REQUEST A CONSULTATION

Ready to contact us for a consultation about your condition and our medical services? The staff of our Global Patient Services is ready to help. Please select the appropriate button to get started.
search-icon
News

Sheba’s Improved Model for Predicting Respiratory Failure in Patients with COVID-19

Respiratory Failure
Analyzing electronic medical records of patients in the MDClone big data platform, the new model will help prevent patient deterioration

Since COVID-19 emerged in 2019, multiple waves have mounted heavy pressure on healthcare systems worldwide, straining existing intake protocols and the control of patient flow. The National Early Warning Score (NEWS) is one of the existing models used to deal with the challenge. However, the model was found to have significant flaws. 

In an effort to find an alternative for NEWS, Sheba analyzed readily available clinical data on MDClone’s big data platform to create a new model. The improved model has shown better performance in 385 patients with COVID-19, of whom 42 required intubation. By enhancing the ability to predict deterioration among COVID-19 patients, Sheba hopes to optimize triage, ensuring as many lives as possible are saved even when the medical center is stretched thin due to the constraints of a COVID-19 outbreak.

Recognized across the globe for advancing healthcare, this latest development follows innovations such as AI-based quick blood testing, a monitoring system for patients in home quarantine, and many other solutions deployed at Sheba during the pandemic.

Related
virtual care
Nuvo Group and Sheba Develop AI-Powered Gestational Diabetes Management Platform
A groundbreaking gestational diabetes pilot was recently completed by the two organizations, deploying Nuvo's INVU platform at Sheba’s Beyond Virtual Hospital, simplifying care for pregnant…
Read More
pain treatment for pancreatic cancer
Breakthrough Study: Radiation Therapy Effective for Pain Management in Cancer Patients
Pancreatic cancer, the 12th most common cancer worldwide, is associated with pain in approximately 75% of patients at the time of diagnosis, and as much…
Read More
arc center
News Nov 30.
Real-Time Detection of Cancerous Biomarkers With Artificial Intelligence
Lung cancer is the most common cancer and accounts for 1.76 million deaths per year worldwide. Non-small cell lung cancer (NSCLC) comprises 85% of all…
Read More