DeepMind uses AI to fight kidney injury through early prediction


UK-based AI research firm DeepMind announced its progress on making an AI-driven system that better detects the Acute kidney injury (AKI) medical condition that kills upwards of 400,000 hospitalized patients in the U.S. and UK annually.

For those who are not familiar with the term, AKI refers to a condition where the kidneys cease to filter waste circulating in the blood. According to sources, AKI could happen very quickly in two days or less and presents a serious threat to hospitalized patients.

In separate partnerships with The Royal Free London NHS Foundation Trust (RFL) and the U.S. Department of Veterans Affairs, DeepMind attempted to explore preventative solutions for AKI through “an algorithm that can predict the presence of AKI up to 48 hours in advance and an app that cuts missed AKI cases from 12.4 percent to 3.3 percent.”

DeepMind’s studies with the RFL included Steams, an app that assists kidney specialists through an interactive algorithm and a compilation of bedside review of medical information.

Overall, the app narrowed the review process of an emergency medical case from several hours to less than 15 minutes, thus giving patients more of a fighting chance. It also reduced the cost of admission by around 17 percent.

“Critically, these early findings from the Royal Free suggest that, in order to improve patient outcomes even further, clinicians need to be able to intervene before AKI can be detected by the current NHS algorithm — which is why our research on AKI is so promising,” wrote DeepMind. “These results comprise the building blocks for our long-term vision of preventative healthcare, helping doctors to intervene in a proactive, rather than reactive, manner.”

DeepMind mentioned that it will replace Streams with Google Health in the coming months. On top of that, the team is working with Google to create new healthcare tools that “address critical patient safety issues and … [potentially] save thousands of lives globally,”

At the U.S. Department of Veterans Affairs, DeepMind used a machine learning model that accurately predicted AKI in patients two days early and identified around 90 percent of severe cases that required dialysis — a method that uses an external machine to purify the blood without the help of a kidney.

DeepMind believes that its AI solution has the flexibility to detect other diseases like sepsis, and it wants to expand on its capability even further.

That said, predicting AKI is not the first time the UK-based AI research firm invested in AI-driven healthcare solutions. In 2018, DeepMind tried to look for early signs of blindness through a trial using an algorithm at the NHS. The company also worked on advanced breast cancer research with universities and more.

Via: VentureBeat