Back to results
US 12,635,877 B2Granted
Methods and systems for quantifying retinal vascular patterns and treatment of disease
AssigneeOhio State Innovation Foundation
InventorsYanhui Ma, Matthew Ohr, Cynthia Roberts
Life of this patent
Estimating the term…
Abstract
The present disclosure relates to methods and systems for quantifying the vascular pattern using ocular imaging that shows vessel profiles for the detection, prevention, and treatment of diabetic retinopathy and other related ocular vascular diseases.
Claims25
A method for treating an ocular vascular disease in a patient in need thereof, the method comprising:
- extracting a retinal vascular feature from a set of images acquired by an imaging modality and obtained from the patient to provide to a computer software;
- carrying out a multi-dimensional quantification comprising layer- and sector-based regions of interest via the computer software to generate a collection of vascular orientation pattern curves;
- analyzing the vascular orientation pattern curves via a trained AI model of the computer software;
- placing the patient into a group based on the analysis of the vascular orientation pattern curve of the patient, wherein said groups comprise:
- i. a normal range, ii. mildly abnormal range, and iii. a severely abnormal range;
- selecting the patient from group ii for further monitoring, and selecting the patient from group iii for treatment,
- wherein the trained AI model is configured to:
- receive a data set comprising ocular or retinal vascular image or object;
- determine a set of orientation vectors at each pixel, or a set of pixels, of a set of pre-defined regions of the ocular or retinal vascular image or object, including a first region and a second region;
- determine a set of first quantification metrics from the set of orientation vectors for the first region;
- and
- determine a set of second quantification metrics from the set of orientation vectors for the second region, wherein the first and second sets of quantification metrics are employed in a diagnosis or a treatment of an ocular or retinal vascular disease.
9 more elements
A system comprising:
- a processor;
- and
- a memory having instructions stored thereon, wherein execution of the instructions by the processor causes the processor to:
- receive a data set comprising ocular or retinal vascular image or object;
- determine a set of orientation vectors at each pixel, or a set of pixels, of a set of pre-defined regions of the ocular or retinal vascular image or object, including a first region and a second region;
- determine a set of first quantification metrics from the set of orientation vectors for the first region;
- and
- determine a set of second quantification metrics from the set of orientation vectors for the second region;
- wherein the first and second sets of quantification metrics are employed in a diagnosis or a treatment of an ocular or retinal vascular disease,
- wherein the system is employed in a set of steps comprising:
- extracting a retinal vascular feature from a set of images acquired by an imaging modality and obtained from the patient to provide to the system;
- carrying out a multi-dimensional quantification comprising layer- and sector-based regions of interest via the system to generate a collection of vascular orientation pattern curves;
- analyzing the vascular orientation pattern curves via a trained AI model of the system;
- placing the patient into a group based on the analysis of the vascular orientation pattern curve of the patient, wherein said groups comprise:
- i. a normal range, ii. mildly abnormal range, and iii. a severely abnormal range;
- and
- selecting the patient from group ii for further monitoring, and selecting the patient from group iii for treatment.
14 more elements
A non-transitory computer-readable medium having instructions stored thereon, wherein execution of the instructions by the processor causes the processor to:
- receive a data set comprising ocular or retinal vascular image or object;
- determine a set of orientation vectors at each pixel, or a set of pixels, of a set of pre-defined regions of the ocular or retinal vascular image or object, including a first region and a second region;
- determine a set of first quantification metrics from the set of orientation vectors for the first region;
- and
- determine a set of second quantification metrics from the set of orientation vectors for the second region;
- wherein the first and second sets of quantification metrics are employed in a diagnosis or a treatment of an ocular or retinal vascular disease
- wherein the computer-readable medium is employed to execute a set of steps when executed by the processor comprising:
- extracting a retinal vascular feature from a set of images acquired by an imaging modality and obtained from the patient to provide to the system;
- carrying out a multi-dimensional quantification comprising layer- and sector-based regions of interest via the system to generate a collection of vascular orientation pattern curves;
- analyzing the vascular orientation pattern curves via a trained AI model of the system;
- placing the patient into a group based on the analysis of the vascular orientation pattern curve of the patient, wherein said groups comprise:
- i. a normal range, ii. mildly abnormal range, and iii. a severely abnormal range;
- and
- selecting the patient from group ii for further monitoring, and selecting the patient from group iii for treatment.
11 more elements
Dependencies are inferred from the claim text.