AI Eye Image Analysis for Non-Invasive Sarcopenia Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting sarcopenia rely on invasive blood tests and manual analysis of fundus images, which are costly and difficult to perform accurately.
Innovation Solution
An electronic device using artificial intelligence machine learning models to analyze fundus, eye, and slit lamp examination images to predict sarcopenia by extracting specific result values and determining sarcopenia occurrence based on these values, with adjustable weights for individual subjects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If invasive blood tests are used to detect sarcopenia biomarkers, then measurement precision is improved, but ease of operation deteriorates and loss of time increases
Solution Approach 1:
The patent replaces the mechanical/invasive blood test system with an optical imaging system. Fundus images are captured non-invasively and processed through machine learning models to detect sarcopenia, eliminating the need for invasive procedures while maintaining diagnostic capability
Solution Approach 2:
The patent introduces fundus images as an intermediary medium to detect sarcopenia. Instead of directly measuring muscle mass through blood tests, the system uses retinal vessel patterns captured in fundus images as a proxy indicator, processed through AI algorithms to infer sarcopenia status
2Measurement precision
If manual analysis of fundus images is performed, then measurement precision is improved, but productivity deteriorates and loss of time increases
Solution Approach 1:
The patent replaces manual visual analysis with an automated machine learning system. Convolutional neural networks and other AI algorithms automatically process fundus images to detect sarcopenia-related patterns, eliminating the need for manual examination while improving consistency and speed
Solution Approach 2:
The system enables self-service automated analysis where the machine learning model independently processes fundus images without human intervention. The AI model performs feature extraction, pattern recognition, and diagnosis automatically, allowing high-volume processing without additional human resources
3Productivity
If comprehensive machine learning analysis is implemented, then productivity is improved, but device complexity increases
Solution Approach 1:
The patent segments the complex analysis task into multiple specialized machine learning models. Different models handle specific aspects such as retinal vessel analysis, optic disc evaluation, and macular assessment, with each model optimized for its specific function. This modular approach manages complexity while maintaining comprehensive analysis capability
Solution Approach 2:
The patent creates a universal machine learning framework that can process multiple types of ophthalmic images (fundus images, OCT images, slit-lamp images) through a common architecture. The system uses shared feature extraction layers and can adapt to different image types, reducing overall system complexity compared to having separate specialized systems for each image type
Data Source
AI summary
Disclosed is an electronic device for predicting sarcopenia, which includes a memory and a processor connected with the memory to execute instructions included in the memory. The processor extracts a first result value as output data for a first machine learning model by using an eye image of a subject as input data for the first machine learning model and determines whether sarcopenia of the subject occurs based on the first result value. The first result value includes an MRD1 value of the subject corresponding to the eye image, an upper eyelid edge location change value, an eye closing speed value, and an eye opening speed value.