London: Scientists have developed a new smartphone app that can accurately detect COVID-19 infection in people’s voices using artificial intelligence (AI). The AI model used in the research is more accurate than rapid antigen test or lateral flow test and is cheaper, faster and easier to use, the researchers said. He said this method could be used in low-income countries where PCR tests are expensive and difficult to distribute. The discovery was presented Monday at the European Respiratory Society International Congress in Barcelona, Spain.
According to the researchers, the AI model is accurate 89 percent of the time, while the accuracy of lateral flow tests varies widely depending on the brand. Furthermore, lateral flow tests are significantly less accurate in detecting COVID-19 infection in people showing no symptoms, he said.
Wafa Aljabwi, a researcher at Maastricht University, Netherlands, said, “These promising results suggest that simple voice recording and fine-tune AI algorithms can potentially achieve high precision in determining which patients have COVID-19.” -19 is an infection.”
“Such tests can be provided at no cost and are easy to interpret. In addition, they enable remote, virtual testing and turnaround times of less than a minute,” Aljawabi said.
The new test could be used, for example, at entry points for large gatherings, capable of rapidly screening populations, the researchers said.
The COVID-19 infection usually affects the upper respiratory tract and vocal cords, causing a change in a person’s voice.
Aljabvi and her supervisors used data from the University of Cambridge’s crowd-sourcing COVID-19 Sounds app, which has 893 audio samples from 4,352 healthy and non-healthy participants, of whom 308 had tested positive for COVID-19.
The app gets installed on the user’s phone. Participants report some basic information about demographics, medical history, and smoking status, and are then asked to record some respiratory sounds.
These included coughing three times, taking deep breaths through the mouth three to five times, and reading a short sentence on the screen three times.
The researchers used a voice analysis technique called mail-spectrogram analysis, which identifies different voice characteristics such as loudness, power and variation over time.
“That way we can decompose many of the properties of the participants’ voices? said Aljabvi.
He said, “In order to differentiate the voices of COVID-19 patients from those who did not have the disease, we created various artificial intelligence models and evaluated how to classify COVID-19 cases. Which one works best in me,” she said.
They found that a model called long-short term memory (LSTM) outperformed other models. LSTM is based on neural networks, which mimic the way the human brain operates and recognizes the underlying relationships in the data.
Its overall accuracy was 89 percent, its ability to correctly detect positive cases, or “sensitivity,” was 89 percent, and its ability to correctly identify negative cases, or “specificity,” was 83 percent, the researchers found.
In another study, Henry Glyde, a PhD student at the University of Bristol, showed that AI can be used through an app called myCOPD to predict when patients with chronic obstructive pulmonary disease (COPD) have their disease outbreaks. It is possible Exacerbation of COPD can be very serious and is associated with an increased risk of hospitalization.
Symptoms include shortness of breath, cough and production of excess mucus.