Structure-Derived Visual Field Priors for Faster Glaucoma Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Visual field tests for glaucoma diagnosis are lengthy, subjective, and prone to patient fatigue, necessitating improvements for reduced test time and accuracy while leveraging structural and functional characteristics of the eye.
Innovation Solution
A system utilizing machine learning and ophthalmic imaging to derive structure-derived visual field priors, optimizing the starting points of functional visual field tests by predicting threshold sensitivity values based on biometric measurements, reducing the need for iterative adjustments and shortening test duration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If standard automated perimetry (SAP) with iterative thresholding is used, then visual field test accuracy is improved, but test duration increases leading to patient fatigue
Solution Approach 1:
The system performs preliminary biometric measurements (OCT, OCTA, fundus imaging) to obtain structural and functional characteristics of the eye before conducting the visual field test. These preliminary data are used to predict expected threshold values, which then serve as optimized starting points for the SAP iterative thresholding process, reducing the number of iterations needed while maintaining accuracy
Solution Approach 2:
The patent introduces biometric data (OCT retinal layer thickness, OCTA blood flow, fundus imaging characteristics) as intermediary information that mediates between the structural eye characteristics and the functional visual field test. This intermediary data enables prediction of threshold values without requiring extensive iterative testing, thus bridging structure-function relationship efficiently
2Measurement precision
If longer visual field tests are administered to ensure accuracy, then measurement precision is improved, but patient cooperation and alertness deteriorate due to fatigue
Solution Approach 1:
The system collects and processes biometric data before the visual field test to predict threshold values in advance. This preliminary action enables the test to start with optimized parameters, reducing the time required to achieve accurate measurements and thereby maintaining patient alertness and cooperation throughout the test
Solution Approach 2:
The system uses the patient's own biometric data (OCT, OCTA, fundus imaging) to generate personalized predictions of their visual field thresholds. This self-service approach tailors the test parameters specifically to each patient's anatomical and functional characteristics, optimizing the test duration and accuracy for that individual patient
3Productivity
If biometric measurements from OCT and OCTA are integrated with visual field testing, then test optimization is improved, but device complexity increases
Solution Approach 1:
The patent creates a multi-functional system where a single integrated platform performs biometric measurements (OCT, OCTA, fundus imaging), processes the data through machine learning algorithms to predict thresholds, and then controls the visual field test presentation. This universal system consolidates multiple functions into one device, improving efficiency while managing complexity through integration
Solution Approach 2:
The system replaces manual test parameter adjustment and interpretation with automated machine learning algorithms that process biometric data and generate predicted threshold values. This substitution of mechanical/manual processes with computational algorithms streamlines the workflow and reduces the operational complexity despite the increased system integration
4Loss of time
If fewer test points are used to reduce test time, then loss of time is reduced, but measurement precision deteriorates
Solution Approach 1:
The system applies local quality optimization by using biometric data to predict which specific test points are most critical for each patient based on their individual anatomical and functional characteristics. Rather than uniformly reducing all test points, the system maintains or emphasizes testing at locations where the patient's biometric data indicates higher risk or greater variability, ensuring precision is preserved where most needed while reducing time at less critical locations
Data Source
Figure 1
Figure 2~3
Figure 4~6
AI summary
System for customizing visual field (VF) tests uses a machine learning model (15) trained on retina images (12A, 12C, 12D), including optical coherence tomography (OCT), optical coherence tomography angiography (OCTA), fundus, and/or fluorescein angiography images. In operation, in preparation for administering a specific VF test (13) to a patient, a retina image of the patient is submitted to the present machine model, which responds by synthesizing a VF prediction for the patient. The synthesized VF may be used to optimize the specific VF test prior to administering it to the patient.