Google’s AI Can Accurately Detect Acute Kidney Injury Before Doctors

Doctors may now be able to predict who is at risk of developing acute kidney injury, before symptoms even arise, thanks to “mindblowing” artificial intelligence.

Acute kidney injury (AKI) is when your kidneys suddenly stop working. Approximately 2 million people worldwide die every year as a result of AKI, a disease for which no effective treatment exists, according to researchers from the University of Pittsburgh School of Medicine.

AKI is known as the “silent killer” because it is hard to detect and often diagnosed late. The condition involves sudden damage or decreased blood flow to the kidneys and can lead to organ failure and death.

However, if treated early, one in three cases are preventable.

Now, scientists at Google-owned DeepMind Health say they have developed artificial intelligence that can spot the killer two days before symptoms are recognized by doctors.

DeepMind’s research analyses 600,000 data points — such as blood tests, heart rate, and blood pressure — to calculate whether someone will develop AKI.

Tests on the health records of 700,000 US participants found that it accurately detected nine in ten of the most serious cases.

The findings of the study were published in the journal Nature.

Currently, there is no standard method of early prediction, with medics relying on guesswork until a patient deteriorates.

Could spot sepsis too

Experts also believe it could the technology be used to detect other deadly conditions, such as sepsis.

Dom King, from DeepMind Health, said: “This progress represents potentially a very significant change in how medicine is practiced and care is delivered.

“A lot of care at the moment is very reactive and this represents the potential to remove the needle to proactive, preventative care.

“It really is mindblowing for me as a doctor that in some way these AI systems are almost doing what an expert physician does.”

The DeepMind team have also trialed a mobile app which analyses patient data and immediately alerts medical teams to high-risk cases.

The tech, which has been piloted at the Royal Free Hospital in London, was four times less likely to miss AKI than standard methods.

And it picked up cases in under 14 minutes, compared to several hours.

Dr. Chris Streather, Royal Free London chief medical officer, said the results were “incredibly encouraging.”

Source: Nypost