Prediction of Critical Medical Resources for Combatting Future Pandemics
MetadataShow full item record
AbstractThe COVID-19 pandemic wrecked an avalanche of resource management disaster on many countries. Preventable deaths occurred due to lack of resources, especially ventilators. One major unforgettable lesson we can learn from the COVID-19 disaster is that proactive planning of ventilators can save a huge number of lives globally in future pandemics. In this study, we aim to address this need by developing a predictive model for ventilators. Using open-source data from ‘Our World in Data’, we employ an ensemble of existing time series analysis techniques and missing data handling strategies to predict ventilators at a population level. A full-scale application of the proposed modelling framework was demonstrated for India, Nigeria, Uruguay and Poland as representative cases of different scenarios. Furthermore, as part of the robustness checks, we test the model’s performance for periods of increased severity (e.g., increased death rate) and reproduction rate during a pandemic with USA, UK, Germany and France as sample cases. We consider the population-based model and implications of the prediction results for a possible extension to ventilator associated other critical medical resources in an ICU unit. This thesis contributes to the existing body of knowledge and methods for predicting ventilators and other critical medical resources that are mostly addressed at local settings. More importantly, the proposed framework can be used to predict resources for COVID-19 like pandemics for any global population level where ICU patients data is scant. In addition to the methodological contribution, this thesis demonstrates the role of evidence-based decision-making in healthcare disaster preparedness and recovery plan.
The following license files are associated with this item:
- Creative Commons