CORNEAL EPITHELIUM SEGMENTATION IN OPTICAL COHERENCE TOMOGRAPHY IMAGES

    公开(公告)号:US20220148187A1

    公开(公告)日:2022-05-12

    申请号:US17453559

    申请日:2021-11-04

    Applicant: Alcon Inc.

    Abstract: The systems and methods described herein provide improved techniques for segmenting corneal epithelium layer. A method includes receiving an optical coherence tomography (OCT) image of an eye; generating, based on the OCT image, a binarized image of the eye; generating, based on the binarized image of the eye and the OCT image, a binary mask of a cornea of the eye; segmenting, based on the binary mask of the cornea of the eye, an anterior cornea of the eye on the OCT image; generating, based on the OCT image and the segmented anterior cornea of the eye, a binary mask for an epithelium layer of the eye; segmenting, based on the binary mask for the epithelium layer of the eye, a Bowman's layer in the cornea of the eye on the OCT image; and causing the segmented anterior cornea and the segmented Bowman's layer data to be used for generation of an epithelium map.

    DEEP LEARNING FOR OPTICAL COHERENCE TOMOGRAPHY SEGMENTATION

    公开(公告)号:US20210192732A1

    公开(公告)日:2021-06-24

    申请号:US17127651

    申请日:2020-12-18

    Applicant: Alcon Inc.

    Abstract: Systems and methods are presented for providing a machine learning model for segmenting an optical coherence tomography (OCT) image. A first OCT image is obtained, and then labeled with identified boundaries associated with different tissues in the first OCT image using a graph search algorithm. Portions of the labeled first OCT image are extracted to generate a first plurality of image tiles. A second plurality of image tiles is generated by manipulating at least one image tile from the first plurality of image tiles, such as by rotating and/or flipping the at least one image tile. The machine learning model is trained using the first plurality of image tiles and the second plurality of image tiles. The trained machine learning model is used to perform segmentation in a second OCT image.

    POLARIZATION SENSITIVE OPTICAL COHERENCE TOMOGRAPHY FOR VISUALIZATION OF VITREOUS OPACITIES

    公开(公告)号:US20250090015A1

    公开(公告)日:2025-03-20

    申请号:US18828214

    申请日:2024-09-09

    Applicant: Alcon Inc.

    Abstract: A system of visualizing a target site in an eye, using a polarization sensitive optical coherence tomography (PS-OCT) device, includes a controller having at least one processor and at least one non-transitory, tangible memory on which instructions are recorded. The target site is one or more vitreous opacities in the vitreous humor of the eye. The controller is configured to receive PS-OCT data and determine at least one parameter corresponding to birefringence properties of collagen fibrils in the vitreous humor based on the PS-OCT data. The at least one parameter includes respective spacing of the collagen fibrils. The controller is configured to determine a respective location of the one or more vitreous opacities when the at least one parameter is outside a predefined range and generate a control signal adapted for guiding a treatment beam at the respective location of the one or more vitreous opacities.

    AUTOMATIC SEGMENTATION OF ANTERIOR SEGMENT OF AN EYE IN OPTICAL COHERENCE TOMOGRAPHY IMAGES

    公开(公告)号:US20220148186A1

    公开(公告)日:2022-05-12

    申请号:US17453550

    申请日:2021-11-04

    Applicant: Alcon Inc.

    Abstract: Provided herein are techniques for automatically segmenting anterior segment of an eye in an optical coherence tomography (OCT) image. A method includes receiving an OCT image of an eye; cropping, based on one or more structures of the eye in the OCT image, the OCT image of the eye into one or more sub-images corresponding to the one or more structures; for each of the one or more sub-images of the OCT image of the eye: generating a background seed and a foreground seed of the sub-image; generating, based on the background and the foreground seeds, and an image segmentation model, a mask for a structure of the one or more structures of the eye included in the sub-image; generating, based on the mask for the structure, one or more contours of the structure included in the sub-image; and displaying the one or more contours on the sub-image.

    CALIBRATION OF IMAGING SYSTEM WITH COMBINED OPTICAL COHERENCE TOMOGRAPHY AND VISUALIZATION MODULE

    公开(公告)号:US20240027180A1

    公开(公告)日:2024-01-25

    申请号:US18354096

    申请日:2023-07-18

    Applicant: Alcon Inc.

    CPC classification number: G01B9/02072 A61B3/102 A61B90/20 G01B9/02091

    Abstract: An imaging system includes a housing assembly having a head unit configured to be at least partially directed towards a target site. An optical coherence tomography (OCT) module and a visualization module are located in the housing assembly and configured to respectively obtain OCT data and visualization data of the target site. The system includes a controller configured to generate a scanning pattern for a region of calibration selected in a calibration target. OCT data of the region of calibration is synchronously acquired. The controller is configured to obtain a projected two-dimensional OCT image of the region of calibration based on the OCT data, as an inverse mean-intensity projection. The controller is configured to register the projected two-dimensional OCT image to a corresponding view extracted from the visualization data, via a cascaded image registration process having a coarse registration stage and a fine registration stage.

    Deep learning for optical coherence tomography segmentation

    公开(公告)号:US11562484B2

    公开(公告)日:2023-01-24

    申请号:US17127651

    申请日:2020-12-18

    Applicant: Alcon Inc.

    Abstract: Systems and methods are presented for providing a machine learning model for segmenting an optical coherence tomography (OCT) image. A first OCT image is obtained, and then labeled with identified boundaries associated with different tissues in the first OCT image using a graph search algorithm. Portions of the labeled first OCT image are extracted to generate a first plurality of image tiles. A second plurality of image tiles is generated by manipulating at least one image tile from the first plurality of image tiles, such as by rotating and/or flipping the at least one image tile. The machine learning model is trained using the first plurality of image tiles and the second plurality of image tiles. The trained machine learning model is used to perform segmentation in a second OCT image.

    Corneal epithelium segmentation in optical coherence tomography images

    公开(公告)号:US12211211B2

    公开(公告)日:2025-01-28

    申请号:US17453559

    申请日:2021-11-04

    Applicant: Alcon Inc.

    Abstract: The techniques described herein provide improved techniques for segmenting corneal epithelium layer. A method includes receiving an optical coherence tomography (OCT) image of an eye; generating, based on the OCT image, a binarized image of the eye; generating, based on the binarized image of the eye and the OCT image, a binary mask of a cornea of the eye; segmenting, based on the binary mask of the cornea of the eye, an anterior cornea of the eye on the OCT image; generating, based on the OCT image and the segmented anterior cornea, a binary mask for an epithelium layer of the eye; segmenting, based on the binary mask for the epithelium layer of the eye, a Bowman's layer in the cornea of the eye on the OCT image; and causing the segmented anterior cornea and the segmented Bowman's layer data to be used for generation of an epithelium map.

    DEEP LEARNING FOR OPTICAL COHERENCE TOMOGRAPHY SEGMENTATION

    公开(公告)号:US20230124674A1

    公开(公告)日:2023-04-20

    申请号:US18068978

    申请日:2022-12-20

    Applicant: Alcon Inc.

    Abstract: Systems and methods are presented for providing a machine learning model for segmenting an optical coherence tomography (OCT) image. A first OCT image is obtained, and then labeled with identified boundaries associated with different tissues in the first OCT image using a graph search algorithm. Portions of the labeled first OCT image are extracted to generate a first plurality of image tiles. A second plurality of image tiles is generated by manipulating at least one image tile from the first plurality of image tiles, such as by rotating and/or flipping the at least one image tile. The machine learning model is trained using the first plurality of image tiles and the second plurality of image tiles. The trained machine learning model is used to perform segmentation in a second OCT image.

Patent Agency Ranking