University of Portsmouth Featured PhD Programmes
Anglia Ruskin University ARU Featured PhD Programmes

Relating neuronal dysfunctions with mental deficits in neurodegeneration: a computational and multimodal imaging study


Neuroscience Institute

Sheffield United Kingdom Biomedical Engineering Neuroscience Computer Science Mathematics Psychology

About the Project

PhD candidates with relevant background are welcome to join Professor Li Su’s research group studying neuronal dysfunctions underlying dementia using a combined computational (such as AI and machine learning) and neuroimaging approach. Professor Li Su is Professor of Neuroimaging in the Flagship Neuroscience Institute at Sheffield University and Principal Investigator in Department of Psychiatry at University of Cambridge.

Background

Neurological diseases such as Parkinson’s disease are often associated with complex mental dysfunctions such as depression, anxiety, cognitive decline, visual perceptual deficits and thought disorders. For example, the prevalence of visual hallucinations is about 80% in dementia with Lewy bodies (DLB), 60% in Parkinson’s disease dementia and 15% in Alzheimer’s disease (AD) dementia and vascular dementia. So, detailed mechanistic models for mental disorders in neurological diseases are vital to develop effective treatments.

Visual hallucinations are particularly difficult to measure and its neurobiological underpinnings remain unclear. However, impairments in multiple neurotransmitter systems have been implicated, e.g. in DLB, loss of basal-forebrain cholinergic neurons, abnormal nicotinic receptor binding, as well as dopamine and noradrenergic dysfunctions. Recently, reduction in visual GABAergic and glutamatergic transmissions and changes in gamma oscillation have also been associated with visual hallucinations and other symptoms in DLB (Khundakar et al., 2016; Kujala et al., 2015). We argue that the molecular level understanding in the brain is still insufficient to explain and understand the mechanistic relationships between underlying neurobiology and mental symptoms.

Integrating computational neuroscience models with neuroimaging showed a strong promise to the above question. Previous study measured synaptic dysfunctions from MEG in patients with distinct monogenic ion channelopathies (Gilbert et al., 2016), determined dynamic basal ganglia – cortical interactions in patients with Parkinson’s disease, and made quantitative predictions on the effect of L-DOPA on functional connectivity (Marreiros et al., 2013). In one of our own modelling studies, we showed the impact of acetylcholine on attention deficits in AD (Mavritsaki et al., 2018). Except our own pilot study, little endeavour has been made to empirically validate models of visual hallucination in these neurological diseases, thus limiting its applicability in informing new pharmacological or non-pharmacological approaches to manage these distressing symptoms.

Hypothesis

The computational model based on neuroimaging data in patients with DLB can simulate behavioural patterns similar to visual hallucinations in DLB. Comparing models derived from patients and healthy controls will provide new insights into disease mechanisms in DLB.

Aims and objectives

1) To build novel neural network models of hallucinations based upon our existing model framework and imaging data. The model parameters relating to neurotransmitters and network connectivity will be determined by MRS and MEG data from the DLB and the control groups.

2) To test new treatments with the computational model, we aim to use the computational model to identify candidate therapeutic targets in drug-repositing and predict effects of simultaneously modulating multiple neurotransmitter systems on the symptomology of DLB in silicon.

Methods

The PhD project will be based on data collected from the Multimodal Imaging in Lewy Body Disorders or MILOS study (IRAS 202332), in which, we have behavioural, MEG, MRI and amyloid PET data from patients with DLB and similarly aged controls. We have also developed a computational model based on the neural mass model (Moran et al., 2013) with neurobiological details such as ion channels, neurotransmitters and different neuronal subpopulations to study the realistic time scales and firing rates in the neural activity. The model successfully simulated the behavioural patterns of both healthy controls and visual hallucinations similar to those found in DLB.

In the current PhD project, we will fit the basic model to neuroimaging data. The parameters associated with inhibitory interneurons will be initially set by the GABA MRS from controls, and the model will be trained to activate the correct object representations contained in the visual inputs, i.e. negligible chances of hallucination. Then, we will reset the model parameters using GABA from the DLB group, as well as dopamine and acetylcholine levels in DLB from the literature, while keeping other model parameters unchanged. We predict that the model will now produce hallucinating states, owing to a failure to constrain object-related representations.

   Based upon the basic model, a biophysically more detailed computational model will be implemented using a modified neural fields model in dynamic causal modeling (DCM) (Friston et al., 2016). The additional model parameters relating to neurotransmitters and network connectivity will be fitted to both GABA MRS and MEG data from the DLB and the control groups separately using Bayesian methods. Specifically, the model based on neurobiologically plausible constraints such as conductance, synaptic latency and neurotransmitters can generate realistic neural activity and predict MEG time series, which can be compared to real MEG data to fit model parameters. Using the computational models, we can also simulate the effect of drug interventions on the occurrence and frequency of hallucinations (Jorgensen WL, 2004). These include drugs that target GABA, for example antiepileptics, and more importantly, the combined effects of antiepileptic drugs and cholinesterase inhibitors (a family of drugs that has already been tested to treat DLB and AD).

Applications are open to students from both the UK and overseas, though we note that due to funding constraints the availability of positions for students with overseas fee status will be more limited. We anticipate competition for these studentships to be very intense. We would expect applicants to have an excellent undergraduate degree in a relevant discipline. We would also expect applicants to have completed or be undertaking a relevant master’s degree to a similar very high standard (or have equivalent research experience).

--

Please complete a University Postgraduate Research Application form available here: https://www.sheffield.ac.uk/postgradapplication/, state the main supervisor in the respective box and select ‘Neuroscience’ as the department.

We expect to carry out 30-minute interviews on 27th April (am, GMT) and 4th May (pm, GMT). If you are shortlisted for interview, we will aim to inform you of this by 23rd April. If you are unable to attend at the specified times, please let us know if we confirm that we would like to interview you.


Funding Notes

EPSRC FUNDED
• 3.5 years PhD studentship commencing October 2021
• UKRI equivalent home stipend rate per annum for 3.5 years
• Tuition fees for 3.5 years
• EPSRC studentships come with a £4,500 Research Training Support Grant over the course of the award.

References

Friston K et al (2016). Bayesian model reduction and empirical Bayes for group (DCM) studies. NeuroImage 128, 413–431.
Franciotti R et al (2006) Cortical rhythms reactivity in AD, LBD and normal subjects: a quantitative MEG study. Neurobiol. Aging. 27, 1100-1109.
Gilbert JR et al (2016). Profiling neuronal ion channelopathies with non-invasive brain imaging and dynamic causal models: Case studies of single gene mutations. Neuroimage, 124, 43-53. doi:10.1016/j.neuroimage.2015.08.057
Jorgensen WL (2004) The Many Roles of Computation in Drug Discovery, Science, 303, 1813-1818.
Khundakar et al (2016) Analysis of primary visual cortex in dementia with Lewy bodies indicates GABAergic involvement associated with recurrent complex visual hallucinations, Acta Neuropathologica Communications, 4:66 DOI 10.1186/s40478-016-0334-3.
Kujala J et al (2015) Gamma oscillations in V1 are correlated with GABAA receptor density: A multi-modal MEG and Flumazenil-PET study, Scientific Reports, 5, doi:10.1038/srep16347
Marreiros AC et al (2013) Basal ganglia–cortical interactions in Parkinsonian patients, NeuroImage, 66: 301-310.
Mavritsaki E, Bowman H, Su L (2018) Attentional deficits in Alzheimer’s disease: investigating the role of acetylcholine with computational modelling. Chapter in Multiple scale models of brain, Springer.

Email Now

Insert previous message below for editing? 
You haven’t included a message. Providing a specific message means universities will take your enquiry more seriously and helps them provide the information you need.
Why not add a message here

The information you submit to University of Sheffield will only be used by them or their data partners to deal with your enquiry, according to their privacy notice. For more information on how we use and store your data, please read our privacy statement.

* required field

Your enquiry has been emailed successfully



Search Suggestions

Search Suggestions

Based on your current searches we recommend the following search filters.



FindAPhD. Copyright 2005-2021
All rights reserved.