Google AI Technology That Can Predict Kidney Disease in Advance

Kidney related diseases are increasing all over the world by epidemic proportions. If we speak about numbers then globally, one in every ten individuals is affected by kidney diseases. The 2016 Global Burden of Disease reports an 87% rise in the global burden of kidney diseases and doubling of deaths between 1990 and 2016 due to kidney damage or failure. The reason is, it the acute or chronic disease happens very quickly and with little warning.

Scientists from Google's DeepMind developed an algorithm to give doctors and patients better chances against a fatal type of kidney failure. The AI detect AKI 48 hours before it occurs. (Image Credit: express.co)

To tackle this issue, a United Kingdom based artificial intelligence company Known as “DeepMind” which is actually owned by Google was doing a research. And now they unveils that their AI technology will help in predicting a deadly kidney condition two days before it occurs.

DeepMind revealed this creation of theirs in a paper published in the journal Nature. This artificial intelligence algorithm use to find signs suggesting a patient is at risk of chronic Kidney disease and predict the condition 48 hours before it occurs.   

As per Professor Jermy Hughes who is currently chair of Kidney research UK said in his statement that, “Such digital technologies will undoubtedly open up new possibilities for timely clinical interventions and treatments to reduce harm to patients.” Importantly the model correctly predicted 9 out of 10 patients whose condition deteriorated so severely that they can required dialysis. The findings in research showed that the Artificial Intelligence was 55.8% accurate in predicting overall Acute Kidney Injury cases whereas in cases which are severe enough to require dialysis, it was 90.2% accurate.

To perform the whole study, researchers used data of 703,782 adults from records of U.S Departments of Veteran Affairs. This records also consist of approx. 600000 health indicators, which helps to determine the signs of Acute Kidney Injuries easily but as they were a lot in numbers, so the team picked around 4000 indicators that are considered potentially danger signs of Kidney damage. Using these indicators, DeepMind algorithm can detect the risk of a kidney failure before it occurs.

Professor of Renal Medicine at the University of Manchester, Donal O’Donoghue said that, “This is a game changer that will advance our care and understanding of acute kidney injury”. Feedback from the qualitative study was positive. This is a major milestone for the DeepMind Health team, who will be carrying this work forward as part of Google Health, led by Dr. David Feinberg.