MANHASSET, N.Y.--(BUSINESS WIRE)--There have been more than 3 million confirmed coronavirus disease 2019 (COVID-19) cases reported across the United States. At the height of the pandemic in the Northeast this spring, hospitals were flooded, and doctors, patients, and patient families were asked to make critical care decisions quickly. In a new perspective paper published today in the Springer Nature journal, Bioelectric Medicine, the Feinstein Institutes for Medical Research discussed how artificial intelligence (AI) and machine learning (ML) tools could be used to aid in making those decisions.
Critical care decision-making tools based on ML, or the study of computer algorithms that improve automatically through experience, have been increasingly available, and some have already been used by clinicians worldwide. Throughout a hospitalized patient’s journey, there are opportunities for ML supported choices based on collected vitals, laboratory results, medication orders, and comorbidities. As noted in the paper, datasets collected from patients rapidly grew during the pandemic, which underscores the need to begin to develop and validate “Emergency ML” models to put into practice.
“The rate of increase in the number of patients admitted, triaged and treated during the COVID-19 crisis was unprecedented these past three months,” said Theodoros Zanos, PhD, senior author of the paper and assistant professor at the Institute of Bioelectric Medicine at Feinstein. “Researching patient data from the pandemic is key to creating cutting-edge AI tools that can provide our frontline clinicians with important information and assist them in making evidence-based recommendations.”
Numed, a well established company in business since 1975 provides a wide range of service options including time & material service, PM only contracts, full service contracts, labor only contracts & system relocation. Call 800 96 Numed for more info.
Through this paper, a team of computer scientists and engineers, as well as emergency medicine, intensive care, and hospitalists, offered insights on how and when ML can inform better clinical decisions. As patients move between a range of clinical settings (outpatient clinics, emergency departments, floor units, intensive care units), health providers make a range of decisions from small to potentially life-altering. Those decisions, along with relevant clinical data and outcomes, could be harnessed to develop novel clinical ML tools.
For example, during the COVID-19 crisis, data collected from the emergency room, such as first labs, information on patients presenting symptoms (or lack thereof), triaging decisions, as well as rolling assessments of beds and staffing needs, can be essential in the process of developing smarter triaging models.