Researchers present the first artificial intelligence tool to detect the probability of COVID-19 # 215058

Washington, December 12: A study in the Journal of Medical Internet Research featured Biocogniv’s new AI-COVID software that can easily predict the likelihood of COVID-19 infection. A team of researchers from the University of Vermont and Cedars-Sinai found high precision in predicting the likelihood of COVID-19 infection using routine blood tests, which can help hospitals reduce the number of referred patients. for the rare PCR tests.

Lead author and assistant professor at the University of Vermont Timothy Plante, MD, MHS said, “Nine months after this pandemic, we now have a better understanding of how to care for COVID-19 patients, but there is still a great neck. bottle in COVID-19 diagnosis with PCR test “.

The PCR test is the current standard diagnosis for COVID-19 and requires specific sampling, such as a nasal swab, and specialized laboratory equipment to function.

“Based on data from more than 100 US hospitals, the national average response time for COVID-19 tests ordered in emergency rooms is greater than 24 hours, far from the expected response time of one hour,” said Biocogniv’s chief operating officer, Tanya Kanigan, Ph.D. Complete blood count and complete metabolic panels are common laboratory tests requested by emergency departments and have a quick turnaround time. These tests provide information about the immune system, electrolytes, kidneys, and liver. The researchers were able to train a model that analyzes changes in these routine tests and assigns a probability that the patient is negative for COVID-19 with high precision.

Jennifer Joe, MD, an emergency physician in Boston, Mass. And Biocogniv’s chief medical officer, said: “AI-COVID takes a few seconds to generate its informational result once these blood tests return, which can then be incorporated by the lab. in his own interpretation of evidence.. “

Cedars-Sinai pulmonary and internal medicine specialist Victor Tapson, MD, says these assistive tools that help doctors rule out possible diagnoses are familiar in emergency medicine.

“For example, a low-D-dimer blood test can help us rule out clots in certain patients, allowing providers to skip expensive and often time-consuming diagnoses, such as chest CT scans,” Tapson said.

The Biocogniv team believes that a secondary benefit of labs incorporating AI-COVID could be the reduction in time for traditional PCR results.

“With the help of AI-COVID, labs could alleviate some of the testing bottlenecks by helping providers better allocate rapid PCR tests to patients who really need them,” said Joe.

The AI-COVID model was validated with real-world data from Cedars-Sinai, as well as geographically and demographically diverse patient encounter data from 22 U.S. hospitals, achieving an area under the curve (or AUC) of 0 , 91 over 1.00.

Biocogniv Chief Scientist George Hauser, MD, a pathologist, said: “This allows the model to achieve a high sensitivity of 95 percent while maintaining a moderate specificity of 49 percent, which is very similar to the performance of other test rules. discard commonly used tests. “

“I am honored to have such an impressive team of University of Vermont and Cedars-Sinai medical scientists as collaborators invalidating this timely model,” said Biocogniv CEO Artur Adib, PhD.

“Artificial intelligence has made considerable progress; now is the time to harness this powerful tool for new advancements in healthcare, and we are pleased to direct it to help laboratories and hospital providers combat the current COVID-19 crisis,” he added Adib.