AI technology used to predict kidney illnesses in advance

Researchers have made more beneficial use of artificial intelligence (AI) by using it to diagnose fatal kidney illnesses some 48 hours before they occur.

Partnering with the US Department of Veteran Affairs (VA), Google’s health and AI group DeepMind unveiled a new technology that can predict if a patient has potentially lethal kidney injuries 48 hours before many of the symptoms can be recognized by the doctors.

Diagnosing acute kidney injuries can be difficult for doctors as the symptoms are not always immediate and clear. During tests, the team said that their algorithms precisely predicted 90% of acute kidney injuries that would usually end up asking for dialysis, that too 48 hours prior, as per CNBC.

Lab-grown blood vessels built from patient’s cells to make dialysis easier

As per DeepMind, the team analyzed information stored electronically from over 100 VA hospitals and reviewed data on hundreds and thousands of patients. However, ensuring privacy, the team removed personal details such as names and social security numbers.

“We’ve been really excited for the potential of using AI to support clinicians moving care from reactive to proactive and preventative,” said Dominic King, DeepMind’s co-founder and clinical lead.

DeepMind also consists of a mobile assistant for clinicians called ‘Streams’. The results showed that via the app, specialists were able to review urgent cases within 15 minutes, process that would take several hours otherwise, reported Engadget.

Moreover, the firm is now also researching on how to deliver these alerts in emergency situations so that the doctors can properly recognize and act on them accordingly. King believes that still some work has to be done for creating a regulatory framework to bring predictive tools to medicine, and also to better understand how they can be delivered in real time.