Precision Medicine DTP - Analysing and recommending personalised care pathways for multimorbid patients with the use of artificial intelligence techniques

Background

Multimorbidity is one of the biggest challenges that health services are facing in the UK and worldwide. Care for people living with multiple conditions accounts for over half of the costs of primary and secondary care in the UK, and the number of these patients is rising.

Organising the care for patients with multimorbidities is challenging. Care delivery is often fragmented, provided by specialists that focus on single conditions rather than the patient as a whole. This often leads to disruptions, gaps, and duplication of care, which may put the patient at risk and result in poor care quality and efficiency.

The Academy of Medical Sciences, the Health Foundation and other bodies recommend the development of new models of care that are person-centred and integrated, and they highlight the role of digital tools to support clinical decision-making [1]. However, there has been limited research to date on computational approaches to the planning of care for multimorbid patients, and a recent literature review revealed that workflow optimisation was not included in existing interventions [2]. Little is also known about current practice, in particular with regards to typical care pathways for patients with multimorbidities: what tasks are involved, to what extend they comply with medical guidelines, how they vary between different patient groups and how effective they are.

At the same time, there is a growing volume of data that we can capitalise on. This includes data from the MyWay digital platform and from SCI-diabetes, which is hailed as one of the best and most complete national diabetes registries in the world. There is now the opportunity to harness the power of this data and to link into wider linked comorbid datasets, to gain an understanding of existing care pathways and use this knowledge to design personalised, integrated care pathways for the precision medicine era.

Artificial Intelligence (AI) methods can be particularly useful to this end. Process mining [3] and model checking techniques have been successfully used for analysing and verifying care pathways over medical guidelines, respectively, while statistical machine learning techniques are appropriate for devising recommendations based on historical data. However, related work has focussed on single conditions, and little is known about how these methods can be applied to patients with multimorbidity.

Aims

The aim of this project is to employ AI techniques to provide insight into the complexities of care for multimorbid patients, as well as decision support towards the design of personalised care pathways that are safe, effective and in accordance with best practice. Focusing on diabetes patients that live with hypertension or kidney disease, the objectives of this research are to:

Discover the care pathways for patients with multimorbidity based on large clinical data with the use of process mining techniques.
Analyse the discovered care pathways in terms of clinical outcome, efficiency and compliance with medical guidelines, so as to identify cases of best practice.
Develop AI algorithms for personalised care planning informed by best practice, evidence and patient characteristics that can be deployed as a decision support tool.
Training Outcomes

The student will receive interdisciplinary training across health data science, with a focus on AI, computational modelling and statistical analysis as applied to integrated care for diabetes and typical comorbidities. They will gain experience in working with data and they will develop innovative solutions by engaging not only with their academic supervisors, but also with industrial partners and the NHS, allowing them to enrich their insights. They will gain expertise in key areas surrounding multimorbidity, while developing a range of soft and transferrable skills that will prepare them for a career in academia or industry.

This MRC programme is joint between the Universities of Edinburgh and Glasgow. You will be registered at the host institution of the primary supervisor detailed in your project selection.

All applications should be made via the University of Edinburgh, irrespective of project location. For those applying to a University of Glasgow project, your application along with any supporting documents will be shared with University of Glasgow.

http://www.ed.ac.uk/studying/postgraduate/degrees/index.php?r=site/view&id=919

Please note, you must apply to one of the projects and you must contact the primary supervisor prior to making your application. Additional information on the application process is available from the link above.

For more information about Precision Medicine visit:
http://www.ed.ac.uk/usher/precision-medicine