Using data collected from patients’ vital signs, University of Florida researchers have designed an artificial intelligence system that can speed up and focus physicians’ decision-making during the critical early stages of hospitalization.
The algorithm works by taking a torrent of data from six vital signs measured within six hours of hospitalization. It then focuses this data into one of four distinct groups, giving clinicians a clearer, more accurate, and more accurate view of a patient’s prognosis and potential medical outcomes. The Results have been published October 13 in the magazine Digital Health PLOS.
He said the approach uses artificial intelligence to analyze patient data faster and more comprehensively than doctors Azra Bihorak, MDSenior Associate Dean for Research Affairs at UF Medical College And UF’s manager Intelligent Critical Care Center. Within hours, the system can identify patients who may be at risk of poor outcomes.
“This system has the potential to speed up clinicians’ decision-making process as well as make it more accurate,” Bihorak said.
These findings are the result of a collaboration between dozens of UF researchers with expertise in surgery, computer science, medicine, anesthesiology, and biomedical engineering.
To evaluate the system, the researchers used a de-identified database of adult patients admitted to UF Health Shands Hospital between 2014 and 2016. The algorithm was validated and tested using data from nearly 100,000 people of all age groups.
When machine learning, a type of artificial intelligence, was applied to early, routine vital sign data, the system identified patients with unique disease categories and distinct clinical outcomes. The patients were then grouped into one of four distinct ‘groups’. Patients assigned to one of the groups showed early signs of hypotension, increased cardiac activity, and low-grade inflammation. While these cases can be severe in their early stages, they also have the potential to resolve them and produce positive outcomes. The algorithm grouped other patients into another group most likely to have chronic renal and cardiovascular disease. The researchers found that they were more likely to die within three years.
The value of the algorithm lies in its ability to quickly collect and analyze multiple data points for patients, Bihorak said. For example, low blood pressure can be an early indicator of various medical problems in the future. When combined with other patient data and analyzed by an algorithm, clinicians have a clearer picture of the patient’s path.
“It’s really like an early warning sign. Within six hours, it can help identify patients who may be at risk of not doing well. It tells us which patients may be at risk of deterioration and who needs more attention right away.”
Next, Bihorak said it is seeking additional grants that will allow the team to further study the system and eventually test its effectiveness in patients currently in hospital. She said it is possible to deploy such a system without much cost.
“This is a simple and elegant solution. It takes the data that is already collected and uses it to its full potential to benefit the patient,” Bihorak said.
Includes colleagues from the UF Intelligent Critical Care Center who have made notable contributions to research Yuanfang Ren, Ph.D.a computer science expert and assistant scientist at the UF School of Medicine. Tyler J. Loftus, MDwho is an assistant professor at Department of Surgery; And the Gilbert R. Abashrash, MDsaid Bihorak, who is a professor and chair of the department of surgery.
The research was supported by multiple grants from the National Institutes of Health, the National Science Foundation and the University of Florida.
Media contact: Doug Bennett, email@example.com, 352-265-9400