Patients rinse their mouth with a saline wash and spit into a vial. This is then examined by a small spectral device that, in simple terms, shines light on the specimen and analyses the reaction to see if it is consistent with COVID-19.

With machine learning it gets more accurate over time.

Eli Schwartz of the Center for Geographic Medicine and Tropical Diseases at Sheba Medical Center, who is leading the trial, said it was easier to use than PCR swabs commonly used to detect COVID-19.

"So far we have very promising results in this new method which will be much more convenient and much cheaper," he said.

The center said in an initial clinical trial involving hundreds of patients, the new artificial intelligence-based device identified evidence of the virus in the body at a 95% success rate.

Amos Panet, an expert in molecular virology at Jerusalem's Hebrew University, said he would like to see more data and comparisons with existing tests before making a final judgment.

The amount of virus present in saliva increases as patients get sicker, he said, and a big challenge is to detect in "people who are borderline".

"It will be a game changer only if we see validation of this technology against the current technology," he said.

Sheba, located just outside Tel Aviv, has partnered with the device's developer, Israeli firm Newsight Imaging, to bring the system to market.

The company said they are in the process of getting regulatory approval. Each test costs less than 25 cents and it expects the device will eventually cost less than $200.

Popular

Kadin urges stability after Constitutional Court ruling

Jokowi says transition to new government to begi

BCA loan disbursement surges 17.1 percent yoy in first quarter