Artificial intelligence (AI) has been racking up the wins in health and medical applications these days. From uncovering different forms of diabetes to predicting heart disease from retina scans, machine learning has uncovered all sorts of hidden knowledge that is already saving lives and reducing healthcare costs. Now comes word of a new partnership that may result in further advances.

DeepMind is the company behind AlphaGo, the AI program that beat the world’s best human Go player. Now a part of Alphabet (parent company of Google), DeepMind seeks ways to apply its analytics engines for practical applications, such as reducing power consumption in Google data centers. The company is also actively exploring health and medical applications. The most recent announcement covers a new project in partnership with the U.S. Veterans Administration (VA). DeepMind and VA researchers are going to use depersonalized electronic health records of 700,000 veteran patients in the hopes of being able to identify the early signs of Acute Kidney Injury (AKI), which is a leading cause of death in hospitals. The onset can be sudden — often without obvious symptoms — and can strike people of any age. The goal of the program is to identify factors that can be used to predict AKI before a patient starts to deteriorate noticeably. With earlier detection, the hope is that more effective treatments can be applied sooner, reducing the loss of life and other complications that can arise.

If the project is successful, the DeepMind and VA researchers hope to apply the same AI techniques to examine other types of deterioration such as infections. These efforts may lead to a reduction of in-hospital deaths, saving lives and money in the process.