Dr N Lone, Prof Saturnino Luz, Dr K Baillie
No more applications being accepted
Funded PhD Project (European/UK Students Only)
About the Project
Background
Even among survivors, critical illness (life-threatening illness requiring life support) can have devastating consequences. Our previous work shows that many patients have severe health problems and disabilities that may last for the rest of their lives.[1] Deciding who may benefit from critical care is therefore challenging. In many cases, it may not be in the patient’s best interests to undergo life-supporting therapies. At present, clinicians and patients/families base the decision on the expectation of a reasonable chance of survival of an acceptable quality to the individual. This project aims to provide concrete data to support this decision-making.
Critically ill patients are defined by significant heterogeneity in terms of the both the type of critical illness and extent of pre-existing ill-health. Whilst the impact of the critical illness on outcomes has been the focus of much research, our understanding of the effects of pre-existing ill health remains superficial. Distinct trajectories of ill-health which relate progression over time to changes in health status are used implicitly on a daily basis in clinical practice. However, current analytical approaches do not fully account for the complexity of pre-existing ill-health trajectories. The time of onset of specific comorbidities in relation to other important factors, including correlations with other comorbid illnesses, changes in functional status and the critical illness event itself, are rarely incorporated.[2] Ageing population demographics will escalate demand for critical care in the near future. There is therefore an urgent need to better understand distinct trajectories and how these affect outcomes in the critically ill.
Electronic health records linked to diverse databases, including critical care, social care and health registries, provide rich data to better characterise distinct health trajectories over longer timespans. The use of analytical approaches such as dynamic Bayesian networks or structural time series models will maximise information from these data.
Aims
The project will take advantage of established, complex health and social care linked datasets in combination with physiological data stored in critical care databases to identify distinct phenotypes of health trajectories of illness, multimorbidity and functional status which impact on patient outcomes in the context of critical illness. The student will bring together advanced statistical methods and machine learning data-driven approaches to achieve this aim.
Specifically, the objectives are to:
Build on existing work in our groups[1,3,4] to prepare and interrogate complex datasets to identify distinct phenotypes of health trajectories in the years before a critical illness episode using standard data-driven clustering approaches to partition the patient population;
Investigate the impact of these phenotypes combined with acute physiological factors on clinically relevant outcomes, including mortality and health care costs, to identify those at highest risk of adverse outcomes;
Apply and evaluate novel methods, developed in our groups, to identify distinct subgroups of patients, including Bayesian network approaches to patient-patient similarity networks derived from an informed variable selection;
Evaluate the reproducibility of trajectory phenotypes in external datasets e.g. Swedish linked datasets.
Training Outcomes
The student will benefit from a supervisory team with expertise in medical informatics/machine learning, clinical epidemiology, and critical care.[1,3,4] An individually tailored training programme in core disciplines relevant to precision medicine will be developed which will build on and complement the student’s prior knowledge/skills. This will include formal training in epidemiology, statistics, computational, and applied machine learning methods. The student will benefit from the vibrant, academic environment in the Usher Institute and the clinical academic department of Critical Care in Edinburgh Royal Infirmary.
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:
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 are encouraged to contact the primary supervisor prior to making your application. Additional information on the application process if available from the link above.
For more information about Precision Medicine visit:
http://www.ed.ac.uk/usher/precision-medicine
Funding Notes
Start: September 2018
Qualifications criteria: Applicants applying for a MRC DTP in Precision Medicine studentship must have obtained, or will soon obtain, a first or upper-second class UK honours degree or equivalent non-UK qualifications, in an appropriate science/technology area.
Residence criteria: The MRC DTP in Precision Medicine grant provides tuition fees and stipend of at least £14,553 (RCUK rate 2017/18) for UK and EU nationals that meet all required eligibility criteria.
Full eligibility details are available: http://www.mrc.ac.uk/skills-careers/studentships/studentship-guidance/student-eligibility-requirements/
Enquiries regarding programme: [Email Address Removed]
References
1. Lone et al. 2016. Five Year Mortality and Hospital Costs Associated With Surviving Intensive Care. American Journal of Respiratory and Critical Care Medicine. 194(2):198-208. DOI: https://doi.org/10.1164/rccm.201511-2234OC
2. Beck et al. Diagnosis trajectories of prior multi-morbidity predict sepsis mortality. Scientific Reports 6, Article number: 36624 (2016) DOI: http://dx.doi.org/10.1038/srep36624
3. Baillie, JK et al. 2017. Analysis of the human monocyte-derived macrophage transcriptome and response to lipopolysaccharide provides new insights into genetic aetiology of inflammatory bowel disease. PLoS Genetics. 2017 Mar 6;13(3). e1006641 DOI: 10.1371/journal.pgen.1006641
4. Moustafa, K, Luz, S & Longo, L. 2017. Assessment of mental workload: A comparison of machine learning methods and subjective assessment techniques. in Human Mental Workload : Models and Applications. DOI: http://dx.doi.org/10.1007/978-3-319-61061-0_3