A novel solution for the early detection of Acute Kidney Injury patients

A2002902

Acute Kidney Injury (AKI) is a growing, deadly epidemic. In the UK alone there are around 615,000 episodes of AKI each year, leading to 100,000 deaths. To put this in perspective, from February 2021–January 2022, there were 11,561,049 cases of COVID-19 in England, leading to 53,356 deaths. 

AKI causes a build-up of waste products in the blood, which affects other organs such as the brain, heart, and lungs. It means long hospital stays for most patients and, while kidney health can recover if AKI is diagnosed and treated quickly, many patients need dialysis. 1 in 5 people admitted to hospital develop AKI, mostly affecting the frail and seriously unwell. In addition, AKI puts patients at increased risk of chronic kidney disease. An NHS-wide AKI alert system has been introduced to address these challenges. 

This innovation goes beyond what is already available to doctors. It has been developed using artificial intelligence (AI) that produced an algorithm. An algorithm is a step-by-step process completed in exactly the same way every time it is performed. This algorithm is based on that currently used in the NHS AKI alert system, which uses the results of blood tests taken as standard patient care. This means that patients do not have to have additional blood tests, and data that is already available is used.  

AKI Predict improves on the NHS algorithm and calculates the risk of a patient developing AKI before they get it, instead of when they already have it. This means doctors could treat patients more quickly, and patients would be more likely to recover.  The aim now is to investigate the effect of other diseases on the development of AKI. This will make AKI-Predict more accurate, and help doctors to understand which diseases are the biggest risk factors for AKI. 

Lead Researcher	Ella Scallan Data Science Research Assistant
Co-Researchers	Dr Andrew Lewington Dr Sergei Krivov Dr Stefan Auer
Host Organisation	University of Leeds
Grant Amount	£3,100
Start Date	01/06/2025
Estimated Duration	11 months
Impact Areas	Health inequalities – Transplantation Innovation & Health Technologies
Tags/key notes	Early Career Researcher Renal and urogenital