By Regina Stracqualursi
As of Dec. 7, 2020, over 100,000 people across the United States were hospitalized with COVID-19, according to the COVID Tracking Project, and over 20,000 COVID-19 patients are currently in an intensive care unit (ICU). As we enter the ninth month of the current pandemic, it has become clear that the condition of COVID-19 patients can rapidly change, making the respiratory disease challenging to treat. If clinicians can better anticipate the needs of their patients battling COVID-19, they may be able to provide a more effective intervention and, in turn, save more lives.
Recently, an algorithm developed by a team of engineers at Rensselaer Polytechnic Institute was able to successfully predict whether or not a COVID-19 patient would need ICU intervention. The algorithm combines imaging data taken from chest computed tomography (CT) scans with demographic information, vital sign data, and blood test results to make the prediction, which could help clinicians better tailor treatments to individual patients.
The research, supported by the National Institutes of Health and published in Medical Image Analysis, is led by Pingkun Yan, an assistant professor of biomedical engineering at Rensselaer. The team also includes Professor Ge Wang and graduate students Hanqing Chao, Xi Fang, and Jiajin Zhang, as well as collaborators from Massachusetts General Hospital.
“[AI] really enables us to analyze a large quantity of data and also extract the features that may not be that obvious to the human eye,” said Yan. To test the algorithm, the research team compared its predictions to the data of 295 patients from three hospitals in three different countries.
In the next phase of their research, the team will combine the algorithm they developed with a similar algorithm Yan created using chest CT scans to predict cardiovascular disease risk. “We know that a key factor in COVID mortality is whether a patient has underlying conditions and heart disease is a significant comorbidity,” Yan said. “How much this contributes to their disease progress is, right now, fairly subjective. So, we have to have a quantification of their heart condition and then determine how we factor that into this prediction.”
Yan hopes that the research could go on to help patients with other lung diseases, in addition to COVID-19. “Assessing their heart disease condition, together with their lung condition, could better predict their mortality risk so that we can help them to manage their condition,” he said.
