Networking News

Artificial Intelligence Algorithm Identifies COVID-19 in CT Scans

Using artificial intelligence to diagnose COVID-19 based on CT scans and clinical data will help hospitals quickly isolate patients and prevent spreading.

Artificial Intelligence

Source: Thinkstock

By Samantha McGrail

- Researchers at Mount Sinai are the first in the country to use artificial intelligence (AI) to detect COVID-19 in patients based on CT scans of the chest combined with clinical data, according to a recent study published in Nature Medicine

Traditionally, COVID-19 lab tests can take up to two days to complete, but artificial intelligence has the potential to analyze large amounts of data quickly.

“At Mount Sinai, we recognized this early and were able to mobilize the expertise of our faculty and our international collaborations to work on implementing a novel AI model using CT data from coronavirus patients in Chinese medical centers,” Zahi Fayad, PhD, director of the biomedical engineering and imaging institute (BMEII) at the Icahn School of Medicine at Mount Sinai, mentioned a press release.

The study used scans of more than 900 patients that Mount Sinai received from institutional collaborators at hospitals in China. The scans included 419 confirmed COVID-19-positive cases and 486 COVID-19-negative scans.

Chest CT scans are a useful tool for evaluating and diagnosing symptomatic patients suspected to have the virus.

Researchers leveraged patients’ clinical information, including blood test results showing any abnormalities in white blood cell counts or lymphocyte counts, as well as their age, sex, and symptoms. 

The group then integrated data from those CT scans with the clinical information to develop an AI algorithm, which copies the workflow a physician uses to diagnose COVID-19 and provides a positive or negative diagnosis prediction, researchers noted. 

Initially, researchers trained the algorithm on data from 626 out of 905 patients, and then tested the model on the remaining 279 patients in the study group to judge the test’s sensitivity. 

The results showed that the algorithm had statistically significantly higher sensitivity, at 84 percent, compared to radiologists’ sensitivity of 75 percent when evaluating the images and clinical data, researchers said. 

Additionally, the algorithm recognized 68 percent of COVID-19-positive cases, when radiologists found these cases negative due to the negative CT appearance. 

“We were able to show that the AI model was as accurate as an experienced radiologist in diagnosing the disease, and even better in some cases where there was no clear sign of lung disease on CT,” Fayad said.

Researchers noted that accurate detection is vital to keep patients isolated if scans don’t show lung disease when patients first present symptoms. And because COVID-19 symptoms often resemble a flu or common cold, the virus can be difficult to diagnose.

While CT scans are not widely used for diagnosis in the US, imaging can still play a critical role because it gives a rapid and accurate diagnosis, Fayad said. 

“The high sensitivity of our AI model can provide a 'second opinion' to physicians in cases where CT is either negative (in the early course of infection) or shows nonspecific findings, which can be common. It's something that should be considered on a wider scale, especially in the United States, where currently we have more spare capacity for CT scanning than in labs for genetic tests,” he said. 

Matthew Levin, MD, director of Mount Sinai’s clinical data science team and a member of the Mount Sinai COVID informatics center, said that this study is extremely important because it proves that an artificial intelligence algorithm can help with early identification of COVID-19. 

“Artificial intelligence algorithm be used in the clinical setting to triage or prioritize the evaluation of sick patients early in their admission to the emergency room. This is an early proof concept that we can apply to our own patient data to further develop algorithms that are more specific to our region and diverse populations,” he said.