Amazon grants AI tech to a Harvard-affiliated teaching hospital

Amazon is now working with a Harvard-affiliated hospital to improve the efficiency of their healthcare systems using AI. The Boston-based hospital Beth Israel Deaconess Medical Center (BIDMC) in union with Amazon is majorly focusing on projects that would increase the cost-effectiveness of daily operations, such as patient schedules, as of now.

“Every minute spent on cumbersome clerical tasks and management add up to millions in lost productivity and directly impacts patient care,” said Dr. John Halamka, executive director of the Health Technology Exploration Center at BIDMC. But, with useful tools being offered by Amazon, it aids BIDMC book operating rooms more precisely; predict when patients are likely to miss appointments with the hospital’s superior specialists; alert the hospital staff in case of any incomplete or missing documentation, and much more. Additionally, Amazon’s software will also help the hospital staff track down the necessary documents of a patient, thus eliminating any delay in procedures as complicated as surgeries, for example.

BIDMC is already a subscriber of Amazon’s cloud-based services. Since 2016, BIDMC has used Amazon services to store and manage a copy of the hospital’s vital data, so that it can almost instantly recover from any disaster that would have wiped out its off-site backup servers. The medical center is currently making use of tools such as Amazon SageMaker, Apache MXNet, Tensorflow and Amazon Comprehend Medical that are all hosted on AWS.

Amazon and BIDMC conducted an analysis of data from surgeries that took place in the hospital since the 1980s. They then developed and tested a scheduling system – over the last two years – which they claim has made the booking of operating tables as easy as booking a restaurant’s. The new system was also able to free up space and increase operating room capacity by 30 percent, the studies reveal.