US 12,133,683 B2
Vision testing via prediction-based setting of initial stimuli characteristics for user interface locations
Mohamed Abou Shousha, Fort Lauderdale, FL (US)
Assigned to University of Miami, Miami, FL (US)
Filed by UNIVERSITY OF MIAMI, Miami, FL (US)
Filed on Aug. 3, 2021, as Appl. No. 17/393,101.
Application 17/393,101 is a continuation in part of application No. 17/343,292, filed on Jun. 9, 2021, granted, now 11,659,987.
Application 17/343,292 is a continuation of application No. 17/083,043, filed on Oct. 28, 2020, granted, now 11,039,742, issued on Jun. 22, 2021.
Prior Publication US 2022/0125299 A1, Apr. 28, 2022
Int. Cl. A61B 3/024 (2006.01); A61B 3/02 (2006.01); A61B 3/028 (2006.01); A61B 3/06 (2006.01); G06N 3/02 (2006.01); G16H 50/20 (2018.01); G16H 50/30 (2018.01)
CPC A61B 3/024 (2013.01) [A61B 3/022 (2013.01); A61B 3/028 (2013.01); A61B 3/06 (2013.01); G06N 3/02 (2013.01); G16H 50/20 (2018.01); G16H 50/30 (2018.01)] 20 Claims
OG exemplary drawing
 
1. A system for dynamically determining stimuli characteristics for vision defect determination during a vision test in a head mounted display, the system comprising:
one or more processors; and
non-transitory computer-readable media storing instructions that, when executed by the one or more processors, cause operations comprising:
obtaining first feedback indicating a binary characteristic under which a user sees one or more first stimuli presented on a user interface, wherein indicating the binary characteristic under which the user sees one or more first stimuli presented on the user interface consists of indicating whether the user sees one or more first stimuli presented on the user interface with a first characteristic or a second characteristic;
generating a feature input based on the first feedback;
inputting the feature input into a prediction model, wherein the prediction model is neural network trained to predict non-binary characteristics for sets of locations and confidence scores associated with the sets of locations based on feedback indicating binary characteristics under which users see one or more stimuli presented on user interfaces;
obtaining, via the prediction model, based on the first feedback, (i) a set of predicted non-binary characteristics for a set of locations of the user interface and (ii) a set of confidence scores associated with the set of locations;
selecting, based on the set of confidence scores, one or more locations of the set of locations that are to be tested during a visual test presentation, the one or more locations being selected over one or more other locations of the set of locations based on the set of confidence scores;
presenting, based on the set of predicted non-binary characteristics for the set of locations of the user interface, one or more stimuli at the selected locations during the visual test presentation;
obtaining second feedback indicating one or more threshold characteristics under which the user sees the one or more stimuli at the selected locations, wherein the one or more threshold characteristics are non-binary characteristics; and
generating visual defect information for the user based on the second feedback.