Improving the presurgical evaluation of patients with refractory focal epilepsy using advanced neuroimaging techniques

Lead Research Organisation: University College London
Department Name: Institute of Neurology

Abstract

Epilepsy is a common neurological condition affecting around 1% of the population (600,000 people in the UK). Frequent seizures carry risk of injury and death and significant psychosocial consequences, affecting relationships, employment and the ability to drive. There are over 20 medications to treat epilepsy, but a third of people with epilepsy continue to have seizures despite medication.

Surgical treatment of epilepsy in which the abnormal part of the brain is removed is an effective technique to cure or improve seizures suitable for some people with epilepsy. The National Hospital for Neurology and Neurosurgery, with the Epilepsy Society and UCL Institute of Neurology, is the largest centre in the UK undertaking epilepsy surgery (100 operations a year) and has an internationally known research programme to improve the identification of patients for surgery and understand and improve the outcomes.

In order for a person to be suitable for surgery, it is necessary to confirm that seizures are arising from one part of the brain and that it is safe to remove this part. This requires many tests including MRI brain scans. Despite advances in technology, around 30% of people with seizures arising from one part of the brain have completely normal MRI scans. Further investigation with other types of scan is expensive, may involve radiation exposure and is not available in most places, unlike MRI. If imaging cannot identify the source of seizures, recording wires may be placed directly onto the brain surface. This is available in only a few places, is invasive, very expensive and carries some risk.

In my previous research, presented and published internationally, I developed MRI techniques to improve the safety of surgery and understand its effects on the brain. I showed preliminary results in which new MRI techniques could identify the source of seizures in patients with normal MRI scans.

In the first part of this project, I will assess several new MRI techniques to determine which are helpful in identifying the source of seizures in people with severe epilepsy and normal standard scans. I will use new ways of acquiring MRI scans and will work with colleagues in the UCL Centre for Medical Image Computing, with whom I have worked for the last 5 years, and the Montreal Neurological Institute to develop new ways to analyze the scans by computer to detect subtle abnormalities not seen by the human eye.

These developments will enable a larger proportion of people with epilepsy to undergo surgery and be cured or significantly improved from epilepsy. MRI may replace some more expensive and less readily available scans. This will simplify and shorten the investigation pathway so surgery can take place more quickly. Since MRI is widely available at all epilepsy surgery centres, the new techniques can be introduced elsewhere to improve assessment and treatment for people with epilepsy. Understanding of these techniques can also be applied to other conditions such as dementia and stroke.

In the second part, I will build on my prior work using MRI to make surgery safer. I used a special type of MRI scan to identify connections in the brain important for vision and displayed these connections to the surgeon when they were operating so they could treat epilepsy without damaging vision and thus allow people to return to driving. At present, there are many limitations with this and analysis takes a long time. I will improve the technique to give more detailed and accurate images of the brain more quickly. The results and software will be made freely available to allow other centres to adopt the technique.

Overall more people will be identified as suitable for surgical treatment, more quickly and at lower cost using techniques that are applicable to all epilepsy surgery centres rather than just at highly specialised units. A greater understanding of these new MRI techniques will allow them to be used in many other conditions.

Technical Summary

Research questions:
1. Can arterial spin labelling (ASL) reliably identify interictal hypoperfusion and determine the laterality and broad localisation of epilepsy in comparison to FDG-PET?
2. Can neurite orientation density and dispersion imaging (NODDI) identify subtle structural abnormalities, particularly malformations of cortical development?
3. Are developmental MRI techniques such as sodium MRI and glucose chemical exchange saturation transfer (GlucoCEST) of benefit?
4. Can voxel-based and more advanced analyses of multimodal MRI data, including both conventional structural scans and these new data detect subtle abnormalities?
5. Is higher resolution diffusion tensor imaging (ZOOM-DTI) beneficial for optic radiation tractography?
6. Can automated seed placement and real time probabilistic tractography improve tractography processing?

Patients with refractory focal epilepsy undergoing presurgical assessment and conventional MRI will be invited to have additional research scans and healthy age/sex-matched controls will undergo the same scans. Standard postsurgical follow up includes clinical assessment and repeat scans at 4 and 12 months.

ASL and NODDI will be processed as in previous literature and software will be developed to perform voxel-based analysis and improve the detection of subtle abnormalities. This will improve the understanding of data from these scans in epilepsy and other conditions and allow a greater proportion of patients to undergo surgery with a cheaper and more streamlined assessment process.

Patients with temporal lobe epilepsy will undergo ZOOM-DTI and tractography data will be used to guide surgery with postoperative assessment of visual outcomes. Atlas-based techniques for automated seed point placement will be further developed for the optic radiation and corticospinal tract. Tractography processing will be implemented on a graphical processing unit (GPU) to enable more accurate results to be available more quickly.

Planned Impact

Improved health and well-being of people with epilepsy

Epilepsy is a major global problem and delays for surgical treatment through failing to consider patients for surgery and to subsequently identify appropriate patients. Improved imaging techniques will enable a greater proportion of patients to be deemed suitable for surgery with consequent improved quality of life. Assessment will occur more quickly, be more convenient with fewer hospital visits and involve less exposure to radiation and risk.

Beneficial economic effects for society

People with severe epilepsy are frequently unable to work, consume significant healthcare resources and many survive on social security benefits. By facilitating a greater number of patients to undergo surgery leading to seizure freedom or marked improvement enables a return to the workforce and increased economic output. It benefits psychosocial factors such as relationships and holding a driving licence.

Improved economy and organisation of healthcare delivery

At present the evaluation pathway for epilepsy surgery can be lengthy and involves nuclear medicine techniques that have limited availability (PET is not available at all centres; SPECT may require patients to go abroad). If these techniques can be avoided in some patients, this has beneficial cost and resources implications for the NHS and increases patient throughput. Utilising MRI techniques that can be implemented on clinical scanners would enable a greater number of centres to perform presurgical assessment.

Such benefits may extend to the treatment of patients with other conditions requiring similar investigation including cerebrovascular disease and dementia/neurodegenerative disorders which are very common important healthcare problems.

Implications for healthcare policymakers and changed practices

Although consideration for epilepsy surgery is enshrined in current NICE guidelines for the treatment of epilepsy, such referral typically occurs late in the disease. By increasing the utility of epilepsy surgery, this can inform guidelines and service provision with the aim to change practice in the UK (and abroad) by considering surgery and the potential benefits at a much earlier stage.

Close working with a charity, the Epilepsy Society

The Epilepsy Society is a charity that is the UK's leading provider of epilepsy services working for the benefit of all people affected by epilepsy. They create a powerful voice for people with epilepsy and actively campaign for their rights and influence policy on healthcare, social reform and welfare. By continuing to work closely with them, the third sector will be aware of research advances to inform the public and policy.

Commercialisation and exploitation

Scanner manufacturers include sequences for diffusion imaging, and more recently arterial spin labelling (ASL), as standard in scanner console software. Tractography processing modules may be available on the scanner or in separate commercial packages. If the novel techniques are beneficial, scanner manufacturers may wish to include optimal sequences for ASL and diffusion acquisitions suitable for NODDI as standard. Processing modules for these techniques and the improved tractography algorithms developed could be included in commercial packages.

Research staff skills

Researchers at the EPSRC Doctoral Training Programme at the Centre for Medical Image Computing will benefit from a link to an active clinical service where advances can rapidly be translated into clinical use and improve the strong link between basic research and clinical application. This will enable research students to recognise the importance and role of clinical translation and how these can be introduced in practice and would form an important part of the training which can be subsequently applied elsewhere as researchers frequently work in the commercial sector following their doctoral degree.

Publications

10 25 50

publication icon
Kochan M (2015) Simulated field maps for susceptibility artefact correction in interventional MRI. in International journal of computer assisted radiology and surgery

publication icon
Duncan JS (2016) Brain imaging in the assessment for epilepsy surgery. in The Lancet. Neurology

publication icon
Vos SB (2018) Evaluation of prospective motion correction of high-resolution 3D-T2-FLAIR acquisitions in epilepsy patients. in Journal of neuroradiology. Journal de neuroradiologie

 
Description Charitable Grant
Amount £60,000 (GBP)
Organisation Toyota Fundation 
Sector Private
Country Japan
Start 12/2017 
End 12/2018
 
Description NIHR North Thames CRN
Amount £100,000 (GBP)
Organisation National Institute for Health Research 
Department NIHR Biomedical Research Centre
Sector Academic/University
Country United Kingdom
Start 08/2016 
End 02/2019
 
Description Research Grant
Amount £113,333 (GBP)
Organisation Fight for Sight 
Sector Charity/Non Profit
Country United Kingdom
Start 10/2015 
End 09/2017
 
Title Hippocampal T2 profiling 
Description Publicly available software scripts to automated determine T2 profiles with the human hippocampus from MRI imaging 
Type Of Material Improvements to research infrastructure 
Year Produced 2017 
Provided To Others? Yes  
Impact In our centre, this has been introduced into routine clinical practice for the evauation of patients for epilepsy surgery in the National Hospital for Neurology & Neurosurgery. It has been made freely available so other centres can implement the same. 
URL https://github.com/sjoerdvos/hippocampal_T2
 
Description CMIC 
Organisation University College London
Department Centre for Medical Image Computing
Country United Kingdom 
Sector Academic/University 
PI Contribution All the data being used for this project is acquired by myself or colleagues. I am providing the clinical input into the development of the imaging methods and developing scripts and writing papers.
Collaborator Contribution Development and implementation of computational techniques for registered pre- and post-operative imaging data; to be used for implementation of tractography results in neuronavigational suite. Implementation of scripts for hippocampal segmentation and T2 mapping.
Impact Multiple papers (in publications), open access Hippocampal segmentation website (Hipposeg) and publicly available scripts for T2 mapping.
Start Year 2010
 
Description GOSH 
Organisation Great Ormond Street Hospital (GOSH)
Department NIHR Great Ormond Street Biomedical Research Centre
Country United Kingdom 
Sector Public 
PI Contribution 1. Collaborative grant awarded from Fight for Sight for assessment of optic radiation tractography for epilepsy surgery guidance in children, this builds on my work in adults with my MRC Clinical Research Training Fellowship 2. Partner in a multi-centre collaborative study on automated detection of FCD in which I am supervising a student.
Collaborator Contribution 1. Funding application, supply ot staff (neurosurgery, ophthalmology, neurophysiology), appointment of post-doc research assistant. 2. Development of original scripts to perform machine learning on imaging data.
Impact None yet
Start Year 2016
 
Description MNI 
Organisation McGill University
Department Montreal Neurological Institute and Hospital
Country Canada 
Sector Hospitals 
PI Contribution Acquisition and supply of imaging data and analysis of data
Collaborator Contribution Development of software algorithms to analysis imaging data.
Impact None yet
Start Year 2017
 
Description Newcastle Peter Taylor 
Organisation Newcastle University
Country United Kingdom 
Sector Academic/University 
PI Contribution Acquisition and supply and imaging data and clinical data and input into analysis and interpretation of results.
Collaborator Contribution Analysis of data, and supervision of a student in this area.
Impact Paper on impact of epilepsy surgery on structural connectome (Neuroimage Clinical 2018)
Start Year 2016
 
Title Automated hippocampal T2 profiling 
Description Automated calculation of hippocampal T2 relaxometry values in an anterior-posterior gradient for detection of hippocampal sclerosis in patients with epilepsy. 
Type Of Technology Software 
Year Produced 2017 
Open Source License? Yes  
Impact In our centre, this has been introduced into routine clinical practice for the evauation of patients for epilepsy surgery in the National Hospital for Neurology & Neurosurgery. It has been made freely available so other centres can implement the same 
URL https://github.com/sjoerdvos/hippocampal_T2
 
Description Computer-aided interpretation of epilepsy imaging 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Professional Practitioners
Results and Impact Talk at hospital neurology department informing about my research
Year(s) Of Engagement Activity 2017
 
Description Epilepsy Neuroimaging Teaching Course, ILAE UK 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Neuroimaging training course for the International League Against Epilepsy (charity), attended by 30 participants from across the globe
Year(s) Of Engagement Activity 2017
 
Description Epilepsy Society 21 years of MRI public video 
Form Of Engagement Activity A broadcast e.g. TV/radio/film/podcast (other than news/press)
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact Participation in a video for emphasis the advances that have been in the 21 years of the MRI Unit, Epilepsy Society
Year(s) Of Engagement Activity 2016
URL https://video.helloeko.com/AqNdpV?autoplay=true
 
Description Epilepsy Society Research Associate Members 
Form Of Engagement Activity Participation in an open day or visit at my research institution
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Supporters
Results and Impact Presentation of my current and planned research at the annual Epilepsy Society Research Associate Members meeting
Year(s) Of Engagement Activity 2016
 
Description Image-guidance and computer-assisted imaging in epilepsy surgery 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Professional Practitioners
Results and Impact Butler Epilepsy Lecture delivered at University of Alberta Hospital to a mixed audience
Year(s) Of Engagement Activity 2017
 
Description Insight Investment Funding pitch 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Industry/Business
Results and Impact Presentation of current research programme as part of a pitch to be "Charity of the Year" for Insight Investments. This was to professional staff members and made available on the Intranet.
Year(s) Of Engagement Activity 2016
 
Description Presentation and panel member at the Brain's Networks, Sahlgrenska Academy, Gothenburg 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Talk "What can clinical neurology learn from the brain's networks and vice versa" and panel member for debates between different disciplines looking at the Brain's Networks
Year(s) Of Engagement Activity 2015
 
Description Talk at the Wessex Neurological Centre 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Professional Practitioners
Results and Impact Talk "Neuroimaging for epilepsy" at the academic meeting of the Wessex Neurological Centre, University of Southampton
Year(s) Of Engagement Activity 2016