SUNRISE CRITICAL INFRASTRUCTURE SERIES

One of the driving concerns during any epidemic is the strain on the healthcare system. As we have seen with the COVID-19 pandemic, hospitals and Intensive care Units (ICUs) can quickly become overwhelmed by cases (1). This is the reason why many initiatives have been developed in different countries by different research teams with the aim of establishing tools and models to effectively predict the demand for resources in pandemics and public health emergencies environments. The purpose of this post is to showcase some examples of such initiatives.

One of them were developed by researcher from Mount Allison University, York University, and Health Canada. They developed two-module model to forecast the effects of relaxation of non-pharmaceutical intervention and vaccine against COVID-19 uptake on daily incidence, and the cascade effects on healthcare demand. They employed a mathematical modelling framework to project COVID-19 epidemic scenarios and quantify healthcare demand. This framework can be modified to consider other infectious diseases [1].

Researchers from Beijing Wuzi University created a demand forecasting method based on the number of current confirmed cases. The number of current confirmed cases were estimated using a bilateral long-short-term memory and genetic algorithm support vector regression (BILSTM-GASVR) combined prediction model. Then, based on the number of infected cases, ICU healthcare resources (healthcare workers, equipment and drugs) demand forecasting models were constructed. Data on the number of COVID-19-infected cases in Shanghai between January 20, 2020, and September 24, 2022, were used to perform a numerical example analysis, obtaining results that were closer to the real values [2].

In USA, Icahn School of Medicine at Mount Sinai researchers developed a machine learning-based risk prioritization tool that predicts ICU transfer within 24 h COVID-19 patients, seeking to facilitate efficient use of care providers’ efforts and help hospitals plan their flow of operations. Time series data, including vital signs, nursing assessments, laboratory data, and electrocardiograms, were used as input variables for training a random forest model. The results obtained showed that the model could be used as a screening tool to identify patients at risk of imminent ICU transfer within 24 hours. This tool could improve the management of hospital resources and patient-throughput planning [3].

During COVID-19 pandemics, the increase of patients had often led to shortages of ICU healthcare resources and a short-term mismatch of supply and demand. These issues have drastically impacted in anti-epidemic frontline healthcare workers and in the treatment outcomes of infected patients. The accurate forecasting of the demand, for hospitals in general and for ICU in particular, healthcare resources can facilitate the rational resource allocation of hospitals under changes in demand patterns, which is crucial for improving the provision of critical care and rescue efficiency [2].

Good results showed by these three initiatives demonstrate that deep learning and artificial intelligence could be very useful in improving the effectiveness of demand forecasting and decision making in the hospital environment, which has a direct impact on the treatment of patients and the efficient use of resources.

Bibliography: 

1. Betti MI, Abouleish AH, Spofford V, Peddigrew C, Diener A, Heffernan JM. COVID-19 Vaccination and Healthcare Demand. Bull Math Biol. 2023 Mar 17;85(5):32. doi: 10.1007/s11538-023-01130-x. PMID: 36930340; PMCID: PMC10021065.
2. Zhang W, Li X. A data-driven combined prediction method for the demand for intensive care unit healthcare resources in public health emergencies. BMC Health Serv Res. 2024 Apr 17;24(1):477. doi: 10.1186/s12913-024-10955-8. PMID: 38632553; PMCID: PMC11022462.
3. Cheng FY, Joshi H, Tandon P, Freeman R, Reich DL, Mazumdar M, Kohli-Seth R, Levin M, Timsina P, Kia A. Using Machine Learning to Predict ICU Transfer in Hospitalized COVID-19 Patients. J Clin Med. 2020 Jun 1;9(6):1668. doi: 10.3390/jcm9061668. PMID: 32492874; PMCID: PMC7356638.                                                                                                                                                                                                                                                                                                                  Written by: Isabel Garcia Merino and Carolina Gutierrez Montero, QUIRONSALUD (QS).