talks

Parameter Inference in Immune Digital Twins: A Stochastic Adaptive Filtering Approach

June 16, 2026 · Invited talk · Washington State University, Pullman, WA, USA · slides

The Digital Twin paradigm relies on three fundamentals: a mechanistic software replica, longitudinal object-specific data, and a dynamic inference framework to iteratively update the model. While the potential for personalized medicine is substantial, the inherent stochasticity of the immune response makes iterative parameter inference a sweeping task. Here we present a data-driven framework for the iterative updating of immune digital twins using stochastic adaptive filters.

Adaptive Observers for Medical Digital Twins: A Framework for Personalized Medicine

February 19, 2026 · Departmental colloquium · University of Idaho, Department of Mathematics and Statistical Science, Moscow, ID, USA · slides

In this colloquium I presented preliminary results of my current work on adaptive observers applied to medical digital twins, enabling virtual replicas of patients to continuously synchronize with real-time physiological data. Unlike traditional batch-based identification algorithms, this approach uses adaptive observers to jointly estimate unmeasurable states and monitor critical biological markers. Applications include predicting HIV trajectories and tracking antimicrobial resistance in personalized medicine.

Understanding SIR-type epidemic behaviors through set-based dynamic analysis

October 10, 2024 · Departmental colloquium · University of Idaho, Department of Mathematics and Statistical Science, Moscow, ID, USA · slides

In this talk I presented my PhD results at the University of Idaho: a set-based dynamic analysis of SIR-type systems, used to discuss realistic implementations of finite-time interval social distancing measures. The work focuses on minimizing the Epidemic Final Size while keeping the Infected Peak Prevalence controlled at all times, and minimizing the side effects of the intervention.