Semi-metric topology of functional brain networks: Sensitivity and specifity in autism spectrum and major depression disorder

Add to your list(s) Download to your calendar using vCal

• Tiago Simas, Department of Psychiatry, University of Cambridge
• Tuesday 09 April 2013, 11:00-12:00
• BCNI seminar room, Sir William Hardy Building, Downing Site.

If you have a question about this talk, please contact Mikail Rubinov.

The main motivation for this study on Autism spectrum and Major Depression disorder was the observation that Autism and Depression were on opposite extremes on a factor meta-analysis on a wide spectrum of psychiatry disorders done by Mana et al. In our study we support this observation through a novel type of analysis on Brain connectivity networks. This type of analysis studies the multi-dimensional topology of networks, which e.g. are embedded in non-metric spaces, through the embedding of its non-metric geometry into a metric geometry. In general those non-metric geometries are classified as semi-metrics, since they violate the triangle inequality for a given distance function. This type of analysis has been applied in other fields such as web mining, bioinformatics, etc., which involves in some way, mathematical relations. In this work, the semi-metric analysis done in functional connectivity brain networks, allow us to analyze the specialization and co-activity of the brain. We start measuring the global level of brain specialization for each group (autism, depression and control groups), where we found group statistical differences. Then we were able to identify specific regions where the patients have statistical different levels of specialization and/or co-activity. The depression group shows us a more specialized left-brain hemisphere and an increased co-activity in the right hemisphere, when compared with the control group. Autism group, in general shows a more specialized right hemisphere and an increase of co-activity on the left hemisphere, when compared with the control group. We then try to correlate the hemisphere specialization with the increase of co-activity on the opposite hemisphere, and found some brain areas that correlate highly in both patients groups (autism, depression) when compared with the control group.

This talk is part of the Brain Mapping Unit Networks Meeting and the Cambridge Connectome Consortium series.

This talk is included in these lists:

• BCNI seminar room, Sir William Hardy Building, Downing Site
• Brain Mapping Unit Networks Meeting and the Cambridge Connectome Consortium
• Department of Psychiatry talks stream

Note that ex-directory lists are not shown.