Developing continuity of care prediction models using Explainable AI for outcomes of care in multimorbid patients

Summary

Machine learning methods have the potential to innovate how the key concept of continuity-of-care can improve the care of patients. This project aims to develop explainable and interpretable approaches for clinicians and patients, to understand how they might be used in the multimorbid patients with heart failure and chronic obstructive pulmonary disease.

How are we doing it?

There will be a review of literature which brings the clinical topics of HF, COPD and continuity-of-care together with where machine learning methods have been used in data science of healthcare. National databases which have collected anonymised care will be used to understand and develop how continuity-of-care models might be applied to the care of HF and COPD patients. This research will be done in conjunction with healthcare professionals and patients.

What happens next?

The PhD project is framed over 3 years. In the first year, it will be to understand the current literature and the healthcare data. In the first and second years will be to use routinely collected care data and develop machine learning methods for continuity-of-care. In all three years there will be engagement with healthcare professionals and in the third year, there will also be work on how new approaches might be embedded in actual care in the future.

Funding

NIHR Exeter BRC.

People Involved

PhD student: Mr. Varun Kulkarni Supervisory Team: Prof. Umesh Kadam, Dr. Aishwaryaprajna, Dr. Ben Owusu and Dr Louise Pealing.