Advancing the neurocomputer

Nanomaterial-based microelectrode arrays for in vitro bidirectional brain-computer interfaces: a review

A bidirectional in vitro brain-computer interface (BCI) directly connects isolated brain cells with the surrounding environment, reads neural signals and inputs modulatory instructions. As a noninvasive BCI, it has clear advantages in understanding and exploiting advanced brain function due to the simplified structure and high controllability of ex vivo neural networks. However, the core of ex vivo BCIs, microelectrode arrays (MEAs), urgently need improvements in the strength of signal detection, precision of neural modulation and biocompatibility. Notably, nanomaterial-based MEAs cater to all the requirements by converging the multilevel neural signals and simultaneously applying stimuli at an excellent spatiotemporal resolution, as well as supporting long-term cultivation of neurons. This is enabled by the advantageous electrochemical characteristics of nanomaterials, such as their active atomic reactivity and outstanding charge conduction efficiency, improving the performance of MEAs. Here, we review the fabrication of nanomaterial-based MEAs applied to bidirectional in vitro BCIs from an interdisciplinary perspective. We also consider the decoding and coding of neural activity through the interface and highlight the various usages of MEAs coupled with the dissociated neural cultures to benefit future developments of BCIs.

Understanding the Impact of Neural Variations and Random Connections on Inference

Recent research suggests that in vitro neural networks created from dissociated neurons may be used for computing and performing machine learning tasks. To develop a better artificial intelligent system, a hybrid bio-silicon computer is worth exploring, but its performance is still inferior to that of a silicon-based computer. One reason may be that a living neural network has many intrinsic properties, such as random network connectivity, high network sparsity, and large neural and synaptic variability. These properties may lead to new design considerations, and existing algorithms need to be adjusted for living neural network implementation. This work investigates the impact of neural variations and random connections on inference with learning algorithms. A two-layer hybrid bio-silicon platform is constructed and a five-step design method is proposed for the fast development of living neural network algorithms. Neural variations and dynamics are verified by fitting model parameters with biological experimental results. Random connections are generated under different connection probabilities to vary network sparsity. A multi-layer perceptron algorithm is tested with biological constraints on the MNIST dataset. The results show that a reasonable inference accuracy can be achieved despite the presence of neural variations and random network connections. A new adaptive pre-processing technique is proposed to ensure good learning accuracy with different living neural network sparsity.