CN-115690060-B - Abnormality determination method and device for eye ultrasonic image and related equipment

CN115690060BCN 115690060 BCN115690060 BCN 115690060BCN-115690060-B

Abstract

The application provides an abnormality determination method, device and related equipment for an eye ultrasonic image, which are used for determining whether the ultrasonic image meets the preset non-artifact image requirement or not by identifying artifact characteristics of the ultrasonic image which are acquired in advance and based on the artifact characteristics, respectively identifying a cornea region, an iris region, a lens region and a ciliary body region in the ultrasonic image if the ultrasonic image meets the preset non-artifact image requirement, respectively acquiring boundary definition characteristics of the cornea region, the iris region, the lens region and the ciliary body region, respectively acquiring position characteristics of boundaries of the iris region and the lens region, and determining whether the ultrasonic image is abnormal or not based on the boundary definition characteristics of the cornea region, the iris region, the lens region and the ciliary body region and the position characteristics of the boundaries of the iris region and the lens. The accuracy of abnormality determination of the eye ultrasound image is improved.

Inventors

ZHENG BIQING
HU SHAN

Assignees

武汉楚精灵医疗科技有限公司

Dates

Publication Date: 20260505
Application Date: 20221107

Claims (9)

1. A method of anomaly determination for an ocular ultrasound image, the method comprising: identifying artifact characteristics of an ultrasonic image acquired in advance, and judging whether the ultrasonic image meets preset non-artifact image requirements based on the artifact characteristics, wherein the ultrasonic image is an ultrasonic image for performing ultrasonic biological microscopic examination on a target part of a patient, and the target part is an eye part; If the ultrasonic image meets the preset non-artifact image requirement, respectively identifying a cornea region, an iris region, a crystalline lens region and a ciliary body region in the ultrasonic image; Respectively acquiring boundary definition characteristics of the cornea region, the iris region, the crystalline lens region and the ciliary body region; Acquiring position features of the boundary of the iris region and the boundary of the lens region; determining whether the ultrasound image is abnormal based on boundary sharpness characteristics of the cornea region, iris region, lens region, and ciliary body region, and position characteristics of a boundary of the iris region and a boundary of the lens; wherein the determining whether the ultrasound image is abnormal based on the boundary sharpness characteristics of the cornea region, the iris region, the lens region, and the ciliary body region, and the position characteristics of the boundary of the iris region and the boundary of the lens comprises: performing weighted fitting on boundary definition characteristics of the cornea region, the iris region, the lens region and the ciliary body region and position characteristics of the boundary of the iris region and the boundary of the lens to obtain an abnormality degree parameter of the ultrasonic image; and determining whether the ultrasonic image is abnormal or not based on the abnormality degree parameter and a preset abnormality degree threshold.
2. The abnormality determination method of an ocular ultrasound image according to claim 1, characterized in that acquiring a boundary-sharpness feature of the cornea region includes: cutting the cornea region by taking a first rectangular frame with a preset size as a boundary to obtain a first cornea image; performing binarization processing on the first cornea image to obtain a processed second cornea image; on the basis of the connected domains, respectively calculating the area of each connected domain in the second cornea image; screening each connected domain based on the area and a preset area threshold value to obtain a target connected domain, wherein the target connected domain comprises a front elastic layer connected domain and a rear elastic layer connected domain; respectively acquiring a first central line of the front elastic layer communicating region and a second central line of the rear elastic layer communicating region, and acquiring a plurality of target distances between the first central line and the second central line; and calculating the variance values of the target distances, and determining the definition characteristic of the cornea region based on the variance values and a preset variance threshold.
3. The abnormality determination method of an ocular ultrasound image according to claim 1, wherein the iris region includes two sub-iris regions, and acquiring a boundary sharpness feature of the iris region includes: Respectively acquiring the boundary perimeter of each sub-iris region in the two sub-iris regions; performing binarization processing on all pixel points on the boundary of each sub-iris region, and acquiring a pixel value set of all the processed pixel points; Screening all the processed pixel points based on a preset pixel value threshold value and the pixel value set to respectively obtain a target pixel point set on the boundary of the sub-iris region; Respectively comparing the pixel point number value of the target pixel point set on the boundary of each sub-iris region with the boundary perimeter of each sub-iris region to obtain the target boundary ratio of each sub-iris region; And determining clear boundary characteristics of the iris areas based on the target boundary ratio of each sub-iris area and a preset boundary ratio threshold.
4. The abnormality determination method of an ocular ultrasound image according to claim 1, characterized in that acquiring a boundary-sharpness feature of the lens region includes: Acquiring a boundary perimeter of the lens region; performing binarization processing on all pixel points on the boundary of the crystalline lens region, and acquiring a pixel value set of all the processed pixel points; Screening all the processed pixel points based on a preset pixel value threshold and the pixel value set to respectively obtain a target pixel point set on the boundary of the crystalline lens region; Comparing the target pixel point set on the boundary of the lens region with the boundary perimeter of the lens region respectively to obtain a target boundary ratio of the lens region; And determining the boundary definition characteristic of the lens region based on the target boundary ratio of the lens region and a preset boundary ratio threshold.
5. The abnormality determination method of an ocular ultrasound image according to claim 1, wherein acquiring a boundary-sharpness feature of the ciliary body region comprises: cutting the ciliary body area by taking a second rectangular frame with a preset size as a boundary to obtain a first ciliary body image; Acquiring a first boundary line of a ciliary body area in the first ciliary body image and a first centroid of the first boundary line; performing binarization processing on the first ciliary body image to obtain a processed second ciliary body image; Acquiring a second boundary line of a ciliary body area in the second ciliary body image and a second centroid of the second boundary line; A euclidean distance of the first centroid and the second centroid is calculated, and a boundary clarity feature of the ciliary body region is determined based on the euclidean distance.
6. The abnormality determination method of an ocular ultrasound image according to claim 1, wherein the iris region includes two sub-iris regions, the acquiring a position feature of a boundary of the iris region and a boundary of the lens region includes: respectively acquiring a first overlapping degree and a second overlapping degree of the boundaries of the two sub-iris areas and the boundaries of the crystalline lens area; and determining the position characteristics of the boundary of the iris region and the boundary of the lens region based on the first overlapping degree, the second overlapping degree and a preset overlapping degree threshold value.
7. An abnormality determination device for an eye ultrasound image, the device comprising: the first identification unit is used for identifying the artifact characteristics of an ultrasonic image acquired in advance, judging whether the ultrasonic image meets the preset non-artifact image requirement or not based on the artifact characteristics, wherein the ultrasonic image is an ultrasonic image for performing ultrasonic biological microscopic examination on a target part of a patient, and the target part is an eye part; The second identification unit is used for respectively identifying a cornea region, an iris region, a crystalline lens region and a ciliary body region in the ultrasonic image if the ultrasonic image meets the preset non-artifact image requirement; a first acquisition unit configured to acquire boundary definition characteristics of the cornea region, the iris region, the lens region, and the ciliary body region, respectively; A second acquisition unit configured to acquire a position feature of a boundary of the iris region and a boundary of the lens region; A first determination unit configured to determine whether the ultrasound image is abnormal based on boundary sharpness characteristics of the cornea region, iris region, lens region, and ciliary body region, and position characteristics of a boundary of the iris region and a boundary of the lens; The first determining unit is further configured to perform weighted fitting on boundary definition features of the cornea region, the iris region, the lens region and the ciliary body region and position features of a boundary of the iris region and a boundary of the lens to obtain an abnormality degree parameter of the ultrasonic image, and determine whether the ultrasonic image is abnormal based on the abnormality degree parameter and a preset abnormality degree threshold.
8. A computer device, the computer device comprising: One or more processors; memory, and One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the abnormality determination method of an ocular ultrasound image of any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that a computer program is stored thereon, which is loaded by a processor to perform the steps in the abnormality determination method of an ocular ultrasound image as claimed in any one of claims 1 to 6.

Description

Abnormality determination method and device for eye ultrasonic image and related equipment Technical Field The application relates to the technical field of auxiliary medical treatment, in particular to an abnormality determination method and device for an eye ultrasonic image and related equipment. Background An ultrasonic biological microscope (Ultrasound Biomicroscope, UBM), which is a basic examination instrument commonly used in ophthalmology, mainly examines the angle of the atria. The method is to use high-frequency ultrasonic imaging technology for examination, and for contact examination, surface anesthetic is used in the examination process, and then a probe is used for contacting the surface of the cornea, so that the position of the atrial angle is measured. But does not belong to invasive examination, does not cause obvious damage to eyes, and does not affect the eyesight and the physiological functions of eyes of patients. However, the present inventors have found that UBM images are prone to distortion, anomalies, which can have some impact on the diagnostic results of the ophthalmic doctor. Disclosure of Invention The application provides an abnormality determination method, an abnormality determination device and related equipment for an eye ultrasonic image, and aims to solve the technical problem of how to efficiently and accurately determine whether the eye ultrasonic image is abnormal or not. In one aspect, the present application provides a method for abnormality determination of an ocular ultrasound image, the method comprising: identifying artifact characteristics of an ultrasonic image acquired in advance, and judging whether the ultrasonic image meets preset non-artifact image requirements based on the artifact characteristics, wherein the ultrasonic image is an ultrasonic image for performing ultrasonic biological microscopic examination on a target part of a patient, and the target part is an eye part; If the ultrasonic image meets the preset non-artifact image requirement, respectively identifying a cornea region, an iris region, a crystalline lens region and a ciliary body region in the ultrasonic image; Respectively acquiring boundary definition characteristics of the cornea region, the iris region, the crystalline lens region and the ciliary body region; Acquiring position features of the boundary of the iris region and the boundary of the lens region; Determining whether the ultrasound image is abnormal based on boundary sharpness characteristics of the cornea region, iris region, lens region, and ciliary body region, and position characteristics of a boundary of the iris region and a boundary of the lens. In one possible implementation of the present application, obtaining the boundary-sharpness feature of the cornea region includes: cutting the cornea region by taking a first rectangular frame with a preset size as a boundary to obtain a first cornea image; performing binarization processing on the first cornea image to obtain a processed second cornea image; on the basis of the connected domains, respectively calculating the area of each connected domain in the second cornea image; screening each connected domain based on the area and a preset area threshold value to obtain a target connected domain, wherein the target connected domain comprises a front elastic layer connected domain and a rear elastic layer connected domain; respectively acquiring a first central line of the front elastic layer communicating region and a second central line of the rear elastic layer communicating region, and acquiring a plurality of target distances between the first central line and the second central line; and calculating the variance values of the target distances, and determining the definition characteristic of the cornea region based on the variance values and a preset variance threshold. In one possible implementation manner of the present application, the iris region includes two sub-iris regions, and the obtaining the boundary definition characteristic of the iris region includes: Respectively acquiring the boundary perimeter of each sub-iris region in the two sub-iris regions; performing binarization processing on all pixel points on the boundary of each sub-iris region, and acquiring a pixel value set of all the processed pixel points; Screening all the processed pixel points based on a preset pixel value threshold value and the pixel value set to respectively obtain a target pixel point set on the boundary of the sub-iris region; Respectively comparing the pixel point number value of the target pixel point set on the boundary of each sub-iris region with the boundary perimeter of each sub-iris region to obtain the target boundary ratio of each sub-iris region; And determining clear boundary characteristics of the iris areas based on the target boundary ratio of each sub-iris area and a preset boundary ratio threshold. In one possible implementation of the present application,