Dear all



I am writing to highlight a PhD opportunity at Lancaster University in the Centre for Health Informatics, Computation and Statistics (http://chicas.lancaster-university.uk). The PhD will be supervised by Dr Benjamin Taylor (https://www.lancaster.ac.uk/staff/taylorb1/) and Professor Jo Knight (https://www.lancaster.ac.uk/people-profiles/joanne-knight) and co-supervised by a highly enthusiastic team of local clinicians and key figures within the NHS.



Due to the competitive nature of the application process, the candidate should have, or expect to receive a first-class undergraduate or a postgraduate qualification at distinction level in Statistics or Data Science.



The title of the project is



Statistical Learning for Early Detection and Prevention of Diabetes



A synopsis is below:



Diabetes remains one of the largest challenges for our health service, and predictions

have suggested that the NHS could be spending 17 billion pounds per year treating it by

2035.



Diabetes is a challenging disease because its consequences are many and varied and

despite there being a known preventive strategy for type-2 diabetes, there is a lack of

pragmatic evidence surrounding the actual impacts of lifestyle interventions in preventing

individuals at a high risk of diabetes becoming clinically diabetic. The number of diabetic

individuals is large, which makes public-scale campaigns difficult (costly and resource

demanding) to implement and better use of existing data, targeting the most at risk

individuals could make a big difference to the long-term clinical outcomes of patients.

Using historical EMIS records including predictors such as BMI, age, sex, HBAIC and

other blood counts as well as the existence of co-morbidities, practice pharmacist

prescribing data and socioeconomic data, this project aims to:



(1) Develop statistical models to identify which of the pragmatically-collected risk factors

(EMIS / prescribing / socioeconomic) are are useful for identifying individuals at risk of pre-

diabetes* and clinical diabetes



(2) To develop statistical models for HBAIC trajectories and evaluate this biomarker as a

potential major component of a risk score calculator (see next)



(3) To produce statistical models of risk as a function of time and develop an risk score

calculator for predicting transition to (a) pre-diabetes and (b) clinical diabetes within a

given time-period.



By involving local key stakeholders, the results of this research will directly influence

diabetes care in the Morecambe bay area and through publications, influence national

practice.



The methodological aspects of the project have potential to be directed towards either machine learning or statistics, or a mix of the two.



The Lancaster University campus is situated in a beautiful 360 acre parkland site at Bailrigg, just 3 miles from Lancaster City Centre. Lancaster University is one of Britain's top universities, with over 12,000 students and 2,500 employees within the Bailrigg campus that is now almost a small town in its own right. For those applicants who enjoy the outdoors, living in Lancaster offers easy access to the Lake District and Yorkshire Dales.



Interested and appropriately qualified applicants should contact Dr Benjamin Taylor ([log in to unmask]<mailto:[log in to unmask]>) or Professor Jo Knight ([log in to unmask]<mailto:[log in to unmask]>) for further information. Please include an up-to-date CV as an attachment.



Kind regards



Ben





References



- Beverley Balkau et al (2008). Predicting Diabetes: Clinical, Biological, and Genetic Approaches

Data from the Epidemiological Study on the Insulin Resistance Syndrome (DESIR). Diabetes Care

2008 Oct; 31(10): 2056-2061. https://doi.org/10.2337/dc08-0368



- Kedir N Turi el al (2017). Predicting Risk of Type 2 Diabetes by Using Data on Easy-to-Measure

Risk Factors. Prev Chronic Dis 2017;14:160244. DOI: http://dx.doi.org/10.5888



- Yang et al (2018) Type 2 diabetes mellitus prediction model based on data mining.

https://doi.org/10.1016/j.imu.2017.12.006



- Alghamdi et al (2017) Predicting diabetes mellitus using SMOTE and ensemble machine learning

approach: The Henry Ford ExercIse Testing (FIT) project.

https://doi.org/10.1371/journal.pone.0179805



- https://diabetestimes.co.uk/diabetes-nhs-costs-could-hit-17-billion/



--



Benjamin Taylor, MSci MSc PhD

Lecturer in Biostatistics

Centre for Health Informatics, Computation and Statistics (http://chicas.lancaster-university.uk)

Faculty of Health and Medicine

B05a Furness Building

Lancaster University

Lancaster

LA1 4YF

UK



Telephone: +44 (0)1524 593499



Email: [log in to unmask]<mailto:[log in to unmask]>

Website: http://www.lancs.ac.uk/staff/taylorb1/





You may leave the list at any time by sending the command



SIGNOFF allstat



to [log in to unmask], leaving the subject line blank.