Lens Ordering System Using Deep Learning for Psychosomatic Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lens treatment systems fail to comprehensively consider various factors such as user psychosomatic conditions and usage scenarios, making it difficult to determine appropriate lens materials and coatings for optimal eye health and comfort.
Innovation Solution
A computer-based lens ordering system that uses user information, including basic details, usage information, and psychosomatic symptoms, to determine and output recommended lens information using a learned model trained with deep learning techniques to select resin material, pigment, and coating types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple factors (usage information and psychosomatic information) are considered to determine lens information, then the quality of lens recommendations is improved, but the complexity of the system increases
Solution Approach 1:
The patent segments the lens determination process into two distinct information types: usage information (what the lens should do) and psychosomatic information (user's physical/mental conditions). This segmentation allows the system to handle multiple factors systematically through separate input categories that are processed independently before integration, reducing overall system complexity while maintaining comprehensive analysis.
Solution Approach 2:
The patent introduces a learned model (neural network) as an intermediary between the input information and output recommendations. This intermediary automatically processes and integrates the multiple input factors (usage and psychosomatic information) through learned relationships, eliminating the need for complex manual rule-based processing and simplifying the system architecture while improving recommendation quality.
2Measurement precision
If a learned model with deep learning techniques is used to process user information, then the accuracy of lens information determination is improved, but the computational resources required increase
Solution Approach 1:
The patent applies preliminary action by pre-training the neural network model offline with large datasets before deployment. The model learns optimal relationships between input information and lens recommendations in advance, so that during actual use, only inference (not full training) is required. This significantly reduces computational resources needed at runtime while maintaining high accuracy.
3Adaptability or versatility
If comprehensive user information is collected to determine lens treatments, then the suitability of lens recommendations is improved, but the difficulty of information processing increases
Solution Approach 1:
The patent transforms diverse user information (usage preferences and psychosomatic conditions) into standardized numerical parameters that the neural network can process. By converting qualitative information into quantitative parameters, the system handles comprehensive user data more easily, improving suitability while reducing processing difficulty through parameter standardization.
Data Source
AI summary
A lens ordering system includes: a computer, the computer being provided with a CPU, a storage unit, and a selection unit that is configured to be controlled by the CPU. The storage unit stores data, configured to enable determination of lens information that includes a resin material type, a pigment type, and a coating type that are used for a lens material, based on user information that includes basic information pertaining to a user, and at least one of usage information pertaining to lens usage desired by the user or psychosomatic information including psychosomatic symptoms and conditions including the eyes of the user. The selection unit determines and outputs the lens information based on the user information acquired pertaining to the user and the data stored in the storage unit.