DeepMind AI Will Be Able To Predict Acute Kidney Injuries 48 Hours Before It Happens

Thanks to computers being able to do thousands of calculations and processes in a short amount of time, it has made our lives a lot more efficient. This is especially true in the medical sector, where thanks to the use of AI, we have seen how it could potentially save lives as it can pick up on diseases and illness before a regular human doctor can.

For example, earlier this year, researchers at Google developed an AI that can detect lung cancer more efficiently, but it seems that they are not stopping there. DeepMind AI, owned by Google’s parent company Alphabet, has developed an algorithm that can detect what is known as Acute Kidney Injury (AKI) 48 hours before it happens, thus having the potential to save lives.

AKI, for those unfamiliar, is a sudden episode of kidney failure where a build up of waste products in your blood makes it hard for your kidneys to do its job, ultimately affecting other organs like your brain, heart, and/or lungs. This AI was developed in partnership with the US Department of Veteran Affairs.

Based on their initial tests of their model, their AKI detection algorithm managed to correctly identified 9 out of 10 patients whose condition worsened to the point where they were in need of dialysis. This is a huge deal as AKI is said to kill about 500,000 people in the US and 100,000 in the UK annually, usually because it is not detected soon enough.

Tyler Lee
Categories: Computers, Medical
Tags: AI (Artificial Intelligence), deepmind, Google