Real-Time Control of a Focus Tunable Lens for Presbyopia Correction Using Ciliary Muscle Biopotentials and Artificial...

Area of Science:

• Ophthalmology
• Biomedical Engineering
• Artificial Intelligence

Background:

• Ageing causes presbyopia, a progressive loss of near vision, significantly impacting quality of life.
• Current corrective methods for presbyopia provide limited compensation and fail to restore dynamic visual focusing.
• Restoring dynamic accommodation is crucial for addressing the functional limitations of presbyopia.

Purpose of the Study:

• To develop and evaluate a closed-loop correction system for presbyopia.
• To utilize biopotential signals from the ciliary muscle and artificial neural networks for real-time accommodative state prediction.
• To enable dynamic focusing capabilities in presbyopic eyes.

Main Methods:

• Non-invasive contact lens electrodes were used to collect ciliary muscle biopotential data.
• A multi-layer perceptron artificial neural network classified the preprocessed biopotential signals.
• A control system adjusted an external focus-tunable lens based on the classifier's output.

Main Results:

• The system achieved a prediction accuracy of 0.79 for the eye's accommodative state.
• F1-scores of 0.78 for accommodation and 0.77 for disaccommodation prediction were recorded.
• Near visual acuity in presbyopic subjects significantly improved, while distance acuity remained stable.

Conclusions:

• The study demonstrates the feasibility of restoring near-natural accommodation in presbyopia using neuromuscular signals and adaptive lens control.
• The developed system shows promise for improving visual function in individuals with age-related vision loss.
• Future work will focus on system validation, dataset expansion, and pre-clinical testing for implantable devices.