Home Health Care AWS grant lets Beth Israel Deaconess Medical Center explore machine learning

AWS grant lets Beth Israel Deaconess Medical Center explore machine learning

112
0
SHARE

With the help of an academic research sponsorship grant from Amazon Web Services, a Harvard Medical School-affiliated teaching hospital has initiated a multi-year program to research how machine learning can improve patient care, according to an Amazon blog post.

The hospital, Beth Israel Deaconess Medical Center in Boston, will utilize AWS machine learning services to uncover data-based solutions to boost patient outcomes and streamline operations.

One of BIDMC’s first projects involved applying machine learning to the hospital environment.

The organization used ML to optimize the schedules of its 41 operating rooms and align them to enhance patient flow.

Another effort harnessed the technology to improve operational flow in operating rooms. Incoming pre-surgical documents will be scanned and processed with TensorFlow on Amazon SageMaker (a machine learning service), which is hosted in BIDMC’s AWS cloud. The goal is to save hospital staff members time. The machine learning process recognizes and inserts consent forms into EHRs, and if a form isn’t found, a signal appears on the EHR to prompt clinicians to follow up with those patients.

“Every minute spent on cumbersome clerical tasks and management adds up to millions in lost productivity and directly impacts patient care,” John Halamka, executive director of the Health Technology Exploration Center at BIDMC and international healthcare innovation professor at Harvard Medical School, said in a statement. “This machine learning research sponsorship will support our commitment to using new and emerging technologies in health care to drive projects that will transform care for patients at BIDMC and around the world.”

BIDMC has also developed a machine learning model built on AWS that can pinpoint where operating room schedule modifications would improve efficiency. The model can predict the outcomes of changes to the schedule as well.

Outside the operating room, the Boston-based medical center is leveraging Amazon SageMaker and the Apache MXNet deep learning API to predict which patients are likely to keep their scheduled office appointments.

Going forward, BIDMC’s projects will explore the level of risk in intensive care units and predict when the hospital will experience an unexpectedly high volume of patients. Using Amazon services, the hospital and research partners will analyze datasets, transfers, referrals and other variables. BIDMC can then use the AWS Cloud to process the data and SageMaker to build machine learning models for predicting where and when hospital space will be available. This process can be applied to the medical center’s emergency department, which typically sees a surge in patient visits in the middle of the week.

Credit: monsitj, Getty Images

Source link

LEAVE A REPLY

Please enter your comment!
Please enter your name here

4 × two =