Computational models of neurodegenerative disease progression
Lead Research Organisation:
University College London
Department Name: Computer Science
Abstract
The project develops new computer science technology for modelling the progression of a disease or developmental process. It pioneers the use of state-of-the-art generative modelling and learning techniques to address this problem. It demonstrates the new approach by addressing questions of intense current interest in neurology: what is the sequence of clinical and pathological decline in two important diseases, Alzheimer's disease (AD) and fronto-temporal dementia (FTD), and how does it vary over the population? The methodological development introduces new and general-purpose techniques in computer science and the experimental work adds fundamental new knowledge in neurology.
The progression is the sequence of events that occurs as the disease or process advances. All diseases have an associated set of symptoms and pathologies. For example, AD causes loss of memory, personality changes, brain shrinkage, and deposits of abnormal proteins. However, other neurological diseases share many of these same occurrences. An additional fundamental characteristic that distinguishes diseases is the order in which the symptoms and pathologies appear. Knowledge, or a model, of this disease progression supports early diagnosis, which can maximize the effect of a treatment. It also provides insight into disease mechanisms that can accelerate development of the treatments. Furthermore, an effective model helps construct robust staging systems, which enable clinicians to tailor treatment and care plans for individual patients: so called "personalized medicine".
Modelling disease progression, however, is a major challenge. First, the sequence of events can vary substantially among patients; monitoring a few individuals closely does not capture the variation over the larger population. Second, such close monitoring is often impossible, because the necessary examinations are too expensive or invasive to perform regularly. Thus, models must come from more cross-sectional data obtained from many patients each making a few irregular visits to a clinic. Very large data sets of this kind are available and contain a wealth of information, but current techniques for mining that information remain crude and do not exploit the available data effectively.
The investigators on this project recently introduced a new computational approach to disease progression modelling: the event-based model. Unlike standard models, it learns the sequence of events directly from a large cross-sectional data set without requiring a-priori staging or ordering of the patients. Preliminary results using small data sets from genetically confirmed disease cohorts demonstrate the uniquely rich description of disease progression the new approach can provide. However, application to larger and less-controlled data sets, where the real interest lies, presents major new challenges.
This project develops the event-based model from proof-of-concept to practical research tool. It then demonstrates the tool focussing on applications in neurological disease, although long-term applicability is much wider. In particular, we construct detailed models of the progression of AD and FTD, their variability over the population, and the influence of factors such as genetic profile. Finally, the project initiates exploration of the wider family of computational models of disease progression and their potential to extract new and fundamental information. For example, we introduce new models that potentially reveal disease subtypes, provide disease-staging systems, and highlight potential causal relationships among events.
The new model-based approach has the potential to revolutionize the way we think about disease progression and thus to make a major impact in diagnosis, disease management, and treatment development for some of the most devastating and widespread medical problems facing us today. The project initiates a long-term effort towards these ends.
The progression is the sequence of events that occurs as the disease or process advances. All diseases have an associated set of symptoms and pathologies. For example, AD causes loss of memory, personality changes, brain shrinkage, and deposits of abnormal proteins. However, other neurological diseases share many of these same occurrences. An additional fundamental characteristic that distinguishes diseases is the order in which the symptoms and pathologies appear. Knowledge, or a model, of this disease progression supports early diagnosis, which can maximize the effect of a treatment. It also provides insight into disease mechanisms that can accelerate development of the treatments. Furthermore, an effective model helps construct robust staging systems, which enable clinicians to tailor treatment and care plans for individual patients: so called "personalized medicine".
Modelling disease progression, however, is a major challenge. First, the sequence of events can vary substantially among patients; monitoring a few individuals closely does not capture the variation over the larger population. Second, such close monitoring is often impossible, because the necessary examinations are too expensive or invasive to perform regularly. Thus, models must come from more cross-sectional data obtained from many patients each making a few irregular visits to a clinic. Very large data sets of this kind are available and contain a wealth of information, but current techniques for mining that information remain crude and do not exploit the available data effectively.
The investigators on this project recently introduced a new computational approach to disease progression modelling: the event-based model. Unlike standard models, it learns the sequence of events directly from a large cross-sectional data set without requiring a-priori staging or ordering of the patients. Preliminary results using small data sets from genetically confirmed disease cohorts demonstrate the uniquely rich description of disease progression the new approach can provide. However, application to larger and less-controlled data sets, where the real interest lies, presents major new challenges.
This project develops the event-based model from proof-of-concept to practical research tool. It then demonstrates the tool focussing on applications in neurological disease, although long-term applicability is much wider. In particular, we construct detailed models of the progression of AD and FTD, their variability over the population, and the influence of factors such as genetic profile. Finally, the project initiates exploration of the wider family of computational models of disease progression and their potential to extract new and fundamental information. For example, we introduce new models that potentially reveal disease subtypes, provide disease-staging systems, and highlight potential causal relationships among events.
The new model-based approach has the potential to revolutionize the way we think about disease progression and thus to make a major impact in diagnosis, disease management, and treatment development for some of the most devastating and widespread medical problems facing us today. The project initiates a long-term effort towards these ends.
Planned Impact
Management and treatment of neurodegenerative diseases is one of the biggest challenges facing medicine today. For example, Alzheimer's disease is the most common form of dementia, and the US alone currently has 5 million sufferers. Their direct medical costs exceed $250 billion and that figure is predicted to rise to $1 trillion by 2050. No disease-modifying drugs are currently available for AD or any other neurodegenerative disease. However, research into possible treatments is intensive and viable options are in the pipeline. To bring such treatments to market requires large-scale clinical trials with appropriate patient cohorts. Lack of early and accurate diagnosis is a major obstacle to these trials. The current gold-standard diagnosis for Alzheimer's disease remains post-mortem histology. The practical alternative remains clinical diagnosis, which has accuracy only around 80%, because the symptoms are easily confused with other dementias and psychiatric disorders.
The earlier and more accurate diagnosis and finer grained patient staging our new models offer will have a major impact both in the development of treatments, by identifying appropriate patient cohorts more precisely, and in designing personalized treatment plans for patients earlier and better. Greater specificity in cohort identification reduces patient numbers required for clinical trials reducing costs. The models also provide new sensitivity to the effects of treatments aiding identification of the most effective treatments more quickly. Both mechanisms accelerate the translation of effective treatments to market.
From an economic point of view in the UK, a treatment that increases by just one year the portion of the average AD patient's life during which they live independently could save in the region of 1 trillion pounds over the next 20 years: a major saving for the UK tax payer. The revenue for pharmaceutical companies marketing such a treatment will also add significantly to the economy. From a societal point of view, benefits of prolonged independence are clear not just for the patients but equally for their families and carers, whose lives are almost as deeply affected.
The general approach we introduce here has wider application to many other diseases where similar savings and revenue-generating opportunities exist.
The earlier and more accurate diagnosis and finer grained patient staging our new models offer will have a major impact both in the development of treatments, by identifying appropriate patient cohorts more precisely, and in designing personalized treatment plans for patients earlier and better. Greater specificity in cohort identification reduces patient numbers required for clinical trials reducing costs. The models also provide new sensitivity to the effects of treatments aiding identification of the most effective treatments more quickly. Both mechanisms accelerate the translation of effective treatments to market.
From an economic point of view in the UK, a treatment that increases by just one year the portion of the average AD patient's life during which they live independently could save in the region of 1 trillion pounds over the next 20 years: a major saving for the UK tax payer. The revenue for pharmaceutical companies marketing such a treatment will also add significantly to the economy. From a societal point of view, benefits of prolonged independence are clear not just for the patients but equally for their families and carers, whose lives are almost as deeply affected.
The general approach we introduce here has wider application to many other diseases where similar savings and revenue-generating opportunities exist.
Publications
Holmes HE
(2016)
Imaging the accumulation and suppression of tau pathology using multiparametric MRI.
in Neurobiology of aging
Holmes HE
(2017)
Comparison of In Vivo and Ex Vivo MRI for the Detection of Structural Abnormalities in a Mouse Model of Tauopathy.
in Frontiers in neuroinformatics
James SN
(2021)
A population-based study of head injury, cognitive function and pathological markers.
in Annals of clinical and translational neurology
James SN
(2018)
Using a birth cohort to study brain health and preclinical dementia: recruitment and participation rates in Insight 46.
in BMC research notes
Jiao J
(2017)
Direct Parametric Reconstruction With Joint Motion Estimation/Correction for Dynamic Brain PET Data.
in IEEE transactions on medical imaging
Jiao J
(2015)
Detail-Preserving PET Reconstruction with Sparse Image Representation and Anatomical Priors.
in Information processing in medical imaging : proceedings of the ... conference
Johnsen SF
(2015)
Detection and modelling of contacts in explicit finite-element simulation of soft tissue biomechanics.
in International journal of computer assisted radiology and surgery
Keshavan A
(2022)
CSF biomarkers for dementia.
in Practical neurology
Kinnunen KM
(2018)
Presymptomatic atrophy in autosomal dominant Alzheimer's disease: A serial magnetic resonance imaging study.
in Alzheimer's & dementia : the journal of the Alzheimer's Association
Kochan M
(2015)
Simulated field maps for susceptibility artefact correction in interventional MRI.
in International journal of computer assisted radiology and surgery
Description | We have developed new mathematical and computational modelling technology for revealing the long-term progression patterns of neurological diseases. It also identifies distinct subtypes within heterogeneous populations. The new technology has led to new insight into Alzheimer's disease. The ideas extend to a wide range of other diseases and applications of the models developed now include Huntington's disease, multiple sclerosis, amyotrophic lateral sclerosis; many others are in the pipeline. The new models potentially offer a powerful new tool for differential diagnosis and prognosis. |
Exploitation Route | Lots of possibilities for developing the basic tools further, which we continue to develop and apply the ideas through follow-on funding associated with it. The ideas have applications beyond neurodegenerative disease to other areas of medicine, such as cancer and infectious diseases, as well as basic development. |
Sectors | Healthcare Pharmaceuticals and Medical Biotechnology |
URL | http://cmic.cs.ucl.ac.uk/pond/ |
Description | Accelerated Magnetic Resonance Imaging for Alzheimer's Disease (ADMIRA) |
Amount | £390,000 (GBP) |
Organisation | Alzheimer's Society |
Sector | Charity/Non Profit |
Country | United Kingdom |
Start | 01/2023 |
End | 12/2025 |
Description | BAND Fellowship |
Amount | $150,000 (USD) |
Organisation | Michael J Fox Foundation |
Sector | Charity/Non Profit |
Country | United States |
Start | 11/2015 |
End | 03/2017 |
Description | Biomarkers Across Neurodegenerative Diseases (BAND2) |
Amount | $149,669 (USD) |
Funding ID | 11042 |
Organisation | Alzheimer's Association |
Sector | Charity/Non Profit |
Country | United States |
Start | 01/2016 |
End | 05/2017 |
Description | CHDI Research Funding Scheme |
Amount | £266,000 (GBP) |
Organisation | CHDI Foundation |
Sector | Charity/Non Profit |
Country | United States |
Start | 06/2016 |
End | 06/2019 |
Description | Computational PLatform for Assessment of Cognition In Dementia (C-PLACID) |
Amount | £1,500,000 (GBP) |
Funding ID | EP/M006093/1 |
Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
Sector | Public |
Country | United Kingdom |
Start | 01/2015 |
End | 12/2017 |
Description | Computational models of neurodegenerative disease progression |
Amount | £599,868 (GBP) |
Funding ID | EP/J020990/1 |
Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
Sector | Public |
Country | United Kingdom |
Start | 09/2012 |
End | 09/2015 |
Description | Delineating impact of COVID-19 infection in high-risk populations |
Amount | $155,800 (USD) |
Organisation | Microsoft Research |
Sector | Private |
Country | Global |
Start | 07/2020 |
End | 08/2021 |
Description | EPSRC Centre for Doctoral Training in Intelligent, Integrated Imaging In Healthcare (i4health) |
Amount | £6,034,274 (GBP) |
Funding ID | EP/S021930/1 |
Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
Sector | Public |
Country | United Kingdom |
Start | 08/2019 |
End | 03/2028 |
Description | EPSRC Doctoral Prize (UCL) |
Amount | £110,000 (GBP) |
Funding ID | Alexandra Young two year post-doctoral fellowship |
Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
Sector | Public |
Country | United Kingdom |
Start | 08/2016 |
End | 08/2018 |
Description | EuroPOND: data-driven models for progression of neurological diseases |
Amount | € 5,500,000 (EUR) |
Funding ID | 666992 |
Organisation | European Commission |
Department | Horizon 2020 |
Sector | Public |
Country | European Union (EU) |
Start | 01/2016 |
End | 12/2019 |
Description | Horizon 2020 |
Amount | € 4,975,862 (EUR) |
Funding ID | 666992 |
Organisation | European Commission H2020 |
Sector | Public |
Country | Belgium |
Start | 01/2016 |
End | 01/2020 |
Description | I-AIM: Individualised Artificial Intelligence for Medicine |
Amount | £838,376 (GBP) |
Funding ID | MR/S03546X/1 |
Organisation | Medical Research Council (MRC) |
Sector | Public |
Country | United Kingdom |
Start | 01/2020 |
End | 12/2024 |
Description | JPND: Early-Detection of Alzheimer's Disease Subtypes |
Amount | € 1,800,000 (EUR) |
Funding ID | MR/T046422/1 |
Organisation | JPND Research |
Sector | Academic/University |
Country | Global |
Start | 11/2020 |
End | 11/2023 |
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 | 04/2024 |
Description | Medical image computing for next-generation healthcare technology |
Amount | £1,500,000 (GBP) |
Funding ID | EP/M020533/1 |
Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
Sector | Public |
Country | United Kingdom |
Start | 05/2015 |
End | 05/2020 |
Description | Network Pump-Priming/Equipment Grant |
Amount | £2,200 (GBP) |
Organisation | Alzheimer's Research UK |
Sector | Charity/Non Profit |
Country | United Kingdom |
Start | 05/2017 |
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 |
Company Name | Queen Square Analytics |
Description | Queen Square Analytics provides brain imaging analytical services for neurological clinical trials. |
Year Established | 2020 |
Impact | Not yet |
Website | https://www.queensquareanalytics.com/ |
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 |