CN117372419A - Method for evaluating proliferation condition of cells on inner limiting membrane based on retina OCT (optical coherence tomography) image - Google Patents
Method for evaluating proliferation condition of cells on inner limiting membrane based on retina OCT (optical coherence tomography) image Download PDFInfo
- Publication number
- CN117372419A CN117372419A CN202311547320.4A CN202311547320A CN117372419A CN 117372419 A CN117372419 A CN 117372419A CN 202311547320 A CN202311547320 A CN 202311547320A CN 117372419 A CN117372419 A CN 117372419A
- Authority
- CN
- China
- Prior art keywords
- limiting membrane
- inner limiting
- gamma
- proliferation
- retina
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000012528 membrane Substances 0.000 title claims abstract description 74
- 238000000034 method Methods 0.000 title claims abstract description 32
- 210000001525 retina Anatomy 0.000 title claims description 47
- 230000035755 proliferation Effects 0.000 title claims description 31
- 238000012014 optical coherence tomography Methods 0.000 title description 36
- 230000003902 lesion Effects 0.000 claims abstract description 44
- 238000009826 distribution Methods 0.000 claims abstract description 41
- 230000004663 cell proliferation Effects 0.000 claims abstract description 23
- 238000011160 research Methods 0.000 claims abstract description 16
- 230000002207 retinal effect Effects 0.000 claims abstract description 9
- 201000010099 disease Diseases 0.000 claims description 23
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 23
- 206010025421 Macule Diseases 0.000 claims description 10
- 238000003745 diagnosis Methods 0.000 claims description 10
- 238000005315 distribution function Methods 0.000 claims description 9
- 230000035945 sensitivity Effects 0.000 claims description 5
- 230000004438 eyesight Effects 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 abstract description 8
- 238000003384 imaging method Methods 0.000 abstract description 4
- 238000012545 processing Methods 0.000 abstract description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 239000003814 drug Substances 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 10
- 238000002310 reflectometry Methods 0.000 description 9
- 210000004027 cell Anatomy 0.000 description 8
- 230000002159 abnormal effect Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 6
- 238000007619 statistical method Methods 0.000 description 6
- 238000011156 evaluation Methods 0.000 description 5
- 208000017442 Retinal disease Diseases 0.000 description 4
- 230000008859 change Effects 0.000 description 4
- 201000007527 Retinal artery occlusion Diseases 0.000 description 3
- 238000003759 clinical diagnosis Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 206010020718 hyperplasia Diseases 0.000 description 3
- 238000011835 investigation Methods 0.000 description 3
- 208000001344 Macular Edema Diseases 0.000 description 2
- 206010025415 Macular oedema Diseases 0.000 description 2
- 208000002367 Retinal Perforations Diseases 0.000 description 2
- 206010064930 age-related macular degeneration Diseases 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 210000003161 choroid Anatomy 0.000 description 2
- 238000013211 curve analysis Methods 0.000 description 2
- 230000000994 depressogenic effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 210000002189 macula lutea Anatomy 0.000 description 2
- 208000029233 macular holes Diseases 0.000 description 2
- 201000010230 macular retinal edema Diseases 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 210000004126 nerve fiber Anatomy 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002062 proliferating effect Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 206010012689 Diabetic retinopathy Diseases 0.000 description 1
- 208000003098 Ganglion Cysts Diseases 0.000 description 1
- 206010021143 Hypoxia Diseases 0.000 description 1
- 206010061218 Inflammation Diseases 0.000 description 1
- 208000005400 Synovial Cyst Diseases 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000003776 cleavage reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007954 hypoxia Effects 0.000 description 1
- 238000011503 in vivo imaging Methods 0.000 description 1
- 230000004054 inflammatory process Effects 0.000 description 1
- 208000028867 ischemia Diseases 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 208000038015 macular disease Diseases 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000004660 morphological change Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000007017 scission Effects 0.000 description 1
- 230000002792 vascular Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000004382 visual function Effects 0.000 description 1
- 210000004127 vitreous body Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10101—Optical tomography; Optical coherence tomography [OCT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20076—Probabilistic image processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30024—Cell structures in vitro; Tissue sections in vitro
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30041—Eye; Retina; Ophthalmic
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Eye Examination Apparatus (AREA)
Abstract
The invention discloses a retinal OCT image-based method for evaluating cell proliferation conditions on an inner limiting membrane, which is applied to the technical field of medical image processing. The method comprises the following steps: collecting OCT images and recording patient information, extracting boundary pixel values and coordinate values of an inner limiting membrane, dividing a research target area range, counting boundary pixel gray values of the inner limiting membrane in the research target area range, drawing a gray histogram, fitting the histogram with normal distribution and Gamma distribution respectively, calculating the difference between peak values of the two distributions to be used for measuring whether cell proliferation lesions exist on the inner limiting membrane, dividing a threshold value according to the trend of a Gamma distribution density function curve, marking lesion positions, and calculating the percentage of cell proliferation area. The invention can be applied to OCT images with various resolutions and contrasts, and can obtain accurate boundary segmentation results of the inner limiting membrane, thereby reducing high requirements of clinical medicine on equipment performance and being beneficial to reducing medical cost.
Description
Technical Field
The invention relates to the technical field of medical image processing, in particular to a method for evaluating cell proliferation conditions on an inner limiting membrane based on retina OCT images.
Background
Optical Coherence Tomography (OCT) has been widely used as a detection means for ophthalmic clinical diagnosis of retinal diseases due to its non-invasive, non-contact and applicability to in vivo imaging. The retina B-scan image obtained by OCT can provide thickness information and morphological characteristics of each layer of retina for clinicians, and has important significance for detecting and evaluating diseases of retina morphological change (such as retina detachment, macular hole, macular edema, and the like).
The inner limiting membrane, which is the innermost layer of the retina, is a layer of unstructured homogeneous membrane. The inner surface is smooth, contacts with vitreous cortex, and the outer surface is wave-shaped and is adjacent to nerve fiber layer and ganglion cell layer. The stimulus of factors such as retinal inflammation, external stimulus, hypoxia, ischemia and the like can lead to the tight adhesion of an inner limiting membrane with vitreous cortex and proliferation tissues, so that tangential traction force of a macular region is increased, and then a series of retinal diseases such as macular holes, macular edema, retinal cleavage and the like are caused, and the retinal diseases are clinically manifested as obvious visual pain, blurring, distortion and deformation, and seriously affect vision and visual functions. Cell proliferation on the inner limiting membrane is a main factor causing the tangential traction force of the macular disease to be increased, and the situation can be improved clinically by adopting an inner limiting membrane stripping technology, so that the early recovery of the macular area structure is promoted. However, the evaluation and diagnosis of the peeling of the internal limiting membrane, the peeling mode, the area size and the like before operation are easy to be controversial.
In all clinical imaging, the most intuitive means to observe the inner limiting membrane is the OCT B-scan image, which can provide the doctor with comprehensive information such as lesion location, lesion area, and lesion extent. However, the highest longitudinal resolution of the currently known OCT systems is insufficient to distinguish its outer boundaries such that an accurate measurement of the thickness of the inner limiting membrane and its proliferating tissue is not possible. The subjective perception of a clinician is only relied on, and the condition degree of cell proliferation on the inner limiting membrane cannot be accurately judged and the lesion position is positioned due to the fact that the human eye is insensitive to the small brightness change of the image, so that the condition of missed detection of early lesions possibly occurs, the misdiagnosis rate is increased, and the treatment of the condition is delayed, and therefore, the quantitative index is provided to have important clinical diagnostic significance for the evaluation and accurate and intelligent lesion positioning method of the condition of cell proliferation on the inner limiting membrane.
In the prior art, US patent 8868155 proposes a quantitative measurement method for OCT imaging and evaluation of retinal tissue to provide prognosis and diagnosis details of related pathological tissues, layering and measuring thickness of tissues of various cell layers of a subject's retina, extracting optical characteristics so as to identify local abnormalities in the retinal structure, only showing significant differences in the mean value statistics of the optical scattering rates of the cell layers, and does not propose a demarcation value clearly distinguishing the presence or absence of diabetic retinopathy and a quantitative index having guiding significance in clinical application, and the evaluation method is susceptible to erroneous results generated by the influence of image quality. US9737205 proposes a method of analyzing retinal OCT image data, constructing a classifier that predicts AMD progression over a given period of time, achieving a risk prediction of AMD progression from dry to wet for a particular patient over a short period of time, but the accuracy of the classifier is only 0.74, and because of variability in individual patients, a large amount of long-term data of individual cases need to be collected to maintain model stability and accuracy, limiting its clinical application. Chinese patent CN113066067 proposes a quantitative detection method based on an OCT image of an ophthalmic retina, in which pixel gray values represent tissue reflectivity, and pixel gray values of longitudinal scan lines in a target area of investigation are analyzed line by line, but positioning of the target area of investigation depends on manual identification, so that automatic identification of a lesion position cannot be realized, and no clear distinction between a demarcation value of a lesion and an evaluation classification index with reference is proposed. Chinese patent CN 103886592 proposes a method for gray level analysis between retina layers based on 3D-OCT, which uses a graph search technique to segment the multi-layer structure of retina, then uses a texture classification method to detect the RAO (retinal artery occlusion) region, and finally performs gray level analysis of retina layers. However, only statistical analysis is performed on the gray mean value and standard deviation of each layer, no referent index and no grading data are given to judge the severity of RAO, and automatic positioning of the focus cannot be realized. An article entitled "Choroidal vascularity index as a measure ofvascular status ofthe choroid: measurements in healthy eyes from apopulation-based study" was designed to formulate the index choroidal blood vessel index (CVI) as a ratio of Luminal Area (LA) to total depressed choroidal area (TCA) for assessing the vascular condition of the depressed central choroid. However, the positioning of the target area of research depends on manual calibration, and images with different sizes have no universality, and abnormal parts of lesions cannot be marked automatically.
In summary, the quantitative statistical analysis of the reflectivity of the local tissue of the retina by combining the image processing method can provide data support for the diagnosis of retinal diseases, but the following problems limit the application of the method in clinical diagnosis: 1. the image sizes, resolutions, and contrast obtained by different OCT systems are widely different, and the quantitative index calculated for the different system images cannot be stabilized within a comparable range. 2. The B-scan image quality is easily affected by signal intensity, so that the image quality is different, and the difference of the average value of the reflectivities of images with different signal intensities is obvious, so that an error conclusion is easy to generate by only carrying out statistical analysis on the reflectivity of tissues. 3. Statistical analysis of tissue reflectivity of patients indicates that the course of disease extent depends on comparison with healthy controls, but independent demarcation values cannot be obtained, clearly distinguishing patients from healthy controls. 4. The identification of the research target area where the focus is located depends on manual identification, and the phenomenon of early lesion missed diagnosis caused by insensitivity of human eyes to small brightness change cannot be avoided basically. Therefore, how to provide a method for evaluating the proliferation status of cells on the inner limiting membrane based on the OCT image of retina is a problem that needs to be solved by those skilled in the art.
Disclosure of Invention
In view of this, the present invention provides a method for evaluating cell proliferation conditions on the inner limiting membrane based on the OCT image of retina, which locates lesions and measures the disease degree by analyzing the pixel gray values of the inner limiting membrane layer in the OCT image of retina, and locates the lesion position and the estimated area size by the divided abnormal data, and provides a referenceable quantitative index for the clinician for enhancing the accuracy of diagnosis.
In order to achieve the above object, the present invention provides the following technical solutions:
a method for evaluating cell proliferation conditions on an inner limiting membrane based on retina OCT images, comprising the following steps:
s1, acquiring OCT tomographic images, recording a plurality of patient fundus data and information, and acquiring pixel gray values of tissues of each layer on retina;
s2, dividing the retina layer structure by using a graph theory and dynamic programming-based method, and extracting boundary pixel values and coordinate values of an inner limiting membrane;
s3, determining the position coordinates of the macula fovea according to the extracted inner limiting membrane coordinates, and dividing a research target area by taking the macula fovea with highest vision sensitivity as the center;
s4, counting the gray value of the boundary pixel of the inner limiting membrane in the range of the research target area, and drawing a gray histogram;
s5, fitting the histogram by using a normal distribution density function and a Gamma distribution density function respectively, and calculating the difference between the peak values of the two distributions and the corresponding gray values;
s6, dividing a threshold according to the trend of the Gamma distribution density function curve, and distinguishing the proliferation and non-proliferation areas on the inner limiting membrane;
s7, marking a lesion cell proliferation area in the B-scan image;
s8, calculating the ratio of the lesion area to the research target area, and measuring the cell proliferation lesion area and the disease degree on the inner limiting membrane.
Optionally, the information collected in S1 includes name, age, sex, date of visit, diagnosis of illness, presence or absence of cell proliferation on the inner limiting membrane.
Optionally, in S3, performing high-order fitting on boundary coordinate values of the inner limiting membrane, searching pits of the curve by using the second derivative, and taking a point with the shortest euclidean distance from the geometric center of the image in all pits as a central point of the macula lutea fovea.
Optionally, the Gamma distribution density function in S5 is:
wherein x represents the distribution range of pixel gray values, k is the shape parameter of the Gamma distribution function, and θ is the scale parameter of the Gamma distribution function;
the normal distribution function is:
where x represents the distribution range of pixel gray values, σ is the standard deviation of normal distribution, and μ is the average value of normal distribution.
Optionally, the peak value 2Max is fit with Gamma in S6 gamma Minimum value Min of gray histogram as expected μ of normal distribution ref Peak 2Max for Gamma fit gamma The symmetrical gray value of (2) is determined as a threshold value for distinguishing a lesion region of the highlight reflection region, and pixels above the threshold value of the lesion region are classified as lesion cell proliferation regions.
Optionally, the threshold value of the lesion area is 2Max gamma -Min ref
Compared with the prior art, the invention provides the method for evaluating the proliferation condition of cells on the inner limiting membrane based on the OCT image of retina, which has the following beneficial effects:
1. the segmentation algorithm adopted by the invention can be applied to OCTBscan images with various resolutions and contrasts, and can obtain accurate boundary segmentation results of the inner limiting membrane; the characteristic information is extracted through the segmentation boundary, and the change of the reflectivity intensity of the inner limiting membrane is further analyzed, so that the analysis of the cell proliferation condition on the inner limiting membrane is not limited to thickness data which cannot be provided by an OCT system due to insufficient resolution, the high requirement of clinical medicine on the equipment performance is reduced, and the medical cost is reduced;
2. the characteristic data self-adaptive threshold value is defined based on the single B-scan image, the method for classifying abnormal data has stability, is not influenced by the imaging quality of an OCT system, the proposed reference index is essentially a difference value, the influence of the amplitude of the data is eliminated, the statistical result is not influenced by the integral brightness of the OCT image any more, and the numerical value can be stably compared in the (3, 10) interval;
3. the proposed index is independent data obtained by analyzing a single B-scan image, obvious differences exist between a patient and a healthy control group through statistical analysis, obvious lesions of an inner limiting membrane in time can be obtained through ROC curve analysis, and the diagnosis accuracy of the index can reach 86.3%; make up the human eye and read the defect that the film is insensitive to the intensity change of reflectivity, assist the clinician to diagnose;
4. according to the method, the macula fovea can be automatically positioned according to the extracted boundary coordinates of the inner limiting membrane, abnormal data are further divided in the calculation process, the lesion positions in the image are indexed according to the abnormal data, the lesion area index is further obtained, the automatic division of the research target area and the automatic identification of the lesion positions are realized, and therefore the difficulty and the workload of a clinician in reading a film are reduced, and the shortage of medical manpower and material resources is saved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart showing a method for evaluating the proliferation status of cells on an inner limiting membrane according to the present invention;
FIG. 2 is a schematic view of a B-scan image of a retina OCT in an embodiment of the present invention; wherein 2a is a healthy control group OCT B-scan image, and 2B is a disease group OCT B-scan image;
FIG. 3 is a schematic view of a retina layering partition in accordance with an embodiment of the present invention; wherein 3a is a retina layering and dividing schematic diagram of a healthy control group, and 3b is a retina layering and dividing schematic diagram of a disease group;
FIG. 4 is a gray level histogram of the inner limiting membrane of the retina in an embodiment of the invention; wherein 4a is a healthy control group retina inner limiting membrane gray level histogram, and 4b is a disease group retina inner limiting membrane gray level histogram;
FIG. 5 is a map of the location of retinal intimal hyperplasia in an embodiment of the present invention; wherein 5a is a map of the proliferation position of the retina inner limiting membrane of the healthy control group, and 5b is a map of the proliferation position of the retina inner limiting membrane of the disease group;
fig. 6 is a graph of the accuracy ROC in an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The embodiment of the invention discloses a method for evaluating the proliferation condition of cells on an inner limiting membrane based on a retina OCT image, which is shown in figure 1 and comprises the following steps:
s1, acquiring OCT tomographic images, recording a plurality of patient fundus data and information, and acquiring pixel gray values of tissues of each layer on retina;
s2, dividing the retina layer structure by using a graph theory and dynamic programming-based method, and extracting boundary pixel values and coordinate values of an inner limiting membrane;
s3, determining the position coordinates of the macula fovea according to the extracted inner limiting membrane coordinates, and dividing a research target area by taking the macula fovea with highest vision sensitivity as the center;
s4, counting the gray value of the boundary pixel of the inner limiting membrane in the range of the research target area, and drawing a gray histogram;
s5, fitting the histogram by using a normal distribution density function and a Gamma distribution density function respectively, and calculating the difference between the peak values of the two distributions and the corresponding gray values;
s6, dividing a threshold according to the trend of the Gamma distribution density function curve, and distinguishing the proliferation and non-proliferation areas on the inner limiting membrane;
s7, marking a lesion cell proliferation area in the B-scan image;
s8, calculating the ratio of the lesion area to the research target area, and measuring the cell proliferation lesion area and the disease degree on the inner limiting membrane.
Further, the information collected in S1 includes name, age, sex, date of visit, diagnosis of illness, presence or absence of cell proliferation on the inner limiting membrane.
And further, in S3, performing high-order fitting on boundary coordinate values of the inner limiting membrane, searching pits of the curve by utilizing a second derivative, and taking a point with the shortest Euclidean distance with the geometric center of the image in all pits as a central point of the macula lutea fovea.
Further, the Gamma distribution density function in S5 is:
wherein x represents the distribution range of pixel gray values, k is the shape parameter of the Gamma distribution function, and θ is the scale parameter of the Gamma distribution function;
the normal distribution function is:
where x represents the distribution range of pixel gray values, σ is the standard deviation of normal distribution, and μ is the average value of normal distribution.
Further, in S6, gamma is used for fitting peak value 2Max gamma Minimum value Min of gray histogram as expected μ of normal distribution ref Peak 2Max for Gamma fit gamma The symmetrical gray value of (2) is determined as a threshold value for distinguishing a lesion region of the highlight reflection region, and pixels above the threshold value of the lesion region are classified as lesion cell proliferation regions.
Further, the threshold value of the lesion area is 2Max gamma -Min ref 。
Further, in one embodiment of the present invention, image acquisition is performed using the apparatus Zeiss CIRRUS HD-OCT 500 with a longitudinal resolution of 5 μm and a lateral resolution of 15 μm, using a high definition 21 line scan mode, with a scan area covering a 6mm by 6mm fundus range. The size of the acquired image is 938 multiplied by 625pix, the OCT tomographic image is shown in figure 2, wherein 2a is a healthy control group OCT B-scan image, 2B is a disease group OCT B-scan image, the inner surface of an inner limiting membrane of the healthy control group is smooth and flat, the inner limiting membrane is not adhered to a vitreous body, an image of a cell proliferation lesion exists on the inner limiting membrane of the disease group, and the marked cell proliferation part shows a pixel point with a high gray value on the image due to the increase of reflectivity; in analyzing the images, only one frame of image taken in one visit of a single case is selected in order to ensure the independence among individual cases, and one of 10, 11, 12 frames of the 21-line scanned images, in which the fovea is obvious, is selected for data processing and analysis in order to locate the positional relationship between the lesion and the fovea.
The inner limiting membrane boundary coordinate and gray value data are extracted by using a retina segmentation algorithm based on graph theory and dynamic programming, wherein the retina layering segmentation result is shown in fig. 3, 3a is a healthy control group retina layering segmentation result, 3b is a disease group retina layering segmentation result, and the marked boundary is a pixel point on each column of the image. Selecting the pixel points at the uppermost layer of each column, performing curve fitting on coordinate values of the pixel points for 20 times by using a least square method, performing second derivative and first derivative on the fitted curve to find pits, and observing that more than one pit exists in the boundary curve of the inner limiting membrane, wherein the center of the macula is positioned at the center position of the image when shooting, so that the point with the shortest Euclidean distance with the geometric center of the image in all the found pits is defined as the center point of the macula, namely the solid line labeling position in fig. 3, and specific coordinate values of the coordinate points are given above the image. The region is divided into a study target region, i.e., a region between the broken lines, by expanding 130 pixel points each to the left and right with the determined coordinates as the center. The selection of the region of interest excludes the high gray value effect of the nerve fiber layer and the selected region is the region of highest visual sensitivity on the retina.
As shown in fig. 4, the gray value histogram of the boundary pixels of the inner limiting membrane in the statistical research target area, wherein 4a is the gray value histogram of the inner limiting membrane of the retina of the healthy control group, and 4b is the gray value histogram of the inner limiting membrane of the retina of the disease group. The gray level histogram of the non-proliferation inner limiting membrane boundary approximates to normal distribution, and the disease group shows high gray level value due to the increase of the reflectivity of the lesion area, so the histogram distribution is more approximate to Gamma distribution with k > 1.
The gray level histogram is fitted by the normal distribution density function and the Gamma distribution density function respectively, so that the difference between the two fitting curves of the healthy control group is small, and the fitting curves of the disease group have obvious difference. This difference is measured by the difference between the corresponding gray values of the two curve peaks diff To reflect it. The analysis considers that the gray level histogram of the disease group consists of a non-proliferation region and a proliferation region, so that the gray level distribution of the disease group should contain high gray level value pixels of normal distribution and proliferation tissue, and the high brightness pixel points representing the proliferation tissue can be separated by determining a threshold value.
Determining a threshold value as the minimum value Min of the gray level histogram through Gamma fitting result ref Regarding the fit peak Max gamma Symmetric gray value 2Max gamma -Min ref Pixel points above the threshold are classified as lesion proliferation regions marked in OCT tomographic images, and the marking result is shown in FIG. 5, wherein 5a is the proliferation position of inner limiting membrane of retina in healthy control group, 5bIs the position of inner limiting membrane hyperplasia of retina of disease group. Calculating the ratio of abnormal data quantity to total data quantity to obtain proliferation area percentage area per The area of the proliferation region in the investigation target region is measured as another quantitative indicator.
In this example, to verify that the index proposed by the present invention has significant differences between the disease group and the healthy control group, OCT images of patients hospitalized in the clinic for 2022 were collected for statistical analysis. All images are judged by a doctor, checked by a professional technician, and divided into a healthy group and a disease group according to whether the internal limiting membrane hyperplasia phenomenon exists in the research target area. The case information is collected simultaneously, including gender, age, left and right eyes and shooting date, and the total of 281 cases of healthy group and 194 cases of disease group, and mean is calculated according to the method diff The healthy group is significantly lower than the disease group (P<0.05). The statistical results are shown in Table 1:
TABLE 1
Finally, the quantization index mean proposed for verification diff The 475 cases of patient data were imported into the SPSS for ROC curve analysis. The results of the analysis are shown in FIG. 6, and the sensitivity is 73.2%, the specificity is 86.5%, and the Youden coefficient is 0.597. Area under ROC curve (AUC) of 0.863, threshold 5.462 defining the presence or absence of significant proliferation, i.e. at mean diff >The probability of the presence of a distinct proliferative lesion on the inner limiting membrane at 5.462 was 86.3%. Description mean diff The quantitative index has good diagnosis accuracy and high reference value in clinical diagnosis. Meanwhile, the method is suitable for OCT systems with various parameters and has stable numerical values, so that the method has extremely high application value.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (6)
1. A method for evaluating the proliferation condition of cells on an inner limiting membrane based on an OCT image of a retina, comprising the steps of:
s1, acquiring OCT tomographic images, recording a plurality of patient fundus data and information, and acquiring pixel gray values of tissues of each layer on retina;
s2, dividing the retina layer structure by using a graph theory and dynamic programming-based method, and extracting boundary pixel values and coordinate values of an inner limiting membrane;
s3, determining the position coordinates of the macula fovea according to the extracted inner limiting membrane coordinates, and dividing a research target area by taking the macula fovea with highest vision sensitivity as the center;
s4, counting the gray value of the boundary pixel of the inner limiting membrane in the range of the research target area, and drawing a gray histogram;
s5, fitting the histogram by using a normal distribution density function and a Gamma distribution density function respectively, and calculating the difference between the peak values of the two distributions and the corresponding gray values;
s6, dividing a threshold according to the trend of the Gamma distribution density function curve, and distinguishing the proliferation and non-proliferation areas on the inner limiting membrane;
s7, marking a lesion cell proliferation area in the B-scan image;
s8, calculating the ratio of the lesion area to the research target area, and measuring the cell proliferation lesion area and the disease degree on the inner limiting membrane.
2. The method according to claim 1, wherein the information collected in S1 includes name, age, sex, date of visit, diagnosis of illness, presence or absence of proliferation of cells on the inner limiting membrane.
3. The method for evaluating the proliferation of cells on the inner limiting membrane based on the retina OCT image according to claim 1, wherein in the step S3, the boundary coordinate values of the inner limiting membrane are subjected to high-order fitting, the pits of the curve are searched by utilizing the second derivative, and the point with the shortest Euclidean distance from the geometric center of the image in all pits is taken as the center point of the macula fovea.
4. The method for evaluating the proliferation status of cells on the inner limiting membrane based on the OCT image of the retina according to claim 1, wherein the Gamma distribution density function in S5 is:
wherein x represents the distribution range of pixel gray values, k is the shape parameter of the Gamma distribution function, and θ is the scale parameter of the Gamma distribution function;
the normal distribution function is:
where x represents the distribution range of pixel gray values, σ is the standard deviation of normal distribution, and μ is the average value of normal distribution.
5. The method for evaluating a proliferation status of cells on an inner limiting membrane based on a retinal OCT image according to claim 1, wherein the peak value 2Max is fit with Gamma in S6 gamma Minimum value Min of gray histogram as expected μ of normal distribution ref Peak 2Max for Gamma fit gamma Is determined as a threshold value that distinguishes a lesion area of a highlight reflection area, higher thanPixels of the threshold value of the lesion region are classified as lesion cell proliferation regions.
6. The method for evaluating a proliferation status of cells on an inner limiting membrane based on a retinal OCT image according to claim 5, wherein the threshold of the lesion area is 2Max gamma -Min ref 。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311547320.4A CN117372419A (en) | 2023-11-20 | 2023-11-20 | Method for evaluating proliferation condition of cells on inner limiting membrane based on retina OCT (optical coherence tomography) image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311547320.4A CN117372419A (en) | 2023-11-20 | 2023-11-20 | Method for evaluating proliferation condition of cells on inner limiting membrane based on retina OCT (optical coherence tomography) image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117372419A true CN117372419A (en) | 2024-01-09 |
Family
ID=89398549
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311547320.4A Pending CN117372419A (en) | 2023-11-20 | 2023-11-20 | Method for evaluating proliferation condition of cells on inner limiting membrane based on retina OCT (optical coherence tomography) image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117372419A (en) |
-
2023
- 2023-11-20 CN CN202311547320.4A patent/CN117372419A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2162054B1 (en) | Pattern analysis of retinal maps for diagnosis of optic nerve diseases by optical coherence tomography | |
| Adal et al. | An automated system for the detection and classification of retinal changes due to red lesions in longitudinal fundus images | |
| Caprioli et al. | Measurement of relative nerve fiver layer surface height in glaucoma | |
| Mwanza et al. | Ability of cirrus HD-OCT optic nerve head parameters to discriminate normal from glaucomatous eyes | |
| Garway-Heath et al. | Quantitative evaluation of the optic nerve head in early glaucoma | |
| Miglior et al. | Clinical ability of Heidelberg retinal tomograph examination to detect glaucomatous visual field changes | |
| CN110875092B (en) | Health big data service method and system based on remote fundus screening | |
| Liu et al. | Analysis of foveal microvascular abnormalities in diabetic retinopathy using optical coherence tomography angiography with projection artifact removal | |
| US9775510B2 (en) | Methods to reduce variance in OCT analysis of the macula | |
| WO2018116321A2 (en) | Retinal fundus image processing method | |
| WO2010131944A2 (en) | Apparatus for monitoring and grading diabetic retinopathy | |
| Na et al. | Progression of retinal nerve fiber layer thinning in glaucoma assessed by cirrus optical coherence tomography-guided progression analysis | |
| Huang et al. | Stability analysis of fractal dimension in retinal vasculature | |
| Chua et al. | Diagnostic ability of individual macular layers by spectral-domain OCT in different stages of glaucoma | |
| Joshi | Analysis of retinal vessel networks using quantitative descriptors of vascular morphology | |
| CN111402184B (en) | Method and system for realizing remote fundus screening and health service | |
| Ganekal | Ganglion cell complex scan in the early prediction of glaucoma | |
| Wang et al. | Deep learning for diagnosing and segmenting choroidal neovascularization in OCT angiography in a large real-world data set | |
| Geng et al. | Trends in the retinal nerve fiber layer thickness changes with different degrees of visual field defects | |
| WO2021046418A1 (en) | Systems and methods for detection and grading of diabetic retinopathy | |
| CN117372419A (en) | Method for evaluating proliferation condition of cells on inner limiting membrane based on retina OCT (optical coherence tomography) image | |
| Kook et al. | Scanning laser polarimetry using variable corneal compensation in the detection of glaucoma with localized visual field defects | |
| CN113781381B (en) | System for discernment chronic kidney disease image | |
| Duan et al. | Evaluation of automated segmentation algorithms for optic nerve head structures in optical coherence tomography images | |
| Ni et al. | Angle closure glaucoma detection using fractal dimension index on SS-OCT images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |