The Defense Ministry will begin testing a voice sampling application to diagnose symptoms of the novel coronavirus, using artificial intelligence to identify the "unique vocal 'fingerprint" of virus carriers.
A team from the ministry's Directorate for Defense Research and Development (DDR&D) will conduct the trial with Tel Aviv-based start-up Vocalis Health, the developer of a smartphone platform that uses vocal biomarkers and big data to analyze population health.
The start-up plans to extend its artificial intelligence-based voice sampling platform to detect virus symptoms related to the patient's respiratory system. The Defense Ministry tests will sample the voices of patients already diagnosed with COVID-19, as well as carriers and the general public.
Vocal information will be used, the ministry said, to improve early diagnosis and also monitor the deterioration or improvement of patient health. Vocal recordings will undergo data analysis using neural networks and preliminary results of the study are expected within four to six weeks.
"The purpose of this study is to develop a unique algorithm based on machine learning and sound characteristics in order to make initial medical assessments and in order to monitor the symptoms and conditions of patients," the ministry said in a statement.
"The advantage of this monitoring system is that it can be conducted from afar, in order to prevent the spread of the disease and overburdening of the national healthcare system."
Vocalis Health CEO Tal Wendrow said the start-up had decided to focus its efforts on finding a solution for the current crisis, by adapting its technology for the coronavirus.
"This is an example of collaboration between leading organizations, working together around the clock to achieve one goal - to find a solution for the healthcare system," Wendrow said. "Collecting the voices is the initial step in finding the solution, and we urge the public to enter voice samples of the website or application."