According to the present application, provided is a computer-implemented method of predicting a clinical activity score for conjunctival hyperemia. The method described in the present application includes: training a conjunctival hyperemia prediction model using a training set; acquiring a first image include at least one eye of a subject and an outer region of an outline of the at least one eye; outputting, by the conjunctival hyperemia prediction model executing on a processor, a first predicted value for a conjunctival hyperemia, a first predicted value for the conjunctival edema, a first predicted value for an eyelid redness, a first predicted value for an eyelid edema, and a first predicted value for a lacrimal edema; and generating a score for the conjunctival hyperemia based on the selected first predicted value for a conjunctival hyperemia.

Systems and methods for vision test and uses thereof are disclosed. A method may be implemented on a mobile device having at least a processor, a camera and a display screen. The method may include capturing at least one image of a user using the camera of the mobile device; interactively guiding the user to a predetermined distance from the display screen of the mobile device based on the at least one image; presenting material on the display screen upon a determination that the user is at the predetermined distance from the display screen; and receiving input from the user in response to the material presented on the display screen. The material presented on the display screen may be for assessing at least one characteristic of the user's vision. Mobile devices and non-transitory machine-readable mediums having machine-executable instructions embodied thereon for assessing a user's vision also are disclosed.

The present disclosure provides methods and an apparatus for imaging and analysing images of presumed protein deposits in the retina, retinal tissue or retinal structures and discloses methods differentiating or classifying these deposits and other optical signals from retinal structures into 1) whether they contain or do not contain classes, of proteins or protein deposits called amyloids or other proteins and/or protein deposits related to neurodegenerative eye and brain disease(s); 2) which type(s) of amyloid or other proteins or protein deposits they contain, as well as 3) whether the form and/or properties of the deposit are associated with a class of diseases or with one or another specific condition(s) (or disease(s)); whether or not this is a disease or class of disease associated with the retina or more generally with the nervous system, including the brain or 4) classified as associated with one or another level of severity of condition(s), or disease(s).

A method of training a convolutional neural network model to predict diabetes from an image of a retina is provided. The method of training a convolutional neural network includes processing a first dataset, wherein processing the first dataset comprises: extracting a circular region from a retinal image, resizing the circular region, cropping the circular region, and placing the circular region onto a black background; training an initial model using a second dataset to yield a first model; training the first model using a third dataset to yield a second model; and training the second model using the first dataset to yield a third model.

Systems and methods are provided for machine learning based monitoring. Image data from a camera is received. On the hardware accelerator, a person detection model based on the image data is invoked. The person detection model outputs first classification result. Based on the first classification result, a person is detected. Second image data is received from the camera. In response to detecting the person, a fall detection model is invoked on the hardware accelerator based on the second image data. The fall detection model outputs a second classification result. A potential fall based on the second classification result is detected. An alert is provided in response to detecting the potential fall.

An example apparatus for adjusting eye gaze in images one or more processors to execute instructions to bidirectionally train a neural network; access a target angle and an input image, the input image including an eye in a first position; generate a vector field with the neural network; and generate a gaze-adjusted image based on the vector field, the gaze-adjusted image including the eye in a second position.

A processor of an ophthalmologic image processing device acquires an ophthalmologic image photographed by an ophthalmologic image photographing device. The processor inputs the ophthalmologic image into a mathematical model trained by a machine learning algorithm to acquire a result of an analysis relating to at least one of a specific disease and a specific structure of a subject eye. The processor acquires information of a distribution of weight relating to an analysis by a mathematical model, as supplemental distribution information, for which an image area of the ophthalmologic image input into the mathematical model is set as a variable. The processor sets a part of the image area of the ophthalmologic image, as an attention area, based on the supplemental distribution information. The processor acquires an image of a tissue including the attention area among a tissue of the subject eye and displays the image on a display unit.

Disclosed herein are methods and apparatus for making a determination about a cataract in an eye in ambient lighting conditions.

A method and system for selecting an intraocular lens, with a controller having a processor and tangible, non-transitory memory. A plurality of machine learning models is selectively executable by the controller. The controller is configured to receive at least one pre-operative image of the eye and extract, via a first input machine learning model, a first set of data. The controller is configured to receive multiple biometric parameters of the eye and extract, via a second input machine learning model, a second set of data. The first set of data and the second set of data are combined to produce a mixed set of data. The controller is configured to generate, via an output machine learning model, at least one output factor based on the mixed set of data. An intraocular lens is selected based in part on the at least one output factor.

An eyes measurement system, a method and a computer-readable medium are provided, including a client device with a measurement application and a cloud processing device, where the cloud processing device receives the subject's eye images uploaded by the measurement application. After pre-processing the eye images, the cloud processing device uses a prediction model to obtain the predicted eye measure of the subject such as an MRD1, an MRD2 and an LF, and presents the predicted eye measure of the MRD1, the MRD2 and the LF to the clinicians as a basis for diagnosis. Therefore, the eye images are taken without restricting to the places, and the prediction model is used to accurately obtain the subject's eye measure, thereby providing the clinicians with a clear basis for diagnosis.