Developing a binary communication protocol between biological neural networks using virtual white matter.

Area of Science:

• Neuroscience
• Biocomputing
• Systems Biology

Background:

• Biological neural networks (BNNs) exhibit distributed computation via interconnected neurons.
• Current in vitro biocomputing often uses isolated cultures, limiting complex processing.
• Scalable biocomputing requires reliable communication between separate biological processing units.

Purpose of the Study:

• To expand the Virtual White Matter (VWM) platform for structured binary communication between distinct neural cultures.
• To enable real-time information exchange for distributed biocomputing architectures.

Main Methods:

• Utilized spatiotemporal electrical stimulation patterns to encode 3-bit data words.
• Employed machine learning for real-time decoding of evoked neural responses.
• Implemented parity-based error correction to improve data transmission fidelity.

Main Results:

• Successfully transmitted 3-bit data packets between separate neural cultures.
• Achieved individual bit decoding accuracies of 75-90% and aggregate word accuracy over 52%.
• Demonstrated enhanced transmission fidelity and reliable word-level communication using error correction.

Conclusions:

• Physically separated neural cultures can reliably exchange structured information via engineered communication frameworks.
• The VWM platform provides a foundation for scalable, distributed biocomputing.
• This work supports in vitro models for multi-network neural computation.