Doheny Eye Institute

University of Southern California

Carl Zeiss Meditec Gmbh

University of California

Srinivas R Sadda

Alexander C Walsh

Paul G Updike

Akihito Uji

Paul Francis Stetson

David E Huang

Ou Tan

Eran Halperin

Nadav Rakocz

Muneeswar Gupta

Jeffrey Chiang

Deep learning methods and systems for detecting biomarkers within optical coherence tomography volumes using such deep learning methods and systems are provided. Embodiments predict the presence or absence of clinically useful biomarkers in OCT images using deep neural networks. The lack of available training data for canonical deep learning approaches is overcome in embodiments by leveraging a large external dataset consisting of foveal scans using transfer learning. Embodiments represent the three-dimensional OCT volume by 'tiling' each slice into a single two dimensional image, and adding an additional component to encourage the network to consider local spatial structure. Methods and systems, according to embodiments are able to identify the presence or absence of AMD-related biomarkers on par with clinicians. Beyond identifying biomarkers, additional models could be trained, according to embodiments, to predict the progression of these biomarkers over time.

Retinal disease biomarker prediction by stacking slices of 3D OCT image to reshape into 2D image and applying trained feature extractor and CNN

In accordance with one aspect of the present invention, an optical coherence tomography-based ophthalmic testing center system includes an optical coherence tomography instrument comprising an eyepiece for receiving at least one eye of a user or subject; a light source that outputs light that is directed through the eyepiece into the user's or subject's eye, an interferometer configured to produce optical interference using light reflected from the user's/subject's eye, an optical detector disposed so as to detect said optical interference; and a processing unit coupled to the detector. The ophthalmic testing center system can be configured to perform a multitude of self-administered functional and/or structural ophthalmic tests and output the test data.

Optical coherence tomography-based ophthalmic testing methods, devices and systems

The disclosure herein provides methods, systems, and devices for improving optical coherence tomography machine outputs through multiple enface optical coherence tomography angiography averaging techniques. The embodiments disclosed herein can be utilized in ophthalmology for employing optical coherence tomography (OCT) for in vivo visualization of blood vessels and the flow of blood in an eye of a patient, which is also known generally as optical coherence tomography angiography (OCTA). The embodiments disclosed herein can use linear registration, affine registration and/or elastic registration to align a plurality of optical coherence tomography angiography images or videos at corresponding superficial vascular layers having well-defined features or landmarks, and to apply the same linear registration, affine registration and/or elastic registration settings and/or data to corresponding deeper tissue layers, such as the choriocapillaris, which generally do not have well-defined features or landmarks, in order to align a plurality of corresponding deeper tissue layers for the purpose of averaging the images or video to produce a clearer and more accurate image or video of the tissue structure at deeper tissue layers.

Systems, methods, and devices for optical coherence tomography multiple enface angiography averaging

In accordance with one aspect of the present invention, an optical coherence tomography instrument comprises an eyepiece for receiving at least one eye of a user is provided; a light source that outputs light that is directed through the eyepiece into the user's eye; an interferometer configured to produce optical interference using light reflected from the user's eye; an optical detector disposed so as to detect said optical interference; and electronics coupled to the detector. The electronics can be configured to perform a risk assessment analysis based on optical coherence tomography measurements obtained using the interferometer. An output device can be electrically coupled to the electronics, and may be configured to output the risk assessment to the user through the output device. The optical coherence tomography instrument can be self-administered, and the eyepiece can be a monocular system or a binocular system.

Optical coherence tomography device, method, and system

Systems and methods for classifying abnormalities within optical coherence tomography images of the eye are presented. One embodiment of the present invention is the classification of pigment epithelial detachments (PEDs) based on characteristics of their internal reflectivity, size and shape. The classification can be based on selected subsets of the data located within or surrounding the abnormalities. Training data can be used to generate the classification scheme and the classification can be weighted to highlight specific classes of particular clinical interest.

Systems and methods for automated classification of abnormalities in optical coherence tomography images of the eye

Growing community of inventors

Pasadena, CA, United States of America

Srinivas R Sadda