Assessing Pattern Recognition Performance of Neuronal Cultures through Accurate Simulation
This work provides a rigorous assessment framework for the pattern recognition capabilities of neuronal cultures, which is important for researchers working on brain-computer interfaces and biological computing.
This paper develops a methodology to assess the pattern recognition performance of neuronal cultures by creating a digital model that accurately reproduces the behavior of real cultures. Using this simulated culture, they performed handwritten digit recognition and rigorously evaluated its performance, also identifying improved simulation parameters for the task.
Previous work has shown that it is possible to train neuronal cultures on Multi-Electrode Arrays (MEAs), to recognize very simple patterns. However, this work was mainly focused to demonstrate that it is possible to induce plasticity in cultures, rather than performing a rigorous assessment of their pattern recognition performance. In this paper, we address this gap by developing a methodology that allows us to assess the performance of neuronal cultures on a learning task. Specifically, we propose a digital model of the real cultured neuronal networks; we identify biologically plausible simulation parameters that allow us to reliably reproduce the behavior of real cultures; we use the simulated culture to perform handwritten digit recognition and rigorously evaluate its performance; we also show that it is possible to find improved simulation parameters for the specific task, which can guide the creation of real cultures.