Using AI to assess chest CT images is valid for making COVID-19 diagnoses, according to a group of researchers at Johns Hopkins Medicine.

The team asserts that recognizing the use of AI expands the role of CT from a means for screening for signs of COVID-19 in the lungs to a tool that can be used to diagnose the virus earlier and enable quicker intervention.

“AI’s power to generate models from large volumes of information — fusing molecular, clinical, epidemiological and imaging data — may accelerate solutions to detect, contain and treat COVID-19,” said the group in a statement.
CT chest imaging, while not currently a routine method for diagnosing COVID-19, excludes other possible causes for COVID-like symptoms and is used to confirm diagnoses made by other means. It also is used to monitor the progress of patients with severe cases of the disease.

The researchers say that applying to CT chest images can unlock “untapped potential” in diagnosing and monitoring COVID-19. Among this potential is risk stratification, the process of categorizing patients for the type of care they receive based on the predicted course of their COVID-19 infection. They also suggest using AI-enhanced CT chest imaging to assess the effectiveness of treatments against the disease, and to model how COVID-19 behaves so that novel, customized therapies can be developed, tested and deployed to guide the immune system against the SARS-CoV-2 virus that causes COVID-19.

“AI may help identify the immunological markers most associated with poor clinical course, which may yield new targets,” they said.

Researchers at Johns Hopkins Medicine are not the only ones touting the use of CT imaging as a COVID-19 diagnostic method. A study out of Wuhan, China in February indicated CT to be more accurate than lab tests in confirming the presence of the virus. Shortages in lab tests have also led some providers throughout the U.S. to use CT instead to make diagnoses.

Their findings were published in Radiology: Artificial Intelligence.

CT Homepage