Digital FPGA implementation of spontaneous astrocyte signalling

摘要

Astrocytes are the most abundant type of glial cells in the central nervous system (CNS). These non-neuronal cells are able to regulate the neurons activity in the different parts of brain tissue by calcium waves generation in its internal space. Moreover, astrocytes interact with neurons and modulate the spiking activity of them. In this paper, a set of piecewise linear estimations of a three-dimensional spontaneous astrocyte model are presented for digital FPGA realization. This leads to achieve a high-speed and low-cost system in large-scale implementation. In this approach, the three-dimensional original model is converted to a two-dimensional one and the hardware overhead have been reduced, significantly due to eliminating the large number of multiplications in the original astrocyte model. Simulation results in MATLAB demonstrate that our method can mimic the original calcium waves in high degree of similarity. To validate our method in case of hardware, the proposed model has been tested and simulated in Modelsim software and also implemented on Spartan3 XC3S50 (TQ144) FPGA board. Hardware realization results show that the proposed model has high similarity by the simulation outputs. Consequently, this reduced-model of astrocyte can be used in large-scale networks because of its low-cost hardware and high-speed system.