A project in preparation
The project aims to develop mathematical digital twins for patients admitted to intensive care with sepsis or septic shock. The 48-month duration in the proposal is a planned duration; implementation dates have not yet been set.
Scientific objective
Starting at ICU admission, connect clinical observations to a patient-specific dynamical model. The research aims to reconstruct unobserved physiological states, estimate parameters and represent trajectories of deterioration, stabilisation or treatment response with quantified uncertainty.
Methods
Dynamical models, data assimilation, inverse problems and uncertainty quantification will be combined to work with heterogeneous, noisy and irregularly sampled data. The observation process itself will be studied, since measurement frequency may reflect the patient’s condition.
The proposed demonstrator aims to provide interpretable representations and traceable summaries for clinicians. This is a research project, rather than a clinically validated tool or an autonomous treatment-prescribing AI.
Coordination and proposed collaborations
The proposal is led by the University of Strasbourg, with Christophe Prud’homme and Julie Helms as scientific coordinators. It envisages collaboration between the University of Strasbourg and Strasbourg University Hospitals, bringing together applied mathematics, health-data science and intensive-care medicine.