We develop uncertainty quantification methods and algorithms to support clinical and industrial decision making, working closely with stakeholders in the health services, government and industry.
CORU staff involved: Christina Pagel, Sonya Crowe, Alex Diaz, Samuel Jackson, Hugh Kinnear
- CHIMERA (2020 - 2024)
CHIMERA: Collaborative Healthcare Innovation through Mathematics, EngineeRing and AI (2020 - 2024)
CHIMERA is one of four EPSRC-funded Mathematical Sciences in Healthcare Hubs announced in March 2020. It is a multidisciplinary Hub which brings together experts in mathematics, statistics, data science and machine learning, with unique, high volume and rich vital signs data sets from both adult and paediatric Intensive Care Units through Project Partnerships with Great Ormond Street Hospital (GOSH) and University College London Hospital (UCLH). CHIMERA will develop new physiology models, and use them to inform clinical decision making. Based in UCL, CHIMERA will also act as a Hub for a network of national and international collaborators spanning mathematical and engineering sciences, as well as critical care and industry partners.
Hospitals collect a wealth of physiological data that provide information on patient health. Full use of this data is significantly limited by its complexity and by a limited mechanistic understanding of the relationship between internal physiology and external measurement. Addressing this challenge requires multidisciplinary collaboration between mathematicians developing new biomechanical models, clinicians who measure and interpret the data to treat patients, and statistical and computational scientists to bridge the two-way translation between model output and real-life data. CHIMERA is designed to foster such collaboration to generate new understanding of physiology, new methods for relating physiology to real time data, and, finally, to translate these into practice, improving outcomes for patients by supporting clinical decision making.
For more information, visit CHIMERA website.
- DATA-CENTRIC (2018 - 2021)
DATA-CENTRIC: Developing AccounTAble Computational ENgineering Through Robust InferenCe (2018 - 2021)
The aim of this EPSRC fellowship is to develop algorithms that are accountable. This means algorithms capable of quantifying the uncertainty arising from computation itself, delivering simulations that are more transparent, traceable and at the same time more efficient. Numerical computations in industry (and hence the models that depend on them) suffer from an inevitable loss of accuracy due to: a) time and cost constraints of running modern high-fidelity computer models, b) simplifying approximations necessary to translate mathematical models into computational models, and c) limited numerical precision inherent to any computer system. Therefore, there is a continuous risk of relying on unverified computational evidence, and the path from modelling to decision-making can be (inadvertently or unwillingly) obscured by the lack of accountability.
DATA-CENTRIC works under the framework of Probabilistic Numerics, an emerging research area that enables decision-makers to monitor, diagnose and control the quality of computer simulations. Probabilistic Numerics treats computation as a statistical problem, thus enriching computation with a probabilistic measure of numerical error. This idea is gathering momentum, especially in the UK. However, theoretical development are still in their early stages and except for a few examples, it has not been applied to solve large-scale industrial problems. Consequently, it has not yet been adopted by industry. DATA-CENTRIC aims at bridging this gap. In particular, it will produce new solutions to industrial problems in Biomechanics and Robust Design. This has the potential of transforming personalised medicine and high-value manufacturing and will open the door to new industrial applications.
YouTube Widget Placeholderhttps://youtu.be/C-vpyb9VyzQ
- Rosetrees Grant (2021 - 2022)
“Leveraging the potential of advanced machine learning techniques to support clinical decision-making during emergency transport of critically ill children to paediatric intensive care”
This project led by the UCL Institute of Health Informatics and Great Ormond Street Hospital brings together CORU, clinicians, data scientists and engineers to tackle a previously unaddressed health need: how to improve the early identification of deterioration and support clinical decision making during the transport of sick children to intensive care, by applying machine learning methods to high-resolution clinical data.
Emergency transport of the country’s sickest children from general hospitals to specialist paediatric intensive care units (PICUs) is performed by retrieval teams, who provide ‘mobile intensive care’ during transport (continuous monitoring of vital parameters such as heart rate, blood pressure and oxygen levels, artificial ventilation and continuous drug infusions). However, although vital signs are measured and displayed at high-frequency (multiple readings per second), clinicians routinely use only a fraction of these data for decision making. Crucial judgements regarding how sick the patient is, the expected course, and which patients are likely to respond to specific treatments, have been traditionally based on qualitative assessment of snapshots or an overall impression based on trends, which can be especially complex in sick children with a wide range of ages and diagnoses.
In this project, we will employ advanced machine learning techniques (e.g. deep neural networks, time series classification) to classify patients based on their “physiotypes”. The aim is to develop
preliminary models to accurately identify patients at high risk of experiencing poor outcome and those likely to respond to specific interventions which will then lead to larger projects to build on this first-in-kind work.