Author ORCID Identifier
https://orcid.org/0000-0001-8779-0911
Biosketch
Saiprasad V. R is a postdoctoral researcher in nonlinear dynamics and epidemic modelling at the Southern University of Science and Technology (SUSTech), Shenzhen. He recently completed his Ph.D. in Physics at SASTRA Deemed to be University, India, where his thesis, “Bridging Scales in Epidemic Modeling: From Individuals to Networks and Ecosystems,” developed multiscale frameworks that link individual-level processes, adaptive contact networks, and eco-epidemiological interactions with data-driven forecasting.
His work combines analytical tools such as stability and bifurcation analysis, Lyapunov exponents, and probability density, based extreme-event diagnostics with numerical simulation and machine-learning methods, including reservoir computing and Echo State Networks for model-free outbreak prediction. These approaches have been applied to problems ranging from vaccination and multi-wave COVID-19 dynamics to rabbit–predator disease systems, adaptive human behaviour on contact networks, and early-stage monkeypox forecasting, leading to publications in EPJ Plus and Physical Review E and multiple manuscripts under review.
Beyond research, Saiprasad has served as a teaching assistant for undergraduate and postgraduate laboratories and programming courses, and has contributed as a project assistant on a DST-CRG-funded project on computing with nano-oscillators, gaining experience in mentoring, scientific communication, and lab management. At SUSTech, he aims to further advance rigorous, data-informed dynamical systems approaches for understanding contagion, complex networks, and other spreading processes, while actively engaging in interdisciplinary collaborations at the interface of physics, mathematics, and public health. His broader research vision is to develop robust multiscale modelling frameworks that can inform evidence-based intervention design and policy for emerging and re-emerging infectious diseases in a rapidly changing world.
Date of Award
17-8-2025
Document Type
Thesis
School
School of Electrical & Electroncis Engineering
Programme
Ph.D.-Doctoral of Philosophy
First Advisor
Dr.V.K.Chandrasekar
Keywords
Epidemic modeling, SEIRV model, Adaptive networks, COVID-19 modeling
Abstract
Epidemics have shaped the trajectory of human history, repeatedly exposing vulnerabilities in healthcare systems and influencing socio-economic dynamics on a global scale. The COVID-19 pandemic, in particular, underscored the pressing need for flexible and responsive modeling framework capable of adapting to dynamic and evolving epidemic scenarios. Motivated by these challenges, this thesis explores the development of innovative epidemic models that integrate classical mathematical modeling techniques, network-based frameworks, and modern data-driven approaches. Each chapter addresses a distinct dimension of epidemic spread, from vaccination dynamics and community coupling to ecological complexity and adaptive human behavior, culminating in a model-free paradigm suitable for early-stage outbreak scenarios.
The thesis begins by extending classical SEIR compartmental models to incorporate waning vaccine-induced immunity, providing a refined understanding of herd immunity thresholds. Through parameter estimation using COVID-19 data from India, the model illustrates how even extensive vaccination efforts may not prevent resurgence without booster strategies. The study quantitatively captures the influence of vaccination rate on long-term equilibrium states and provides critical thresholds for controlling disease spread. Building upon this foundation, the next focus lies in modeling multi-wave epidemics in spatially coupled communities. Here, a logistic influx term is added to the susceptible population to mimic the periodic emergence of new viral variants. Bifurcation analysis reveals how vaccination influences the system’s transition across periodic, endemic, and disease-free regimes. Extension to a two-patch system with dispersal shows rich dynamical features including birhythmicity and multistability, underscoring the complex role of mobility in epidemic control.
The third contribution investigates eco-epidemiological systems with seasonal forcing, inspired by Rabbit Hemorrhagic Disease. A predator-prey infection model is analyzed under varying seasonal amplitudes and frequencies, uncovering the emergence of extreme events (EEs)—rare, large-amplitude outbreaks. Chaos and periodicity are examined through Lyapunov exponents and Probability Density Functions (PDFs) of prey population maxima provide a more robust measure of extreme events than conventional threshold based approaches. The work emphasizes how seasonality shapes ecological and epidemiological transitions, providing insights into biodiversity conservation and disease management. Shifting to the network paradigm, the fourth chapter introduces a novel adaptive network model in which the contact structure co-evolves with epidemic prevalence. The reduction in an individual’s maximum contact capacity during an outbreak is modeled using a logistic decay function, reflecting adaptive behaviour in response to disease prevalence. Using CoMix survey data from countries like Belgium and the UK, the model is validated and analyzed using both effective-degree ODEs and Gillespie simulations. Results show that early, strong contact reduction can significantly suppress epidemic peaks, highlighting the necessity for adaptive, timely interventions.
Finally, the thesis proposes a model-free approach for early outbreak prediction, focusing on the monkeypox epidemic. An Echo State Network (ESN), a type of reservoir computing architecture, is trained on real-time data to forecast outbreak trends. While achieving high predictive accuracy, the approach is paired with powerlaw scaling analysis to enhance interpretability. This hybrid framework offers a balance between purely data driven forecasts and mechanistic understanding, which is especially beneficial during the early stages of an epidemic when key parameters are yet to be identified. Overall, this thesis presents a multi-scale, multidisciplinary modeling framework that addresses key limitations in classical epidemic models. By progressively incorporating vaccination effects, spatial coupling, ecological complexity, adaptive behavior, and machine learning, the work enhances our ability to understand, predict, and mitigate epidemic outbreaks under diverse real-world conditions. These insights hold substantial implications for public health planning and intervention design in a rapidly changing world.
Recommended Citation
VR, Saiprasasd Mr, "Bridging Scales in Epidemic Modeling From Individuals to Networks and Ecosystems" (2025). Theses and Dissertations. 160.
https://knowledgeconnect.sastra.edu/theses/160