Anyone who has recently kept a bedside vigil in a hospital’s intensive care unit (ICU) is likely to have been impressed by just how much information is now gathered on a critically ill patient. Sophisticated monitors surround the bed, the electronic screens spewing out a torrent of data. That information can be vital, particularly for those near death. But it also poses a critical challenge for today’s hurried nurses and physicians: how to make sense of all that data.
To help out, researchers at the University of Pennsylvania Medical Center are testing a “smart” ICU technology that collects and analyzes a patient’s vital signs. The artificial-intelligence system produces a 3-D graph that could make it easier for a clinician to quickly pick up any warning signs. The smart ICU uses adapted off-the-shelf software for neural networks and fuzzy logic that allows it to analyze several measurements simultaneously, looking for dangerous trends; it can also “learn” the patterns of a patient, including their ideal vital signs.
“It’s not intended to replace a physician,” says C. William Hanson, section chief of anesthesia/critical care at the medical center and developer of the technology. “It’s intended to act as an intelligent assistant that’s vigilant.” Patients in the ICU, says Hanson, are “monitored for everything we can monitor electronically, and there’s very sophisticated minute-by-minute information” on everything from respiratory rates to blood flow. But actual analysis of the data “is, in reality, totally unsophisticated,” says Hanson. “There’s hidden information in a lot of that stuff.”
By uncovering that information, the smart ICU will, if it works, flag dangerous conditions soon after they develop. For example, says Hanson, blood accumulating around the heart can be a serious problem for some patients, but it’s often not noticed until the individual’s blood pressure significantly falls. In fact, there are early warning signs, including a series of interrelated, predicting signals, such as a drop in blood pressure and an increase in heart rate. While “a really good clinician” would likely notice those indicators, says Hanson, that “presupposes that there’s a really good, experienced clinician-and that they’re at the bedside.”
Preliminary real-life tests of the system are under way. Hanson notes that the medical center’s ICU is already installing a PC at every bedside that will collect information from the monitors and act as an electronic spreadsheet. “In the world of information technology, that’s a fairly trivial thing to do,” says Hanson. “There are a lot more interesting things the PCs can do. There’s still a lot of smartening up that can go on in the ICU.”
In today’s busy hospitals, a more intelligent ICU can only be a good idea.
This new data poisoning tool lets artists fight back against generative AI
The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models.
Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist
An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.
Data analytics reveal real business value
Sophisticated analytics tools mine insights from data, optimizing operational processes across the enterprise.
The Biggest Questions: What is death?
New neuroscience is challenging our understanding of the dying process—bringing opportunities for the living.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.