Feinstein Institutes for Medical Research scientists published in Nature Communications results of a two-and-a-half-year-long study using 35,000 diverse patients to develop a new digital clinical support tool
When the Coronavirus disease 2019 (COVID-19) pandemic gripped the nation, doctors scrambled to keep up with their patients’ constantly changing and unpredictable outlook. The Feinstein Institutes for Medical Research scientists saw early potential in using diverse patient population data from Northwell Health – New York’s largest health system –and developed a clinical tool to better inform frontline staff on a patient’s prognosis and severity of the disease. Two-and-half-years later, pulling from 35,000 COVID-19 inpatients, researchers published results in Nature Communications detailing how data from basic blood work and other vitals could accurately project a patient’s outcomes and aid in clinical care decision-making.
Latest NaturalAI Insights: InspireXT Announces Acquisition Of NaturalAI – A Conversational Artificial Intelligence Platform To Expand Its Solution Portfolio
“By harnessing data and developing a real-time auto-updating clinical tool, we set out to create a tool that accounts for these developments and helps clinicians make the decisions they need to deliver better care.”
The research, led by Theo Zanos, PhD, associate professor at the Feinstein Institutes’ Institute of Health System Science and Institute of Bioelectronic Medicine, accounted for rapid changes in patient condition and outcomes across different COVID-19 waves and the variants that caused them, along with the emergence of vaccines and treatments, and created a self-monitoring, auto-updating artificial intelligence (AI) model clinical support tool. Dr. Zanos and his lab analyzed electronic health record (EHR) data from patients hospitalized across 13 Northwell Health hospitals between April 2020 and May 2022.
“COVID-19 was one of the most dynamic diseases we’ve witnessed in modern history and information about how to care for patients was constantly evolving,” said Dr. Zanos, senior author of the paper. “By harnessing data and developing a real-time auto-updating clinical tool, we set out to create a tool that accounts for these developments and helps clinicians make the decisions they need to deliver better care.”
AI News: Infobip Creates AI-powered Chatbot for Uber
Current predictive models for COVID-19 patients do not account for real-time shifts in patient characteristics and outcomes. In Dr. Zanos’ new model, the proposed framework continuously monitors the predictive performance and is automatically updated when it detects performance drifts. This framework was crucial to provide clinicians accurate predictions for 28-day survival in COVID-19 patients throughout all stages of the pandemic and can be applied similarly on other dynamic diseases.
The model uses only five early and commonly collected patient data points upon hospitalization; age, serum urea nitrogen, lactate, serum albumin and red cell distribution width. The model remained accurate across the over two-year study, occurring across four waves and three dominant variants (Alpha, Delta and Omicron) and performs equally regardless of race, ethnicity and gender.
“This important study harnessed data analytics and technology to develop novel insights into a new illness,” said Kevin J. Tracey, MD, president and CEO of the Feinstein Institutes and Karches Family Distinguished Chair in Medical Research. “Dr. Zanos’ strategy provides a new model to study COVID-19 as a guide to clinical decision-making and better outcomes.”
In addition to the published study that details the validation and techniques to build the tool, the team is providing the medical and scientific community with the open-source code of the framework, the clinical data used to derive these models, as well as an online calculator (mlmd.org/nocos) that allows users to plug in information and view the tool’s functionality.
Latest Aithority Insights : NVIDIA Raises the Standard of Low Code DevOps with the NVIDIA AI Enterprise 2.1
[To share your insights with us, please write to sghosh@martechseries.com]
Comments are closed.