The algorithm, developed by Maccabi and AI health leader Medial EarlySign, has identified approximately 40,000 patients at highest risk of COVID-19 complications for fast-tracked testing
TEL AVIV, Israel – April 22, 2020 – Maccabi Healthcare Services, Israel’s leading HMO with 2.4 million members, announced today the deployment of a new AI-powered algorithm that identifies individuals estimated to be at the highest risk of severe COVID-19 complications due to pre-existing conditions and other health factors. The new algorithm has already identified the top 2% of highest-risk patients (approximately 40,000 people), following analysis of all Maccabi patients’ anonymized electronic health records (EHRs).
The algorithm was developed by Medial EarlySign and the Kahn-Sagol-Maccabi Research and Innovation Institute. Medial EarlySign is a technology leader in machine learning-based solutions that aid in the early detection and prevention of high-burden diseases. The company is currently in advanced negotiations with prominent medical systems in the United States who are interested in the algorithm as part of their COVID-19 healthcare protocols.
“The world is currently at war with COVID-19 and our algorithm, developed together with EarlySign, will help us fight the virus effectively,” said Ran Sa’ar CEO of Maccabi Healthcare Services. “The algorithm and the fast-tracked testing it enables will reduce the number of severe COVID-19 cases and help save lives. This innovation demonstrates Maccabi’s unique development and implementation capabilities and our robust big data.”
When an individual flagged by the algorithm as high-risk contacts a nurse or doctor to report COVID-19-like symptoms, the system will automatically notify the medical professional that the patient is in the high-risk group. The patient will then be sent for immediate testing. Tests are performed at designated Maccabi facilities, drive-in stations or, if necessary, in the patient’s home. This allows for medical procedures to begin as quickly as possible following a positive diagnosis, helping to limit the spread of the virus.
The algorithm identifies high-risk patients through analysis of dozens of routine medical factors, including:
- Respiratory disease such as pneumonia, bronchiolitis, and influenza
- Hospital admission history
- Weight and BMI
- Medications prescribed for respiratory illnesses or conditions, such as asthma and cough
- Heart disease
- Smoking history
- Digestive disease
The new algorithm enables Maccabi to objectively select the patients who should be fast-tracked for COVID-19 testing and classifies them according to three levels of estimated risk. Maccabi’s COVID-19 medical task force can also use the risk levels as part of its decision-making on optimal hospitalization options for each patient – home hospitalization, designated hotels, or hospital admission – and the necessary frequency of follow-ups for each patient.
“As one of the largest HMOs in the world with 2.4 million members and 27 years’ worth of electronic health records, Maccabi’s algorithm relies on big data from one of the largest and highest-quality collections of anonymized EHRs in the world,” said Prof. Varda Shalev, director of the Kahn-Sagol-Maccabi Research and Innovation Institute. “Early identification of those at greatest risk is crucial to supporting healthcare professionals and to flattening the curve of the pandemic.”
“In building the new algorithm for COVID-19, we utilized numerous aspects of our Flu Complications AlgoMarker,” said Dr. Jeremy Orr, CEO of Medial EarlySign. “Based on anonymized data collected from millions of people treated by Maccabi, our models were adapted and optimized according to existing medical knowledge and known risk factors relating to COVID-19. The data from Maccabi is being continually updated, allowing us to further improve this essential new algorithm.”
# # #