Edinburgh Research Explorer

Using a Combination of a Mixture Model and Topological FDR in the Context of Presurgical Planning

Research output: Contribution to conferencePoster

Original languageEnglish
Publication statusPublished - 2011
Event17th Annual Meeting of the Organization for Human Brain Mapping - Centre des Congrès de Québec, Québec, Canada
Duration: 26 Jun 201130 Jun 2011

Conference

Conference17th Annual Meeting of the Organization for Human Brain Mapping
CountryCanada
CityQuébec
Period26/06/1130/06/11

Abstract

Functional Magnetic Resonance Imaging (fMRI) can be used in presurgical planning to establish which parts of the brain are functionally eloquent so that they can be avoided in surgical procedures such as tumour resection. In contrast to cognitive neuroscience where fMRI scans are averaged over multiple healthy controls, in clinical use only a single scan is available for analysis. This poses challenges, since data obtained from single subjects can be very noisy and the signal strength can vary between patients due to their anatomy, time of scanning and degree of cooperation. Because of this variation some labs use different thresholds selected manually for every patient. To avoid this potentially subjective approach, we propose a new method that adapts to changing signal strength while still maintaining sound statistical properties.

Event

17th Annual Meeting of the Organization for Human Brain Mapping

26/06/1130/06/11

Québec, Canada

Event: Conference

ID: 18241231