Biomimetic computational modelling of retinal ganglion cell responses to contact lens designs with refractive and opaque non-refractive features
- 1nthalmic Pty Ltd, Suite L2, Level 3, 2A Lord Street, Botany, Sydney, NSW 2019 Australia
- 2School of Optometry and Vision Science, UNSW, Sydney, Australia
- 3School of Clinical Medicine, UNSW, Sydney, Australia
Abstract
Purpose: To computationally evaluate how single vision (SV), concentric bifocal (BF), and a contact lens with non-refractive opaque elements (AR) lens designs modulate retinal ganglion cell (RGC) activity.
Methods: Wide-field retinal simulations were generated using a schematic myopic eye corrected with SV, BF, and AR designs across three stimuli (Lena image, logMAR chart, dynamic gratings), three pupil diameters, and three residual accommodative error configurations. For each configuration, retinal image montages emulating blink-induced on-eye lens movement over one second were processed through a multilayer biomimetic virtual retina model to simulate RGC spike trains, quantified using median absolute deviation (MAD) and up-crossing rate (UCR).
Results: Across stimuli, pupils and residual‑error configurations, SV lens design exhibited an initial transient peak in RGC firing followed by rapid decay to a steady, low-variability tonic state (MAD range: static 29.68 to 49.21, dynamic 402.74 to 409.83 kHz). The BF lens design produced increased temporal variability and burst-like activity relative to SV, that depended on pupil and residual error (MAD range: static 36.05 to 72.14, dynamic 384.37 to 413.26 kHz; UCR up to static 11.25%, dynamic 16.25%). AR lens design yielded strong and sustained phasic, burst-like activity (MAD range: static 68.64 to 129.01, dynamic 418.25 to 454 kHz; UCR up to static 20%, dynamic 18.75%), reflected in elevated MAD and UCR values.
Conclusions: In this computational model, both BF and AR lens designs produced measurable changes in simulated RGC temporal activity relative to SV, with AR inducing greater and more configuration-independent modulation. These findings must be considered hypothesis-generating only; they are consistent with the proposition that temporal encoding of RGC response may contribute to myopia control, however, they do not establish causation. Empirical validation in prospective clinical studies is warranted before any clinical inference can be drawn.
Supplementary Information: The online version contains supplementary material available at https://doi.org/10.1186/s12886-026-05150-8.
Related
- Computational modelling of decision-making in suicide attempters
- [Re] Computational Modelling of Glucose Uptake by SGLT1 and Apical GLUT2 in the Enterocyte
- Computational modelling of acetylcholine deficiency-mediated dysfunction in the basal ganglia-thalamocortical network related to Parkinson’s disease dementia
- Human lung organoid modelling of tissue-resident antiviral T cell responses
- An anatomy-guided whole-heart framework for computational modelling of the complete human cardiac conduction system
BibTeX
@article{bakaraju2026biomimetic,
title = {{Biomimetic computational modelling of retinal ganglion cell responses to contact lens designs with refractive and opaque non-refractive features}},
author = {Ravi C. Bakaraju and Praveen K. Bandela},
journal = {BMC Ophthalmology},
year = {2026},
month = jul,
doi = {10.1186/s12886-026-05150-8},
url = {https://doi.org/10.1186/s12886-026-05150-8},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Biomimetic computational modelling of retinal ganglion cell responses to contact lens designs with refractive and opaque non-refractive features
AU - Ravi C. Bakaraju
AU - Praveen K. Bandela
PY - 2026
DA - 2026/07/30
JO - BMC Ophthalmology
DO - 10.1186/s12886-026-05150-8
UR - https://doi.org/10.1186/s12886-026-05150-8
AB - Purpose: To computationally evaluate how single vision (SV), concentric bifocal (BF), and a contact lens with non-refractive opaque elements (AR) lens designs modulate retinal ganglion cell (RGC) activity. Methods: Wide-field retinal simulations were generated using a schematic myopic eye corrected with SV, BF, and AR designs across three stimuli (Lena image, logMAR chart, dynamic gratings), three pupil diameters, and three residual accommodative error configurations. For each configuration, retinal image montages emulating blink-induced on-eye lens movement over one second were processed through a multilayer biomimetic virtual retina model to simulate RGC spike trains, quantified using median absolute deviation (MAD) and up-crossing rate (UCR). Results: Across stimuli, pupils and residual‑error configurations, SV lens design exhibited an initial transient peak in RGC firing followed by rapid decay to a steady, low-variability tonic state (MAD range: static 29.68 to 49.21, dynamic 402.74 to 409.83 kHz). The BF lens design produced increased temporal variability and burst-like activity relative to SV, that depended on pupil and residual error (MAD range: static 36.05 to 72.14, dynamic 384.37 to 413.26 kHz; UCR up to static 11.25%, dynamic 16.25%). AR lens design yielded strong and sustained phasic, burst-like activity (MAD range: static 68.64 to 129.01, dynamic 418.25 to 454 kHz; UCR up to static 20%, dynamic 18.75%), reflected in elevated MAD and UCR values. Conclusions: In this computational model, both BF and AR lens designs produced measurable changes in simulated RGC temporal activity relative to SV, with AR inducing greater and more configuration-independent modulation. These findings must be considered hypothesis-generating only; they are consistent with the proposition that temporal encoding of RGC response may contribute to myopia control, however, they do not establish causation. Empirical validation in prospective clinical studies is warranted before any clinical inference can be drawn. Supplementary Information: The online version contains supplementary material available at https://doi.org/10.1186/s12886-026-05150-8.
SN - 1471-2415
ER -