Statistical and Machine Learning Methods to Predict Glucocorticoid Toxicity Outcomes for Patients with Immune-Mediated Inflammatory Diseases

Summary

Immune-Mediated Inflammatory Diseases (IMIDs) represent a group of disorders which cause inflammation of various parts of the body due to the triggered immune system response. They affect more than 2 million people worldwide and make a serious impact on families, resulting in an additional burden on the health system.

For example, Polymyalgia Rheumatica (PMR) and Giant Cell Arthritis (GCA) are very common types of autoimmune rheumatic disorders, one of the groups of IMIDs. PMR affects muscles around the neck, shoulders and hips, while GCA affects arteries, usually in the human head. These two disorders are mostly observed in the elderly and cause such symptoms as severe headaches, tiredness, vision problems, loss of weight, etc.

Currently, there are no cures for IMIDs. However, certain medications are used to help with the symptoms. Glucocorticoids (GCs) are a class of steroid hormones predominantly used in the UK to reduce inflammation in IMIDs patients. However, it has been shown that the use of high doses of GCs is associated with an increased risk of Cardiovascular Disease (CVD). With recent advancements in data-driven techniques researchers look for patterns and hidden relationships between patients’ data and the risk/prediction of particular diagnoses. For example, in [1] a higher risk of CVD was confirmed for patients with 6 immune-mediated diseases on lower doses of GCs. In [2] are demonstrated models which aggregate predictions of CVD risk from a set of ML methods.

In this project, the student will work with the Clinical Practice Research Datalink (CPRD). The data contain demographics, lifestyle, medication, and diagnosis (type of IMIDs and CVDs) and are linked with hospital records (HES, https://digital.nhs.uk/data-and-information/data-tools-and-services/data-services/hospital-episode-statistics) and mortality data taken from the Office of National Statistics (ONS, https://www.ons.gov.uk/atoz?query=mortality&size=10). The student will test how Statistical and Machine Learning methods help with the prediction of CVDs outcome/severity for IMIDs patients who are taking GC medication. The student will look into methods to identify important features which influence CVDs outcome/severity and work on the development of predictive ML-based models. The methodology is to be validated using UK Biobank data (https://www.ukbiobank.ac.uk/).

References

[1] Pujades-Rodriguez M, Morgan AW, Cubbon RM, Wu J (2020) Dose-dependent oral glucocorticoid cardiovascular risks in people with immune-mediated inflammatory diseases: A population-based cohort study. PLOS Medicine 17(12): e1003432. https://doi.org/10.1371/journal.pmed.1003432

[2] Alaa, A. M., Bolton, T., Di Angelantonio, E., Rudd, J., & van der Schaar, M. (2019). Cardiovascular disease risk prediction using automated machine learning: A prospective study of 423,604 UK Biobank participants. PloS one, 14(5), e0213653. https://doi.org/10.1371/journal.pone.0213653

Applicants to research degree programmes should normally have at least a first class or an upper second class British Bachelors Honours degree (or equivalent) in an appropriate discipline. The criteria for entry for some research degrees may be higher, for example, several faculties, also require a Masters degree. Applicants are advised to check with the relevant School prior to making an application. Applicants who are uncertain about the requirements for a particular research degree are advised to contact the School or Graduate School prior to making an application.

The minimum English language entry requirement for research postgraduate research study is an IELTS of 6.0 overall with at least 5.5 in each component (reading, writing, listening and speaking) or equivalent. The test must be dated within two years of the start date of the course in order to be valid. Some schools and faculties have a higher requirement.