US approves first AI-based device to identify hidden signs of COVID-19 in asymptomatic carriers

The U.S. Food and Drug Administration (FDA) has authorised the first AI-based device to identify hidden signs of COVID-19 in asymptomatic carriers, reports the agency’s press office.

The Tiger Tech COVID Plus Monitor is a bandage-like patch that uses light sensors and a small computer processor to check viral biomarkers, such as hypercoagulation — a common COVID-19 abnormality that can lead to increased blood clotting.

Tiger Tech COVID Plus Monitor. Data: Tiger Tech Solutions, Inc.

The device’s sensors collect pulsatile blood-flow signals over three to five minutes. The processor then extracts key information from the measurements and runs them through a machine-learning model.

Final results, including whether the biomarkers are positive, are displayed in different colours.

Clinical trials of the bandage’s effectiveness were conducted in medical facilities and schools. In hospital settings it correctly identified COVID-19 biomarkers in 98.6% of patients.

It also identified the absence of control markers in people with negative tests — at around 94.5%. According to the FDA, school trials confirmed these results.

The agency stressed that the device is not intended to diagnose the coronavirus infection and does not replace a standard test. It also will not assist in treating patients with symptoms. The bandage should serve as an additional diagnostic measure, especially in potentially asymptomatic cases.

Developers of the device and the FDA say that using multiple methods in combination is useful for limiting the virus’s spread in public places, including healthcare facilities, schools, workplaces, parks, stadiums and airports.

Earlier in 2021, researchers presented a machine-learning algorithm that could determine, with up to 90% probability, whether an uninfected person would die from COVID-19.

Researchers say their development will help hospitals allocate resources more effectively to assist patients with the coronavirus.

In February, an AI algorithm estimated that the number of COVID-19 cases in the United States was three times higher than official statistics claimed.

In March, British researchers created a screening tool for COVID-19 that can diagnose the virus with 98% accuracy by analyzing the sound of a cough using AI.

Subscribe to ForkLog’s channel on YouTube!