WIM no. 17 Spring 2020

מכון ויצמן למדע Mapping corona in stealth mode Predict Corona fuses data on symptoms, geography to predict outbreaks in the making W eizmann Institute scientists have developed a method for monitoring, identifying, and predicting geographical clusters in which the coronavirus is prevalent and will likely spread, by tracking symptoms throughout the Israeli public—both healthy and sick individuals. The method is at the core of the Predict Corona project, spearheaded by Institute’s Prof. Eran Segal and Prof. Benjamin Geiger, together with the Hebrew University’s Prof. Yuval Dor. Predict Corona is being rolled out to the entire Israeli public in coordination with Israel’s Ministry of Health and the country’s HMOs, and is being adopted now by over 10 countries, including the U.S., Italy, and Spain. “Viral spread occurs in clusters of infection,” says Prof. Segal. “Therefore, early identification of clusters may facilitate various actions aimed at slowing down the spread of the virus.” The scientists describe their method in a fast-tracked publication in Nature Medicine. AI in the service of public health Predict Corona is a brief, one-minute voluntary survey that asks Israelis to share their geographical location (including street and neighborhood), and any symptoms associated with COVID-19 —while protecting patient privacy—to assess infection patterns for SARS-CoV-2, the virus that causes COVID-19. The questionnaire tracks the development of virus-induced symptoms, and the analysis relies on Big Data algorithms and artificial intelligence. Already, the scientists have detected a significant increase in symptoms reported by the public in areas where verified coronavirus patients are known to have been. The mapping may enable health authorities to concentrate anti-contagion efforts on areas in which an outbreak and spread of the virus is predicted—while allowing them to ease measures in areas where an outbreak is not expected. The initiative relies on this daily questionnaire, which can be accessed in five languages (Hebrew, Arabic, English, Russian, and Amharic). Together with Prof. Ran Balicer of the Clalit Research Institute (affiliated with Clalit Health Services, Israel’s largest HMO) and other researchers, the scientists called on other countries to implement the strategy. The interest was immediate, and countries including the U.S., Italy, Spain, India, Luxembourg, Malaysia, Germany, and Britain, are using the tool. Prof. Segal and researchers from the U.S. are spearheading the formation of an international consortium that will share methods, insights, and summary data to build predictive tools and compare across all countries. Companies such as Intel, as well as academic researchers and members of the Israeli military, have reached out to offer their assistance and are already contributing to this effort. g Prof. Benny Geiger g Prof. Eran Segal A framework for identifying regional outbreak and spread of COVID-19 from one-minute population-wide surveys Weizmann MAGAZINE Coronavirus Response