Technion delves deeper into oximetric data to improve COVID-19 treatment
The researchers have developed a new set of tools to analyze the data collected by oximeters, which monitor oxygen saturation levels in a patient's blood.
By DONNA RACHEL EDMUNDSMedical worker is seen at the intensive care unit (ICU) of Jinyintan hospital in Wuhan, the epicentre of the novel coronavirus outbreak(photo credit: REUTERS)
Information gathered by oximeters used to monitor patients with COVID-19 could prove to help doctors predict deterioration in a patient's condition, researchers at the Technion-Israel Institute of Technology and Rambam Health Care Campus have said.The researchers have developed a new set of tools to analyze the data collected by oximeters, which monitor oxygen saturation levels in a patient's blood. Low oxygen levels in blood indicate low oxygen in the body's tissues, which can ultimately lead to organ failure.Oximeters are currently used in two ways to monitor patients with coronavirus. Firstly, individuals with mild cases or suspected cases are being supplied with low-cost commercial oximeters to self-monitor their oxygen saturation levels at home. However, a number of the devices used in this setting, even those which have CE or FDA approval, have limited accuracy and therefore don't produce data of a level that is suitable for research analysis.Secondly, and in the context of the research, more importantly, patients in the intensive care unit are constantly monitored with oximeters but no algorithms have yet been developed capable of analyzing the data generated over time. It is here that the toolbox developed by the laboratory for Artificial Intelligence in Medicine in the Technion Faculty of Biomedical Engineering comes in.Master’s student Jeremy Levy and Assistant Professor Joachim Behar, together with Dr. Ronit Almog and Dr. Danny Eytan from Rambam Health Care Campus have developed a range of oximetry biomarkers to analyze the data generated over time by the oximeters, in the hope that it will lead to better monitoring of ICU patients and allow doctors to more accurately predict when a patient is starting to deteriorate, allowing for earlier intervention in their treatment."The hypothesis is that we will be able to identify patterns of deterioration of COVID-19 patients from their oximetry physiological time series. We developed analysis tools (digital biomarkers) to uncover such patterns," Behar told The Jerusalem Post. "We welcome medical collaborations that may contribute such oximetry data from COVID-19 patients so that we may analyse and elaborate such data driven prognosis algorithms."So far the researchers haven't set a time frame for when they expect to see results, but are focused on gathering as much data as possible in order to build up a comprehensive picture.
