Epidemic spreading dynamics in large-scale brain modeling: clinical applications

Epidemic spreading dynamics in large-scale brain modeling: clinical applications

Fecha: 12 de marzo de 2025

Hora: 12:00h

Lugar: Seminario del Centro de Instrumentación Científica de la UGR

Ponente: Ana Paula Millán, investigadora «Ramón y Cajal» en el Instituto Carlos I.

Contacto:

Parte del ciclo de charlas sobre neurociencia SPIN

Computational models of brain dynamics provide new insights into the prognosis of neurological disorders such as epilepsy or Alzheimer's Disease (AD). The use of abstract models such as epidemic spreading allows for simplified pictures that, by focusing on the behavior of interest, enable interpretability. At the same time, these modelling schemes have often lower data requirements, more in line with the reality of clinical data. I will present recent results regarding the clinical applicability of epidemic spreading models to epilepsy surgery and the propagation of pathological proteins in AD.
In the case of epilepsy, for patients with drug-resistant epilepsy the treatment of choice is epilepsy surgery. However, up to 50% of the patients continue to have seizures one year after the resection. In order to aid presurgical planning and predict postsurgical outcome in a patient-by-patient basis, we developed the ESSES (Epidemic Spreading Seizure and Epilepsy Surgery framework ) framework of individualized computational models. ESSES combines a simple epidemic spreading dynamic -namely the Susceptible-Infected-Recover (SIR) model- to describe seizure propagation, with patient-specific connectivity derived from magnetoencephalography (MEG) and epileptogeneity maps integrating all presurgical information available. We showed in a retrospective study (N = 15) that this simple dynamic is enough to reproduce the main aspects of seizure propagation as recorded via invasive electroencephalography (iEEG) [2,3]. Remarkably , the SIR model parameters that best describe the iEEG seizure patterns correspond to the critical transition between the percolating and absorbing phases [2,3,4], and the similarity between the iEEG and ESSES-modelled seizures predicted surgical outcome (area under the curve AUC = 0.73). We validated the clinical use of ESSES with a blind, independent pseudo-prospective study (N = 34), for which iEEG data (highly invasive and not always part of the presurgical evaluation) was not required. Using ESSES to find optimal resection strategies [1,2], we found that their size was smaller (AUC = 0.65) for patients with good outcome, indicating intrinsic differences in the presurgical data of patients with good and bad outcome [4]. The actual resection also had a larger overlap with the optimal one (AUC = 0.64), and a larger effect decreasing modelled seizure propagation (AUC = 0.78) for patients with good outcome [4]. Overall, ESSES could correctly predict 75% of NSF and 80.8% of SF cases pseudo-prospectively.

Spreading dynamics are also found in the case of AD: the pathological protein τ propagates over the brain associated with the progression of the disease. The mechanisms driving this propagation, and the paths it follows, remain unclear. In a recent cross-sectional study, we made use of epidemic spreading dynamics to investigate which spreading pathways -that is, whether either functional or structural connections, or simply spatial closeness- drive τ propagation [5]. We simulated τ propagation over each type of brain connectivity and compared the simulated τ depositions with 18F-flortaucipir PET binding potentials at several stages of the AD continuum. We found that functional connectivity yield the best predictions of actual τ propagation along the AD continuum. These results suggest that, in addition to

structural connectivity, functional connectivity plays an important role in τ spreading, and highlight that neuronal dynamics play a key role in promoting this pathological process.
These studies showcase the potential of epidemic spreading models in large-scale brain modeling, opening new roads for clinical applications.

*References: *[1] I.A. Nissen et al. Scient. Rep.. 11.1 (2021). [2] A.P. Millán et al. Scient. Rep. 12.1 (2022). [3] A.P. Millán et al. Netw. Neurosci. 1.55 (2022). [4] A.P. Millán et al. Network Neuroscience 8.2 (2024). [5] D. Schoonhoven et al. Brain awad189 (2023).