Intraocular Lens Selection Using AI Error Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current intraocular lens (IOL) calculation formulas diverge significantly at specific ranges of input parameters, leading to clinical dilemmas in selecting the most accurate formula for individual eyes.
Innovation Solution
The use of a three-dimensional super surface that represents ideal portions of multiple IOL selection formulas, combined with a neural network to estimate errors in lens power calculations, allows for the selection of the most accurate intraocular lens power based on ocular measurement parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple IOL calculation formulas are used to improve accuracy across different eye types, then the reliability of lens power determination is improved, but the device complexity and difficulty of selection increase
Solution Approach 1:
The patent segments the selection of IOL calculation formulas by creating a three-dimensional super surface that divides the parameter space into distinct regions. Each region corresponds to the optimal performance range of a specific IOL formula (e.g., SRK/T, Hoffer Q, Holladay I, Haigis). By segmenting the formula selection based on measured parameter values (axial length, corneal power, ACD), the system eliminates the complexity of manual formula selection while maintaining high accuracy across diverse eye types.
Solution Approach 2:
The patent introduces a three-dimensional super surface as an intermediary between the measured ocular parameters and the IOL formula selection. This super surface acts as a mediator that automatically determines which formula is most appropriate for given parameter values, eliminating the need for clinicians to directly compare and select from multiple formulas. The super surface translates complex formula performance characteristics into an intuitive visual tool that guides formula selection.
2Ease of operation
If a single IOL calculation formula is used to simplify the selection process, then the ease of operation is improved, but the measurement precision and reliability decrease at specific parameter ranges
Solution Approach 1:
The patent creates a universal three-dimensional super surface that incorporates the optimal performance characteristics of multiple IOL formulas into a single tool. This super surface serves multiple functions: it identifies the best formula for any given set of ocular parameters, visualizes the performance characteristics of different formulas, and guides clinicians in making accurate formula selections. By making the selection tool universal rather than formula-specific, the system maintains ease of operation while ensuring precision across all parameter ranges.
3Manufacturing precision
If IOL formulas are adjusted for specific parameter ranges to improve accuracy, then the manufacturing precision of the calculation model is improved, but the device complexity increases
Solution Approach 1:
The patent transitions from two-dimensional formula adjustments (modifying formulas for specific parameter ranges) to a three-dimensional solution space. The three-dimensional super surface plots axial length, corneal power, and ACD as three independent dimensions, creating a volumetric representation of formula performance. This dimensional expansion allows the system to maintain high precision across all parameter combinations without requiring complex adjustments to individual formulas, as the super surface naturally captures the optimal formula selection across the entire three-dimensional parameter space.
Data Source
AI summary
Aspects of the present invention may include systems and methods for intraocular lens selection using a formula and a deep learning machine to estimate the error of the formula. In an aspect, the disclosure provides a method for intraocular lens selection. The method may include obtaining at least two ocular measurement parameters and a lens selection parameter for an eye. The method may include determining an intraocular lens power based on a formula using the at least two ocular measurement parameters. The method may include determining an estimated error of the formula using a deep learning machine trained on verified post-operative results using intraocular lenses selected by the formula. The method may include adjusting the lens selection parameter based on the estimated. The method may include redetermining the intraocular lens power based on the formula and the adjusted lens selection parameter.