This story is over 5 years old.

'Medical Robot Assistants' Are Helping Nurses Schedule Tasks in the Labor Ward

A robot developed at MIT makes suggestions to help nurses juggle scheduling logistics in the labor ward.

In the event of a c-section, if a robot suggested who should perform it, would you listen? Ninety percent of the time, nurses and physicians did.

A robot programmed by MIT's Computer Science and Artificial Intelligence Lab (CSAIL) suggests where to move patients and who should perform caesarian sections. The team from CSAIL thinks robots are most effective in helping with one of the most "complex" tasks in the labor ward of the hospital: scheduling.


"The aim of the work was to develop artificial intelligence that can learn from people about how the labor and delivery unit works, so that robots can better anticipate how to be helpful or when to stay out of the way—and maybe even help by collaborating in making challenging decisions," says MIT professor Julie Shah, senior author on two papers based on CSAIL's research. One paper focused on the robot's decision making the Beth Israel hospital labor ward and was presented at the recent Robotics: Science and Systems (RSS) Conference at the University of Michigan; the other paper, which focused on the robot's capabilities in a navy simulation, was presented at this week's International Joint Conference on Artificial Intelligence (IJCAI). Both papers were published through the Interactive Robotics Group at MIT.

In Boston's Beth Israel labor ward, where nurses have to coordinate a dozen other nurses to juggle upwards of 20 patients and 20 hospital rooms, and make split-second decisions about when to perform a C-section, the complexity of scheduling proved to be a viable task to pawn off on the robot.

The program developed by CSAIL not only suggests what would be a good decision in the workplace, but also what would be a bad idea, as well. For instance, in several scenarios, nurses asked the robot, "What is a good decision?" and it would respond with recommendations about where to place a patient or who should perform a c-section. Nurses would then follow up with the question, "What is a bad decision?" and the robot would provide an alternate, and presumably unpreferable, suggestion in the same exact format as the good decision: "A bad decision would be to place a scheduled cesarean section patient in room 14 and have nurse Kristen take care of her."


The CSAIL robot, which learns from human workers to help assign and schedule tasks, can be used in a variety of fields, from medicine to the military.

The same system was also tested in a video game simulating missile-defense scenarios in the Navy.

The researchers trained the robot how to make scheduling and logistics decisions by looking at several possible decisions and comparing them to those that are simultaneously not made in the decision-making moment, creating a dynamic scheduling policy.

In the video game simulation, the robot system sometimes even outperformed human experts on reducing attacks and bringing costs down.

The robotic system was tested out on a Nao robot, who presented successful suggestions in the hospital 90 percent of the time. With this technology, decision making can become faster, as solutions to complex logistics spread among various workplaces.