CN111861917A - Choroidal OCT image enhancement method and device based on signal reverse compensation - Google Patents
Choroidal OCT image enhancement method and device based on signal reverse compensation Download PDFInfo
- Publication number
- CN111861917A CN111861917A CN202010660531.9A CN202010660531A CN111861917A CN 111861917 A CN111861917 A CN 111861917A CN 202010660531 A CN202010660531 A CN 202010660531A CN 111861917 A CN111861917 A CN 111861917A
- Authority
- CN
- China
- Prior art keywords
- oct
- signal
- attenuation
- image
- choroidal
- 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
- 238000000034 method Methods 0.000 title claims abstract description 18
- 210000003161 choroid Anatomy 0.000 claims abstract description 49
- 238000012937 correction Methods 0.000 claims description 22
- 238000012545 processing Methods 0.000 claims description 16
- 238000003384 imaging method Methods 0.000 claims description 14
- 230000002238 attenuated effect Effects 0.000 claims description 11
- 238000012800 visualization Methods 0.000 claims description 9
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 230000003993 interaction Effects 0.000 claims description 5
- 230000003287 optical effect Effects 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 5
- 230000005855 radiation Effects 0.000 claims description 5
- 230000002708 enhancing effect Effects 0.000 claims description 2
- 239000000523 sample Substances 0.000 claims 9
- 210000004204 blood vessel Anatomy 0.000 abstract description 11
- 230000007547 defect Effects 0.000 abstract 1
- 238000012014 optical coherence tomography Methods 0.000 description 52
- 238000013135 deep learning Methods 0.000 description 12
- 230000011218 segmentation Effects 0.000 description 10
- 210000001519 tissue Anatomy 0.000 description 10
- 210000003583 retinal pigment epithelium Anatomy 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 239000000049 pigment Substances 0.000 description 6
- 210000000873 fovea centralis Anatomy 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 230000002792 vascular Effects 0.000 description 5
- 230000003044 adaptive effect Effects 0.000 description 4
- 230000000875 corresponding effect Effects 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 4
- 210000003786 sclera Anatomy 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 208000028867 ischemia Diseases 0.000 description 3
- 238000002372 labelling Methods 0.000 description 3
- 238000000926 separation method Methods 0.000 description 3
- 238000010521 absorption reaction Methods 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000004402 high myopia Effects 0.000 description 2
- 208000002780 macular degeneration Diseases 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000011002 quantification Methods 0.000 description 2
- 238000002310 reflectometry Methods 0.000 description 2
- 210000001525 retina Anatomy 0.000 description 2
- 206010012689 Diabetic retinopathy Diseases 0.000 description 1
- 206010061818 Disease progression Diseases 0.000 description 1
- 206010025421 Macule Diseases 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 206010064930 age-related macular degeneration Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000000149 argon plasma sintering Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000001775 bruch membrane Anatomy 0.000 description 1
- 210000003986 cell retinal photoreceptor Anatomy 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000005750 disease progression Effects 0.000 description 1
- 230000036040 emmetropia Effects 0.000 description 1
- 210000000981 epithelium Anatomy 0.000 description 1
- 230000004438 eyesight Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 230000031700 light absorption Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000002503 metabolic effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 235000015097 nutrients Nutrition 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 230000036284 oxygen consumption Effects 0.000 description 1
- 230000037361 pathway Effects 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 230000002207 retinal effect Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000006496 vascular abnormality Effects 0.000 description 1
Images
Classifications
-
- G06T5/70—
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
- A61B3/102—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for optical coherence tomography [OCT]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
- A61B3/12—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for looking at the eye fundus, e.g. ophthalmoscopes
- A61B3/1225—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for looking at the eye fundus, e.g. ophthalmoscopes using coherent radiation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
- G06T11/008—Specific post-processing after tomographic reconstruction, e.g. voxelisation, metal artifact correction
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- 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
- G06T2207/20048—Transform domain processing
- G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]
-
- 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
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
-
- 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
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Abstract
A choroid OCT image enhancement method and device based on signal reverse compensation are disclosed, wherein a choroid image signal acquired by a system is preprocessed, and a reverse signal attenuation compensation model is established based on the idea of OCT backscattering signal reverse compensation, so that the choroid signal is reversely compensated, and the signal-to-noise ratio between blood vessels and non-blood vessel tissues is enhanced, thereby solving the defect that the boundary of the choroid blood vessel OCT image is difficult to segment.
Description
Technical Field
The invention relates to the technical field of OCT (optical coherence tomography), in particular to a choroidal OCT (optical coherence tomography) image enhancement method and a choroidal OCT image enhancement device based on signal reverse compensation.
Background
The choroid is a highly vascularized and pigment-rich tissue located between the retina and sclera. The main function is to provide oxygen and nutrients to the RPE and outer retina, which delivers approximately 90% of the retinal oxygen consumption, necessary to maintain the high metabolic activity of the outer retinal photoreceptor cells; the choroid is the only pathway for its exchange of material, particularly in the macular foveal avascular zone. Many diseases are closely related to the abnormality of the vascular structure of the choroid, such as age-related macular degeneration, high myopia macular degeneration, diabetic retinopathy and the like. Therefore, it is of great significance to realize quantitative analysis of choroidal vessels.
The frequency domain optical coherence tomography technology can realize three-dimensional imaging of the fundus, and complete choroidal vascular tissues and stromal tissues can be displayed through an averaged image enhancement technology or a high-penetrability sweep light source, so that a favorable imaging technology is provided for choroidal vascular analysis. However, the current system does not solve the problem of poor signal-to-noise ratio of the blood vessel in the deep part of the choroid caused by backscattering attenuation, thereby causing the difficulty of blood vessel boundary segmentation. For this reason, most of the existing algorithms for quantitatively analyzing choroidal vessels in OCT images only adopt the traditional image processing method, but because the choroid is below the RPE, light passes through the RPE and then is attenuated due to pigment absorption, and meanwhile, the choroid is rich in pigment, so that the light passes through the choroid and is attenuated in light signals at the depth, and the boundary between the choroid and the pigment epithelium layer and the boundary between the choroid and the sclera are blurred.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a choroid OCT image enhancement method and device based on signal reverse compensation.
The technical solution adopted by the invention is as follows: a choroidal OCT image enhancement method based on signal inverse compensation comprises the following steps:
(1) and (3) acquiring data: acquiring a fundus image by using OCT, preprocessing the image including appropriate cutting, and reserving an OCT choroid intensity map;
(2) reverse attenuation compensates for choroidal signals: by extracting the attenuation principle and rule of the scattered light of the choroid and constructing a signal compensation and enhancement algorithm, the visualization and the contrast of the choroid image can be improved.
The step (2) of inversely attenuating and compensating the choroidal signal specifically comprises the following steps: by extracting the attenuation principle and the attenuation rule of the choroidal scattered light and constructing a signal compensation and enhancement algorithm, the visualization and the contrast of a choroidal image can be improved, the attenuation correction processing algorithm of an OCT signal comprises two steps of attenuation compensation and image enhancement contrast for light respectively, and in an OCT system, photoelectric signals interfered between a reference arm and a sample arm can be represented by the following formula:
where k is the wave number, the collected signal is divided into m wave numbers with equal interval, rho is the photoelectric conversion efficiency of OCT detector, and S [ k ]m]Refers to the radiation energy of the light source in the corresponding wave band, and Δ x is the optical path difference between the reference arm and the sample arm, R RAnd RSReflectance of the reference arm and sample arm, respectively;
interaction term H [ k ] between reference arm and sample armm]Can be obtained from the above formula (1):
the reflectance profile function s (z) in the depth direction can be obtained by performing inverse discrete fourier transform on the above equation (2):
discretizing the formula, and performing attenuation correction compensation on the OCT signal of each pixel point:
where N is the number of pixels of the a scan, and α can be adjusted according to the tissue, the premise that the above equation (3) holds is that it is assumed that most of the beam energy is attenuated within the imaging depth range, and the attenuation outside the imaging depth range is negligible.
Performing exponentiation operation on the original signal subjected to attenuation correction compensation to enhance image contrast, wherein the signal intensity of each pixel is as follows:
s in formula (5)ac(z) is the attenuation corrected signal.
A choroidal OCT image enhancement device based on signal inverse compensation comprises the following modules,
an input module: for inputting an OCT choroidal intensity map;
an image processing module: constructing a signal compensation and enhancement algorithm by extracting a choroid scattered light attenuation principle and rule for improving visualization and contrast of a choroid image;
an output module: for outputting the image after removing the artifact by the attenuation correction.
The image processing module comprises the following algorithm models: by extracting the attenuation principle and the attenuation rule of the choroidal scattered light, the attenuation correction processing algorithm of the OCT signals improves the visualization and the contrast of a choroidal image, the attenuation correction processing algorithm of the OCT signals comprises two steps of attenuation compensation and image contrast enhancement respectively, and in the OCT system, photoelectric signals interfered between a reference arm and a sample arm can be represented by the following formula:
where k is the wave number, the collected signal is divided into m wave numbers with equal interval, rho is the photoelectric conversion efficiency of OCT detector, and S [ k ]m]Refers to the radiation energy of the light source in the corresponding wave band, and Δ x is the optical path difference between the reference arm and the sample arm, RRAnd RSReflectance of the reference arm and sample arm, respectively;
interaction term H [ k ] between reference arm and sample armm]Obtained from the above formula (1):
obtaining a reflectance profile function s (z) in the depth direction by performing inverse discrete fourier transform on the above expression (2):
discretizing the formula, and performing attenuation correction compensation on the OCT signal of each pixel:
where N is the number of pixels of the a scan, and α can be adjusted according to the tissue, the premise that the above equation (3) holds is that it is assumed that most of the beam energy is attenuated within the imaging depth range, and the attenuation outside the imaging depth range is negligible.
Performing exponentiation operation on the original signal subjected to attenuation correction compensation to enhance image contrast, wherein the signal intensity of each pixel is as follows:
s in formula (5)ac(z) is the attenuation corrected signal.
The invention has the beneficial effects that: the invention provides a choroid OCT image enhancement method and device based on signal reverse compensation.
Drawings
FIG. 1 is a flow chart of the technical solution of the present invention.
Fig. 2 is OCT fundus image acquisition and pre-processing.
FIG. 3 is a schematic diagram of choroidal signal attenuation correction and image contrast de-enhancement; the left image is the original image, and the right image is the image after attenuation correction and artifact removal.
FIG. 4 is a comparison of the inner and outer boundaries of a conventional dynamic programming algorithm and a deep learning automatic segmentation choroid; wherein the left image is the result of the traditional algorithm, and the right image is the result of the deep learning algorithm.
FIG. 5 is a diagram of adaptive threshold separation of choroidal vessels from non-vessels; wherein the upper image is high myopia, and the lower image is emmetropia.
Fig. 6 is a diagram of choroidal vessels after three-dimensional reconstruction.
Detailed Description
The invention will be better explained with reference to the drawings and some specific embodiments.
(1) Data acquisition and preprocessing:
the fundus image is acquired using current commercial instruments or self-constructed OCT, and the image is pre-processed, including appropriate cropping, leaving an OCT choroidal intensity map, as shown in figure 2.
(2) Reverse attenuation compensates for choroidal signals:
the signal received by the OCT detector is a backscattered and reflected signal. Due to the fact that loss occurs in light scattering of a certain wavelength caused by the influence of RPE and choroid self-pigment on light absorption, by extracting the attenuation principle and rule of choroid scattered light, a signal compensation and enhancement algorithm is constructed, and visualization and contrast of a choroid image can be improved.
The attenuation correction processing algorithm of the OCT signal comprises two steps of attenuation compensation and image enhancement contrast of light respectively. In an OCT system, the photoelectric signal of the interference between the reference arm and the sample arm can be represented by the following formula:
where k is the wave number, the collected signal is divided into m wave numbers with equal interval, rho is the photoelectric conversion efficiency of OCT detector, and S [ k ]m]Refers to the radiation energy of the light source in the corresponding wave band, and Δ x is the optical path difference between the reference arm and the sample arm, R RAnd RSThe reflectivity of the reference arm and the sample arm, respectively. Interaction term H [ k ] between reference arm and sample armm]Can be obtained by the formula (1):
the reflectance profile function s (z) in the depth direction can be obtained by performing inverse discrete fourier transform on equation (2):
while equations (1) - (3) hold the assumption that no energy attenuation occurs in the beam during propagation. However, the energy attenuation of the light beam occurs during the propagation process, so we need to improve the above formula. When the light beam penetrates the tissue sample, a small portion is converted to heat and the remainder is scattered, where the light beam is attenuated in the form of absorption. Assuming that the local attenuation of the beam is only related to the scattering of light and is backscattered with a fixed proportion, it is possible to calculate that the local attenuation of the beam at depth Z is proportional to the reflectivity R and the backscatter constant α at that location.
In practical situation, after the direct data processing of OCT, the attenuated signal D is obtainedAInstead of the original signal s (z). This is a main cause of occurrence of a dark shadow in strongly attenuated tissues such as blood vessels and pigments, i.e., a cause of occurrence of an artifact. To correct forThe signal is attenuated and the attenuation term needs to be eliminated, while the artifacts are also removed. Discretizing the formula, and performing attenuation correction compensation on the OCT signal of each pixel point:
Where N is the number of pixels of the A-scan, and α can be adjusted according to the organization. The above equation holds if it is assumed that most of the beam energy is attenuated within the imaging depth range, and the attenuation outside the imaging depth range is negligible.
In the later stage, the invention performs exponentiation operation on the original signal after attenuation correction compensation to enhance the image contrast, and the signal intensity of each pixel is as follows:
in the formula Sac(z) is the attenuation corrected signal, as shown in FIG. 3.
(3) Deep learning intelligent segmentation:
accurate identification of the inner and outer choroidal boundaries is an important step in accurately quantifying three-dimensional choroidal vascular indicators.
The suprachoroidal boundary is defined as the boundary between Bruch's membrane and the Retinal Pigment Epithelium (RPE), and the infrachoroidal boundary is defined as the boundary between the choroid and the sclera. Because the RPE layer is represented as a high signal band in the OCT image, the automatic division of the suprachoroidal boundary can be well realized based on the traditional shortest path graph theory algorithm; however, the contrast difference of the boundary between the choroid and the sclera in the OCT image makes it difficult to automatically segment the lower boundary by the shortest path graph theory algorithm. Deep learning is the most important breakthrough in the field of artificial intelligence, and makes a great breakthrough in the field of computer vision, and the efficiency of the algorithm based on deep learning is obviously better than that of the traditional algorithm. The invention establishes a deep learning segmentation model to realize automatic segmentation of the upper and lower choroidal boundaries, and comprises the following specific steps:
Semi-automatically labeling the inner and outer boundaries of the choroid of thousands of OCT images based on a traditional shortest path graph theory algorithm, and accurately describing the upper and lower boundaries of the choroid in an image; the labeled choroidal image set was as 8: 2, randomly dividing the training set and the testing set.
Inputting the training image into an open-source deep learning neural network model, setting a training optimization algorithm as a random gradient descent method (SGD), setting an algorithm learning ratio to be 1.0e-5, setting an iteration momentum to be 0.9, setting an iteration cost function to be a Dice coefficient (dicecoefficient), setting an iteration pass number (epoch) to be 150, and setting a batch sample size (batch size) to be 8. In addition, necessary enhancement processing (such as translation, rotation, and inversion) is performed on the input image, and the robustness of the model is improved. Wherein, the Dice coefficient calculation formula is as follows:
x denotes a choroidal boundary prediction set, Y denotes a choroidal boundary labeling set, | X | + | Y | denotes an intersection or overlap between the two sets, | X | + | Y | denotes the total amount of the two. The larger the Dice coefficient is, the higher the similarity of the two sets is, and the more accurate the model is; when the prediction set and the annotation set are completely the same, the Dice coefficient is 1; when the prediction set is not correlated with the annotation set, the Dice coefficient is 0. In the model training process, a Dice coefficient greater than 0.95 is set as a target function.
Secondly, inputting the images of the test set into the deep learning neural network model, and calculating a Dice coefficient and a boundary error between the output result of the choroid boundary and the labeling set to evaluate the segmentation performance of the deep learning neural network model. The results are shown in FIG. 4 and Table 1.
TABLE 1 efficiency of deep learning and segmentation of OCT image choroid boundary
(4) Adaptive thresholding to separate choroidal vessels from non-vessels: choroidal vessels and stroma exhibit different characteristics in OCT images, with vessels dominated by low-intensity signals and stroma dominated by high-intensity signals. Since the brightness of the choroidal signal in the OCT image is affected by the RPE layer, the type of instrument, the focusing condition during operation, and other relevant factors, the separation of choroidal vessels and stroma based on the fixed threshold method is theoretically limited and has poor universality. In contrast, interference caused by overall shift of the brightness of the choroidal signal can be better avoided based on the adaptive threshold, and automatic separation of choroidal vessels and stroma can be better realized, and the flow of the method is as follows:
in the OCT image, a square frame having a length and a width of 2 × w +1 pixel blocks is used as a local window. The coordinate (x, y) is the geometric center of the square frame, the luminance information of all pixel blocks in the frame is counted, and the mean value m (x, y) and the variance s (x, y) are obtained. According to the parameter k, an in-frame threshold value T (x, y) can be obtained, and the calculation formula is as follows:
T(x,y)=m(x,y)+k*s(x,y) (7)
Secondly, according to the threshold value T (x, y) in the frame, all pixel blocks in the frame can be subjected to binarization processing, as follows:
where i and j are relative coordinates characterizing the relative geometric center (x, y) of the pixel block, Ax+i,y+jRepresenting the luminance of a block of pixels with coordinates (x + i, y + j), Ax+i,y+jBinarized data representing a pixel block of coordinates (x + i, y + j).
And thirdly, sequentially moving the local windows to realize binarization processing of the brightness matrix A of all pixel blocks in the OCT image so as to obtain a binarization matrix B. The results are shown in FIG. 5.
(5) Three-dimensional overall and each region quantification index:
only characteristic information on a certain cross section of the choroid can be acquired based on a single OCT image, but the disease progression of the disease is difficult to accurately evaluate by a single OCT image analysis mode due to the uneven spatial distribution of certain choroid lesions. The invention images the eyeground in a radial scanning mode to obtain choroid three-dimensional space data, and performs registration and reconstruction on the image to realize choroid three-dimensional space reconstruction. In the OCT image, the fovea centralis is taken as a reference point, the position of the fovea centralis is horizontally translated to ensure that the horizontal coordinates of the fovea centralis in the space coordinate in all the OCT images are the same, the three-dimensional reconstruction of the image is realized, and the overall quantitative index and each area quantitative index which can represent choroidal ischemia are calculated according to the distribution and the proportion of blood vessels in the three-dimensional space in the image. The method comprises the following specific steps:
Acquisition of an image: imaging is carried out by taking the fovea of the fundus macula as a center in a radial scanning mode, and m pieces of choroid two-dimensional cross section data are obtained.
Marking characteristic points: since all OCT images contain the feature of the fovea maculata, the present invention marks the fovea maculata as a feature point of the OCT images for subsequent image registration.
Image registration: the fovea centralis in all OCT images has the same abscissa in space coordinates by horizontally translating the position of the fovea centralis, so that the images are registered.
Reconstructing a choroid three-dimensional space: extracting a choroid region template M (x, y, z) in the registered image based on the upper and lower choroid boundaries obtained by the deep learning neural network model, and automatically separating blood vessels and matrixes by combining an adaptive threshold method to obtain a binarized three-dimensional space choroid structure matrix V (x, y, z).
Establishing and testing indexes: indexes such as choroidal vessel volume CVV, choroidal non-vessel volume SV, choroidal vessel index CVI and choroidal ischemia index CII are established for quantification. The calculation formula of each index is as follows:
wherein P isx,Py,PzRespectively, characterize the physical geometry of a voxel in three-dimensional space along the x, y and z axes. n, m and k represent the number of pixels along the x, y and z axes, respectively, in the three-dimensional spatial matrix.
The invention tests the repeatability of the index by repeatedly carrying out OCT imaging on the same subject for 2 times by the same operator, and compares the repeatability with the two-dimensional corresponding index, and the test result is shown in the following table.
0-6mm | Left side 3-6mm | Left side 1-3mm | 0-1mm | Right side 1-3mm | The right side is 3-6mm | ||
CVV | Is perpendicular to | 0.995 | 0.989 | 0.987 | 0.985 | 0.963 | 0.964 |
Level of | 0.994 | 0.981 | 0.98 | 0.99 | 0.96 | 0.989 | |
Three-dimensional | 0.998 | 0.997 | 0.998 | 0.997 | 0.998 | 0.999 | |
CVI | Is perpendicular to | 0.883 | 0.736 | 0.815 | 0.814 | 0.714 | 0.731 |
Level of | 0.865 | 0.664 | 0.798 | 0.864 | 0.786 | 0.541 | |
Three-dimensional | 0.991 | 0.986 | 0.965 | 0.975 | 0.983 | 0.982 | |
CII | Is perpendicular to | 0.883 | 0.736 | 0.815 | 0.814 | 0.714 | 0.731 |
Level of | 0.865 | 0.664 | 0.798 | 0.864 | 0.786 | 0.541 | |
Three-dimensional | 0.991 | 0.986 | 0.965 | 0.975 | 0.983 | 0.982 | |
SV | Is perpendicular to | 0.968 | 0.86 | 0.962 | 0.871 | 0.927 | 0.934 |
Level of | 0.935 | 0.694 | 0.964 | 0.939 | 0.91 | 0.768 | |
Three-dimensional | 0.994 | 0.977 | 0.989 | 0.990 | 0.994 | 0.992 |
Therefore, the repeatability of the three-dimensional index is obviously superior to that of the two-dimensional index. In the two-dimensional indexes, the repeatability of the vertical two-dimensional index is superior to that of the horizontal two-dimensional index.
After preprocessing a choroid image signal acquired by a system, establishing a reverse signal attenuation compensation model based on the idea of reverse compensation of an OCT (optical coherence tomography) backscattering signal, reversely compensating the choroid signal, enhancing the signal-to-noise ratio between blood vessels and non-blood vessel tissues, and further performing intelligent segmentation on the inner and outer boundaries of the choroid by adopting a segmentation method based on deep learning; on the basis of boundary segmentation, an improved self-adaptive threshold method is further adopted to automatically separate out three-dimensional choroidal vessels and non-vascular tissues, and the overall quantitative index and each region quantitative index capable of representing choroidal ischemia are calculated according to the distribution and the proportion of the vessels in the three-dimensional space in an image. The method is suitable for all OCT systems which have high penetration on tissues and can acquire choroid and images thereof, and has the advantages of automation, strong universality and high-precision reflection of choroid three-dimensional blood vessel abnormality.
The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may occur to those skilled in the art without departing from the principle of the invention, and are considered to be within the scope of the invention.
Claims (4)
1. A choroidal OCT image enhancement method based on signal inverse compensation is characterized by comprising the following steps:
(1) and (3) acquiring data: acquiring a fundus image by using OCT, preprocessing the image including appropriate cutting, and reserving an OCT choroid intensity map;
(2) reverse attenuation compensates for choroidal signals: by extracting the attenuation principle and rule of the scattered light of the choroid, a signal compensation and enhancement algorithm is constructed, and the visualization and the contrast of the choroid image are improved.
2. The method for enhancing the choroidal OCT image based on inverse signal compensation according to claim 1, wherein said step (2) of inversely attenuating the compensated choroidal signal comprises the steps of: by extracting the attenuation principle and the attenuation rule of the choroidal scattered light, the attenuation correction processing algorithm of the OCT signals improves the visualization and the contrast of a choroidal image, the attenuation correction processing algorithm of the OCT signals comprises two steps of attenuation compensation and image contrast enhancement respectively, and in the OCT system, photoelectric signals interfered between a reference arm and a sample arm can be represented by the following formula:
Where k is the wave number, the collected signal is divided into m wave numbers with equal interval, rho is the photoelectric conversion efficiency of OCT detector, and S [ k ]m]Refers to the radiation energy of the light source in the corresponding wave band, and Δ x is the optical path difference between the reference arm and the sample arm, RRAnd RSReflectance of the reference arm and sample arm, respectively;
interaction term H [ k ] between reference arm and sample armm]Obtained from the above formula (1):
obtaining a reflectance profile function s (z) in the depth direction by performing inverse discrete fourier transform on the above expression (2):
discretizing the formula, and performing attenuation correction compensation on the OCT signal of each pixel:
where N is the number of pixels of the a scan, and α can be adjusted according to the tissue, the premise that the above equation (3) holds is that it is assumed that most of the beam energy is attenuated within the imaging depth range, and the attenuation outside the imaging depth range is negligible.
Performing exponentiation operation on the original signal subjected to attenuation correction compensation to enhance image contrast, wherein the signal intensity of each pixel is as follows:
s in formula (5)ac(z) is the attenuation corrected signal.
3. A choroidal OCT image enhancement device based on signal inverse compensation is characterized by comprising the following modules,
an input module: for inputting an OCT choroidal intensity map;
An image processing module: constructing a signal compensation and enhancement algorithm by extracting a choroid scattered light attenuation principle and rule for improving visualization and contrast of a choroid image;
an output module: for outputting the image after removing the artifact by the attenuation correction.
4. The apparatus as claimed in claim 1, wherein the image processing module comprises the following algorithm models: by extracting the attenuation principle and the attenuation rule of the choroidal scattered light, the attenuation correction processing algorithm of the OCT signals improves the visualization and the contrast of a choroidal image, the attenuation correction processing algorithm of the OCT signals comprises two steps of attenuation compensation and image contrast enhancement respectively, and in the OCT system, photoelectric signals interfered between a reference arm and a sample arm can be represented by the following formula:
where k is the wave number and the acquired signal is divided into m equally spaced wavesNumber, rho is the photoelectric conversion efficiency of the OCT probe, Skm]Refers to the radiation energy of the light source in the corresponding wave band, and Δ x is the optical path difference between the reference arm and the sample arm, RRAnd RSReflectance of the reference arm and sample arm, respectively;
interaction term H [ k ] between reference arm and sample arm m]Obtained from the above formula (1):
obtaining a reflectance profile function s (z) in the depth direction by performing inverse discrete fourier transform on the above expression (2):
discretizing the formula, and performing attenuation correction compensation on the OCT signal of each pixel:
where N is the number of pixels of the a scan, and α can be adjusted according to the tissue, the premise that the above equation (3) holds is that it is assumed that most of the beam energy is attenuated within the imaging depth range, and the attenuation outside the imaging depth range is negligible.
Performing exponentiation operation on the original signal subjected to attenuation correction compensation to enhance image contrast, wherein the signal intensity of each pixel is as follows:
s in formula (5)ac(z) is the attenuation corrected signal.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010660531.9A CN111861917A (en) | 2020-07-10 | 2020-07-10 | Choroidal OCT image enhancement method and device based on signal reverse compensation |
PCT/CN2020/139964 WO2022007353A1 (en) | 2020-07-10 | 2020-12-28 | Method and apparatus for enhancing choroid oct image on basis of signal reverse compensation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010660531.9A CN111861917A (en) | 2020-07-10 | 2020-07-10 | Choroidal OCT image enhancement method and device based on signal reverse compensation |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111861917A true CN111861917A (en) | 2020-10-30 |
Family
ID=73153384
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010660531.9A Pending CN111861917A (en) | 2020-07-10 | 2020-07-10 | Choroidal OCT image enhancement method and device based on signal reverse compensation |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN111861917A (en) |
WO (1) | WO2022007353A1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113160160A (en) * | 2021-04-12 | 2021-07-23 | 哈尔滨医科大学 | Calcification identification method based on OCT light attenuation image |
WO2022007353A1 (en) * | 2020-07-10 | 2022-01-13 | 温州医科大学 | Method and apparatus for enhancing choroid oct image on basis of signal reverse compensation |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102458221A (en) * | 2009-06-02 | 2012-05-16 | 佳能株式会社 | Image processing apparatus, control method thereof, and computer program |
CN104755908A (en) * | 2012-07-27 | 2015-07-01 | 统雷有限公司 | Agile imaging system |
CN107374583A (en) * | 2017-05-31 | 2017-11-24 | 执鼎医疗科技(杭州)有限公司 | One kind eliminates pseudo- image method and storage medium and imaging system in OCT blood vessel imagings |
CN107615005A (en) * | 2015-05-28 | 2018-01-19 | 赛莱特私人有限公司 | The D spectral domain opticals imaging device of high-resolution 3 and method |
CN107862661A (en) * | 2017-11-06 | 2018-03-30 | 郑州轻工业学院 | A kind of optical coherence tomography system method for correcting image |
CN109730633A (en) * | 2018-12-28 | 2019-05-10 | 中国科学院宁波工业技术研究院慈溪生物医学工程研究所 | Choroidal artery angiographic method and equipment based on optical coherence tomography swept-volume |
CN110060261A (en) * | 2019-04-17 | 2019-07-26 | 北京理工大学 | A kind of blood vessel segmentation method based on Optical coherence tomography imaging system |
CN110693457A (en) * | 2019-10-14 | 2020-01-17 | 浙江大学 | Tissue activity detection method and system based on optical coherence technology |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110101362B (en) * | 2019-04-29 | 2021-12-17 | 中国科学院宁波工业技术研究院慈溪生物医学工程研究所 | Method for removing image noise related to OCT and OCTA |
CN111862114A (en) * | 2020-07-10 | 2020-10-30 | 温州医科大学 | Choroidal three-dimensional blood vessel imaging and quantitative analysis method and device based on optical coherence tomography system |
CN111861917A (en) * | 2020-07-10 | 2020-10-30 | 温州医科大学 | Choroidal OCT image enhancement method and device based on signal reverse compensation |
-
2020
- 2020-07-10 CN CN202010660531.9A patent/CN111861917A/en active Pending
- 2020-12-28 WO PCT/CN2020/139964 patent/WO2022007353A1/en active Application Filing
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102458221A (en) * | 2009-06-02 | 2012-05-16 | 佳能株式会社 | Image processing apparatus, control method thereof, and computer program |
CN104755908A (en) * | 2012-07-27 | 2015-07-01 | 统雷有限公司 | Agile imaging system |
CN107615005A (en) * | 2015-05-28 | 2018-01-19 | 赛莱特私人有限公司 | The D spectral domain opticals imaging device of high-resolution 3 and method |
CN107374583A (en) * | 2017-05-31 | 2017-11-24 | 执鼎医疗科技(杭州)有限公司 | One kind eliminates pseudo- image method and storage medium and imaging system in OCT blood vessel imagings |
CN107862661A (en) * | 2017-11-06 | 2018-03-30 | 郑州轻工业学院 | A kind of optical coherence tomography system method for correcting image |
CN109730633A (en) * | 2018-12-28 | 2019-05-10 | 中国科学院宁波工业技术研究院慈溪生物医学工程研究所 | Choroidal artery angiographic method and equipment based on optical coherence tomography swept-volume |
CN110060261A (en) * | 2019-04-17 | 2019-07-26 | 北京理工大学 | A kind of blood vessel segmentation method based on Optical coherence tomography imaging system |
CN110693457A (en) * | 2019-10-14 | 2020-01-17 | 浙江大学 | Tissue activity detection method and system based on optical coherence technology |
Non-Patent Citations (1)
Title |
---|
李培等: "傅里叶域光学相干层析成像技术的研究进展", 《中国激光》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022007353A1 (en) * | 2020-07-10 | 2022-01-13 | 温州医科大学 | Method and apparatus for enhancing choroid oct image on basis of signal reverse compensation |
CN113160160A (en) * | 2021-04-12 | 2021-07-23 | 哈尔滨医科大学 | Calcification identification method based on OCT light attenuation image |
CN113160160B (en) * | 2021-04-12 | 2022-07-08 | 哈尔滨医科大学 | Calcification identification method based on OCT light attenuation image |
Also Published As
Publication number | Publication date |
---|---|
WO2022007353A1 (en) | 2022-01-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111862114A (en) | Choroidal three-dimensional blood vessel imaging and quantitative analysis method and device based on optical coherence tomography system | |
Hormel et al. | Artificial intelligence in OCT angiography | |
Chen et al. | DN-GAN: Denoising generative adversarial networks for speckle noise reduction in optical coherence tomography images | |
US9418423B2 (en) | Motion correction and normalization of features in optical coherence tomography | |
US10194866B2 (en) | Methods and apparatus for reducing artifacts in OCT angiography using machine learning techniques | |
Hassan et al. | Deep learning based joint segmentation and characterization of multi-class retinal fluid lesions on OCT scans for clinical use in anti-VEGF therapy | |
US8721077B2 (en) | Systems, methods and computer-readable medium for determining depth-resolved physical and/or optical properties of scattering media by analyzing measured data over a range of depths | |
US20120070049A1 (en) | Image processing apparatus, control method thereof, and computer program | |
US11854199B2 (en) | Methods and systems for ocular imaging, diagnosis and prognosis | |
CN103348359A (en) | 3D retinal disruptions detection using optical coherence tomography | |
CN111861917A (en) | Choroidal OCT image enhancement method and device based on signal reverse compensation | |
Wu et al. | Geographic atrophy segmentation in SD-OCT images using synthesized fundus autofluorescence imaging | |
EP3417401B1 (en) | Method for reducing artifacts in oct using machine learning techniques | |
Eghtedar et al. | An update on choroidal layer segmentation methods in optical coherence tomography images: a review | |
Garcia-Marin et al. | Patch-based CNN for corneal segmentation of AS-OCT images: Effect of the number of classes and image quality upon performance | |
Tan et al. | Three-dimensional choroidal vessel network quantification using swept source optical coherence tomography | |
Marciniak et al. | Measurement of retina vessels by segmentation of images reconstructed from optical coherence tomography data | |
Stankiewicz et al. | Volumetric segmentation of human eye blood vessels based on OCT images | |
US20230298236A1 (en) | Pre-processing of OCT B-scans for OCT Angiography | |
Kim | Dynamic Biomechanical Characteristics of the Optic Nerve Head by OCT Imaging | |
Cao et al. | A novel method of detecting and locating the macula based on morphological features fitting | |
Chen et al. | Segmentation and Visualization of Drusen and Geographic Atrophy in SD-OCT Images | |
EP4252179A1 (en) | Quality maps for optical coherence tomography angiography | |
Jorgensen et al. | Dynamic contour model for aligning and segmenting OCT B-scans | |
CN116710959A (en) | Use of multiple subvolumes, thicknesses and curvatures for OCT/OCTA data registration and retinal landmark detection |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201030 |
|
RJ01 | Rejection of invention patent application after publication |