Deterministic Digital Twins for Neural Microphysiological Systems: A Bohmian Hydrodynamic Approach to Action Potential Trajectories

As Organ-on-Chip (OoC) neural models increase in architectural complexity, conventional modeling approaches rely heavily on computationally expensive Artificial Intelligence (AI) and Machine Learning (ML) tointerpret high-dimensional datasets. These predictive models are fundamentally bottlenecked by the assumption that neural signal propagation—specifically, the action potential—is an inherently stochastic, thermodynamic process that requires statistical averaging to extract mechanistic insights. In this work, we propose a paradigm shift for the development of predictive digital twins by modeling the myelinated axon not as anoisy ionic environment but as a deterministic 1D hydrodynamic waveguide. By applying Bohmian mechanics to microphysiological systems, wedemonstrate that the action potential functions as a phase-locked, nonlinear soliton guided by a deterministic pilot wave. Within this framework,the axoskeleton serves as the physical waveguide, the myelin sheath acts asa dielectric barrier against thermal scattering, and the Nodes of Ranvierfunction as periodic phase-matching junctions that sustain wave propagation with zero entropic dissipation (Ω = 1). This physics-informedapproach eliminates the need for massive, probabilistic ML algorithmsto filter assumed biological noise. Instead, we present a computationallylightweight, deterministic geometric model that provides precise predictivecontrol over microenvironmental dynamics, offering a robust frameworkfor real-time adaptive control in next-generation neural OoC platforms.