Computer scientists at Surrey have developed innovative software which will allow clinicians to detect Acute Kidney Injury, enabling better disease management and improved quality of life for patients.
The research, funded by the Medical Research Council (MRC), is a collaboration between academics within Surrey’s Department of Computer Science and the Department of Clinical and Experimental Medicine, along with clinicians at East Kent Hospitals University NHS Foundation Trust.
Episodes of Acute Kidney Injury (AKI) can potentially cause irreversible damage to the kidney function, so detecting when they happen is important for predicting a patient’s long term health. This data supports clinicians’ understanding of Chronic Kidney Disease (CKD) – a disease which affects around 3 million people in England and costs the NHS an estimated £2 billion a year
The AKI detection algorithm currently in use is not always accurate, especially when employed in primary care (GP surgeries) where the patient is tested infrequently (only once every three months on average). Based on a novel algorithm developed at Surrey – the Surrey AKI Detection Algorithm (SAKIDA) – the new software makes it possible to reliably detect and understand AKI episodes using irregular and sparse data.
The software is part of a larger Surrey project funded by the MRC, aimed at developing advanced statistical models for predicting CKD using machine learning techniques. During the project, Surrey has collaborated closely with East Kent Hospitals University NHS Foundation Trust, which has provided the real patient data used for the modelling.
Dr Norman Poh, who is leading the project within the Department of Computer Science, explains: “Patients are less likely to have their kidney function measured when they are well, leading to sampling bias which is very difficult for most modelling systems to handle.
“This SAKIDA model is a vital component in our overall work in the field of CKD. Patients with CKD have increased risk of death from cardiovascular disease and End Stage Kidney Failure, leading to dialysis and kidney transplant.”
The research team is currently working on another algorithm-based software which uses a novel ‘broken sticks’ technique to enable clinicians to predict how quickly CKD is progressing in a patient. The ultimate aim will be to use this technique to detect not only CKD but also a range of other measurements such as blood pressure, blood sugar levels and Body Mass Index, enabling clinicians to see how CKD progression is affected by other factors.
The paper, Automatic Detection of Acute Kidney Injury Episodes from Primary Care Data, was published in the 2016 IEEE Symposium on Computational Intelligence in Healthcare and e-health. The software is available for download.
The project is an example of Surrey’s growing focus on delivering future healthcare through interdisciplinary, collaborative, technology-driven research. Find out about our ground-breaking Innovation for Health programme.