CN116250801B - A method and system for measuring blood oxygen saturation based on ocular images - Google Patents
A method and system for measuring blood oxygen saturation based on ocular imagesInfo
- Publication number
- CN116250801B CN116250801B CN202310257704.6A CN202310257704A CN116250801B CN 116250801 B CN116250801 B CN 116250801B CN 202310257704 A CN202310257704 A CN 202310257704A CN 116250801 B CN116250801 B CN 116250801B
- Authority
- CN
- China
- Prior art keywords
- image
- images
- blood vessels
- oxygen saturation
- blood
- 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.)
- Active
Links
Classifications
-
- 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
- A61B3/14—Arrangements specially adapted for eye photography
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0033—Features or image-related aspects of imaging apparatus, e.g. for MRI, optical tomography or impedance tomography apparatus; Arrangements of imaging apparatus in a room
- A61B5/004—Features or image-related aspects of imaging apparatus, e.g. for MRI, optical tomography or impedance tomography apparatus; Arrangements of imaging apparatus in a room adapted for image acquisition of a particular organ or body part
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0077—Devices for viewing the surface of the body, e.g. camera, magnifying lens
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration or pH-value ; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid or cerebral tissue
- A61B5/1455—Measuring characteristics of blood in vivo, e.g. gas concentration or pH-value ; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid or cerebral tissue using optical sensors, e.g. spectral photometrical oximeters
- A61B5/14551—Measuring characteristics of blood in vivo, e.g. gas concentration or pH-value ; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid or cerebral tissue using optical sensors, e.g. spectral photometrical oximeters for measuring blood gases
- A61B5/14555—Measuring characteristics of blood in vivo, e.g. gas concentration or pH-value ; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid or cerebral tissue using optical sensors, e.g. spectral photometrical oximeters for measuring blood gases specially adapted for the eye fundus
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; 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 OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Public Health (AREA)
- Heart & Thoracic Surgery (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Veterinary Medicine (AREA)
- Radiology & Medical Imaging (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Pathology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- General Engineering & Computer Science (AREA)
- Ophthalmology & Optometry (AREA)
- Computational Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Quality & Reliability (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Algebra (AREA)
- Operations Research (AREA)
- Computational Linguistics (AREA)
Abstract
The embodiment of the invention provides a blood oxygen saturation measuring method and system based on an eye image, and belongs to the technical field of image processing. The method comprises the steps of collecting image information of the same eye of a user under two different shooting wavelengths, respectively obtaining an image with a longer shooting wavelength and an image with a shorter shooting wavelength, positioning respective video disc positions of the two images, performing blood vessel segmentation on each image after performing global registration on the video disc positions of the two images to obtain corresponding blood vessel segmentation images, obtaining segmentation information of arterial blood vessels and venous blood vessels based on the two blood vessel segmentation images, and calculating blood oxygen saturation of the eyes of the user currently based on the segmentation information of the arterial blood vessels and the venous blood vessels. The method solves the problem that the blood oxygen saturation measurement cannot be applied due to low accuracy of distinguishing the ocular blood vessels.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a blood oxygen saturation measuring method based on an eye image and a blood oxygen saturation measuring system based on the eye image.
Background
Blood oxygen saturation is the percentage of the volume of oxyhemoglobin in the blood that is bound by oxygen to the volume of total hemoglobin that can be bound, i.e., the concentration of blood oxygen in the blood, which is an important physiological parameter of the respiratory cycle. Accurate measurement of blood oxygen saturation is a constant concern and research problem in the medical related field. The method widely used at present is to measure the blood oxygen saturation based on an optical density measurement method, and the absorption spectrum of hemoglobin (protein carrying oxygen in blood) is greatly different in two states, namely an oxygen-containing state and a non-oxygen-containing state thereof. In short, the blood oxygen saturation is different, the projected color is also changed, and based on the characteristics, the blood oxygen saturation measurement through the blood vessel image can be realized.
The human eye includes a large number of blood vessels which, because of their ocular structure, make an ideal blood oxygen saturation measurement target (color characteristics are obvious), the accuracy of blood oxygen saturation is very high if blood oxygen saturation measurement can be performed based on the ocular vessels. However, the vein and artery are crossed, which makes the distinction of blood vessels difficult. Because the arterial and venous blood vessel blood oxygen saturation criteria are not the same, the blood oxygen saturation measurements obtained are also not reliable if the venous and arterial blood vessels cannot be accurately distinguished. At present, the eye blood vessel distinction mainly depends on manual labeling, but the labor cost and the identification accuracy required by the mode are not ideal, and the method is the biggest obstacle for preventing the blood oxygen saturation measurement scheme based on the eye image. Aiming at the problem that the blood oxygen saturation measurement cannot be applied due to low accuracy of distinguishing the ocular blood vessels in the existing method, a new blood oxygen saturation measurement method based on the ocular image needs to be created.
Disclosure of Invention
The embodiment of the invention aims to provide a blood oxygen saturation measuring method and system based on an eye image, which at least solve the problem that blood oxygen saturation measurement cannot be applied due to low accuracy of distinguishing blood vessels of eyes in the existing method.
In order to achieve the above purpose, the first aspect of the present invention provides a blood oxygen saturation measurement method based on an eye image, which includes collecting image information of the same eye of a user under two different photographing wavelengths, respectively obtaining an image with a longer photographing wavelength and an image with a shorter photographing wavelength, positioning respective optic disc positions of the two images, performing global registration of the two images based on the optic disc positions of the two images, performing vessel segmentation on each image to obtain a corresponding vessel segmentation image, obtaining segmentation information of arterial vessels and venous vessels based on the two vessel segmentation images, and calculating blood oxygen saturation of the eye of the current user based on the segmentation information of the arterial vessels and the venous vessels.
Alternatively, the two different photographing wavelengths are 570+ -5 nm and 600+ -5 nm, respectively.
Optionally, the positioning of the positions of the video discs of the two images comprises positioning the video disc area in the images based on yolov algorithm and fitting the oval outline of the video disc through external moment, wherein the external moment is a detection frame of yolov algorithm.
Optionally, the global registration of two images based on the disc positions of the two images comprises the steps of taking the image with longer shooting wavelength as a target, moving the image with shorter shooting wavelength for registration, including 1) taking the difference between the center coordinates of the disc positions of the two images to obtain DeltaX and DeltaY, and 2) moving the image with shorter shooting wavelength based on DeltaX and DeltaY, including moving the image with shorter shooting wavelength along an X axis and a Y axis by (+/-) (|DeltaX|+2) and (+ |DeltaY|+2) respectively, wherein the specific movement distance is determined by an evaluation parameter between the image with shorter shooting wavelength and the image with longer shooting wavelength in the moving process, and the evaluation parameter is as follows: wherein, tau is an evaluation parameter, delta is the pixel difference value of the two images; and screening out the image coordinates corresponding to the minimum evaluation parameters in the moving process as final coordinates of the image with shorter shooting wavelength.
The method comprises the steps of obtaining a corresponding blood vessel segmentation image by carrying out blood vessel segmentation on each image, wherein the blood vessel segmentation image comprises the steps of respectively preprocessing two images by adopting Gaussian filtering, segmenting vein blood vessels in the preprocessed image with longer shooting wavelength based on a U2Net algorithm to serve as the blood vessel segmentation image of the image with longer shooting wavelength, enhancing the preprocessed image with shorter shooting wavelength based on a defogging algorithm to obtain the enhanced image with shorter shooting wavelength, and segmenting all the vein blood vessels in the enhanced image with shorter shooting wavelength based on the U2Net algorithm to serve as the blood vessel segmentation image of the image with shorter shooting wavelength.
Optionally, the two-image-based blood vessel segmentation image is used for obtaining segmentation information of arterial blood vessels and venous blood vessels, wherein the blood vessel segmentation image of an image with longer shooting wavelength is used as segmentation information of venous blood vessels, the blood vessel segmentation image of an image with shorter shooting wavelength is registered to the blood vessel segmentation image of an image with longer shooting wavelength, and the blood vessel segmentation image mask image of an image with longer shooting wavelength is subtracted from the blood vessel segmentation image mask image of an image with shorter shooting wavelength to obtain arterial blood vessel segmentation image which is used as segmentation information of arterial blood vessels.
The method comprises the steps of extracting frameworks of vein blood vessels and artery blood vessels, obtaining central lines of the vein blood vessels and the artery blood vessels, traversing the central lines of the vein blood vessels and the artery blood vessels based on preset step length, respectively obtaining vascular boundary pixel values of the vein blood vessels and the artery blood vessels, identifying a yellow spot position in an image with longer shooting wavelength, constructing a rectangular coordinate system by taking a connecting line of a central point of a visual disc position and a central point of the yellow spot position in the image with longer shooting wavelength as an abscissa and taking the central point of the visual disc as an origin, respectively calculating the vascular blood oxygen saturation of four quadrant regions in the rectangular coordinate system and the vascular blood oxygen saturation of the artery blood vessels based on the vascular boundary pixel values of the vein blood vessels and the artery blood oxygen saturation, and carrying out average calculation on the vascular blood oxygen saturation of the vein blood vessels and the arterial blood oxygen saturation of each quadrant to obtain the vascular blood oxygen saturation of the current user.
Optionally, the range of the constructed rectangular coordinate system is an area which is expanded outwards by a distance of 2r-5r by taking an origin as a circle center, wherein r is the radius of the video disc.
The invention provides a blood oxygen saturation measuring system based on eye images, which comprises an acquisition unit, a processing unit and a detection unit, wherein the acquisition unit is used for acquiring image information of the same eye of a user under two different shooting wavelengths, respectively obtaining images with longer shooting wavelengths and images with shorter shooting wavelengths, positioning the respective optic disc positions of the two images, the processing unit is used for carrying out global registration on the two images based on the optic disc positions of the two images, then carrying out blood vessel segmentation on each image to obtain corresponding blood vessel segmentation images, obtaining segmentation information of arterial blood vessels and venous blood vessels based on the two blood vessel segmentation images, and the detection unit is used for calculating the blood oxygen saturation of the eyes of the current user based on the segmentation information of the arterial blood vessels and the venous blood vessels.
In another aspect, the present invention provides a computer readable storage medium having instructions stored thereon, which when executed on a computer, cause the computer to perform the above-described method for measuring blood oxygen saturation based on an eye image.
Through the technical scheme, the method is used for dividing the veins and the arteries based on the display difference of the veins and the arteries in the images with different shooting wavelengths, and a registration method for ensuring the accuracy of division is provided. The blood oxygen saturation measurement is carried out on the basis of veins and arteries obtained through accurate segmentation, the measurement accuracy of the blood oxygen saturation is ensured by a way of improving the blood vessel segmentation accuracy, and the problem that the blood oxygen saturation measurement cannot be applied due to the fact that the ocular blood vessel differentiation accuracy is low in the prior art is solved.
Additional features and advantages of embodiments of the invention will be set forth in the detailed description which follows.
Drawings
The accompanying drawings are included to provide a further understanding of embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain, without limitation, the embodiments of the invention. In the drawings:
FIG. 1 is a flow chart of steps of a method for measuring blood oxygen saturation based on an eye image according to an embodiment of the present invention;
FIG. 2 is a schematic view of the state of an eye blood vessel at different wavelengths according to one embodiment of the present invention;
FIG. 3 is a schematic view showing the effects of capturing a long-wavelength image according to an embodiment of the present invention;
FIG. 4 is a schematic view showing the effects of capturing a shorter wavelength image of a segmented blood vessel image according to one embodiment of the present invention;
fig. 5 is a system configuration diagram of an oxygen saturation measurement system based on an eye image according to an embodiment of the present invention.
Detailed Description
The following describes specific embodiments of the present invention in detail with reference to the drawings. It should be understood that the detailed description and specific examples, while indicating and illustrating the invention, are not intended to limit the invention.
Blood oxygen saturation is the percentage of the volume of oxyhemoglobin in the blood that is bound by oxygen to the volume of total hemoglobin that can be bound, i.e., the concentration of blood oxygen in the blood, which is an important physiological parameter of the respiratory cycle. Accurate measurement of blood oxygen saturation is a constant concern and research problem in the medical related field. The method widely used at present is to measure the blood oxygen saturation based on an optical density measurement method, and the absorption spectrum of hemoglobin (protein carrying oxygen in blood) is greatly different in two states, namely an oxygen-containing state and a non-oxygen-containing state thereof. In short, the blood oxygen saturation is different, the projected color is also changed, and based on the characteristics, the blood oxygen saturation measurement through the blood vessel image can be realized.
The human eye includes a large number of blood vessels which, because of their ocular structure, make an ideal blood oxygen saturation measurement target (color characteristics are obvious), the accuracy of blood oxygen saturation is very high if blood oxygen saturation measurement can be performed based on the ocular vessels. However, the vein and artery are crossed, which makes the distinction of blood vessels difficult. Because the arterial and venous blood vessel blood oxygen saturation criteria are not the same, the blood oxygen saturation measurements obtained are also not reliable if the venous and arterial blood vessels cannot be accurately distinguished. At present, the eye blood vessel distinction mainly depends on manual labeling, but the labor cost and the identification accuracy required by the mode are not ideal, and the method is the biggest obstacle for preventing the blood oxygen saturation measurement scheme based on the eye image.
Aiming at the problem that the blood oxygen saturation measurement cannot be applied due to low accuracy of distinguishing the blood vessels of the eyes in the existing method, the scheme of the invention provides a novel blood oxygen saturation measurement method based on the eye images. The blood oxygen saturation measurement is carried out on the basis of veins and arteries obtained through accurate segmentation, the measurement accuracy of the blood oxygen saturation is ensured by a way of improving the blood vessel segmentation accuracy, and the problem that the blood oxygen saturation measurement cannot be applied due to the fact that the ocular blood vessel differentiation accuracy is low in the prior art is solved.
Fig. 1 is a flowchart of a method for measuring blood oxygen saturation based on an eye image according to an embodiment of the present invention. As shown in fig. 1, an embodiment of the present invention provides a blood oxygen saturation measurement method based on an eye image, the method including:
and S10, collecting image information of the same eye of a user under two different shooting wavelengths, respectively obtaining an image with a longer shooting wavelength and an image with a shorter shooting wavelength, and positioning the positions of the video discs of the two images.
In particular, the scheme of the invention also executes an optical density measurement method based on image analysis to measure the blood oxygen saturation, so that the object of the invention is eye image information. The vein and artery blood vessels have larger color distinction, the vein and artery distinction can be processed through image acquisition under different wavelengths by utilizing the characteristic and the corresponding relation between the color and the wavelength, and the artery and vein distinction is carried out based on the distinguished image information so as to obtain accurate vein targets and artery targets. As shown in fig. 2, wherein fig. 1 is an eye image at 570 wavelengths and fig. 2 is an eye image at 600 wavelengths. It is evident from a comparison of the two images that all vessels in the 570 wavelength image are realistic, whereas only venous vessels in the 600 wavelength image are prominent. Theoretically, arterial vessels can be obtained based on the difference between all images and only images of veins.
Preferably, the two different shooting wavelengths are 570+ -5 nm and 600+ -5 nm, respectively.
Specifically, the former is to record all of the veins, and the latter is to record veins therein.
Preferably, the positioning of the respective optic disc positions comprises positioning the optic disc region in the image based on yolov algorithm and fitting the oval outline of the optic disc through external moment, wherein the external moment is a detection frame of yolov algorithm.
Specifically, although the artery and vein segmentation can be performed through two images with different wavelengths, it cannot be ensured that the two images photographed must be matched in a global scope, that is, if the accurate registration of the two images cannot be achieved, the arterial vessel obtained based on the difference between the two images cannot be achieved. The width of the ocular blood vessel is smaller, and accurate registration of the two images needs to be ensured, so that the coordinates of the actually same point in the two images can be ensured to be the same under the same coordinate system. Based on this, the solution of the invention proposes a corresponding registration solution. Because the disc information of the same eye is unchanged, two image registration can be performed with the disc as a reference, based on which disc positioning is preferred.
At present, the video disc positioning is generally two schemes, namely searching the brightest area in a picture, finding the center of the brightest area is the center of the video disc, the scheme is easy to find the position when shooting is irregular and has reflection, and determining the radius of the video disc is inaccurate, the video disc is segmented through deep learning, the center and the radius are found according to the segmentation result, the method is completely dependent on the accuracy of video disc segmentation, the boundary of part of the video disc of the picture is unclear, the segmentation effect is difficult to guarantee, and deviation can also exist when the center is determined. The scheme of the invention detects the position of the video disc through yolov algorithm, detects the external moment of the frame, namely the video disc, fits the oval video disc through the external moment, and the center, namely the center of the oval, has the radius of the distance from the center of the oval to the long side of the oval.
And step S20, performing global registration on the two images based on the video disc positions of the two images, and performing vessel segmentation on each image to obtain a corresponding vessel segmentation image.
In particular, it is known that, in order to achieve a precise segmentation of the venous and arterial blood vessels by means of two images, a precise registration of the two images must be ensured. The scheme of the invention uses a video disc as a reference object, takes an image with longer shooting wavelength as a target, and moves the image with shorter shooting wavelength for registration, and comprises the steps of taking the difference between the center coordinates of the video disc positions positioned by two images to obtain delta X and delta Y, and moving the image with shorter shooting wavelength based on the delta X and the delta Y, wherein the image with shorter shooting wavelength moves (+/-) (|delta X|+2) and (+/-) (|delta Y|+2) along the X axis and the Y axis respectively, and the specific moving distance is determined by the evaluation parameters between the image with shorter shooting wavelength and the image with longer shooting wavelength in the moving process, and the evaluation parameters are as follows:
wherein, tau is an evaluation parameter, delta is the pixel difference value of the two images; And screening out the image coordinates corresponding to the minimum evaluation parameters in the moving process as final coordinates of the image with shorter shooting wavelength.
The structural similarity (Structural Similarity, SSIM) is an index for measuring the similarity of two images. The index was first proposed by the image and video engineering laboratory (Laboratory for Image and Video Engineering) at the university of texas, osding. Two images used in SSIM, one being an uncompressed undistorted image and the other being a distorted image. As an implementation of the structural similarity theory, the structural similarity index defines structural information from the perspective of image composition as being independent of brightness, contrast, reflects properties of object structures in a scene, and models distortion as a combination of three different factors of brightness, contrast, and structure. The mean value is used as an estimate of brightness, the standard deviation is used as an estimate of contrast, and the covariance is used as a measure of the degree of structural similarity. According to the scheme, traversing movement is carried out on the (+/-) (|DeltaX|+2) and (+/-) (|DeltaY|+2), evaluation parameters after each moving point are counted, the smaller the evaluation parameters are, the higher the matching degree of the evaluation parameters and the evaluation parameters is, and the registration of two images can be accurately realized by finding the smallest evaluation parameter.
Further, the method for obtaining the blood vessel segmentation images by carrying out blood vessel segmentation in the corresponding images based on the respective images comprises the steps of carrying out Gaussian filtering processing on the two images respectively to obtain preprocessed images, and segmenting vein blood vessels in the preprocessed images with longer shooting wavelength based on a U2Net algorithm as the blood vessel segmentation images of the images with longer shooting wavelength, as shown in figure 3;
Further, as shown in FIG. 4, the image with shorter shooting wavelength after pretreatment is enhanced based on defogging algorithm to obtain the image with shorter shooting wavelength after enhancement, and all the vessels are segmented in the image with shorter shooting wavelength based on U2Net algorithm to be used as vessel segmentation image of the image with shorter shooting wavelength.
Specifically, since the image noise of the directly acquired image is large, which can count the influence on the pixels of the subsequent vessel segmentation, and noise needs to be removed, image preprocessing needs to be performed on both images to eliminate the image noise. The image denoising processing is carried out based on Gaussian filtering, the Gaussian filtering is linear smoothing filtering, and the method is suitable for eliminating Gaussian noise and widely applied to the noise reduction process of image processing. Each pixel in the image is scanned by a template (or convolution, mask), and the weighted average gray value of the pixels in the neighborhood determined by the template is used to replace the value of the central pixel point of the template.
After the image preprocessing is completed, the display characteristics of the blood vessels in the image are obvious as known from fig. 2, and the blood vessels can be simply segmented by using a U2Net algorithm based on the characteristics. However, since the brightness of the image with a shorter shooting wavelength is low, if the blood vessel is divided directly, it is easy to cause a division error due to a darker scene, and the image with a longer shooting wavelength does not have the problem. In order to accurately segment the blood vessels in the image having a short imaging wavelength, it is necessary to perform an enhancement operation for the image in advance. The scheme of the invention is based on the defogging algorithm to enhance the preprocessed image with shorter shooting wavelength, and obtain the enhanced image with shorter shooting wavelength. Such as using MSCNN algorithm, dehazeNet algorithm, dark channel prior algorithm, etc., without limitation.
And step S30, obtaining segmentation information of the arterial blood vessel and the venous blood vessel based on the two blood vessel segmentation images.
Specifically, after step S20, two blood vessel segmentation images are obtained, which are respectively a venous blood vessel segmentation image corresponding to an image with a longer shooting wavelength and a full blood vessel image segmentation image corresponding to an image with a shorter shooting wavelength. Where venous vessel segmentation information has been obtained, it is necessary to obtain segmentation information of arterial vessels based on both. Specifically, a blood vessel segmentation image of an image with a longer shooting wavelength is used as segmentation information of vein blood vessels, the blood vessel segmentation image of an image with a shorter shooting wavelength is registered to the blood vessel segmentation image of the image with a longer shooting wavelength, and an arterial blood vessel segmentation image is obtained as segmentation information of arterial blood vessels based on a blood vessel segmentation image mask image of the image with the shorter shooting wavelength minus a blood vessel segmentation image mask image of the image with the longer shooting wavelength.
And S40, calculating the blood oxygen saturation of the eyes of the current user based on the segmentation information of the arterial blood vessel and the venous blood vessel.
The method comprises the steps of extracting skeletons of vein blood vessels and artery blood vessels, obtaining central lines of the vein blood vessels and the artery blood vessels, traversing the central lines of the vein blood vessels and the artery blood vessels based on preset step length, obtaining blood vessel boundary pixel values of the vein blood vessels and the artery blood vessels respectively, identifying a macula position in an image with longer shooting wavelength, constructing a rectangular coordinate system by taking a connecting line of a central point of a visual disk position and a central point of a macula position in the image as an abscissa and taking the central point of the visual disk as an origin, respectively calculating vein blood vessel oxygen saturation and artery blood vessel oxygen saturation of four quadrant regions in the rectangular coordinate system based on the blood vessel boundary pixel values of the vein blood vessels and the artery blood vessel, and carrying out average calculation based on the vein blood vessel oxygen saturation and the artery blood vessel oxygen saturation of each quadrant to obtain the vein blood vessel oxygen saturation and the artery blood vessel oxygen saturation of eyes of a current user.
In the embodiment of the invention, skeleton extraction is performed according to the segmented arteries and veins, skeleton intersection points are searched, and the obtained skeleton intersection points are divided into different sections according to the intersection points for processing, so that the influence of inaccurate value at the intersection points is eliminated. And then according to the small image formed by the frame, the pixel value on the boundary of the blood vessel is obtained by the transition value of the pixel mean value from left to right, and compared with the boundary pixels of the blood vessel which are directly segmented conventionally, the method can ensure that the obtained pixel is necessarily the value on the blood vessel, avoid errors caused by inaccurate segmentation of the blood vessel, and exclude the reflection point on the blood vessel. The venous blood vessel pixels are distributed in a U shape, and the arteries are distributed in a W shape.
Specifically, the whole blood vessel is traversed, points are taken along the way, each point is found out, a rectangle (for example, 20 x 10 pixels in size) is cut along the gradient direction of the blood vessel, the rectangle is used as a recognition frame, and the boundary of the blood vessel is found in the recognition frame. In the identification frame, the transition condition of the pixel gray values in the rectangle from left to right is collected, and boundary pixel points are found according to the gradient descending condition.
Preferably, the rectangular coordinate system is constructed in the range of expanding the area by 2r-5r outwards with the origin as the center, wherein r is the radius of the video disc.
Further, the center point is determined by performing a macula position location based on yolov. Establishing a rectangular coordinate system according to 600 wavelength macula lutea and the position of the optic disc, setting a 2-5-time radius area of the optic disc by taking the optic disc as the center of a circle, dividing four quadrants into an upper nasal quadrant, a lower nasal quadrant, an upper temporal quadrant and a lower temporal quadrant, respectively calculating the blood oxygen saturation of each quadrant, and then carrying out average calculation based on the venous blood oxygen saturation and the arterial blood oxygen saturation of each quadrant to obtain the venous blood oxygen saturation and the arterial blood oxygen saturation of the eyes of the current user.
The principle of calculating the blood oxygen saturation is optical density measurement, and the 570-wavelength image and 600-wavelength image of the scheme of the invention are taken as examples, and the Optical Density (OD) represents the attenuation of the emergent light intensity I relative to the incident light intensity I0. The algorithm scans the intravascular pixel points to find out the intravascular minimum light intensity I, then determines the extravascular average brightness I0 according to the blood vessel diameter, and then calculates the optical density according to the formula OD=log (I0/I). Finally, the optical density ratio ODR is calculated according to the formula odr=od 600/OD 570. Because the optical density ratio is directly related to the blood oxygen saturation (SO 2), the corresponding blood oxygen saturation can be directly obtained based on a table look-up mode after the optical density ratio is obtained.
According to the scheme, the 4 quadrants are arranged to respectively calculate the blood oxygen saturation, so that calculation errors caused by regional differences can be avoided, the calculation regions are reduced, the same calculation conditions in each calculation region can be ensured, the calculation accuracy is ensured, and finally, the whole blood oxygen saturation is obtained through each region, so that the calculation accuracy of the whole blood oxygen saturation is improved.
Fig. 5 is a system configuration diagram of an oxygen saturation measurement system based on an eye image according to an embodiment of the present invention. As shown in fig. 5, the embodiment of the invention provides a blood oxygen saturation measuring system based on eye images, which comprises an acquisition unit, a processing unit and a detection unit, wherein the acquisition unit is used for respectively acquiring image information of the same eye of a user under two different shooting wavelengths and positioning respective video disc positions, the processing unit is used for carrying out global registration on the two images based on the video disc positions of the two images and carrying out blood vessel segmentation in corresponding images based on the respective images to obtain respective blood vessel segmentation images, the segmentation information of arterial blood vessels and venous blood vessels is obtained based on the blood vessel segmentation images of the two images, and the detection unit is used for calculating the blood oxygen saturation of the eyes of the current user based on the segmentation information of the arterial blood vessels and the venous blood vessels.
The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium stores instructions, and when the computer is run on the computer, the computer is caused to execute the blood oxygen saturation measuring method based on the eye image.
Those skilled in the art will appreciate that all or part of the steps in a method for implementing the above embodiments may be implemented by a program stored in a storage medium, where the program includes several instructions for causing a single-chip microcomputer, chip or processor (processor) to perform all or part of the steps in a method according to the embodiments of the invention. The storage medium includes a U disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, an optical disk, or other various media capable of storing program codes.
The alternative embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the embodiments of the present invention are not limited to the specific details of the above embodiments, and various simple modifications may be made to the technical solutions of the embodiments of the present invention within the scope of the technical concept of the embodiments of the present invention, and all the simple modifications belong to the protection scope of the embodiments of the present invention. In addition, the specific features described in the above embodiments may be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the various possible combinations of embodiments of the invention are not described in detail.
In addition, any combination of the various embodiments of the present invention may be made, so long as it does not deviate from the idea of the embodiments of the present invention, and it should also be regarded as what is disclosed in the embodiments of the present invention.
Claims (7)
1. A method for measuring blood oxygen saturation based on an eye image, the method comprising:
Acquiring image information of the same eye of a user under two different shooting wavelengths, respectively obtaining an image with a longer shooting wavelength and an image with a shorter shooting wavelength, and positioning the positions of the video discs of the two images;
performing global registration of the two images based on the video disc positions of the two images, and performing vessel segmentation on each image to obtain corresponding vessel segmentation images,
The method comprises the steps of carrying out vascular segmentation on each image to obtain corresponding vascular segmented images, preprocessing the two images by Gaussian filtering, segmenting vein blood vessels in the preprocessed image with longer shooting wavelength based on a U2Net algorithm to serve as vascular segmented images of the image with longer shooting wavelength, enhancing the preprocessed image with shorter shooting wavelength based on a defogging algorithm to obtain an enhanced image with shorter shooting wavelength, segmenting all the vein blood vessels in the enhanced image with shorter shooting wavelength based on the U2Net algorithm, and serving as vascular segmented images of the image with shorter shooting wavelength;
The method comprises the steps of obtaining segmentation information of arterial blood vessels and venous blood vessels based on blood vessel segmentation images of two images, registering blood vessel segmentation images of images with shorter shooting wavelengths to blood vessel segmentation images of images with longer shooting wavelengths, subtracting blood vessel segmentation image mask images of images with longer shooting wavelengths from blood vessel segmentation image mask images of images with shorter shooting wavelengths to obtain arterial blood vessel segmentation images as segmentation information of arterial blood vessels, wherein the blood vessel segmentation images of images with longer shooting wavelengths are taken as segmentation information of venous blood vessels;
obtaining segmentation information of arterial blood vessels and venous blood vessels based on the two blood vessel segmentation images;
calculating the blood oxygen saturation of the eyes of the current user based on the segmentation information of the arterial blood vessel and the venous blood vessel;
The method comprises the steps of extracting frameworks of vein blood vessels and artery blood vessels, obtaining central lines of the vein blood vessels and the artery blood vessels, traversing the central lines of the vein blood vessels and the artery blood vessels based on preset step length, respectively obtaining vascular boundary pixel values of the vein blood vessels and the artery blood vessels, identifying a yellow spot position in an image with longer shooting wavelength based on yolov algorithm, taking a connecting line of a central point of a video disc position and a central point of the yellow spot position in the image with longer shooting wavelength as a horizontal coordinate, and taking the central point of the video disc as an origin, constructing a rectangular coordinate system, respectively calculating and obtaining vascular oxygen saturation of each single vein blood vessel and each single artery blood vessel based on the vascular boundary pixel values of the vein blood vessels and the arterial blood vessels by utilizing an optical density measurement method, further respectively calculating the vascular oxygen saturation of four quadrant regions in the rectangular coordinate system, and further respectively calculating average values of the vascular oxygen saturation of the vein blood vessels and the arterial blood oxygen saturation of each quadrant, and obtaining the vascular oxygen saturation of the current eye of the user.
2. The method of claim 1, wherein the two different photographing wavelengths are 570+ -5 nm and 600+ -5 nm, respectively.
3. The method of claim 1, wherein locating the respective disc positions of the two images comprises:
And positioning a video disc area in the image based on yolov algorithm, and fitting the oval outline of the video disc through external moment, wherein the external moment is a detection frame of yolov algorithm.
4. The method of claim 1, wherein performing global registration of two images based on the disc positions of the two images includes moving the image with the shorter shooting wavelength for registration with the longer shooting wavelength as a target, and includes:
1) Obtaining by taking difference of central coordinates of video disc positions of two images And;
2) Based onAndMoving an image of a shorter shooting wavelength, comprising:
moving the image with shorter shooting wavelength along the x-axis and the y-axis respectively AndThe specific moving distance is determined by an evaluation parameter between an image with a shorter shooting wavelength and an image with a longer shooting wavelength in the moving process, wherein the evaluation parameter is as follows: Wherein, the method comprises the steps of, As an evaluation parameter; Is the pixel difference of two images; Structural similarity of two images;
and screening out the image coordinates corresponding to the minimum evaluation parameters in the moving process as final coordinates of the image with shorter shooting wavelength.
5. The method of claim 1, wherein the rectangular coordinate system is constructed in a range of expanding the region by a distance of 2r-5r outwards with the origin as the center;
where r is the radius of the disc.
6. An eye image based blood oxygen saturation measurement system for performing the eye image based blood oxygen saturation measurement method of any one of claims 1-5, the system comprising:
the acquisition unit is used for acquiring image information of the same eye of a user under two different shooting wavelengths, respectively obtaining an image with a longer shooting wavelength and an image with a shorter shooting wavelength, and positioning the positions of the video discs of the two images;
A processing unit for:
Performing global registration of the two images based on the video disc positions of the two images, and then performing vessel segmentation on each image to obtain a corresponding vessel segmentation image;
obtaining segmentation information of arterial blood vessels and venous blood vessels based on the two blood vessel segmentation images;
And the detection unit is used for calculating the blood oxygen saturation of the eyes of the current user based on the segmentation information of the arterial blood vessel and the venous blood vessel.
7. A computer readable storage medium having instructions stored thereon, which when run on a computer causes the computer to perform the eye image based blood oxygen saturation measurement method of any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310257704.6A CN116250801B (en) | 2023-03-16 | 2023-03-16 | A method and system for measuring blood oxygen saturation based on ocular images |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310257704.6A CN116250801B (en) | 2023-03-16 | 2023-03-16 | A method and system for measuring blood oxygen saturation based on ocular images |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116250801A CN116250801A (en) | 2023-06-13 |
| CN116250801B true CN116250801B (en) | 2025-12-12 |
Family
ID=86684310
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310257704.6A Active CN116250801B (en) | 2023-03-16 | 2023-03-16 | A method and system for measuring blood oxygen saturation based on ocular images |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116250801B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117351009B (en) * | 2023-12-04 | 2024-02-23 | 江苏富翰医疗产业发展有限公司 | Method and system for generating blood oxygen saturation data based on multispectral fundus image |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113269737A (en) * | 2021-05-17 | 2021-08-17 | 西安交通大学 | Method and system for calculating diameter of artery and vein of retina |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0871045A (en) * | 1994-09-02 | 1996-03-19 | Canon Inc | Fundus examination device |
| EP1259791B1 (en) * | 2000-05-02 | 2013-11-13 | Cas Medical Systems, Inc. | Method for non-invasive spectrophotometric blood oxygenation monitoring |
| JP4854389B2 (en) * | 2006-06-15 | 2012-01-18 | 株式会社トプコン | Spectral fundus measuring apparatus and measuring method thereof |
| JP6922152B2 (en) * | 2015-10-21 | 2021-08-18 | 株式会社ニデック | Ophthalmology analyzer, ophthalmology analysis program |
| CN108803994B (en) * | 2018-06-14 | 2022-10-14 | 四川和生视界医药技术开发有限公司 | Retinal blood vessel management method and retinal blood vessel management device |
| US12193815B2 (en) * | 2019-08-12 | 2025-01-14 | Oregon Health & Science University | Systems and methods for capillary oximetry using optical coherence tomography |
| CN112716446B (en) * | 2020-12-28 | 2023-03-24 | 深圳硅基智能科技有限公司 | Method and system for measuring pathological change characteristics of hypertensive retinopathy |
-
2023
- 2023-03-16 CN CN202310257704.6A patent/CN116250801B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113269737A (en) * | 2021-05-17 | 2021-08-17 | 西安交通大学 | Method and system for calculating diameter of artery and vein of retina |
Non-Patent Citations (2)
| Title |
|---|
| 先永利.基于眼底相机的双波长视网膜血氧仪若干问题研究及应用.《中国博士学位论文全文数据库 (工程科技Ⅱ辑)》.2018,(第2018年第09期),C030-4. * |
| 基于眼底相机的双波长视网膜血氧仪若干问题研究及应用;先永利;《中国博士学位论文全文数据库 (工程科技Ⅱ辑)》;20180915(第2018年第09期);C030-4 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116250801A (en) | 2023-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111340789B (en) | Fundus retina blood vessel identification and quantification method, device, equipment and storage medium | |
| Abbasi-Sureshjani et al. | Automatic detection of vascular bifurcations and crossings in retinal images using orientation scores | |
| CN108961334B (en) | A method for measuring retinal vessel wall thickness based on image registration | |
| CN104102899B (en) | Retinal vessel recognition methods and device | |
| CN111738992B (en) | Lung lesion area extraction method, device, electronic equipment and storage medium | |
| CN116250801B (en) | A method and system for measuring blood oxygen saturation based on ocular images | |
| CN110598635A (en) | Method and system for face detection and pupil positioning in continuous video frames | |
| Zhang et al. | Optic disc localization by projection with vessel distribution and appearance characteristics | |
| US20250054149A1 (en) | Method of measuring palpebral fissure height, and a device and a storage medium for the same | |
| Michelson et al. | Functional imaging of the retinal microvasculature by scanning laser Doppler flowmetry | |
| EP4574017A1 (en) | Quantitative evaluation method for conjunctival congestion, apparatus, and storage medium | |
| CN116630237A (en) | Image quality detection method and related device, electronic equipment and storage medium | |
| CN112712521B (en) | Automatic positioning method of fundus optic disk based on global gradient search and storage medium thereof | |
| CN110874597B (en) | Blood vessel feature extraction method, device and system for fundus image and storage medium | |
| Jiang et al. | Blood vessel tracking in retinal images | |
| CN117045178B (en) | An intelligent image quality monitoring method and system for fluorescence endoscopic imaging | |
| CN118446979A (en) | Blood vessel accurate positioning method and system for optimizing blood vessel diameter | |
| US10083507B2 (en) | Method for the analysis of image data representing a three-dimensional volume of biological tissue | |
| KR102257998B1 (en) | Apparatus and method for cell counting | |
| CN117351009B (en) | Method and system for generating blood oxygen saturation data based on multispectral fundus image | |
| Roy et al. | A novel computer aided quantification method of focal arteriolar narrowing using colour retinal image | |
| CN111210436A (en) | Lens segmentation method, device and storage medium | |
| CN121053119B (en) | A deep learning-based method, system, and terminal for bronchoscopic image processing. | |
| HK40023085A (en) | Method, device, equipment and storage medium for recognizing and quantifying fundus retinal blood vessels | |
| Fraz et al. | A model based approach for vessel caliber measurement in retinal 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |