Abstract #0434
Definition of connection significance using probabilistic tractography
Morris D, Embleton K, Parker G
University of Manchester
Probabilistic diffusion tractography methods conventionally define probability in terms of the frequency with which a particular connection is observed during a Monte Carlo process. This is not an ideal representation of connection probability as it does not explicitly take into account the possibility that the observed frequency of connection could have been observed by chance. This leads to distance-related artefacts in the probability of connection maps. We demonstrate that the use of a null distribution of connection, together with appropriate statistical analysis, allows for the definition of connection significance from a seed voxel and also corrects for distance-related artefacts.