# US 11,298,017 B2 · Medical analysis system > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,298,017 B2 - Document ID: US11298017B2 - Status: Granted - Kind: B2 - Application number: 16/454,903 - Filed: 2019-06-27 - Published: 2020-12-31 - Granted: 2022-04-12 - Earliest claimed: 2019-06-27 - Applicants: Bao Tran - Inventors: Bao Tran - Web page: https://patentgrep.com/patent/US11298017B2 ## CPC - A61B 3/0025: {characterised by electronic signal processing, e.g. eye models} - A61B 3/14: Arrangements specially adapted for eye photography - A61B 3/0008: {provided with illuminating means (A61B3/117, A61B3/14 take precedence)} - A61B 3/0016: {Operational features thereof} - A61B 3/107: for determining the shape or measuring the curvature of the cornea - A61B 3/16: for measuring intraocular pressure, e.g. tonometers - G16H 30/40: for processing medical images, e.g. editing - G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems - G06T 2207/20084: Artificial neural networks [ANN] - G06T 2207/30041: Eye; Retina; Ophthalmic - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract Systems and methods are disclosed to inspect an eye includes capturing an eye image using a mobile device camera; extracting features of the eye; applying a deep learning neural network to detect potential eye damage; and reporting the potential eye damage for treatment, such as those from laser pointers, among others. ## Claims ### Claim 1 A method to inspect an eye with a mobile device, comprising: capturing an eye image using a mobile device camera coupled to a processor; capturing gyroscope or accelerometer readings as image metadata to aid in orientation normalization and image registration by the processor; extracting features of the eye using the image corrected with image metadata and clinical information from a population including medical history by the processor; applying the extracted features to a deep learning neural network to detect potential eye damage by the processor; reporting the potential eye damage for treatment. ### Claim 2 (depends on 1) The method of claim 1, comprising providing an adapter to couple the mobile device to the eye. ### Claim 3 (depends on 1) The method of claim 1, comprising positioning one or more light emitters or light pipes to carry light from the mobile device in an adapter. ### Claim 4 (depends on 1) The method of claim 1, comprising capturing stereo images of the eye with a second camera on the mobile device and using data from sensors on an adapter for image processing. ### Claim 5 (depends on 1) The method of claim 1, comprising applying a conditional GAN to learn image patterns. ### Claim 6 (depends on 1) The method of claim 1, comprising generating features and applying the features to detect similar eye conditions. ### Claim 7 (depends on 6) The method of claim 6, comprising retrieving treatment or diagnosis information from the detected similar eye conditions. ### Claim 8 (depends on 1) The method of claim 1, comprising generating historical feature vectors from one or more eye examinations of a patient, training the deep learning neural network with the historical feature vectors along with eye images, and applying the trained deep learning neural network to diagnose the eye. ### Claim 9 A method to inspect an eye with a mobile device with a gyroscope or an accelerometer, comprising: capturing an eye image using a mobile device camera coupled to a processor; capturing gyroscope or accelerometer readings as image metadata to aid in orientation normalization and image registration by the processor; extracting features of the eye using the image corrected with image metadata and clinical information including medical history by the processor; and detecting third party intentional laser damage on the eye using the deep learning network. ### Claim 10 (depends on 1) The method of claim 1, comprising providing a similarity search for the eye image and retrieving historical treatment information from similar cases as precedential treatment data. ### Claim 11 (depends on 1) The method of claim 1, comprising displaying from a database eye images similar to the eye image captured by the mobile device camera to provide treatment information from similar cases. ### Claim 12 (depends on 1) The method of claim 1, wherein the mobile device camera comprises an optical zoom lens with focusing adjustment capability to image internal eye targets at different distances from the lens. ### Claim 13 (depends on 12) The method of claim 12, comprising controlling the optical zoom lens to focus on a posterior of the eye; providing a neural network trained to focus on structures on the posterior of the eye; and imaging the posterior of the eye. ### Claim 14 (depends on 1) The method of claim 1, wherein the mobile device provides background lighting for imaging cataract in the eye with retro illumination. ### Claim 15 (depends on 1) The method of claim 1, wherein the mobile device determines intraocular pressure (IOP) with the deep learning neural network trained on mapping lens curvature to IOP. ### Claim 16 (depends on 15) The method of claim 15, comprising calibrating the IOP with a tonometer in an adapter. ### Claim 17 (depends on 1) The method of claim 1, wherein the mobile device images a posterior of the eye with solid state lighting units and light conditioning optics. ### Claim 18 (depends on 17) The method of claim 17, wherein the mobile device emits light with narrow spectral bandwidth, broad spectral bandwidth, visible spectrum, or invisible spectrum. ### Claim 19 (depends on 1) The method of claim 1, comprising placing light sources and sensors in an adapter positioned between the mobile device camera and the eye. ### Claim 20 A method to inspect an eye with a mobile device with a gyroscope or an accelerometer, comprising: capturing an eye image using a mobile device camera coupled to a processor; capturing gyroscope or accelerometer readings as image metadata to aid in orientation normalization and image registration by the processor; extracting features of the eye using the image corrected with image metadata and clinical information including medical history by the processor; applying the extracted features to a deep learning neural network to detect potential eye damage by the processor; reporting the potential eye damage for treatment; and wherein the adapter enables two or more cameras in the mobile device to image the eye.