Method of and system for automated machine-assisted detection of ocular disease conditions in human eyes captured using visible illumination light sources and digital camera systems
Digital imaging processing system and methods for automatically recognizing ocular diseases including dry eye disease, and other ocular conditions in the eyes of humans by automatically (i) processing 2D in vivo digital images of the human eyes formed, captured and detected using a visible-wavelength operating digital camera system, and (ii) using machine-vision image processing subsystems (e.g. engines) to automatically detect the presence and location of specific objects in the ocular surfaces in the human eye, and confirm that specific conditions have been satisfied during image processing to support automated recognition of a specific ocular disease and other ocular conditions indicative of ocular pathology.
1 . A cloud-based system network for (i) machine-vision supported recognition, treatment and management of ocular conditions detected in the one or both eyes of human beings being treated as patients, specifically for dry eye disease (DED) conditions detected in the eyes of the human beings and severity of the detected DED conditions, and (ii) delivery of ocular care to the human beings as patients situated remotely on the Earth, including prescription of treatment and assessment of treatment response, said cloud-based system network comprising:
a plurality of mobile image capturing systems, selected from the group consisting of mobile smartphone camera systems, mobile camera systems and mobile image capturing devices, each said mobile image capturing system having an integrated camera subsystem with image formation optics and an optical sensor for forming and detecting said digital images using visible-wavelength sources of light illumination, and a graphical user interface (GUI) display surface supporting communication and augmented reality (AR) with a patient and/or user whose eyes are to be monitored using said integrated camera subsystem, and cared for using services supported over said cloud-based system network;
a data center supporting web, application and database servers connected to infrastructure of the Internet, to support functions and services to be delivered to patients and users by said cloud-based system network including an automated machine-vision driven ocular disease (OD) recognition engine;
a geolocation (GPS) tracking and linking subsystem for GPS-specified location tracking of systems and subsystems and workstations deployed on said cloud-based system network;
wherein said automated machine-vision driven ocular disease (OD) recognition engine comprises:
an automated digital image processing server realized within said data center, supporting processing of said digital images of said human eyes, and one or more clinical ophthalmic decision support subsystems realized as workstations, tablet computers and/or laptop workstations; and
a database subsystem for storing data and maintaining data records for patients and users across the cloud-based system network;
wherein, in order to realize automated recognition of dry eye disease (DED) conditions present in human eyes, said automated machine-vision driven ocular disease (OD) recognition engine supports and enables one or more methods of automated detection and/or measurement selected from the group consisting of:
an automated detection and/or measurement of hyperemia of the human eyes;
an automated detection and/or measurement of tear film in the human eyes;
an automated detection and/or measurement of oily lid margin in the human eyes;
an automated detection and/or measurement of capped meibomian glands in the human eyes;
an automated detection and/or measurement of conjunctivochalasis in the human eyes;
an automated detection and/or measurement of eyelash crusting in the human eyes; and
an automated detection and/or measurement of eyelid redness in the human eyes;
wherein said automated machine-vision driven ocular disease (OD) recognition engine comprises a deep neural network operating to process a series of digital images of the human eye formed, captured and detected by said mobile image capturing system, so as to enable automated machine-assisted recognition of (i) detected ocular disease (OD) conditions in the human eye including dry eye disease (DED) conditions and the severity thereof, and also the automated treatment and management of the detected ocular disease (OD) conditions;
wherein said automated machine-vision driven ocular disease (OD) recognition engine includes deep-learning machine-vision models employing multi-layer convolutional networks trained using digital images of eyes captured using visible wavelengths of light illumination, to provide an end-to-end system for processing and recognizing input digital images captured by said mobile image capturing system and indicating particular ocular diseases (OD), including dry eye disease (DED);
wherein said automated machine-vision ocular disease (OD) recognition engine further processes patient symptom data, and other ocular/patient factors, that contribute to specific grading of the severity of the dry eye disease (DED) condition recognized by said automated machine-vision driven ocular disease (OD) recognition engine;
wherein machine-based logic packages are installed and operative within said automated machine-vision driven ocular disease (OD) recognition engine and include (i) diagnostic logic for determining ocular diseases (OD) based on disease diagnosis input factors provided to as input to said automated machine-vision driven ocular disease (OD) recognition engine, (ii) treatment logic for determining prescription of one or more of the ocular disease (OD) treatment and management recommendations for each ocular disease (OD) condition automatically determined by said automated machine-vision driven ocular disease (OD) recognition engine, and (iii) management logic for managing treatment of each recognized ocular disease condition recognized by said automated machine-vision driven ocular disease (OD) recognition engine; and
wherein said one or more clinical ophthalmic decision support subsystems are operably connected to the Internet infrastructure, for use by eye care professionals, including doctors and medical technicians, for the purpose of reviewing and managing patient records maintained by said cloud-based system network and providing patient decision support in connection with delivery of dry eye disease (DED) diagnosis, treatment and management care services to human patients situated remotely on the Earth.
2 . The cloud-based system network of claim 1 , wherein each said mobile image capturing system is a mobile smartphone camera system, and said automated machine-vision driven ocular disease (OD) recognition engine supports practice of a digital imaging processing method for automatically recognizing dry eye disease (DED) conditions in humans comprising the steps of:
(a) automatically processing 2D in vivo digital images of the human eyes formed, captured and detected using visible-wavelength sources of light illumination and said mobile image capturing system;
(b) using said automated machine-vision driven ocular disease (OD) recognition engine for automatically processing said digital images and automatically detecting presence and location of specific objects in ocular surfaces in the human eye; and
(c) confirming that specific conditions have been satisfied during said image processing to support automated dry eye disease (DED) condition recognition.
3 . The cloud-based system network of claim 1 , wherein said automated machine-vision driven ocular disease (OD) recognition engine supports displaying an augmented reality (AR) overlay GUI on the display screen of said mobile image capturing system for guiding capture of said digital images of the human eyes using visible wavelengths of light and said mobile image capturing system, and processing the digital images of the human eyes to detect and recognize dry eye disease (DED) conditions in the patient's eyes.
4 . The cloud-based system network of claim 1 , wherein said automated machine-vision driven ocular disease (OD) recognition engine comprises a deep neural network for building a deep-learning machine-vision recognition system capable of recognizing specific categories of ocular conditions (OD), including dry eye disease (DED), from a set of digital ocular images formed, captured and detected by said mobile image capturing system.
5 . The cloud-based system network of claim 4 , wherein said deep neural network operating to process a series of digital images of the human eye formed, captured and detected by said mobile image capturing system, so as to enable automated machine-assisted recognition of (i) dry eye disease (DED) conditions in the human eye and the severity thereof, and (ii) prescription of treatment of the dry eye disease (DED) conditions.
6 . The cloud-based system network of claim 4 , said deep neural network operating to process a series of digital images of the human eye formed, captured and detected by said mobile image capturing system, so as to enable automated machine-assisted recognition of (i) dry eye disease (DED) conditions in the human eye and the severity thereof, and (ii) prescribed treatment of said dry eye disease (DED) conditions, and (iii) assessment of said prescribed treatment.
7 . The cloud-based system network of claim 1 , wherein said automated machine-vision driven ocular disease (OD) recognition engine uses patient symptom data to automatically recognize a grade/severity of the dry eye disease (DED) condition(s) experienced by the patient.
8 . The cloud-based system network of claim 1 , wherein said mobile image capturing system is a mobile smartphone system configured for use by a patient to receive a dry eye disease (DED) diagnosis and course of prescribed treatment generated by said automated machine-vision driven ocular disease (OD) recognition engine.
9 . The cloud-based system network of claim 1 , wherein said automated machine-vision driven ocular disease (OD) recognition engine is configured for use in automatically monitoring and supporting decisions and prescription updates for effective treatment of the dry eye disease (DED) condition automatically recognized by said automated machine-vision driven ocular disease (OD) recognition engine.
10 . The cloud-based system network of claim 1 , wherein said automated machine-vision driven ocular disease (OD) recognition engine is configured for use in continuously monitoring the patient's DED condition and response to prescribed treatment, using said mobile image capturing system for forming and detecting digital images of patient's eyes that are processed by said automated machine-vision driven ocular disease recognition engine using visible-wavelength sources of light illumination.
11 . The cloud-based system network of claim 1 , wherein said automated machine-vision driven ocular disease (OD) recognition engine is configured so that compliance factors are continuously monitored, updated and supplied to the cloud-based system network so as to automatically measure prescribed treatment compliance and support decisions and prescription updates for effective treatment of a recognized dry eye disease (DED) condition.
12 . The cloud-based system network of claim 1 , wherein said automated detection and measurement of hyperemia (i.e. conjunctival injection) of the human eyes employs one or more methods selected from the group consisting of a method of detecting conjunctival injection, a method of detecting subconjunctival hemorrhage, a method of detecting episcleritis/scleritis, and a method of detecting superior limbic keratoconjunctivitis;
wherein said automated detection and measurement of tear film in the human eyes employs a method of detecting tear meniscus height;
wherein said automated detection and measurement of oily lid margin in the human eyes employs a method of detecting meibomian gland dysfunction;
wherein said automated detection and measurement of capped meibomian glands in the human eyes employs a method of detecting meibomian gland dysfunction;
wherein said automated detection and measurement of conjunctivochalasis in the human eyes employs a method of detecting conjunctivochalasis;
wherein said automated detection and measurement of eyelash crusting in the human eyes employs a method of detecting blepharitis; and
wherein said automated detection and measurement of eyelid redness in the human eyes employs a method of detecting chalazion/stye, a method of detecting eyelid cysts and a method of detecting preseptal cellulitis.
13 . The cloud-based system network of claim 1 , wherein said automated machine-vision driven ocular disease (OD) recognition engine employs convolutional neural networks (CNNs) configured and trained for use and deployment on said cloud-based system network, and designed to support and enable automated detection and measurement of dry eye disease (DED) condition in the human eyes photographically represented in digital images of human eyes formed, captured and detected by said mobile image capturing system.