CN-122004747-A - Method, system and storage medium for photographing anterior segment structure

CN122004747ACN 122004747 ACN122004747 ACN 122004747ACN-122004747-A

Abstract

The application provides a method, a system and a storage medium for photographing an anterior ocular segment structure, wherein the method comprises the steps of acquiring a B-Scan image acquired by scanning anterior ocular segment optical coherence tomography equipment in the process of moving a machine head, identifying whether the B-Scan image has an anterior ocular segment structure or not based on gray distribution characteristics of each preset column in the B-Scan image, extracting a region where a light beam potentially appears from the B-Scan image and calculating column projection of the region after the anterior ocular segment structure is identified, determining the background of the region based on multi-scale morphological operation and calculating contrast distribution of the column projection relative to the background, judging whether the B-Scan image is aligned with the anterior ocular segment according to peak significance of the contrast distribution, and triggering the scanning anterior ocular segment optical coherence tomography equipment to execute photographing operation when the alignment is judged. The application can efficiently and accurately complete shooting of the anterior segment structure of the eye.

Inventors

SUN WEIYE
XU SHUKUAN
WANG KEZHI

Assignees

执鼎医疗科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260306

Claims (10)

1. A method for anterior ocular segment structure photography, the method comprising: Acquiring a B-Scan image acquired by the optical coherence tomography equipment of the front section of the sweep frequency in the moving process of the machine head; Identifying whether the B-Scan image has a pre-ocular segment structure or not based on gray scale distribution characteristics of each preset column in the B-Scan image; after the anterior ocular segment structure is identified, extracting a region where a light beam potentially appears from the B-Scan image and calculating a column projection of the region; determining a background of the region based on a multi-scale morphological operation, and calculating a contrast distribution of the column projections relative to the background; judging whether the B-Scan image is aligned with the anterior ocular segment according to the peak significance of the contrast distribution; and triggering the sweep front section optical coherence tomography equipment to execute shooting operation when the sweep front section optical coherence tomography equipment is judged to be aligned.
2. The method of claim 1, wherein the identifying whether the B-Scan image has a pre-ocular structure based on gray scale distribution characteristics of each preset column in the B-Scan image comprises: Performing a column analysis on each preset column pixel data of the B-Scan image to screen out a target column having a strong reflection structure; Performing a peak analysis on each target column to screen out a dual peak column containing two peaks therefrom; judging whether the B-Scan image has a pre-eye anterior segment structure according to whether the distribution of the first peak position of all the double-peak columns in the image accords with a preset arch mode.
3. The method of claim 2, wherein performing a column analysis on each preset column of pixel data of the B-Scan image to screen out target columns for which strong reflection structures are present comprises: For each preset column of pixel data of the B-Scan image, acquiring a gray peak value of the preset column and a binary threshold value calculated by a maximum inter-class variance method; dividing the pixels of the column by using the binary threshold value to determine the average value of the pixels of the background area and the number of the pixels of the foreground area; Judging whether the difference value between the gray peak value and the pixel mean value of the background area is larger than a preset gray difference threshold value and the ratio of the number of pixels of the foreground area to the total number of pixels of the column is larger than a preset proportion threshold value, if so, judging that the preset column is a target column of a strong reflection structure; If not, judging that the preset column is not a target column of the strong reflection structure; Traversing all preset columns of the B-Scan image to screen out all target columns.
4. A method according to claim 3, wherein said performing a peak analysis on each target column to screen out a dual peak column containing two peaks therefrom comprises: truncating the pixel gray values of the corresponding target columns based on the bipartite threshold values to obtain initial peak distribution; processing the initial peak distribution to merge adjacent peaks and remove isolated noise peaks to obtain a processed peak distribution; And determining the number of connected areas based on the processed peak distribution, and screening the target column with the number of the connected areas of two as a double-peak column.
5. The method of claim 4, wherein determining whether the B-Scan image has a anterior ocular segment structure according to whether a distribution of first peak positions of all bi-peak columns in the image conforms to a preset arch pattern comprises: Acquiring the ordinate positions of the first wave crest of all the double wave crest lines in the B-Scan image, and taking the maximum value in all the ordinate positions as an arch vertex coordinate; Determining an ascending trend index based on the ordinate of each point on the left side of the arch vertex coordinate, and determining a descending trend index based on the ordinate of each point on the right side of the arch vertex coordinate; if the upward trend index and the downward trend index both meet the corresponding preset trend thresholds, the distribution of the first wave peaks of all the double wave peak columns in the image accords with a preset arch mode, and a front eye section structure appears in the B-Scan image; Otherwise, no anterior ocular segment structure appears in the B-Scan image.
6. The method of claim 1, wherein the extracting the region of potential occurrence of the light pillar from the B-Scan image and calculating a column projection of the region comprises: determining a lateral coordinate range in which the light beam is allowed to appear in the B-Scan image based on the position of the identified anterior ocular segment structure in the B-Scan image and a preset range; And intercepting an image subarea corresponding to the transverse coordinate range from the B-Scan image, and carrying out pixel gray level average processing on the image subarea along the longitudinal direction to obtain column projection of the area.
7. The method of claim 6, wherein the calculating a contrast distribution of the column projection relative to the background comprises: calculating the difference value between each point gray value in the column projection and the corresponding point gray value in the background based on the column projection and the background; dividing the difference value by the sum of the gray value of the corresponding point in the background and the unit offset to obtain the contrast distribution of the column projection relative to the background.
8. The method of claim 7, wherein determining whether the B-Scan image has been aligned with a anterior ocular segment based on the peak saliency of the contrast distribution comprises: Determining a contrast peak value based on the contrast distribution, and calculating a background mean value and a background standard deviation based on the background; Calculating the peak significance of the contrast distribution based on the contrast peak, the background mean and the background standard deviation; When the peak significance level is greater than a preset judging threshold value, the B-Scan image is aligned to the anterior ocular segment; when the peak significance level is not greater than a preset judgment threshold value, the B-Scan image is not aligned with the anterior ocular segment.
9. A system for photographing anterior segment structures of eyes is characterized by comprising an acquisition module, an identification module and a photographing module, wherein, The acquisition module is used for acquiring a B-Scan image acquired by the optical coherence tomography equipment of the front section of the sweep frequency in the moving process of the machine head; The identifying module is used for identifying whether the anterior ocular segment structure of the B-Scan image appears or not based on the gray distribution characteristics of each preset column in the B-Scan image; The shooting module is used for extracting a region where a light beam potentially appears from the B-Scan image and calculating column projection of the region after the structure of the front eye section is identified, determining the background of the region based on multi-scale morphological operation, calculating contrast distribution of the column projection relative to the background, judging whether the B-Scan image is aligned with the front eye section according to the peak significance degree of the contrast distribution, and triggering the scanning front section optical coherence tomography device to execute shooting operation when the B-Scan image is judged to be aligned.
10. A computer-readable storage medium, on which a computer program is stored which can be run on a processor, characterized in that the computer program, when being executed by the processor, implements a method for capturing a structure of the anterior segment of the eye according to any of claims 1 to 8.

Description

Method, system and storage medium for photographing anterior segment structure Technical Field The present application relates to the field of photographing technologies, and in particular, to a method, a system, and a storage medium for photographing an anterior segment structure of an eye. Background The sweep-frequency anterior segment optical coherence tomography is a key technology for high-resolution imaging of anterior segment structures of eyes such as cornea, anterior chamber, crystalline lens and the like in ophthalmic clinic, and the microstructure information of biological tissues is obtained through a coherent light interference principle, so that important basis is provided for diagnosis and curative effect evaluation of diseases such as cornea lesions, glaucoma, cataract and the like. In clinical examination, the shooting process of the front section of the scan usually depends on an operator to manually control the movement of the machine head of the equipment, and judges whether shooting conditions are met or not by observing an imaging interface in real time, and finally, shooting operation is triggered. In the existing shooting process, the judgment of whether the anterior ocular segment structure exists or not is completely dependent on the visual observation of an imaging interface by an operator. On the one hand, an operator needs to continuously observe the image change in the three-dimensional space moving process of the machine head, frequently adjusts the machine head position to find a proper shooting angle, so that single-case examination consumes longer time, especially in a scene of low patient coordination degree or batch screening, and if judgment is delayed, the best shooting time can be missed, and the discomfort of the patient and the examination time can be further increased. On the other hand, subjective judgment of whether the center of the cornea is centered or not is different from subjective judgment of the middle alignment standard, so that unstable imaging quality is easily caused, and the accuracy of subsequent diagnosis is affected. Disclosure of Invention In order to efficiently and accurately complete photographing of an anterior segment structure of an eye, embodiments of the present application provide a method, a system, and a storage medium for photographing of an anterior segment structure of an eye. In a first aspect, the present embodiment provides a method for capturing an anterior segment structure of an eye, the method comprising: Acquiring a B-Scan image acquired by the optical coherence tomography equipment of the front section of the sweep frequency in the moving process of the machine head; Identifying whether the B-Scan image has a pre-ocular segment structure or not based on gray scale distribution characteristics of each preset column in the B-Scan image; after the anterior ocular segment structure is identified, extracting a region where a light beam potentially appears from the B-Scan image and calculating a column projection of the region; determining a background of the region based on a multi-scale morphological operation, and calculating a contrast distribution of the column projections relative to the background; judging whether the B-Scan image is aligned with the anterior ocular segment according to the peak significance of the contrast distribution; and triggering the sweep front section optical coherence tomography equipment to execute shooting operation when the sweep front section optical coherence tomography equipment is judged to be aligned. In some of these embodiments, the identifying whether the B-Scan image has an anterior ocular segment structure based on gray scale distribution characteristics of each preset column in the B-Scan image comprises: Performing a column analysis on each preset column pixel data of the B-Scan image to screen out a target column having a strong reflection structure; Performing a peak analysis on each target column to screen out a dual peak column containing two peaks therefrom; judging whether the B-Scan image has a pre-eye anterior segment structure according to whether the distribution of the first peak position of all the double-peak columns in the image accords with a preset arch mode. In some of these embodiments, the performing a column analysis on each preset column of pixel data of the B-Scan image to screen out a target column in which a strong reflective structure is present comprises: For each preset column of pixel data of the B-Scan image, acquiring a gray peak value of the preset column and a binary threshold value calculated by a maximum inter-class variance method; dividing the pixels of the column by using the binary threshold value to determine the average value of the pixels of the background area and the number of the pixels of the foreground area; Judging whether the difference value between the gray peak value and the pixel mean value of the background area is larger than a p