Facebook, NYU develop new AI software using chest x-rays to predict patients’ COVID needs
Facebook researchers have created a new artificial intelligence solution to predict the chance a COVID-19 case may worsen or require additional treatment, such as oxygen.
NYU Langone Health teamed up with Facebook AI Research to develop three machine learning models that they say may help physicians avoid discharging patients too early and plan for potential oxygen needs.
“Our model using sequential chest x-rays can predict up to four days (96 hours) in advance if a patient may need more intensive care solutions, generally outperforming predictions by human experts,” the organization said in a blog post Friday.
The team trained its system on two datasets: one containing 26,838 chest x-rays from nearly 5,000 patients and another smaller set of non-COVID chest images.
Both organizations emphasized their models are not products, but rather “research solutions” designed to help hospitals with resource planning during the ongoing pandemic.
“We have been able to show that with the use of this AI algorithm, serial chest radiographs can predict the need for escalation of care in patients with COVID-19,” William Moore, MD, a professor of radiology at NYU Langone Health, said in the blog post. “As COVID-19 continues to be a major public health issue, the ability to predict a patient’s need for elevation of care—for example, ICU admission—will be essential for hospitals.”
This isn’t the first radiology project developed by the social media giant and New York-based health system. The pair has been working on a separate AI system to enable five-minute MRI scans.
Back in August, the team published a study in the American Journal of Roentgenology detailing their MR imaging-focused solution
Matt joined Chicago’s TriMed team in 2018 covering all areas of health imaging after two years reporting on the hospital field. He holds a bachelor’s in English from UIC, and enjoys a good cup of coffee and an interesting documentary.