EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

jaxon: a differentiable, GPU-native simulator for peripheral-nerve fiber models

David Lung, Max Haberbusch

SoftwareX · 2026-12 · 36 · 103196

DOI: 10.1016/j.softx.2026.103196 ↗

Kurz gefasst

Fachartikel in der Zeitschrift SoftwareX, 2026-12. Verfasst von David Lung und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 15 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

jaxon is an open-source, fully differentiable and GPU-native reimplementation of the canonical peripheral-nerve fiber models in JAX/Jaxley: the myelinated McIntyre–Richardson–Grill (MRG) and Sweeney axons and the unmyelinated Sundt and Rattay C-fibers. It reproduces NEURON’s extracellular mechanism through a custom backward-Euler coupled intracellular/periaxonal double-cable solver, agreeing with PyFibers-wrapped NEURON on 9 9 . 6 % of 9 4 3 activation-threshold configurations within 1 % and matching conduction velocity to machine precision in the myelinated models. Because the entire forward model is expressed in JAX, it is both vectorized—simulating whole fiber populations in parallel and reaching a geometric-mean ∼ 8 2 0 × speedup at 𝑁 = 1 0 0 , 0 0 0 fibers on a single GPU—and differentiable, so extracellular-stimulation parameters (per-contact amplitudes, waveform shape, and electrode position) can be optimized directly through the cable equation rather than grid-searched. jaxon slots into existing peripheral-nerve modeling pipelines as a gradient-enabled, population-scale replacement for the NEURON forward solver.

Zitieren

APA

Lung, D., & Haberbusch, M. (2026). jaxon: a differentiable, GPU-native simulator for peripheral-nerve fiber models. SoftwareX, 36, 103196. https://doi.org/10.1016/j.softx.2026.103196

Vancouver

Lung D, Haberbusch M. jaxon: a differentiable, GPU-native simulator for peripheral-nerve fiber models. SoftwareX. 2026;36:103196. doi:10.1016/j.softx.2026.103196

BibTeX
@article{Lung2026,
  author = {Lung, David and Haberbusch, Max},
  title = {jaxon: a differentiable, GPU-native simulator for peripheral-nerve fiber models},
  journal = {SoftwareX},
  year = {2026},
  volume = {36},
  pages = {103196},
  doi = {10.1016/j.softx.2026.103196}
}
Kostenlos starten →