End-to-end learning of safe stimulation parameters for cortical neuroprosthetic vision.

Journal: Journal of neural engineering
Published Date: Jul 22, 2025

Abstract

Direct electrical stimulation of the brain via cortical visual neuroprostheses is a promising approach to restore basic sight for the visually impaired by inducing a percept of localized light called 'phosphenes'. Apart from the challenge of condensing complex sensory information into meaningful stimulation patterns at low temporal and spatial resolution, providing safe stimulation levels to the brain is crucial.We propose an end-to-end framework to learn optimal stimulation parameters (amplitude, pulse width and frequency) within safe biological constraints. The learned stimulation parameters are passed to a biologically plausible phosphene simulator which takes into account the size, brightness, and temporal dynamics of perceived phosphenes.Our experiments on naturalistic navigation videos demonstrate that constraining stimulation parameters to safe levels not only maintains task performance in image reconstruction from phosphenes but consistently results in more meaningful phosphene vision, while providing insights into the optimal range of stimulation parameters.Our study presents a stimulus-generating encoder that learns stimulation parameters (1) satisfying safety constraints, and (2) maximizing the combined objective of image reconstruction and phosphene interpretability with a highly realistic phosphene simulator accounting for temporal dynamics of stimulation. End-to-end learning of stimulation parameters this way enables enforcement of critical biological safety constraints as well as technical limits of the hardware at hand.

Authors

  • Burcu Küçükoğlu
    Department of Machine Learning and Neural Computing, Radboud University Donders Institute for Brain Cognition and Behaviour, Thomas van Aquinostraat 4, Nijmegen, Gelderland, 6525 GD, Nijmegen, GE, 6525 GD, NETHERLANDS.
  • Bodo Rueckauer
    Department of Machine Learning and Neural Computing, Radboud University Donders Institute for Brain Cognition and Behaviour, Thomas van Aquinostraat 4, Nijmegen, GE, 6500 GL, NETHERLANDS.
  • Jaap de Ruyter van Steveninck
    Department of Machine Learning and Neural Computing, Radboud University Donders Institute for Brain Cognition and Behaviour, Thomas van Aquinostraat 4, Nijmegen, GE, 6525 GD, NETHERLANDS.
  • Maureen van der Grinten
    Netherlands Institute for Neuroscience, Meibergdreef 47, Amsterdam, 1105 BA, NETHERLANDS.
  • Yağmur Güçlütürk
    Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, The Netherlands.
  • Pieter Roelfsema
    Department of Vision & Cognition, Netherlands Institute for Neuroscience, Amsterdam, Netherlands.
  • Pieter R Roelfsema
    Department of Vision & Cognition, Netherlands Institute for Neuroscience, 1105BA Amsterdam, Noord Holland, The Netherlands; Department of Integrative Neurophysiology, Center for Neurogenomics and Cognitive Research, Vrije Universiteit Amsterdam, 1981 HV Amsterdam, Noord Holland, The Netherlands; and Department of Computer Science, Rijksuniversiteit Groningen, 9747 AG Groningen, The Netherlands p.roelfsema@nin.knaw.nl.
  • Umut Güçlü
    Radboud University, Donders Institute for Brain, Cognition and Behaviour, Nijmegen, The Netherlands. Electronic address: u.guclu@donders.ru.nl.
  • Marcel van Gerven
    Computational Cognitive Neuroscience Lab, Department of Artificial Intelligence, Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, Netherlands.
  • Marcel A J van Gerven
    Radboud University, Donders Institute for Brain, Cognition and Behaviour, Nijmegen, The Netherlands.

Keywords

No keywords available for this article.

External Resources

Popular Topics
Recent Journals