Advanced neuroimaging techniques such as magnetic resonance imaging (MRI), positron emission tomography (PET), electroencephalogram (EEG) and magnetoencephalography (MEG) help us to visualise structural, molecular and functional changes in the brain. These brain scans can improve our understanding, diagnosis and treatments of neurological and psychiatric diseases such as Alzheimer’s disease, Parkinson’s disease, stroke and depression. Neuroimaging plays an increasingly important role in early detection of neurodegenerative diseases at stages when treatments can be most effective.
We have developed several analysis software packages for neuroimaging data, such as the Matlab RSA Toolbox.
In recent years, AI has obtained human-level performance in many areas such automatic face or speech recognition and has even surpassed human intelligence in situations such as playing chess. In our research, AI is used to analyse and recognise neuroimaging and clinical data. Another classes of AI models can be seen as ‘virtual’ patients capturing their cognitive dysfunctions on computer simulations. By testing the models with neuroimaging data from real patients, we have obtained deeper knowledge about not only ‘what’ has gone wrong in patients’ brains but ‘why’ they experience the symptoms such as recurrent visual hallucinations. Ultimately, the models allow us to develop and test new drug treatments (in-silico) before they are experimented on humans speeding up drug developments.
This is a list of current projects that the AICN group is running or collaborating in.