretical and computational modeling to observation of real systems.

This combined approach has not, to our knowledge, been the main focus of any US or European conference held to date. The core ISPG group is one that is uniquely suited to address these questions. As a whole, the group has considerable expertise in clinical EEG interpretation, signal processing, computer modeling, and acquisition of high spatiotemporal resolution human and animal datasets. In addition, it has historically attracted researchers from the engineering community especially those able to and engaged in development of devices for treatment of epilepsy.

The questions we now seek to address, however, require the addition of epilepsy neuroscience expertise to the group’s skill set. While there is some epilepsy neuroscience expertise within the group, there is far more outside it. We therefore enlist scientists from outside the traditional ISPG with established publication records directly relevant to our topic areas as speakers and participants. Additionally, we plan to recruit junior faculty, residents pursuing careers as physician-scientists, and postdoctoral and graduate students who are interested in delving into this challenging area to attend and contribute posters.

International Workshop on Seizure Prediction (IWSP) series

The ISPG group has held, to date, five international workshops approximately every 18 months, alternating between the U.S. and Europe. The workshops have addressed topics encompassing seizure prediction, seizure generation, and seizure control.

Previous conferences in this series:

  • IWSP5 Dresden, Germany in September 2011
    organized by Ronald Tetzlaff, Christian Elger, and Klaus Lehnertz
  • IWSP3 Freiburg, Germany in September 2007
    organized by Andreas Schulze-Bonhage, Jens Timmer, and Björn Schelter
    (conference book)
  • IWSP2 Bethesda, MD, USA in April 2006
    organized by Leonidas Iasemidis, Chris Sackellares, Randall Stewart, and Joseph Pancrazio
  • IWSP1 Bonn, Germany in April 2002
    organized by Klaus Lehnertz and Brian Litt
    (summary paper)


Bahia Resort
San Diego



home | background | committees | program | registration | abstracts | sponsors | Bahia Resort Hotel | San Diego | other links | contact us