Improving predictive validity of tumour markers: Developing and validating predictive models using a large United Kingdom primary care database
A 4 year, fully funded (UK tuition fees and stipend) project commencing in October 2021. This PhD studentship is delivered as part of the prestigious Industrial Cooperative Awards in Science & Technology (iCASE) EPSRC programme.
This studentship covers the tuition fee at the UK fee rate (£4,500) and a stipend for 4 years at the standard research council rate (at £15,609 in 2021/22). The stipend will be topped up by an additional £3,000 per year, taken from the financial contribution by NPL. Additional funds are also available for training, and travel costs for visits to the NPL, other project partners and academic conferences. Please visit the UKRI website for more information and eligibility criteria.
Funding sourceThis is a project jointly funded by the University of Surrey and National Physical Laboratory (NPL) as part of the iCASE EPSRC programme.
In this project, the student will use routinely collected NHS data from one if the UK’s largest primary care databases Orchid to study the usage and predictive validity of tumour markers.
Over-diagnosis (false positives) and under-diagnosis (false negatives) of cancers are sources of significant burden for patients and the NHS (cost and resources). Tumour markers measured from blood are increasingly used but have limited accuracy and predictive validity. The student will develop and validate predictive models for more accurate diagnoses of cancers using routinely collected NHS healthcare data. We will improve the understanding of the predictive value of tumour markers in relation to clinically relevant outcomes – e.g. the need for treatment. In the future, this could improve the usage of NHS resources and patient care.
Primary care and hospitals are developing large clinical datasets that could become powerful tools to support clinical decisions. This is a rapidly evolving area because the quality and volume of healthcare data is increasing. The University of Surrey with NPL, and the tow strategic partners (University of Oxford and STFC) will provide, academic environment, expertise, access to data and innovative approaches that are required for the successful delivery of the project.
This is a collaborative project. The student will be predominately based in the Faculty of Health and Medical Science, University of Surrey with three months a year placement at NPL.
The student will develop skills in:
- Curation and analysis of large routine healthcare databases
- Predictive modelling
- Novel applications of machine learning to healthcare data
Related linksSupervisory team Dr Agnieszka Lemanska, Faculty of Health and Medical Sciences, University of Su… Dr Nadia Smith, National Physical Laboratory (NPL), Teddington, UK
- Applications are assessed on an individual basis, taking into account academic qualifications and relevant experience. Applicants must have a UK first class honours degree, or its international equivalent, in an appropriate subject; or a UK 2:1 honours degree plus a UK Master’s degree, or their international equivalents; or relevant qualifications and experience.
- Programming skills. Experience of R will be an advantage
- Strong interest in healthcare and in cancer
- Experience in the analysis of healthcare data and a healthcare background will be an advantage
- Excellent communication and organisational skills
- Ability to work independently and as part of a team
- Previous research publication track record will be an asset
This studentship is only available to UK students.
IELTS requirements: IELTS: 7.0 overall with 6.5 in each band.
How to apply
Applications should be submitted through the University of Surrey website. Applicants are welcome to contact the primary supervisor Dr Agz Lemanska, for informal enquiries before applying for the studentship.
Documents Required for Application:
- All degree certificates and transcripts
- Your curriculum vitae
- A research statement - outlining any areas of interest, motivations and previous experience
- Two references