Computer-implemented method and system for determining optical properties of eye
Disclosed is a computer-implemented method for determining optical properties of an eye. The computer-implemented method comprises collecting one or more than one images of the eye. Moreover, the computer-implemented method comprises analyzing the collected one or more than one images using a trained neural network and determining the optical properties of the eye based on the analysis of the one or more than one images. The method provides accessible, cost-efficient, user-friendly, efficient and personalized diagnostics to the users by using the trained neural network.
1 . A computer-implemented method ( 100 ) for determining optical properties of an eye, the computer-implemented method comprising:
detecting a distance of an image capturing unit on a handheld mobile device relative to a position to the eye;
using a guidance algorithm, monitoring a first position of the eye in the image capturing unit's view;
using the guidance algorithm, monitoring a second position of the image capturing unit relative to a light source;
when the detected distance, the first position of the eye in the image capturing unit's view, and the second position relative to the light source, satisfy desired parameters, automatically collecting one or more than one images of the eye from the mobile device using the image capturing unit;
analyzing the collected one or more than one images using a trained neural network; and
determining the optical properties of the eye based on the analysis of the one or more than one images.
2 . The computer-implemented method ( 100 ) according to claim 1 , wherein the one or more than one images are collected from a user device ( 204 ), or a server ( 208 ).
3 . The computer-implemented method ( 100 ) according to claim 1 , further comprising labelling and storing, in the server ( 208 ), the collected one or more than one images, wherein the collected one or more than one images are stored in a folder ( 210 ) associated with a user, wherein the folder has a first subfolder ( 212 ) specific for the one or more than one images of a left eye of the user, a second subfolder ( 214 ) specific for the one or more than one images of a right eye of the user, and a database ( 216 ) of the user.
4 . The computer-implemented method ( 100 ) according to claim 1 , further comprising pre-processing the collected one or more than one images for one or more than one of: a noise removal, a distortion correction, a red-eye effect detection, and a resolution correction.
5 . The computer-implemented method ( 100 ) according to claim 1 , further comprising activating an image capturing unit ( 206 ) of the user device, arranged relative to the eye, to obtain the one or more than one images when a red-eye effect is detected by the image capturing unit.
6 . The computer-implemented method ( 100 ) according to claim 1 , wherein analyzing the collected one or more than one images comprises one or more than one of: detecting patterns on the one or more than one images, determining a distance between the image capturing unit ( 206 ) and the eye, and determining shifts and tilts between the image capturing unit and eye axes, wherein the distance between the image capturing unit ( 206 ) and the eye is measured using a measurement tool.
7 . The computer-implemented method ( 100 ) according to claim 1 , wherein the trained neural network is a convolutional neural network.
8 . The computer-implemented method ( 100 ) according to claim 1 , further comprising employing artificial intelligence algorithms and machine learning tools for detecting the one or more than one characteristic features of the one or more than one images, wherein the machine learning tools are employed for sorting, systematizing, analyzing the one or more than one images and output a diagnosis based thereon, and the artificial algorithms employ the machine learning output and data derived from observation by a specialist to confirm the diagnosis.
9 . The computer-implemented method ( 100 ) according to claim 1 , further comprising screening users based on a pre-defined threshold.
10 . The computer-implemented method ( 100 ) according to claim 1 , further comprising providing a report of the analysis of the one or more than one images, wherein the report comprises one or more than one of: a defect, a probability of a measurement error, and an advice regarding a physical consultation with a specialist.
11 . A system ( 200 ) for determining optical properties of an eye, the system comprising a handheld mobile device comprising a processor ( 202 ) configured to:
detect a distance of an image capturing unit on the handheld mobile device relative to a position to the eye;
monitor a first position of the eye in the image capturing unit's view;
monitor a second position of the image capturing unit relative to a light source;
when the detected distance, the first position of the eye in the image capturing unit's view, and the second position relative to the light source, satisfy desired parameters, automatically collect one or more than one images of the eye from the mobile device using the image capturing unit;
analyze the collected one or more than one images using a trained neural network; and
determine the optical properties of the eye based on the analysis of the one or more than one images.
12 . The system ( 200 ) according to claim 11 , further comprising
a user device ( 204 ) having an image capturing unit ( 206 ) for obtaining the one or more than one images; and
a server ( 208 ) configured to store the collected one or more than one images in a folder ( 210 ) associated with a user, wherein the folder has a first subfolder ( 212 ) specific for the one or more than one images of a left eye of the user, a second subfolder ( 214 ) specific for the one or more than one images of a right eye of the user, and a database ( 216 ) of the user.
13 . The system ( 200 ) according to claim 11 , further comprising a measurement tool for measuring a distance between the image capturing unit ( 206 ) and the eye.
14 . A computer program product comprising a non-transitory computer-readable storage medium having computer-readable instructions stored thereon, the computer-readable instructions being executable by a computing device comprising a processor ( 202 ) to execute instructions that cause the processor to:
detect a distance of an image capturing unit on a handheld mobile device relative to a position to the eye;
monitor a first position of the eye in the image capturing unit's view;
monitor a second position of the image capturing unit relative to a light source;
when the detected distance, the first position of the eye in the image capturing unit's view, and the second position relative to the light source, satisfy desired parameters, automatically collect one or more than one images of the eye from the mobile device using the image capturing unit;
analyze the collected one or more than one images using a trained neural network; and
determine the optical properties of the eye based on the analysis of the one or more than one images.