Rapid elimination of COVID-19 infection using AI, EKG technology-ScienceDaily

AI-enhanced EKG was able to detect COVID-19 infection in the test, with 37% positive predictions (infected people) and 91% negative predictions (non-infected people). When additional normal control subjects were added to reflect the 5% prevalence of COVID-19-similar to the actual population-the negative prediction jumped to 99.2%.The survey results are published at Mayo Clinic Minutes..

The incubation period of COVID-19 is 10 to 14 days, which is longer than other common viruses. Many people show no signs of infection and can unknowingly endanger others. In addition, the time and clinical resources required for current testing methods are significant and access can be an issue.

“Positive verification using smartphone electrodes will make diagnosing COVID infections even easier and highlight what we can do with international cooperation,” said Paul of Rochester Cardiology at May York Clinic. Friedman, MD, said. Dr. Friedman is the senior author of this study.

With the realization of the global health crisis, stakeholders around the world have come together to develop tools that can address the need to quickly, non-invasively and cost-effectively eliminate the presence of acute COVID-19 infections. This study, which includes data from racially diverse populations, was conducted through a global volunteer consortium across four continents and 14 countries.

“Lessons from this Global Working Group show what is feasible, and the need for industry and academia members to use their own EKG systems, electronic health records, and variables from multiple centers. We are now partnering to solve complex problems with how data is collected and transferred. Accessing your own data, “said Suraj Kapa, ​​MD, an electrocardiologist at May York Clinic. “Through this collaboration, sophisticated relationships and data processing frameworks can support the development and validation of new algorithms in the future.”

Researchers have selected patients with EKG data since the diagnosis of COVID-19 was confirmed by genetic testing for the SARS-Co-V-2 virus. These data were control-matched with similar EKG data from patients not infected with COVID-19.

Researchers trained AI with over 26,000 EKGs and used nearly 4,000 to validate their measurements. Finally, the AI ​​was tested on 7,870 previously unused EKGs. With each of these sets, the prevalence of COVID-19 was approximately 33%.

Next, over 50,000 regular EKGs were added to accurately reflect the actual population, reaching a prevalence of COVID-19 of 5%. This has increased the negative predictive value of AI from 91% to 99.2%.

Mayo Clinic engineer in cardiology, Zachi Attia, Ph.D. Explains that prevalence is a variable in the calculation of positive and negative predictors. Specifically, as the prevalence decreases, the negative predictive value increases. Dr. Attia is the co-lead author with Dr. Kapa.

“Accuracy is one of the biggest hurdles in determining the value of a COVID-19 test,” says Dr. Attia. “We need to know not only the sensitivity and specificity of the test, but also the prevalence of the disease. Adding additional control EKG data is important to show how the prevalence of the disease fluctuates. The incidence of illness at different stages of the pandemic varies widely and affects how the test is performed. “

“This study showed the presence of a biological signal of EKG consistent with COVID-19 infection, but included patients with many illnesses. This is a hopeful signal, but smartphone-based. This should be positively tested and confirmed in asymptomatic people using the electrodes of, which can actually be used in the fight against pandemics, “says Dr. Friedman. “Research is currently underway to address the issue.”

About this study

This study was designed and devised by Mayo Clinic researchers, and the work was a charitable donation from the Lerer Family Charitable Foundation Inc. and voluntary contributions from participating physicians and hospitals around the world. Partially made possible by the support. To fight the COVID-19 pandemic. Technical support was donated by GE Healthcare, Philips and Epiphany Healthcare for the transfer of EKG data.

Story source:

material Provided by May York Clinic.. Original written by Terry Malloy. Note: The content can be edited in style and length.