Visual Field Test Assistance Using Bayesian Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current visual field tests for diagnosing diseases like glaucoma are burdensome for both patients and testing personnel, requiring extensive luminance changes and reaction assessments at multiple test points, leading to inefficiencies in inferring visual field changes.
Innovation Solution
A visual field test assistance apparatus that uses Bayesian inference with a mixture distribution to estimate current visual sensitivity at each test point, updating prior distributions with real-time data and providing inference results to reduce the number of necessary tests and improve efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional visual field testing methods are used to test visual sensitivity at multiple test points, then diagnostic accuracy is improved, but testing burden and time consumption increase significantly
Solution Approach 1:
The system performs preliminary Bayesian inference calculations using prior distributions of visual sensitivity data from multiple cases before the actual test. This pre-computation of probability distributions allows the system to predict likely visual sensitivity values at untested points, reducing the number of actual measurements needed while maintaining diagnostic accuracy
Solution Approach 2:
The system implements iterative feedback by updating the prior distribution with actual measurement results as they are obtained during testing. Each new measurement refines the Bayesian inference for remaining untested points, allowing the system to adaptively reduce testing points based on accumulated information while preserving measurement precision
2Measurement precision
If traditional visual field testing methods are used to test visual sensitivity at multiple test points, then diagnostic accuracy is improved, but the burden on patients and testing personnel increases
Solution Approach 1:
The system performs automated Bayesian inference calculations and automatically updates probability distributions based on measurement results. This self-service capability reduces the need for manual computation and complex decision-making by testing personnel, while the adaptive selection of test points reduces the overall burden on patients
Solution Approach 2:
The Bayesian inference system acts as an intermediary between raw measurements and diagnostic conclusions. It processes measurement data through probability distributions to generate inferred visual sensitivity values, simplifying the workflow for personnel and reducing the number of direct patient interactions required
3Productivity
If Bayesian inference with mixture distribution is used to estimate visual sensitivity, then the number of necessary tests is reduced, but calculation complexity increases
Solution Approach 1:
The system pre-computes and stores prior distributions of visual sensitivity parameters from multiple historical cases before actual testing begins. This preliminary preparation of probability models allows the Bayesian inference during testing to proceed more efficiently by leveraging pre-processed statistical information rather than calculating from scratch
Solution Approach 2:
The system uses mixture distributions that replicate the statistical characteristics of visual sensitivity across multiple cases. By copying and adapting these pre-established probability distributions to the current patient's data, the system achieves accurate inference with reduced computational burden compared to developing entirely new models
Data Source
AI summary
A visual field test assistance apparatus infers a current visual field status at each test point. A predetermined mixture distribution is applied to a prior-distribution, and parameters regarding the predetermined mixture distribution are learned with reference to time-series information of visual field test results obtained in the past regarding a plurality of cases. The visual field test assistance apparatus calculates an inference test result vector representing an inference result of visual sensitivity to be obtained by the visual field test at each test point, by Bayesian inference, and performs an information output process on the basis of the inference test result vector obtained by the inference.