Engineering a Biological Neural Network for Neuromorphic Computing

We introduce a Bio-adaptive Processing Unit (BPU) based on a two-reservoir microtunnel Brain-on-Chip with electrophysiological readout. We fabricated and validated three independent devices, demonstrating that human stem cell-derived ( Ngn2+ hiPSCs) cortical neurons extend axons over 1200 µ m, forming robust, long-range connections. Spikes detected within the tunnels exhibited 8-fold higher firing rates and spike amplitudes compared to reservoir areas. Notably, deferred seeding biased 70% of propagating spikes, enabling directed axonal conduction within the living substrate, with spikes reaching velocities of ∼ 0.814 ± 0.8m/s from the older to the younger network by day 28. These results confirm reproducible neuron-electrode interfacing and reliable axonal routing, establishing the BPU’s viability for engineering neural networks in neurobiologically-based neuromorphic platforms. These results offer a promising alternative to silicon-based neuromorphic computing architectures by directly leveraging energy-efficient biological neurons.