Sam Harrison

Code

PROFUMO: Inference of PRObabilistic FUnctional MOdes from fMRI data

Links

ORCiD: 0000-0002-5886-2389

Google Scholar Profile

Publons: 1241796

research overview

The main aim of my research at Oxford is to develop novel techniques that identify brain networks by examining the spontaneous activity of the brain at rest, primarily working with Steve Smith and Mark Woolrich in the FMRIB & OHBA Analysis Groups.

To this end, we have developed PROFUMO, a Bayesian framework for inferring resting-state networks from fMRI data. The code is available at https://git.fmrib.ox.ac.uk/samh/profumo. Crucially, we aim to do this in a way that captures for the variability of these networks across subjects. In our model, this variability can either manifest itself as differences in the spatial location or in the temporal dynamics of these networks.

We are using cutting edge data from the Human Connectome Project to investigate how we can relate this variability in spontaneous brain activity to both behavioural and genetic factors, and have done extensive work with Janine Bijsterbosch on how to interpret subject variability.

More recently, Rezvan Farahibozorg has been doing amazing work looking at extending PROFUMO to very large-scale datasets like the UK Biobank.

BIOGRAPHY

I graduated from Cambridge with an MEng in Information and Computer Engineering. Upon completion, I began the Life Sciences Interface Doctoral Training Centre at Oxford, where the focus on mathematical and computational techniques for the biomedical sciences naturally led onto a DPhil and postdoc in neuroimaging analysis techniques. I was based at FMRIB & OHBA between October 2012 and February 2018.

Between April 2018 and May 2020, I had a postdoctoral fellowship based in the Translational Neuromodeling Unit (TNU) at ETH Zürich. There, I looked at advanced Bayesian inference techniques for effective connectivity, as well as models for physiological noise in fMRI.

As of October 2020, I am based at the New Zealand Brain Research Institute, looking at predictors of Apathy in Parkinson's with Campbell Le Heron. You can find out more about my research there at my NZBRI webpage.