Coventry University Featured PhD Programmes
Birkbeck, University of London Featured PhD Programmes
Imperial College London Featured PhD Programmes
University of Warwick Featured PhD Programmes
Cardiff University Featured PhD Programmes

(MRC DTP) Can we use techniques such as machine learning to improve the risk prediction of infection-related complications in order to optimise antibiotic prescribing?


Project Description

Antimicrobial resistance is a major public health concern and overuse of antibiotics is a major factor in its development. One strategic aim of the Government’s approach to minimise this is to conserve and steward the effectiveness of existing treatments by optimising antibiotic prescribing practices. A recent study by the research group evaluated the outcomes in a large group of patients in primary care with a common infection. It found that patients at very low risk of infection-related complications were as likely to receive an antibiotic as those at high risk, indicating an important need for better targeting of antibiotics. There is considerable potential to use routinely collected electronic health records (EHRs) to estimate the risks of major clinical outcomes (such as hospital admission) which could then be used by clinicians to target treatment. The conventional approach for risk prediction has been to use regression modelling based on known risk factors [1]. There is now major interest in using machine learning techniques. It has been reported that machine learning models can outperform conventional models in predicting risks using EHRs [2]. However, the application of these approaches to EHRs is challenging due to clinicians recording and coding information very differently. Recent research in our group has found that the accuracy of risk predictions that are based on conventional models varies considerably between different clinical sites [3]. Conventional risk prediction models that use EHRs also provide estimates strongly dependent on modelling decisions [4]. The objective of this PhD is to assess how we can improve the risk prediction of infection-related complications. Better evidence on these risks can support clinicians and patients in their decision making whether an antibiotic needs to be prescribed and their diagnosis of serious infections. The practical outcome of this PhD is to implement machine learning methods such as random forests and neural networks in large EHR datasets and to critically evaluate whether these models provide robust and transparent risk predictions that are clinically useful. This PhD will provide training in the quantitative skills of statistics / machine learning. The PhD candidate will work in a multidisciplinary team involving statisticians, data scientists, informaticians, clinicians and social scientists. There will be ample training opportunities within the University of Manchester for PhD students (including statistics, machine learning, data science and data management).

https://www.britanalytics.uk/

https://www.research.manchester.ac.uk/portal/en/researchers/tjeerd-van-staa(bbc14791-bd1b-4f86-b388-dcdc6bef0cd9).html

Entry Requirements:
Applications are invited from UK/EU nationals only. Applicants must have obtained, or be about to obtain, at least an upper second class honours degree (or equivalent) in a relevant subject.

Funding Notes

This project is to be funded under the MRC Doctoral Training Partnership. If you are interested in this project, please make direct contact with the Principal Supervisor to arrange to discuss the project further as soon as possible. You MUST also submit an online application form - full details on how to apply can be found on the MRC DTP website View Website

As an equal opportunities institution we welcome applicants from all sections of the community regardless of gender, ethnicity, disability, sexual orientation and transgender status. All appointments are made on merit.

References

1. Munoz-Price LS, Frencken JF, Tarima S, Bonten. Handling Time-dependent Variables: Antibiotics and Antibiotic Resistance. Clin Infect Dis. 2016 15;62(12):1558-1563.
2. Weng SF, Reps J, Kai J, Garibaldi JM, Qureshi N. Can machine-learning improve cardiovascular risk prediction using routine clinical data? PLoS One 2017; 12: e0174944.
3. Li Y, Sperrin M, Belmonte M, Pate A, Ashcroft DM, van Staa TP. Do population-level risk prediction models that use routinely collected health data reliably predict individual risks? Sci Rep. 2019 Aug 2;9(1):11222.
4. Pate A, Emsley R, Ashcroft DM, Brown B, van Staa T. The uncertainty with using risk prediction models for individual decision making: an exemplar cohort study examining the prediction of cardiovascular disease in English primary care. BMC Med. 2019 Jul 17;17(1):134. 1368-8.

Email Now

Insert previous message below for editing? 
You haven’t included a message. Providing a specific message means universities will take your enquiry more seriously and helps them provide the information you need.
Why not add a message here
* required field
Send a copy to me for my own records.

Your enquiry has been emailed successfully





FindAPhD. Copyright 2005-2019
All rights reserved.