They manage to detect 80% of coronavirus cases through voice analysis
A study with Artificial Intelligence (AI) in Spain has achieved a success of 80% in the detection of infections by coronavirus through the technology of speech recognition. Systems like these could be used to detect some health problems almost immediately in patients, as explained by its creators to the EFE Agency.
The AI company has collaborated in the investigation Biometric Vox, the hospital Baracaldo crosses, in Vizcaya, and Easter Sunday-Figal, Head of Cardiology at the Virgen de la Arrixaca Hospital in Murcia, as well as the Basque Government Department of Health, its BioCruces Bizkaia Research Institute and the Basque Foundation for Innovation and Health Research.
The study has been developed in two phases: referral and validation. In the first, the system I analyzed voices of some infected patients and of others who had not had the disease. In the second, the software was applied to people for whom data was not available, reaching a percentage of success or “sensitivity” of 80%.
Fast detection
In an interview with EFE, Pascual-Figal noted that this study emerged early in the pandemic, when antigen testing and PCR they were not yet widespread. Therefore, they opted to focus on this situation the technology that they had previously devised together with Biometric Vox to patients with heart attacks.
The cardiologist has indicated that, like people suffering from acute coronary ailments, those infected with coronavirus suffer alterations in the voice undetectable to the human ear, but not to AI.
Although the massive implantation of antigen tests and PCR has stolen prominence from this method, Pascual-Figal has highlighted the importance that this type of AI can have in the rapid diagnosis of acute and serious illnesses, when health care immediate it is fundamental to saving lives.
According to the cardiologist, voice analysis could join, among other things, emergency calls. “It will be one more part of the analysis of a call made by an emergency operator to decide the necessary response,” said Pascual-Figal, recalling thatthe machines already carry out other tasks like blood and urine tests and even help interpret X-rays and scans.
The role of technology
Pascual-Figal has underlined that technology can become the basis for 70-90% of medical diagnosis and treatment. Even so, he clarifies, the analysis made by the health specialist.
In addition, the researcher has stated that the most important aspect in health is prevention and that society must be aware that good habits in its daily life are the basis for avoiding diseases.
–
