Portable device for presbyopia correction with optoelectronic lenses driven by pupil response

Presbyopia, the age-related decline in the eye's ability to accommodate, affects millions worldwide. This reduction in accommodation range is primarily due to a loss of elasticity in the eye lens and increased stiffness in the lens nucleus, as evidenced by research from Fisher (1971), Glasser and Campbell (1998), and Heys et al. (2004). While children can accommodate over 10 diopters (D), this ability diminishes throughout life, becoming noticeable in the 40s and significantly impacting near vision by age 50. Current presbyopia correction methods, such as progressive spectacles, multifocal contact lenses, and intraocular lenses, have limitations. These solutions can be invasive, lack comfort or flexibility, and often compromise either distance or near vision, reducing contrast and offering only a limited range of defocus positions. This paper introduces a non-invasive, dynamic, continuous focus range solution that mimics the natural lens's accommodation capabilities. Using tunable opto-electronic lenses controlled by a real-time pupil tracking system integrated into a smartphone, this Dynamic Auto-Accommodation Glasses system aims to provide a smooth, comfortable, and adaptive visual experience across a wide range of viewing distances.

Existing opto-electronic devices for presbyopia correction have been explored. Some utilize tunable lenses with laser-based distance sensors, while others incorporate commercial pupil tracking components. However, these systems often lack portability or are not truly mobile solutions, posing challenges in real-world applications. Previous research by Mompeán et al. (2016) explored presbyopia correction using optoelectronic lenses driven by pupil size, while other studies like Jarosz et al. (2018, 2019) and Padmanaban et al. (2018, 2019) investigated different approaches, highlighting the ongoing quest for a comfortable and effective solution. The proposed Dynamic Auto-Accommodation Glasses aim to address the limitations of existing systems by offering a fully integrated and portable solution.

The Dynamic Auto-Accommodation Glasses system comprises a pair of opto-electronic tunable lenses (Optotune, model EL-16-40-TC-VIS-20D), two USB infrared cameras (Misumi), an infrared LED array, a micro-USB hub, a battery (300mAh, 3.7V), and a Samsung Galaxy S7 smartphone serving as the processing unit and controller. The smartphone's Samsung Exynos 8890 processor processes 320x240 pixel images at approximately 24 frames per second, enabling real-time pupil tracking and lens control. The system uses a custom 3D-printed frame to house the components for comfortable wear. Image processing and pupil tracking utilize a Starburst algorithm, implemented in OpenCL for optimal performance on the smartphone's GPU. The algorithm's stages involve corneal reflection identification, Top-hat transformation, thresholding, dilation, pupil border point search, and RANSAC ellipse fitting. The system's OpenCL implementation incorporates several optimizations including vectorized memory accesses, local memory usage, native math operations, kernel fusion, and reduced-precision floating-point data types to enhance speed and efficiency. Gaze distance is calculated using binocular pupil tracking and trigonometry (formulas 1-5). A one-time calibration procedure establishes the initial pupil positions, and then the system dynamically adjusts the lens power based on continuous pupil position updates. To ensure a smooth and comfortable experience, the system incorporates a smooth accommodation change with a speed of 50 D/s, mimicking the natural accommodation response of the human eye (Figure 4). Visual acuity tests were conducted on eight presbyopic subjects (average age 62.5, range 52-78) at various distances (5m, 1m, 0.5m, 0.3m, 0.25m, 0.2m) comparing visual acuity under three conditions: far correction with trial lenses, near correction with trial lenses, and using the Dynamic Auto-Accommodation Glasses. A paired t-test was used to analyze the results comparing visual acuity across the different distances and conditions. The study was conducted with informed consent obtained from all participants following the Declaration of Helsinki, with ethical approval from the University of Murcia Research Ethics Committee.

The Dynamic Auto-Accommodation Glasses demonstrated significant improvements in visual acuity across a wide range of distances (5m to 20cm) in presbyopic subjects. The smartphone-based processing achieved near real-time performance at approximately 24 frames per second. Visual acuity remained consistently high across all distances tested while using the glasses, unlike the far correction (steadily decreased visual acuity with decreased object distance) and near correction (best acuity only in the 2-3.3 D range) conditions. Paired t-tests showed statistically significant differences in visual acuity at 0.5m (2D) between far correction (M = 0.59, SD = 0.06) and glasses use (M = 0.78, SD = 0.14), p = 0.003291; at 5m (0.2D) between no glasses (M = 0.27, SD = 0.04) and glasses use (M = 0.80, SD = 0.13), p = 0.0000314; and at 0.2m (5D) between no glasses (M = 0.37, SD = 0.09) and glasses use (M = 0.83, SD = 0.15), p = 0.00004284. The system's smooth accommodation change (50 D/s) contributed to a comfortable user experience. The response time of the system is approximately 200ms, enabling a seamless transition between different focusing distances (Figure 5).

The study's findings demonstrate the effectiveness of the developed Dynamic Auto-Accommodation Glasses in providing dynamic and automatic presbyopia correction across a range of distances. The system's high performance, enabled by the OpenCL implementation on the smartphone's GPU, allows for comfortable real-time adaptation. This approach addresses limitations of existing presbyopia correction methods by providing a non-invasive, wearable device that offers continuous focus adjustment and superior visual acuity across various distances. This technology has significant implications for improving the quality of life for individuals with presbyopia. The significant improvement in visual acuity observed across different distances suggests the potential for replacing the need for multiple pairs of glasses, simplifying daily activities for presbyopic individuals. The success in achieving such performance on a smartphone platform underscores the potential for widespread accessibility.

This study successfully demonstrated a portable, smartphone-powered system for dynamic presbyopia correction. The Dynamic Auto-Accommodation Glasses provided significant improvements in visual acuity compared to conventional corrective methods, across a wide range of distances. The optimized OpenCL implementation and smooth accommodation response contributed to a comfortable user experience. Future work could focus on improving the device's size, battery life, field of view, and addressing any remaining issues with large power adjustments. Further research could also explore personalized calibration procedures to optimize performance for individual users.

Current limitations include a relatively small field of view due to the size of the opto-electronic lenses, a bulky design that may not be comfortable for prolonged use, and a limited battery life of less than 8 hours. While these limitations are largely addressable through design improvements, some challenges like enlarging the lens aperture require further technological advancements. Additionally, there is a potential for oscillations when the control system fails, although proper control procedures minimized this issue. Finally, users may experience some magnification changes during large optical power adjustments.