Israeli Scientists Use AI to Detect COVID-19 Clusters

AI will identify SARS-CoV-2 coronavirus hotspots and predict COVID-19 outbreaks.

Posted Mar 17, 2020

Coffee/Pixabay
Source: Coffee/Pixabay

The COVID-19 pandemic is growing at an alarming rate globally. The actual number of those infected worldwide with the SARS-CoV-2 coronavirus that causes the COVID-19 disease is unknown for a variety of reasons, including the lack of widespread testing. To confirm a case of COVID-19 requires accurate testing and the timely reporting of the figures. However, testing for the disease remains a pressing issue for many countries as test kits are in limited supply.

On March 16, 2020, Israeli researchers have launched an artificial intelligence (AI) initiative in efforts to rapidly identify, monitor, and predict COVID-19 disease clusters in hopes of slowing the transmission of the SARS-CoV-2 coronavirus.

The initiative is led by professor Eran Segal at the Weizmann Institute of Science in Rehovot in collaboration with Weizmann professor Benny Geiger and Hebrew University professor Yuval Dor. The concept is straight-forward; apply the pattern-recognition capabilities of AI machine learning towards big data sets of self-reported data from the participating public. The current population of Israel is an estimated 9,092,000 based on the Central Bureau of Statistics, as reported by Ynet in September of 2019. As of March 16, 2020, there are over 50,000 Israelis currently in quarantine according to The Times of Israel.

The researchers hope that people in Israel will complete the questionnaire once daily so that there is enough data to enable the AI algorithm to rapidly detect existing disease hotspots and to provide snapshots of the number of those with symptoms by area. The goal is to have a machine learning algorithm predict future hotspots before an outbreak happens.

The online questionnaire is currently available in both Hebrew and English languages, and it takes less than a few minutes to fill out online. Along with basic demographic questions, the questionnaire asks for symptoms (cough, fatigue, muscle pain, shortness of breath, runny nose or nasal congestion, diarrhea, nausea, vomiting), any existing diagnoses of conditions (diabetes, hypertension, ischemic heart disease, asthma, chronic lung or kidney disease), cigarette smoking habits (if any), current body temperature, and isolation status.  

The researchers plan to evaluate the effectiveness of the pilot project. If the artificial intelligence project is deemed successful, the researchers plan to create a smartphone app for data collection.

By using artificial intelligence, the researchers hope to provide insights on the efficacy of social distancing measures and the impact it has towards reducing the number of symptomatic people. If the pilot is successful, it may serve as a tool to assist public health leaders in making more informed decisions on where social distancing measures need to be strengthened in order to fight the spread of COVID-19.

Copyright © 2020 Cami Rosso All rights reserved.