A system and method for providing visual tests

Abstract

Embodiments of the invention are directed towards systems, methods and computer program products for providing improved eye tests. Such tests improve upon current eye tests, such as visual field tests, by incorporating virtual reality, software mediated guidance to the patient or practitioner such that more accurate results of the eye tests are obtained. Furthermore, through the use of one or more trained machine learning or predictive analytic systems, multiple signals obtained from sensors of a testing apparatus are evaluated to ensure that the eye test results are less error-prone and provide a more consistent evaluation of a user's vision status. As it will be appreciated, such error reduction and user guidance systems represent technological improvements in eye tests and utilize non-routine and non-conventional approaches to the improvement and reliability of eye tests.

What is claimed is: 
     
         1 . A virtual, augmented and mixed reality system for conducting visual tests of a subject, the system comprising:
 a virtual, augmented and mixed reality goggles configured to provide a visual test and virtual assistant to the subject;   at least one subject sensor configured to monitor the subject during administration of the visual test;   a processor, configured by code executing therein to:
 cause the goggles to display one of a plurality of pre-determined visual tests to the subject, wherein each of the plurality of pre-determined visual tests includes one or more operational states and at least one virtual assistant provided within a visual field displayed to the subject; 
 while administering one of a plurality of pre-determined visual tests, provide the output of the at least one subject sensor and a current operational state of the displayed one of a plurality of visual tests to a pre-trained neural network, wherein the pre-trained neural network is configured to output an instruction set in response to the current operational state and the sensor output; 
 wherein the instruction set causes the virtual assistant to provide one or more action prompts to the subject that instructs the subject to a corrective physical action during administration of the visual test. 
   
     
     
         2 . The system of  claim 1 , wherein the visual test is selected from:
 A visual field test, a visual acuity test, a contrast sensitivity test, a tonometry test, an ophthalmic phototherapy test; a pupillometry test; an autorefraction test; an eye movement test, a color test, and visual therapy.   
     
     
         3 . A method of administering a visual test to a subject, the method comprising:
 providing one of a plurality of visual tests, each visual test having a plurality of operational states, to a display device incorporated within a goggle, the display device configured to provide a virtual reality scene to the subject;   receiving sensor data from one or more sensors disposed within the goggles, wherein at least one sensor is configured to track the eye-movements of the subject;   generating a virtual assistant avatar within the virtual augmented or mixed reality scene displayed to the subject;   providing the received sensor data and current operational state of the displayed one of a plurality of visual tests to a pre-trained neural network, wherein the pre-trained neural network is configured to correlate the sensor and operation state input values with an instruction set, wherein the instruction set includes one or more audiovisual prompts for display by the virtual assistant; and   receiving the instruction set and causing the virtual assistant to provide the one or more audiovisual action prompts to the subject, wherein the audiovisual action prompt is correlated with a correction action that instructs the subject to a corrective physical action during administration of the visual test.   
     
     
         4 . A system for a visual field test comprising, at least one data processor comprising a virtual reality, an augmented reality or a mixed reality engine and at least one memory storing instruction which is executed by at least one data processor, at least one data processor configured to:
 (i) present a virtual assistant in virtual reality, augmented reality or mixed reality, wherein the virtual assistant presents to a patient a set of instructions for the visual field test,   (ii) provide a set of n stimuli, wherein n is greater than or equal to 2, each stimulus having a (a) specified size, (b) shape and (c) luminance, wherein the luminance of the stimulus is greater than the luminance of a background;   (iii) receive from the patient at least one response when the patient views at least one stimulus, wherein the response comprises clicking the response button, a verbal response, a sound command or the objective analysis of the anterior segment of the eye;   (iv) repeat steps (ii) to (iii) at least y times, where y is greater than 2, until the patient indicates that the lowest stimulus intensity has been seen;   (v) repeat steps (i) to (iv), wherein the explanation in step (i) is modified based on the patient's response to provide a second set of instructions selected from a library of instructions if a percentage of responses in step (iii) labeled as correct is less than the percentage expected to be correct based on a historical value for the patient's retinal sensitivity score or an estimated percentage of correct choices based on a probability score; or   (vi) calculate a visual field score if a percentage of responses in step (iii) labeled as correct is greater than or equal to the percentage expected to be correct based on a historical value for the patient's retinal sensitivity score or an estimated percentage of correct choices based on a probability score.   
     
     
         5 . The system of  claim 4 , further comprising a virtual reality, an augmented reality or a mixed reality headset. 
     
     
         6 . The system of  claim 4 , wherein the set of instructions comprises a patient guide or an explanation of the visual field test. 
     
     
         7 . The system of  claim 4 , wherein the virtual reality comprises augmented reality or mixed reality. 
     
     
         8 . The system of  claim 4 , wherein the set of instructions in step (i) further comprises noting the presentation of all the stimuli. 
     
     
         9 . The system of  claim 8 , wherein the set of instructions further comprises providing information to the patient on timing and sequence of the visual field test. 
     
     
         10 . The system of  claim 4 , wherein the set of instructions further comprises an explanation of responses of the patient. 
     
     
         11 . The system of  claim 4 , wherein the set of instructions comprises a verbal explanation. 
     
     
         12 . The system of  claim 4 , wherein the set of instructions comprises a pictorial explanation providing a set of actions and vector movements showing a position of a stimuli wherein:
 (i) the stimulus blink and move indicating to the position of the stimulus; and   (ii) the stimulus change their appearance by blinking, glowing, changing the color, hue or intensity to attract the attention of the patient being tested.   
     
     
         13 . The system of  claim 4 , wherein the virtual assistant uses eye-tracking to reposition the visual field stimulus matrix to avoid the effect of “fixation losses”.
 (i) The virtual assistant turns the eye-tracking cameras n milliseconds before showing a stimulus. 
 (ii) The virtual assistant uses the eye-tracking data to detect the actual gaze position. 
 (i) virtual assistant changes the stimulus matrix to synchronize the center of the stimulus matrix with the actual optical axis or gaze position. 
 
     
     
         14 . The system of  claim 4 , wherein the virtual assistant is humanoid in appearance. 
     
     
         15 . The system of  claim 4 , wherein the virtual assistant is an avatar. 
     
     
         16 . The system of  claim 4 , wherein the virtual assistant is a cartoon character. 
     
     
         17 . The system of  claim 4 , wherein the virtual assistant is presented in three dimensions. 
     
     
         18 . The system of  claim 4 , wherein the stimulus are selected from the group consisting of circular stimuli of all Goldman sizes, sinusoidal bars and circular stimuli of different colors. 
     
     
         19 . The system of  claim 4 , wherein the virtual assistant uses neural networks (NN) and a decision tree to evaluate the inputs from patient, sensors and system state and provide a subsequent action, wherein the neural network and decision tree are trained by:
 (i) a training database, wherein the training database includes, for each member of a training population comprised of visual field tests taken by users, an assessment dataset that includes at least data relating to a respective user response to the visual field set and or a sensor input and or a system state.   (ii) a visual field score of the respective test; a training system including an expert system module configured to determine correlations between the respective user responses, sensor inputs and system state to the visual field test and the visual field score of each member of the training population.   (iii) a user testing platform configured to provide a user with a current visual field test and receive user input regarding responses to the current visual field test; an analysis system communicatively coupled to the training system and the user testing platform, the computer system adapted to receive the user input responses generated in response to the current visual field test and to assign a visual field score for the testing platform user using the correlations obtained from the training system.   
     
     
         20 . The system of  claim 1 , wherein the pre-trained neural network are trained by:
 (i) providing a first training dataset to a first neural network, wherein the training dataset is stored in a training database, wherein the training dataset includes, for each member of a training population comprised of users of one or more visual tests, an assessment dataset that includes at least data relating to at least one sensor measurement of a respective user in response to an operational state of the visual test;   wherein the operation state includes at least a success state and a fail state;   (ii) training the first neural network to determine correlations between the respective assessment data set and the operational state of the visual test for each member of the training population;   (iii) providing a second neural network with a second training data set, wherein the second training data set includes one of a plurality of pre-determined operational states of the visual test and one or more corrective subject instructions to change a fail state to a success state;   (iv) training the second neural network to determine correlations between the operational state of the visual test and the corrective instructions to change fail states to success states;   (v) training a third neural network by providing the assessment data to the first neural network, and providing the output of the first neural network to the second neural network as an input, so as to determine the correlation between the assessment dataset and a corrective action to change an associated fail operational state; and   (vi) outputting the third neural network as a trained neural network.