Medical image computing for next-generation healthcare technology

Lead Research Organisation: University College London
Department Name: Computer Science

Abstract

At the heart of this platform is a vision of the future of medical technology enabled by the increasing availability of rich and diverse data from large-scale data collection initiatives. The vision is to exploit this data mass to enrich sparse data acquired at point of care. For example, a single clinical MRI or ultrasound scan, together with subject-specific clinical data (age, sex, symptoms, genetics) can index a centralised data mass to infer likely diagnosis, prognosis, and treatment outcomes by matching to similar individuals about whom much more is known. The same paradigm further enables cheap and/or small and/or low-power devices able to acquire data in non-specialist locations, such as portable or hand-held scanners, or tiny imaging devices in surgical instruments.

Platform funding will maintain our world-leading activities in a range of medical image computing topics, support alignment of currently distinct strands of work for mutual short-term benefit, and develop key enabling technology and demonstrators of our long-term vision. Our current work develops state of the art imaging technology, image analysis techniques, and mathematical and computational models that maximise the information contained in and derived from large data sets. We also develop a range of automated diagnostic systems and surgical support systems that can demonstrate the benefits of the data-driven paradigm.

Platform funding supports the career development of the applicants, a pool of associated academic staff, talented post-doctoral researchers, and students coming through an associated EPSRC Centre for Doctoral Training in Medical Imaging. The platform provides opportunities for all to develop new ideas crossing boundaries between different topics, while contributing to a central long-term vision that supports a variety of future research careers. Competitive allocation of resource under close mentorship of senior colleagues instils early-career researchers with essential academic skills required for successful future careers. It further enhances the career progression of our senior staff by supporting opportunities to acquire new skills, establish new collaborations, explore commercialisation, and recruit the best staff from around the world when opportunities arise.

Planned Impact

Medical image computing is a linchpin of modern medicine and biomedical research. It provides the mathematical and computational models and algorithms that link biomedical image data to an understanding of tissue-and-body structure and function in health and disease. This is the foundation for current and future drug/treatment discovery, development, and deployment.

Our work impacts most directly on neurological diseases and cancer. The UK dementia platform estimates annual socioeconomic costs of dementia in Britain at around £17B. A treatment prolonging independent life of dementia patients by just one year would save around £1B per year in care costs as well as boosting the UK economy through revenue from the treatment if realised through its pharmaceutical industry and thriving community of related SMEs. Annual costs of cancer have similar scale; our work in early diagnostics and treatment delivery offers similar socio-economic impact.

A key impact of our long-term vision of medical technology is to make accurate diagnostic and prognostic testing widely accessible in non-specialist scenarios. For example, measurements from a handheld scanner at a local doctor's surgery, pharmacy, or field hospital, coupled with clinical data link to a central information resource to obtain suggestions of diagnosis, prognosis, treatment assignment and/or further testing strategy. This reduces the cost of patient examinations, reduces the need for invasive tests, and decreases the delay in moving from assessment to treatment. Ultimately for the patient, this leads to cheaper, faster and more accurate diagnosis, more reliable prognosis, reduction of side effects from diagnostic tests, and better-targeted treatment and care. Economically, it stimulates industry to develop and market new drugs and devices for assessment and treatment. This in turn reduces the burden on the healthcare system by prolonging independent good-quality life.

Publications

10 25 50

publication icon
Arjas A (2020) Estimation of dynamic SNP-heritability with Bayesian Gaussian process models. in Bioinformatics (Oxford, England)

publication icon
Arridge S (2019) Solving inverse problems using data-driven models in Acta Numerica

publication icon
Arridge S (2020) Networks for Nonlinear Diffusion Problems in Imaging. in Journal of mathematical imaging and vision

publication icon
Battiston M (2018) Fast and reproducible in vivo T mapping of the human cervical spinal cord. in Magnetic resonance in medicine

 
Description This is a platform grant used to support a wide range of projects mostly led by early career researchers. Here are specific findings/outputs that have emerged from those projects so far, which I will add to as the project continues.

Enrico Kaden: development of Spherical Mean Technique (Kaden et al MRM 2016, NIMG 2016) and release of open-source software: https://ekaden.github.io.

Laura Panagiotaki: development of 3D mold for aligned in-vivo and post-operative imaging of prostate MR and histology (Bailey et al Frontiers in Oncology 2017).
Exploitation Route We are continuing to allocate resource within the grant using the same strategy and exploring some more adventurous allocation, which we will report back on at later stages.
Sectors Healthcare,Pharmaceuticals and Medical Biotechnology

 
Description A multi-modality, surgical planning and guidance system to improve the up-take of laparoscopic liver resection
Amount £1,444,811 (GBP)
Funding ID II-LA-1116-20005 
Organisation National Institute for Health Research 
Sector Public
Country United Kingdom
Start 01/2018 
End 12/2020
 
Description Actuated Robotic Imaging Skins
Amount £2,780,000 (GBP)
Organisation Royal Academy of Engineering 
Sector Charity/Non Profit
Country United Kingdom
Start 10/2019 
End 09/2029
 
Description Developing single cell resolution 3D models of immune surveillance in cancer
Amount £165,263 (GBP)
Funding ID NS/A000069/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 01/2018 
End 12/2020
 
Description Developing single-cell resolution 3D models of immune surveillance in cancer
Amount £487,000 (GBP)
Funding ID NS/A000069/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 01/2018 
End 12/2020
 
Description EPSRC Doctoral Prize
Amount £110,000 (GBP)
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 01/2015 
End 12/2017
 
Description EPSRC Doctoral Prize Fellowship
Amount £100,000 (GBP)
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 09/2016 
End 09/2018
 
Description EPSRC Early career fellowship
Amount £1,000,000 (GBP)
Funding ID EP/N021967/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 07/2016 
End 06/2021
 
Description Early Career Fellowship
Amount £1,239,250 (GBP)
Funding ID EP/P012841/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 04/2017 
End 03/2022
 
Description Enabling Clinical Decisions From Low-power MRI In Developing Nations Through Image Quality Transfer
Amount £1,035,545 (GBP)
Funding ID EP/R014019/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 02/2018 
End 01/2021
 
Description Enabling clinical decisions from low-power MRI in developing nations through image quality transfer
Amount £1,020,000 (GBP)
Funding ID EP/R014019/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 02/2018 
End 01/2021
 
Description FP7-PEOPLE, Marie Curie Action, Intra-European Fellowship
Amount € 231,283 (EUR)
Funding ID 627025 
Organisation European Commission 
Sector Public
Country European Union (EU)
Start 01/2015 
End 02/2017
 
Description Gold-standard assessment of prostate cancer MRI accuracy
Amount $25,000 (AUD)
Organisation Sydney Catalyst Central Office 
Sector Charity/Non Profit
Country Australia
Start 06/2017 
End 05/2018
 
Description IGT Network+
Amount £50,000 (GBP)
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 06/2017 
End 01/2018
 
Description Image Quality Transfer
Amount £60,000 (GBP)
Organisation Microsoft Research 
Sector Private
Country Global
Start 10/2015 
End 09/2018
 
Description Learning MRI and histology image mappings for cancer diagnosis and prognosis
Amount £774,000 (GBP)
Funding ID EP/R006032/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 12/2017 
End 01/2020
 
Description LonDownsPREVENT: A longitudinal study of the mechanisms of cerebral amyloid angiopathy and neurodegeneration in Down syndrome to inform AD prevention
Amount £1,015,308 (GBP)
Funding ID MR/S011277/1 
Organisation Medical Research Council (MRC) 
Sector Public
Country United Kingdom
Start 04/2019 
End 04/2023
 
Description MedCity Collaborate to Innovate
Amount £100,000 (GBP)
Organisation MedCity 
Sector Public
Country United Kingdom
Start 04/2017 
End 04/2018
 
Description Network Pump-Priming/Equipment Grant
Amount £2,200 (GBP)
Organisation Alzheimer's Research UK 
Sector Charity/Non Profit
Country United Kingdom
Start 06/2017 
End 06/2018
 
Description Social Science Plus
Amount £4,000 (GBP)
Organisation University College London 
Sector Academic/University
Country United Kingdom
Start 02/2017 
End 07/2017
 
Description UCL Knowledge and Innovation fund
Amount £15,000 (GBP)
Organisation UCL Business 
Sector Private
Country United Kingdom
Start 03/2017 
End 10/2017
 
Description Maximizing the quality of life of lung cancer survivors: conventional radiotherapy versus proton therapy 
Organisation University of Texas
Department M. D. Anderson Cancer Center
Country United States 
Sector Academic/University 
PI Contribution Agreement to share methodologies for future studies.
Collaborator Contribution Agreement to share data for future studies.
Impact Outputs: invited seminar; blog post. This collaboration is multidisciplinary, involving computer scientists, physicists and radiation oncologists.
Start Year 2017
 
Title MISST - Microstructure Imaging Sequence Simulation ToolBox 
Description Microstructure Imaging Sequence Simulation Toolbox (MISST) is a practical diffusion MRI simulator for development, testing, and optimisation of novel MR pulse sequences for microstructure imaging. MISST is based on a matrix method approach and simulates the signal for a large variety of pulse sequences and tissue models. Its key purpose is to provide a deep understanding of the restricted diffusion MRI signal for a wide range of realistic, fully flexible scanner acquisition protocols, in practical computational time. 
Type Of Technology Software 
Year Produced 2015 
Open Source License? Yes  
Impact MISST has been used in several research studies presented at recent conferences. 
 
Title Spherical Mean Technique 
Description It implements a model-based imaging technique for deriving microstructural maps from MRI. 
Type Of Technology Software 
Year Produced 2016 
Open Source License? Yes  
Impact It's quite new, but already being used in a range of clinical studies. Impact will come and we'll report back when it does. 
URL https://ekaden.github.io
 
Description Blog post: Visit to MD Anderson Cancer Center, Houston, US 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Postgraduate students
Results and Impact Blog post detailing the outcomes of a visit to an international clinical partner to discuss collaboration opportunities.
Year(s) Of Engagement Activity 2017
URL http://catveigablog.wordpress.com/2017/10/09/ucl-global-engagement-funds-201718-visit-to-md-anderson...
 
Description Organisation of Early Career Workshop 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Postgraduate students
Results and Impact The goal of the UCL Centre for Medical Image Computing (CMIC) Early Career workshop is to provide to early career researchers a clear and broad view on different possible career paths available to graduates. Our goal was to deliver a successful workshop where participants feel comfortable discussing their long-term goals, expectations and concerns, while hearing from peers that followed similar and different paths. The workshop was targeted at post-doctoral researchers and final year PhD students. A total of 35 participants participated in this one-day event, which took place on the 8th November 2018.

We have invited UCL/CMIC alumni and collaborators from a mix of background and career stages. These include academic (research and teaching), large industries, small and medium companies, start-ups, governmental and public sector careers. The event gathered 6 speakers and other 11 mentors. The event was divided into two types of session: plenary talks and small group discussions. For the small group discussions, the participants and mentors were divided in groups to discuss different career paths (academic, industry and other) in a ratio 3:1 for participants and mentors.

We have received positive feedback from attendees and invitees, that found the event very useful and appreciated the opportunity both to stop and thinking about their career paths and to network with CMIC researchers, alumni and collaborators.
Year(s) Of Engagement Activity 2018
 
Description Organisation of Workshop on Digital Histology Reconstruction 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Postgraduate students
Results and Impact The workshop was one day long and included prestigious speakers from UK, Europe and the US; mostly from academia, but also from industry. It also served to increase the visibility of newly appointed early stage lecturers of CMIC, such as Dr. Laura Panagiotaki and myself. The workshop was also a great training event, and had 40 registered attendants from CMIC. It also served to solidify the interest group in histology/MRI started set up by myself, and has been the starting point for new collaborations, e.g., with Prof. Karla Miller from Oxford (who's now helping us with MR acquisition), or Dr. Alain Pitiot (I've served as an examiner in the viva voce examination of one of his students). I am also drafting a perspective article that I'll submit in collaboration with the rest of the speakers. In addition, there's a number of grants and publications that are associated with this workshop, and have benefitted from it indirectly:
CRUK/EPSRC. "Developing single-cell resolution 3D models of immune surveillance in cancer". NS/A000069/1. 1/1/2018-31/12/2020. Total: £487K. PI: M. Jansen, CoPIs: Y. Yuan, D. Treanor (Warwick), D. Alexa
EPSRC. "Learning MRI and histology image mappings for cancer diagnosis and prognosis". EP/R006032/1. 1/12/2017-30/1/2020. Total: £774K. PI: D. Alexander, CoPIs: S. Punwani, L. Panagiotaki, I. Kokkinos, D. Hawkes, A. Freeman, T. Mertzanidou.
J.E. Iglesias, M. Modat, L. Peter, A. Stevens, R. Annunziata, T. Vercauteren, E. Lein, B. Fischl, S. Ourselin: "Joint registration and synthesis using a probabilistic model for alignment of MRI and histological sections", under review.
J. Pichat, J.E. Iglesias, S. Nousias, T. Yousry, S. Ourselin, M. Modat: "Part-to-whole Registration of Histology and MRI using Shape Elements", ICCV Bioimage Computing Workshop, 107-115
Year(s) Of Engagement Activity 2017
URL https://www.eventbrite.com/e/workshop-on-digital-histology-reconstruction-tickets-34512655281
 
Description The TADPOLE Challenge 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact We organised a community challenge to predict future progression of ageing subjects to Alzheimer's disease. The TADPOLE challenge: https://tadpole.grand-challenge.org. The proposals should identify what gaps the BRC could fill to realise the proposed strategy. We obtained funding from three charities: Alzheimer's Association, Alzheimer's Society, and Alzheimer's Research UK. Each provided £10K to use as prizes. We offered various categories of prizes to different groups including full-time researchers, undergraduate teams, high-school teams. The endeavour was reported in the scientific press: http://www.alzforum.org/news/community-news/tadpole-challenge-seeks-best-predictors-alzheimers. And very high profile e.g. obtaining large numbers of views on all the broadcasts, see e.g. https://www.youtube.com/watch?v=mZj-sYm7pXg&feature=youtu.be. It did a lot to raise awareness of the challenges in AD and how computer science, statistics, etc, can help, especially among school kids and their teachers.
Year(s) Of Engagement Activity 2017,2018
URL https://tadpole.grand-challenge.org
 
Description Workshop on diffusion MRI meets diffusion MRS. Combining DW-MRI and DW-MRS: a multi-scale approach to microstructure imaging 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact Title: Diffusion MRI meets diffusion MRS. Combining DW-MRI and DW-MRS: a multi-scale approach to microstructure imaging

Organizers: Marco Palombo and Hui Zhang

Topic: development of new methods for brain microstructure non-invasive imaging

Aim: to create a stimulating forum where experts in diffusion-weighted MRI (DW-MRI) and spectroscopy (DW-MRS) techniques can discuss the best way to combine two techniques of complementary scale to improve the non-invasive brain tissue microstructure characterization. To scope out the key research challenges and opportunities associated with this multi-scale approach to microstructure imaging.

Motivation: Up to now, diffusion-weighted magnetic resonance community has been working primarily on the development of either DW-MRI or DW-MRS techniques separately, without exploiting the complementarity between them to create a new unified technique which can perform much better than both on their own. We want to push the community to change its point of view, considering combining the two techniques of complementary scale to improve the non-invasive brain tissue microstructure characterization.

Key invitees: Prof. F Barkhof (ION/UCH, UK); Dr. S Bisdas (ION/UCH, UK); Dr. S Punwani (ION/UCH, UK); Prof. S. Lehéricy (ICM, France); Dr. F Branzoli (ICM, France); Prof. I Ronen (LUMC, Netherlands); Dr. M Nilsson (Lund University, Sweden); Dr. J Valette (CEA/MIRCen, France).

Attendees: over 50 students, researchers and professors from UCL; KCL; Cambridge University; Oxford University; Imperial College; Crick's Institute; NHS.

Outcomes:

1) Review paper on the combination of DW-MRI and DW-MRS, in collaboration with Julien Valette (CEA Fontenay-aux-Roses in Paris, France), Itamar Ronen (LUMC in Leiden, the Netherlands) and Noam Shemesh (Champalimaud Centre for the Unknown in Lisbon, Portugal), published on NeuroImage 2017 (https://doi.org/10.1016/j.neuroimage.2017.11.028); 

2) Collaboration with Francesca Branzoli and Stephane Lehericy (ICM in Paris, France): two abstracts submitted to ISMRM 2018; two papers in preparation; consolidation of long-term collaboration. 

3) Collaboration with Noam Shemesh (Champalimaud Centre for the Unknown in Lisbon, Portugal): two abstracts submitted at the ISMRM 2018; two papers in preparation; establishment of new long-term collaboration. 

4) Invited speaker at an equivalent workshop on combining DW-MRI and DW-MRS, 10-12 October 2018, in ICM in Paris, France, organized by Julien Valette, Itamar Ronen and Francesca Branzoli. The aim of the meeting will be to consolidate the collaborations established in the UCL's workshop and to discuss future projects.
Year(s) Of Engagement Activity 2017