CN111292311B

CN111292311B - Method and system for calculating volume of macular center avascular zone based on OCTA retinal image

Info

Publication number: CN111292311B
Application number: CN202010094886.6A
Authority: CN
Inventors: 陈强; 徐秋卓
Original assignee: Nanjing University of Science and Technology
Current assignee: Nanjing University of Science and Technology
Priority date: 2020-02-17
Filing date: 2020-02-17
Publication date: 2022-08-09
Anticipated expiration: 2040-02-17
Also published as: CN111292311A

Abstract

The invention discloses a macular center avascular zone volume calculation method and a macular center avascular zone volume calculation system based on an OCTA (optical coherence tomography) retinal image, wherein the method comprises the following steps: collecting an OCTA retina image; setting a restricted area containing a macular central avascular zone, namely an FAZ zone; leveling the limited area; sequentially generating a plurality of frames of projection images with the thickness of one pixel in the axial direction in the flattened limiting area, and labeling an FAZ area on each frame of projection image and recording as a label image; reducing the label image to the original layer position before the limitation area is not leveled; and counting the number of label pixels in the restored label image, thereby obtaining the volume of the FAZ area. The FAZ volume obtained by the method is a brand-new analysis index parameter provided for the OCTA image, and has important significance for subsequent FAZ qualitative and quantitative analysis.

Description

Method and system for calculating volume of macular center avascular zone based on OCTA retinal image

Technical Field

The invention relates to a volume parameter calculation method, in particular to a macular center avascular region volume calculation method and system based on an OCTA (optical coherence tomography) retinal image.

Background

The OCTA retina image is an optical coherence tomography blood vessel imaging image, which is characterized in that the frequency amplitude of OCT repeatedly scanned at the same position is divided into a plurality of sections by using a frequency division amplification decoherence blood vessel imaging technology (SSADA) and analyzed by using a decorrelation method, the OCT scanning image is processed, and blood vessels of retina and choroid are reconstructed on the coronary surface. It can effectively present the blood vessel forms of each layer of retina and choroid. In the foveal region of the retina, there is an avascular zone surrounded by superficial and deep retinal vascular plexuses, called the macular central avascular zone (FAZ). Clinical trials have shown that OCTA images can be used to measure the extent of the macular central avascular zone (FAZ). The OCTA image is three-dimensional, but the current FAZ quantitative index is based on the area under OCTA two-dimensional projection, and the specific measurement method is as follows: based on the superficial and deep retinal vascular plexus projection maps, the outline of the avascular zone is outlined to obtain the FAZ area. The two-dimensional measurement mode ignores the form change of the FAZ in the axial direction, cannot reflect the influence of the thickness change of the retina layer structure on the size of the FAZ, and is not beneficial to the visual and comprehensive observation of the FAZ range by a doctor. Therefore, the extension of FAZ dimension measurement to three-dimensional volume parameters is of great significance for clinical evaluation.

Disclosure of Invention

The invention aims to provide a method and a system for calculating the volume of an avascular zone in the center of macula lutea, which can acquire FAZ three-dimensional volume parameters.

The technical solution for realizing the purpose of the invention is as follows: a macular central avascular zone volume calculation method based on an OCTA retinal image, the method comprising the steps of:

step 1, collecting an OCTA retina image;

step 2, setting a limit area containing a macular central avascular area, namely an FAZ area;

step 3, leveling the limited area;

step 4, sequentially generating a plurality of frames of projection images with the thickness of one pixel in the axial direction in the flattened limiting area, and labeling an FAZ area on each frame of projection image and recording as a label image;

step 5, the label image is reduced to the original layer position before the limitation area is not leveled;

and 6, counting the number of label pixels in the restored label image, thereby obtaining the volume of the FAZ area.

Further, step 2 sets a restricted area including a macular central avascular zone, i.e., an FAZ zone, and the specific process includes:

step 2-1, segmenting the lower boundary of an inner limiting membrane ILM and an outer plexiform layer OPL of an OCTA retina image;

step 2-2, ILM is taken as the upper boundary of the confinement region, and the part l μm below the lower boundary of OPL is taken as the lower boundary of the confinement region.

Further, the value of l is a lower boundary value of the deep vascular complex DVC defined by the oca image acquisition device.

Further, the step 3 of flattening the limited area specifically comprises the following steps:

step 3-1, acquiring a three-dimensional global lowest point of an OPL lower boundary, and taking a line coordinate at a position of l micrometers below the lowest point as a lower boundary reference;

step 3-2, moving each row in the restricted area to a lower boundary reference, realizing lower boundary alignment of each row, and obtaining a primary leveling restricted area;

step 3-3, acquiring an upper boundary three-dimensional global highest point of the primary leveling limiting area, and taking the highest point as an upper boundary reference;

and 3-4, converting each column in the preliminary leveling limiting area into a fixed length, wherein the fixed length is the distance length between the lower boundary reference and the upper boundary reference.

Further, in step 5, the label image is restored to the original layer position before the limitation area is not leveled, and the specific process includes:

and (4) reversely executing the reverse operation of the step (3) to restore the label image to the label image which accords with the original layer structure distribution of the OCTA retina image.

Further, in step 6, the volume of the FAZ region is obtained by the following formula:

V＝n×p

in the formula, V denotes a volume of the FAZ region, n denotes the number of label pixels, and p denotes a resolution parameter of the image.

A macular central avascular zone volume calculation system based on an OCTA retinal image, the system comprising:

the acquisition module is used for acquiring an OCTA retina image;

the region setting module is used for setting a limiting region containing a macular center avascular region, namely an FAZ region;

the region transformation module is used for flattening the limited region;

the label image generation module is used for sequentially generating a plurality of frames of projection images with the thickness of one pixel in the axial direction in the flattened limit area, and labeling an FAZ area on each frame of projection image and recording as a label image;

the image restoration module is used for restoring the label image to an original layer position before the limitation area is not leveled;

and the volume calculation module is used for counting the number of label pixels in the restored label image so as to obtain the volume of the FAZ area.

Further, the region setting module includes:

a segmentation unit configured to segment an inner limiting membrane ILM and an outer plexiform layer OPL lower boundary of the OCTA retinal image;

and the region setting unit is used for taking the ILM as the upper boundary of the limiting region and taking the position which is l mu m below the lower boundary of the OPL as the lower boundary of the limiting region.

Further, the region transformation module includes:

the first reference setting unit is used for acquiring a three-dimensional global lowest point of the lower boundary of the OPL and taking a line coordinate at a position l mu m below the lowest point as a lower boundary reference;

the first transformation unit is used for moving each column in the restricted area to a lower boundary reference, realizing lower boundary alignment of each column and obtaining a preliminary leveling restricted area;

the second benchmark setting unit is used for acquiring an upper boundary three-dimensional global highest point of the primary leveling limit area and taking the highest point as an upper boundary benchmark;

and the second transformation unit is used for interpolating each column in the preliminary leveling limiting area into a fixed length, wherein the fixed length is the distance length between the lower boundary reference and the upper boundary reference.

Compared with the prior art, the invention has the following remarkable advantages: 1) the original FAZ two-dimensional projection is expanded in the axial direction through image flattening and the inverse operation of the image flattening, so that the FAZ three-dimensional form is reconstructed, and a calculation result of the FAZ volume is obtained; 2) the FAZ structure is described from a three-dimensional layer for the first time, a certain amount of evaluation indexes of the FAZ volume is provided, a calculation method with strong operability is provided, and a new breakthrough is provided for relevant clinical diagnosis and scientific research work.

The present invention is described in further detail below with reference to the attached drawing figures.

Drawings

FIG. 1 is a flow chart of a method for calculating the volume of the central avascular region of the macula based on an OCTA retinal image in one embodiment.

FIG. 2 is a schematic diagram of upper and lower boundaries of a bounding region in one embodiment.

FIG. 3 is a diagram illustrating flattening of the lower boundary of the confinement region in one embodiment.

FIG. 4 is a diagram illustrating border flattening above a bounding region in one embodiment.

Fig. 5 is a schematic diagram of labeling a FAZ area in a frame of C-scan (projection image) after flattening the restricted area in an embodiment, where (a) is an original projection image and (b) is a projection image after labeling the FAZ area.

Fig. 6 is a schematic diagram illustrating a correspondence relationship between a C-scan label and a B-scan label image in an embodiment, where (a) is a three-dimensional view of a C-scan (x-y view) labeling process, and (B) is a B-scan (x-z view) label image of a frame after the label is completed.

FIG. 7 is a label graph after a horizon has been restored in one embodiment.

Fig. 8 is a schematic three-dimensional structure of a FAZ volume label in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of and not restrictive on the broad application.

In one embodiment, in combination with fig. 1, there is provided a method for calculating a macular central avascular zone volume based on an OCTA retinal image, the method comprising the steps of:

step 1, collecting an OCTA retina image;

step 3, leveling the limited area;

Further, in one embodiment, the step 2 of setting a restricted area containing a macular central avascular zone, i.e. an FAZ zone, includes:

Further, in one embodiment, the value of l is a lower boundary value of the deep vascular complex DVC defined by the OCTA image capturing device or the medical literature.

Exemplarily and preferably, in one embodiment, the l μm is 10 μm.

Further, in one embodiment, the step 3 of leveling the limited area includes the following specific steps:

step 3-3, acquiring an upper boundary three-dimensional global highest point of the primary leveling limit area, and taking the highest point as an upper boundary reference;

Further preferably, in one embodiment, each column in the preliminary flattening limit area is transformed into a fixed length in step 3-4, specifically by linear interpolation.

Further, in one embodiment, the step 5 of restoring the label image to the original layer position before the limitation area is not leveled includes:

Further, in one embodiment, the volume of the FAZ region is determined in step 6 by the following formula:

V＝n×p

In one embodiment, there is provided a macular center avascular zone volume calculation system based on an OCTA retinal image, the system comprising:

the acquisition module is used for acquiring an OCTA retina image;

the region transformation module is used for leveling the limited region;

Further, in one embodiment, the area setting module includes:

Further, in one embodiment, the region transformation module includes:

the first transformation unit is used for moving each column in the restricted area to a lower boundary reference, realizing the alignment of the lower boundary of each column and obtaining a primary leveling restricted area;

and a second transformation unit, configured to interpolate each column in the preliminary flattening limit region to a fixed length, where the fixed length is a distance length between the lower boundary reference and the upper boundary reference.

Illustratively, as a specific example, in one embodiment, the three-dimensional OCTA retinal image size acquired by the OCTA imaging device is 640 × 400 × 400 pixels with a pixel resolution of 3 μm × 15 μm × 15 μm. As shown in fig. 2, the limit region where the FAZ exists is manually divided into an ILM and an OPL lower boundary, and the ILM is taken as the upper boundary of the limit region, with 10 μm below the OPL lower boundary. And acquiring a three-dimensional global lowest point of the lower boundary of the OPL, and taking a line coordinate 10 microns below the lowest point as a lower boundary reference. Each column within the restricted area is moved to the lower boundary datum alignment and the result is shown in figure 3 after the lower boundary of the restricted area is leveled. A three-dimensional global highest point of the upper boundary of the restricted area after the lower boundary is flattened is obtained, the highest point is used as an upper boundary reference, each line of linear interpolation in the restricted area is a fixed length, the fixed length is a distance length between the upper boundary reference and the lower boundary reference, and a flattened result of the upper boundary of the restricted area is shown in fig. 4. After the limiting area is leveled, the projection drawings are taken from top to bottom in sequence, a certain frame of projection drawing C-scan is selected as shown in (a) in fig. 5, the contour of FAZ is manually marked on the C-scan, then the filling operation is carried out on the inside of the contour to obtain a white label in (B) in fig. 5, and the corresponding relation between the C-scan marking and the B-scan label image is shown in (a), (B) and (C) in fig. 6. The label image is restored to the original layer position to obtain the label image as shown in fig. 7. Counting the number of label pixels in the label image, multiplying the number of pixels by a resolution parameter to obtain the FAZ volume of the data of 0.014947mm ³ . The three-dimensional structure of the FAZ volume label is shown in fig. 8.

In summary, the FAZ structure is described from a three-dimensional layer for the first time, the form change of the FAZ in the axial direction is shown, the influence of the thickness change of the retina layer structure on the size of the FAZ is integrated, and a certain amount of evaluation index of the FAZ volume is provided. Compared with the original FAZ two-dimensional projection area index, the three-dimensional FAZ volume is beneficial for doctors to intuitively and comprehensively observe the FAZ range. The FAZ volume calculation method provided by the invention is simple and easy to operate, and has important significance for subsequent researches such as FAZ qualitative and quantitative analysis and the like.

Claims

1. A method for calculating the volume of the central avascular zone of the macula based on an OCTA retinal image, the method comprising the steps of:

step 1, collecting an OCTA retina image;

step 2, setting a limit area containing a macular central avascular area, namely an FAZ area; the specific process comprises the following steps:

step 2-2, taking ILM as the upper boundary of the limiting region, and taking a position l mu m below the lower boundary of the OPL as the lower boundary of the limiting region;

step 3, leveling the limited area; the specific process comprises the following steps:

step 3-4, each column in the preliminary leveling limiting area is converted into a fixed length, and the fixed length is the distance length between the lower boundary reference and the upper boundary reference;

step 5, the label image is reduced to the original layer position before the limitation area is not leveled; the specific process comprises the following steps:

reversely executing the reverse operation of the step 3, and restoring the label image into a label image which accords with the original layer structure distribution of the OCTA retina image;

2. The method of claim 1, wherein the value of l is a deep vascular complex DVC lower boundary value defined by the OCTA image acquisition device.

3. The OCTA retinal image-based macular central avascular zone volume calculation method of claim 1, wherein each column in the preliminary flattening limit zone is transformed into a fixed length in step 3-4, specifically by linear interpolation.

4. The method of claim 1, wherein the step 6 of determining the volume of the FAZ region is performed by the following formula:

V＝n×p

5. A macular central avascular zone volume calculation system based on an OCTA retinal image, the system comprising:

the acquisition module is used for acquiring an OCTA retina image;

the area setting module is used for setting a limit area containing a macular central avascular area, namely an FAZ area; the specific process comprises the following steps:

the region transformation module is used for flattening the limited region; the specific process comprises the following steps:

the image restoration module is used for restoring the label image to an original layer position before the limitation area is not leveled; the specific process comprises the following steps:

reversely executing the reverse operation of the region transformation module, and restoring the label image into a label image which accords with the original layer structure distribution of the OCTA retina image;

6. The system of claim 5, wherein the region setting module comprises:

a segmentation unit for segmenting the inner limiting membrane ILM and the outer plexiform layer OPL lower boundary of the OCTA retina image;

7. The system of claim 5 or 6, wherein the region transformation module comprises: