Retinal Image Blur Analysis for Accessible Eye Refraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining eye refraction require complex and expensive devices, limiting access to a wide range of the population, and necessitate qualified personnel.
Innovation Solution
A method using a smartphone or tablet to determine eye refraction by acquiring retina images, analyzing blur levels with a convolutional neural network, and applying blind deconvolution to reduce blur, without additional optical components or augmented reality displays.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional autorefraction methods are used, then measurement precision is improved, but device complexity and cost increase significantly
Solution Approach 1:
The patent uses a standard camera to capture an optical copy (image) of the retina instead of requiring complex specialized equipment. The retinal image serves as a proxy for direct refraction measurement, allowing refraction determination through image analysis rather than complex optical instrumentation
Solution Approach 2:
The patent replaces complex mechanical/optical autorefraction systems with a digital image processing system. Instead of using specialized optical instruments with multiple light sources and cameras, the invention uses a standard camera capturing retinal images that are then processed algorithmically to determine refraction features
2Measurement precision
If traditional autorefraction methods are used, then measurement precision is improved, but ease of operation deteriorates due to requiring qualified personnel
Solution Approach 1:
The system performs self-service by automatically capturing retinal images and processing them through algorithms to determine refraction features without requiring trained operators. The device handles the entire measurement and analysis process autonomously, making it operable by anyone regardless of medical training
Solution Approach 2:
The patent replaces the need for trained personnel with an automated digital processing system. Instead of requiring qualified individuals to interpret complex optical measurements, the system uses image processing algorithms to automatically extract refraction information from retinal images
3Measurement precision
If traditional autorefraction methods are used, then measurement precision is improved, but accessibility deteriorates due to limited availability
Solution Approach 1:
The patent uses a universal camera device that can perform multiple functions including retinal imaging and refraction measurement. This multi-functional approach allows a single device to serve both as a standard camera and as a medical diagnostic tool, greatly expanding accessibility compared to specialized equipment available only in clinics
Solution Approach 2:
By using retinal images as a copy or proxy for direct refraction measurement, the system enables refraction assessment through a medium (standard camera images) that is universally accessible, thereby expanding adaptability and accessibility to diverse settings and populations
4Measurement precision
If traditional autorefraction methods are used, then measurement precision is improved, but loss of time increases due to complex procedures
Solution Approach 1:
The patent captures refraction information indirectly through retinal images, which can be obtained rapidly with a standard camera. This copying approach eliminates time-consuming procedures associated with traditional autorefraction, allowing quick acquisition of measurement data while maintaining precision through subsequent image analysis
Data Source
AI summary
A system and method for determining a refraction feature of an eye of a subject using an image-capture device, the method including acquiring at least one picture of the retina of the eye of the subject, and determining the refraction feature based on a blur level of the acquired picture of the retina.