National and Kapodistrian University of Athens
Μαθηματική μοντελοποίηση της δυναμικότητας του ιού SARS-CoV-2 εντός και μεταξύ ξενιστών, μέσω υπολογιστικών προσεγγίσεων, γενετικών αλγορίθμων και εξελικτικών αλγορίθμων βελτιστοποίησης
Abstract
dc:descriptionThe COVID-19 pandemic significantly affected public health and social life in Greece, leading to successive waves of morbidity with distinct epidemiological characteristics depending on the prevailing SARS-CoV-2 variant. Within the framework of this dissertation, an extended compartmental model was developed to study the transmission dynamics of the Delta and Omicron SARS-CoV-2 variants in Greece. At the same time, the usefulness of viral load, as reflected in the cycle threshold (Ct) values of molecular diagnostic tests, was evaluated as a potential biomarker for predicting disease severity and monitoring the evolution of pandemic waves. The epidemiological model was developed to incorporate key transmission and clinical progression parameters, such as incubation and recovery times, outcome probabilities, as well as non-pharmaceutical intervention factors (mask use, social distancing, vaccination), in order to realistically depict the pandemic dynamics across its different phases. The model was parameterized twice using appropriately developed fitting and optimization algorithms during the 4th and 5th pandemic waves in Greece, which correspond to the periods of Delta and Omicron variant predominance, according to official data from the National Public Health Organization. Model fitting enabled the estimation and comparative evaluation of critical epidemiological parameters. Compared to the Delta variant period, the Omicron variant was associated with a 73% increase in the effective contact rate, in contrast to the proportion of symptomatic cases requiring hospitalization, which decreased by 42%. Additionally, the proportion of asymptomatic infections was estimated to have increased by 27%, with a fourfold increase in their detection frequency. Regarding non-pharmaceutical interventions, mask use was estimated to have increased by 41% during the Omicron variant period, while adherence to social distancing measures remained at similar levels between the two time periods. Within the framework of the conducted simulations, it was highlighted that more than 5,000 deaths could have been prevented if mask use and social distancing had been 20% more effective during the surge of the Delta and Omicron variants. Furthermore, the dynamic evolution of selected SARS-CoV-2 variants, both within and between hosts, was investigated using biostatistical and analytical methods, on a statistically significant sample collected at NIMTS during the first five pandemic waves in Greece. The sample includes variables recording basic demographic characteristics and cycle threshold (Ct) values from 1,041 COVID-19 positive patients, combined with critical epidemiological indicators such as the number of deaths, daily cases, the basic and effective reproduction number (R₀/Rt), the stringency index of the measures, and the positivity rate. The analysis of basic statistical measures (mean, median, skewness) of Ct values revealed significant changes reflecting both the course of the pandemic waves and the disease dynamics. During the Delta variant predominance period, statistically significantly lower mean and median Ct values were observed compared to the Alpha and Omicron variants, along with increased skewness in their distribution. During this period, despite the low positivity rate, increased hospitalizations and deaths from the disease were recorded, which may be related to the higher viral load. Moreover, during the escalation phases of the pandemic waves, statistically significantly lower Ct values were observed compared to recession periods, indicating the usefulness of viral load as a dynamic epidemiological indicator for the evolution of pandemic waves. Finally, the geometric mean of hospitalizations in regular wards and ICUs showed a negative correlation with Ct values throughout the pandemic waves, suggesting that higher viral load is associated with greater disease severity in hospitalized patients. The analysis of Ct values emerges as an accessible and useful tool for epidemiological surveillance, as it provides information on viral transmission dynamics both within and between hosts. The development and utilization of such biomarkers and epidemiological parameters, combined with the application of appropriate models and computational approaches, can substantially contribute to public health decision-making, strengthening resource management and the strategic planning of interventions in critical periods such as the COVID-19 pandemic.
Degree
thesis:*- Grantor dc:publisher
- National and Kapodistrian University of Athens
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Liossi, Sofia
- Λιόση, Σοφία
Subjects
dc:subject × 30- COVID-19
- SARS-CoV-2
- Κύκλοι κατωφλίου (Ct)
- Επιδημιολογικό μοντέλο
- Δυναμική μετάδοση
- Ιικό φορτίο
- Βελτιστοποίηση
- Προσαρμογή
- Γενετικοί αλγόριθμοι
- Cycle threshold (Ct)
- Epidemiological model
- Transmission dynamics
- Viral load
- Optimization
- Fitting
- Genetic algorithms
- Μαθηματικά
- Φυσικές Επιστήμες
- Μοντελοποίηση και Προσομοίωση
- Στατιστική και Πιθανότητες
- Άλλες Ιατρικές Επιστήμες
- Ιατρική και Επιστήμες Υγείας
- Ιατρική, άλλοι τομείς
- Mathematics
- Natural Sciences
- Modeling and Simulation
- Statistics and Probability
- Other Medical Sciences
- Medical and Health Sciences
- Medicine, miscellaneous
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/59767