Systems and methods for measuring visual function maps

Patent metadata

Field	Value
Publication number	US-10925481-B2
Application number	US-201916290654-A
Country	US
Kind code	B2
Filing date	Mar 1, 2019
Priority date	Mar 2, 2018
Publication date	Feb 23, 2021
Grant date	Feb 23, 2021

Abstract

A system can include a non-transitory memory to store machine readable instructions and data, and a processor to access the non-transitory memory and execute the machine-readable instructions. The machine-readable instruction can include a global module that can be programmed to generate a visual field map (VFM) model that can include a set of visual function map parameters for an entire visual field for a subject. The global module can be programmed to update the set of visual function map parameters corresponding to updating a shape of the VFM based on subject response data generated during each administration of a vision test to a subject.

First claim

What is claimed is:

1. A system comprising:
a non-transitory memory to store machine readable instructions and data;
a processor to access the memory and execute the machine-readable instructions, the machine-readable instructions comprising:
a global module programmed to generate a visual field map (VFM) model comprising a set of visual function map parameters for all retinal locations of a visual field for a subject, wherein the global module is programmed to update the set of visual function map parameters corresponding to updating a shape of the VFM based on subject response data generated during each administration of a vision test to a subject;
a switch module programmed to evaluate the performance of the global module and make a determination as to whether to switch to a local module; and upon switching, calculate independent parameters for each retinal location of the visual field based on the set of visual function map parameters in the global module; and
the local module programmed to update the independent parameters based on the subject response data generated during a subsequent administration of the vision test to the subject to provide an assessment of visual function at each retinal location of the visual field for the subject.

2. The system of claim 1 , wherein the global module is programmed to generate a visual function map across all retinal locations in the visual field, wherein each visual function map comprises a corresponding set of parameters.

3. The system of claim 2 , wherein the visual function is one of but not limited to light sensitivity, contrast sensitivity, visual acuity, contrast sensitivity function, color sensitivity, stereo vision, temporary frequency sensitivity, motion sensitivity, reading speed, and crowding.

4. The system of claim 2 , wherein the global module is programmed to generate a prior probability density function for each combination of the visual function map parameters of each visual function.

5. The system of claim 4 , wherein the prior probability density function is one of an uninformative prior, a weakly informative prior, or an informative prior.

6. The system of claim 4 , wherein the global module is programmed to update the prior probability density function, for each combination of the visual function map parameters according to a Bayes rule based on the response data to generate a posterior probability density function for each combination of the visual function map parameters.

7. The system of claim 6 , wherein the global module is programmed to select a stimulus parameter from a stimulus parameter space to control a subsequent administration of the vision test to the subject based on the updated probability density function for each combination of the visual function map parameters of each visual function.

8. The system of claim 7 , wherein the global module is programmed to iteratively update the probability density function for each combination of the visual function map parameters of each visual function based on the subject response data generated during each administration of the vision test to the subject.

9. The system of claim 8 , wherein the global module is programmed to select the stimulus parameter from among a plurality of stimulus parameters that optimizes an expected information gain on each combination of the visual function map parameters of each visual function.

10. The system of claim 9 , wherein the global module is programmed to select the stimulus parameter based on the posterior probability density function for each combination of the visual function map parameters of each visual function and based on an expected response by the subject according to a given amount of vision tests.

11. The system of claim 4 , wherein the global module is programmed to obtain a prior according to one of (i) a structure-based prior, (ii) a prior derived from statistical or machine learning, (iii) a prior derived from a hierarchical adaptive method, and (iv) a prior from a previous visual function map.

12. The system of claim 1 , wherein to calculate the independent parameters for each retinal location comprises generating a prior distribution of visual function for each retinal location in the visual field based on the posterior probability distribution for each combination of the visual function map parameters of each visual function generated by the global module.

13. The system of claim 1 , wherein the switch module is programmed to evaluate one of a rate of information gain and improvement in information gain by the global module relative to an information criterion.

14. The system of claim 13 , wherein the switch module is programmed to generate the prior probability distribution on measures of visual function at each retinal location based on a result of the evaluation.

15. The system of claim 13 , wherein to evaluate comprises comparing a total expected information gain (TEI) in a subsequent vision test with an average TEI of a plurality of prior administered vision tests to the subject.

16. The system of claim 1 , wherein updating the independent parameters comprises assessing the visual function at each retinal location of the visual field based on an adaptive testing procedure.

17. The system of claim 16 , wherein the adaptive testing procedure comprises a Bayesian adaptive procedure.

18. The system of claim 16 , wherein the adaptive testing procedure comprises a Bayesian active learning procedure.

19. The system of claim 1 , wherein the local module is programmed to evaluate the visual function at each retinal location and an expected information gain across all retinal locations and stimuli parameters in the stimulus space to determine a given stimulus for a subsequent administration of the vision test.

20. The system of claim 19 , wherein the local module is programmed to terminate evaluating the visual function at each retinal location of the visual field based on the adaptive procedure according to an evaluation criterion.

21. The system of claim 1 , further comprising:
a vision stimulation device configured to administer the vision test to the subject in response to one of the global module and the local module.

22. The system of claim 21 , wherein the processor comprises one or more graphics processing unit (GPUs).

23. The system of claim 22 , further comprising an eye tracking unit configured to record an eye movement of the subject during the vision test and generate eye movement data characterizing the eye movements of the subject.

24. The system of claim 23 , wherein the eye movement data are part of the subject response data.

25. The system of claim 21 , further comprising a structural imaging device and a recording device configured to operate in cooperation with the processor.

Assignees

• Ohio State Innovation Foundation
• Adaptive Sensory Tech Inc

Inventors

• Lu Zhong-Lin
• Xu Pengjing
• Lesmes Luis
• Yu Deyue

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G16H50/20

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title

• A61B3/0025 Primary

  characterised by electronic signal processing, e.g. eye models · CPC title

• A61B3/063

  for testing light sensitivity, i.e. adaptation · CPC title

• G06N20/00

  Machine learning · CPC title