NEURON simulation in Python using real time stimulus

[ikasid15]

Hi everyone,

I am working on a group project at Imperial College London which involves real-time control of a biomimetic robot.

This project is divided in 2 parts. The first part involves taking visual input from an animation which is translated to values from 0 to 1 per pixel of input.
The second part is a multi-compartmental neuron model created in NEURON simulation environment in Python.

I need to connect these 2 parts by feeding real-time data as stimulus to the neuron model and thus I was wondering whether you would be able to indicate what would be the best mechanism in NEURON to do that in order the model to be as efficient and flexible as possible.

Currently, I am running the NEURON simulation as many times as the frames of input I am receiving.
I am using the Intfire2() point process for each pixel of input, I am connecting these point neurons to the multi-compartmental neuron models and I am updating the ib value of the Intfire2() point processes depending on the pixel value. Do you think that this is the best approach?

Thank you in advance.
Yannis