An introduction to genetic algorithms
An introduction to genetic algorithms
Optimization via Intermittency with a Self-Organizing Neural Network
Neural Computation
Spike-timing-dependent plasticity in small-world networks
Neurocomputing
Journal of Computational and Applied Mathematics
Simple model of spiking neurons
IEEE Transactions on Neural Networks
| Hi-index | 0.01 |
A simulation study has been done to address how the connectivity structure of the brain complex network relates to its dynamics and its behavioral properties. In this study, we intend to find plausible candidates of structural organization for an epileptic network capable of exhibiting its characteristic dynamics/behavior. For this purpose, a network of hippocampus CA3 neuronal population was implemented and evolved structurally with a constrained genetic algorithm (CGA). This optimization technique was favored by maximizing synchronization among excitable neurons, simulating seizure focus. On the other hand it was constrained to conserve some prescribed characteristics by penalizing infeasible structures in search space. Meanwhile some structural and dynamical aspects of the network were investigated to characterize the evolution path as well as the structure/dynamic to which the network ultimately reaches. The results of our simulations show the following: (1) a network with this evolutionary rewiring process is convergent to a synchronized state where seizures can arise. (2) Numerical analysis of the dynamical origin of seizures, calculated in each generation of CGA, indicates a critical point where global synchrony among multiple neuronal clusters may occur. (3) Different graph topologies more or less have similar endings. Indeed, the search for optimal patterns of connectivity in physiological or pathological models of brain activity may shed light on some of issues that must be dealt with in real neuronal systems; e.g. the control of seizures in epileptogenesis. The present paper tries to step forward this aim.