Current Portfolio
- Epilepsy
- Bone disease
- Nanotoxicity testing
- Abdominal pain studies
- Brain cell imaging
- Huntington's disease
- Breast Cancer
- Viral encephalitis
- Chronic pain in arthritis and fibromyalgia
- Bacterial infection in cystic fibrosis
- Advances in human neuroscience
- Neurofibromatosis
- Hypersecretion of mucus in asthma
- Premature birth
- A clearer picture of pain relief
- Tissue engineering of human liver
- Creating knockout tissues
- Skin cancer
- Magnetoencephalography (MEG)
Quick Links
Brain cell imaging
 |
2010 – 2013 Research Assistant Fellowship Development and experimental validation of an information theoretic approach to the multimodal integration of human EEG and fMRI dataDr. Andrew P. Bagshaw |
The computerised representation of information about external stimuli and internal cognitive processes in neural systems is traditionally studied using invasive electrophysiological recordings in animals such as non-human primates and small animals. This can include inserting electrodes into cats' brains and monitoring their brain activity when not under anasthesia.
Recently, non-invasive techniques have been developed for use in humans, of which combined EEG-fMRI (Electroencephalography-functional Magnetic Resonance Imaging) holds considerable promise in terms of its potential spatiotemporal resolution. However, the best way to integrate EEG and fMRI data is not clear. The proposed projects aims at adapting an information theoretic framework developed in the study of neuronal population coding for non-invasive human brain imaging. The project comprises three stages:
i) The first stage will lay the theoretical and computational foundations to apply information theoretic concepts to simultaneously acquired EEG and fMRI data; in particular, an EEG-fMRI specific information breakdown scheme will be developed and its estimation facilitated using analytical results of Gaussian mixtures models and numerical simulations based on linear Gaussian models.
ii) The second stage will validate the developed methodology of the first stage using a basic visual stimulation paradigm. The spatiotemporal dynamics of information representation of basic visual features (contrast, spatial frequency) will be determined from combined EEG-fMRI data.
iii) The third stage the newly developed framework will be applied to an advanced cognitive-behavioural paradigm enabling the information theoretic characterization of the human perceptual decision system.
This project will lay important groundwork for the application of information theoretic concepts to non-invasive human brain imaging data. It will facilitate the integration of findings from invasive animal studies with human imaging data; furthermore, it will provide insight into the representation of information in the healthy human brain in both simple and complex cognitive tasks, a necessary prerequisite to understand its pathological alteration in neurodegenerative diseases such as epilepsy, depression, schizophrenia, Alzheimer’s and Parkinson’s disease.