A carotid stenosis-induced microcirculation disorder detection system

Abstract

The present application relates to the technical field of medical image processing, and discloses a system for detecting microcirculation disorder caused by carotid stenosis, comprising: a data acquisition module for acquiring a superficial retinal layer image and a deep retinal layer image of a person to be detected; an image analysis module for extracting blood vessel parameters after sequentially performing binarization and skeletonization on the superficial retinal layer image and the deep retinal layer image respectively; and a detection module for obtaining a carotid ultrasound measurement index based on the blood vessel parameters by using a regression model of the blood vessel parameters and the carotid ultrasound measurement index, and determining the microcirculation disorder category to which the person to be detected belongs. The system can non-invasively detect the state of microcirculation of a patient with carotid stenosis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical image processing, in particular to a system for detecting microcirculation disorder caused by carotid artery stenosis. BACKGROUND

[0002] The statements in this section merely provide background information related to the present application and do not necessarily constitute the prior art.

[0003] The main feature of carotid artery stenosis (CAS) is carotid intima atherosclerosis and plaque formation, which is an important cause of chronic cerebral ischemic disease, and the incidence is about 20-30% of all ischemic stroke patients. With the aggravation of population aging, the incidence of asymptomatic carotid artery stenosis has increased significantly, which is insidious in onset and has a slow and continuous impact on the microcirculation system, until chronic hypoperfusion injury occurs in the brain, leading to persistent cognitive and neurological dysfunction. Therefore, early detection of microcirculation changes caused by CAS is crucial for early diagnosis, risk stratification, early intervention and disease management of hypoperfusion brain injury.

[0004] At present, the commonly used examination methods for CAS in clinical practice include carotid ultrasound, computed tomography angiography, magnetic resonance angiography, digital subtraction angiography imaging technology, molecular imaging technology, such as 18F-fluorodeoxyglucose (FDG) PET / CT, which can detect carotid plaque properties, vascular stenosis degree and vessel wall stiffness, and early detection of carotid stenosis or carotid plaque, but it cannot evaluate the changes of microvessels caused by carotid stenosis. The diagnostic criteria of brain MRI and CT for cerebral hypoperfusion mainly rely on morphological changes, which are relatively lagging behind compared with the changes of cerebral vascular structure. Positron emission computed tomography and single photon emission computed tomography can detect brain perfusion, but they are expensive and complex, and are not widely used in clinical practice. Transcranial Doppler can measure intracranial arterial blood flow and blood flow velocity, but the skull bone mineralization degree of the elderly is high, which seriously attenuates the ultrasound, and the positioning accuracy and repeatability of transcranial Doppler are poor. The measurement results have only reference value for judging the perfusion changes caused by carotid stenosis. Although non-invasive fundus photography, ocular ultrasound examination, retinal oxygen saturation detector and fluorescein and / or indocyanine green fundus angiography examination can be used to evaluate the ocular blood circulation, they can only provide qualitative and estimated information, and cannot accurately show the subtle changes of retinal anatomical structure. SUMMARY

[0005] In order to solve the problems of the prior art, the present application provides a system for detecting microcirculation disorder caused by carotid artery stenosis, which can non-invasively detect the state of microcirculation of carotid artery stenosis patients through superficial retinal layer images and deep retinal layer images.

[0006] In a first aspect, the present application provides a carotid stenosis-induced microcirculation disorder detection system;

[0007] A carotid stenosis-induced microcirculation disorder detection system comprises:

[0008] A data acquisition module is configured to acquire a superficial retinal layer image and a deep retinal layer image of a subject to be detected.

[0009] An image analysis module is configured to extract blood vessel parameters by sequentially performing binarization and skeletonization on the superficial retinal layer image and the deep retinal layer image, respectively.

[0010] A detection module is configured to obtain carotid ultrasound measurement indicators based on the blood vessel parameters and a regression model of the blood vessel parameters and the carotid ultrasound measurement indicators, and determine a microcirculation disorder category to which the subject to be detected belongs.

[0011] Further, the image analysis module is further configured to perform standardization and cropping on the superficial retinal layer image and the deep retinal layer image, respectively, in sequence before binarization.

[0012] Further, the blood vessel parameters comprise average blood vessel density, skeleton density, and fractal dimension.

[0013] Further, the fractal dimension is obtained based on the binarized image by using a box counting method.

[0014] Further, the carotid ultrasound measurement indicators comprise carotid intima-media thickness, carotid artery plaque area, and carotid artery diameter.

[0015] Further, the regression model is a linear regression model.

[0016] Further, the system further comprises a model construction module configured to construct the regression model of the blood vessel parameters and the carotid ultrasound measurement indicators, and specifically configured to:

[0017] acquire blood vessel parameters and carotid ultrasound measurement indicators of a plurality of samples, and group the samples;

[0018] based on the blood vessel parameters and the carotid ultrasound measurement indicators of all samples, perform partial correlation analysis of the carotid ultrasound measurement indicators and the blood vessel parameters according to the sample grouping, and construct the regression model of the blood vessel parameters and the carotid ultrasound measurement indicators according to the partial correlation analysis result.

[0019] Further, the superficial retinal layer image and the deep retinal layer image are obtained by scanning the macular fovea of the subject to be detected as the center by using an optical coherence tomography blood vessel imaging technology.

[0020] In a second aspect, the present application also provides an electronic device comprising:

[0021] a memory for non-transitorily storing computer readable instructions; and

[0022] a processor for running the computer readable instructions,

[0023] wherein the computer readable instructions, when run by the processor, perform the following steps:

[0024] obtaining a superficial retinal layer image and a deep retinal layer image of a subject to be detected;

[0025] extracting a blood vessel parameter after binarization and skeletonization of the superficial retinal layer image and the deep retinal layer image in sequence, respectively;

[0026] based on the blood vessel parameter, obtaining a carotid artery ultrasound measurement index by using a regression model of the blood vessel parameter and the carotid artery ultrasound measurement index, and determining a microcirculation disorder to which the subject to be detected belongs.

[0027] In a third aspect, the present application also provides a storage medium for non-transitorily storing computer readable instructions, wherein the non-transitory computer readable instructions, when executed by a computer, perform the following steps:

[0028] obtaining a superficial retinal layer image and a deep retinal layer image of a subject to be detected;

[0029] extracting a blood vessel parameter after binarization and skeletonization of the superficial retinal layer image and the deep retinal layer image in sequence, respectively;

[0030] based on the blood vessel parameter, obtaining a carotid artery ultrasound measurement index by using a regression model of the blood vessel parameter and the carotid artery ultrasound measurement index, and determining a microcirculation disorder to which the subject to be detected belongs.

[0031] Compared with the prior art, the present application has the following beneficial effects:

[0032] The microcirculation disorder detection system caused by carotid artery stenosis of the present application can sensitively detect the progress of carotid artery stenosis and the related changes of microcirculation by using the superficial retinal layer image and the deep retinal layer image of the subject to be detected, can directly and non-invasively reflect the state of microcirculation of the patient with carotid artery stenosis, and can simply, sensitively and repeatedly evaluate the progress degree of microcirculation disorder. BRIEF DESCRIPTION OF DRAWINGS

[0033] The drawings constituting a part of the specification of the present application are used to provide a further understanding of the present application, the illustrative embodiments of the present application and the description thereof are used to explain the present application, and do not constitute an improper limitation on the present application.

[0034] Figure 1 A carotid stenosis caused by microcirculation disorder detection system structure diagram for example one;

[0035] Figure 2 A surface retinal layer image schematic diagram for example one;

[0036] Figure 3 A deep retinal layer image schematic diagram for example one;

[0037] Figure 4 A binarized image schematic diagram of the surface retinal layer image for example one;

[0038] Figure 5 A binarized image schematic diagram of the deep retinal layer image for example one;

[0039] Figure 6 A skeleton diagram schematic diagram of the surface retinal layer image for example one;

[0040] Figure 7 A skeleton diagram schematic diagram of the deep retinal layer image for example one. DETAILED DESCRIPTION

[0041] It should be noted that the following detailed description is exemplary in nature and is intended to provide further description of the application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0042] The embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0043] All data acquisition of the present embodiment is based on the compliance with laws and regulations and user consent, and the legal application of data.

[0044] Term explanation:

[0045] Optical coherence tomography angiography (OCTA): a new non-invasive, high-resolution blood vessel network in vivo imaging technology, using red blood cells moving in capillaries as a contrast mechanism, tracking retinal blood flow to clearly image retinal superficial, deep retinal and choroidal capillary microcirculation structure, and can provide high-resolution three-dimensional microvessel image information for healthy people and patients with various retinal and choroidal diseases.

[0046] Carotid artery stenosis (CAS): the main feature is that the intima atherosclerosis and plaque formation lead to the stenosis of carotid artery lumen, which is an important cause of chronic cerebral ischemic disease, has a high incidence in the middle-aged and elderly, and mostly occurs in the carotid bifurcation and the initial segment of the internal carotid artery.

[0047] Fractal dimension (FD) is a measure of the irregularity of complex shapes, reflecting the effectiveness of complex shapes occupying space, and describing the intrinsic shape of the object. The FD of the vascular tree is a measure of the scale-invariant vascular branching pattern, which has been applied to assess muscle, myocardial ischemia, skin, mesenteric and tumor angiogenesis microvascular perfusion. The FD of retinal microcirculation is a comprehensive measure of the natural self-similar pattern of retinal vascular structure, which can be reduced due to the loss of microvessels or the occlusion of small caliber vessels.

[0048] Embodiment one

[0049] The embodiment provides a microcirculation disorder detection system caused by carotid artery stenosis.

[0050] As shown in Figure 1 , a microcirculation disorder detection system caused by carotid artery stenosis comprises:

[0051] A data acquisition module is configured to acquire a superficial retinal layer image and a deep retinal layer image of a to-be-detected person. The superficial retinal layer image and the deep retinal layer image are both obtained by scanning the macular fovea of the to-be-detected person as the center by using an optical coherence tomography blood vessel imaging technology.

[0052] Specifically, the data acquisition module uses the optical coherence tomography blood vessel imaging technology (for example, the RTVue-XR optical coherence tomography instrument of Optovue) to scan the macular area twice in the x-axis and y-axis (6x6mm) with the macular fovea of the to-be-detected person as the center, and collect high-definition images (signal intensity greater than 40) of the superficial and deep retinal capillary networks. The superficial retinal layer (SRL) image is an image from 3μm below the internal limiting membrane to 15μm below the internal limiting membrane. The deep retinal layer (DRL) image is an image from 15μm below the internal limiting membrane to 70μm below the internal limiting membrane. Before entering the image analysis module, it is checked and confirmed that the blood vessels in the superficial retinal layer image and the deep retinal layer image are well connected and have no obvious artifacts. Figure 2 and Figure 3 As shown in

[0053] An image analysis module is configured to extract blood vessel parameters from the superficial retinal layer image and the deep retinal layer image after binarization and skeletonization, respectively. The blood vessel parameters include average vessel density (VD), skeleton density (SD) and fractal dimension (FD). The image analysis module is specifically configured to:

[0054] (1) Construct a macro using Image J image processing software to standardize, crop and binarize the superficial retinal layer image and the deep retinal layer image, respectively, to obtain a binarized image.

[0055] The specific process of binarization is as follows:

[0056] Set the automatic threshold for the picture with black background. The command in Image J is setAutoThreshold(“Default dark”);

[0057] After setting the automatic threshold, obtain the lower limit and upper limit of the threshold of each picture. The command in Image J is getThreshold(lower, upper);

[0058] Subtract 5 from the lower limit, and then set the threshold with the new upper limit and lower limit. The command in Image J is setThreshold(lower-5, upper);

[0059] Set the image as black background. The command in Image J is setOption(“BlackBackground”, false);

[0060] Obtain the binarized image. The command in Image J is run(“Convert to Mask”).

[0061] (2) Based on each binarized image, calculate the blood vessel parameters:

[0062] Measure the vessel density (VD), which represents the proportion of the blood vessel network area in the total image area (excluding the area without blood vessels) in the planar image (binarized image), i.e. measure the vessel density = blood vessel network area / (total image area - area without blood vessels); as shown in Figure 4 and Figure 5 The binarized images of the superficial retinal layer image and the deep retinal layer image, respectively.

[0063] A skeleton graph of the binarized image is acquired, and a blood vessel skeleton density (SD) is calculated based on the skeleton graph. Specifically, based on the binarized image, a blood vessel skeletonization is performed to obtain the skeleton graph, which can be used to measure the blood vessel skeleton density (SD). The command in Image J is run("Skeletonize"), and SD represents the total blood vessel length of the measured image (binarized image). As shown in FIGS. Figure 6 and Figure 7 The skeleton graphs of the superficial retinal layer image and the deep retinal layer image are shown, respectively.

[0064] The fractal dimension is obtained based on the binarized image using the box counting method. Specifically, the fractal dimension (FD) of the binarized image is calculated using the box counting method of the Fractalyse software. The FD reflects the complexity of the blood vessel branching, and a high FD value indicates a high blood vessel branching density.

[0065] The detection module is configured to obtain the carotid ultrasound measurement index based on the blood vessel parameter and the relationship (regression model) between the blood vessel parameter and the carotid ultrasound measurement index, and determine the microcirculation disorder category (normal, intimal thickening or plaque) to which the to-be-detected person belongs based on the carotid ultrasound measurement index.

[0066] The carotid ultrasound measurement index includes carotid intima-media thickness (CIMT), carotid artery (CCA) plaque area, and carotid artery diameter.

[0067] Specifically, the regression model is a linear regression model.

[0068] The model construction module is configured to construct the relationship between the blood vessel parameter and the carotid ultrasound measurement index, and specifically configured to:

[0069] (1) For people over 40 years old, after excluding patients with diabetes, hypertension, hypercholesterolemia, smoking, cerebrovascular or cardiovascular disease history, and taking other disease treatment drugs, a comprehensive ophthalmic examination is performed. Patients without eye history, trauma or surgery history, without macular edema, with best corrected visual acuity ≥0.3 recorded in the minimum resolution log MAR, with refractive error not exceeding ±6.0D, with intraocular pressure ≤21mmHg, with optic disc cup / disk ratio <0.4 and bilateral symmetry are selected as samples, and 138 samples are obtained.

[0070] (2) Grouping of samples: The samples were divided into normal and abnormal groups according to the maximum value of the intima-media thickness (IMT) of the bilateral carotid artery. The evaluation index was: normal group (normal intima group): IMT <1.0 mm; abnormal group included intima-media thickening group (intima thickening group) and plaque group, intima-media thickening group was defined as intima-media thickness ≥1.0 mm, and plaque group was defined as focal intima-media thickness ≥1.5 mm. Among the 138 eyes, 72 eyes (52.17%) had CIMT thickening, and about 32 (23.19%) patients had carotid artery plaque.

[0071] (3) The data acquisition module and image analysis module were called to obtain the vascular parameters of each sample and the carotid ultrasound measurement indexes (carotid intima-media thickness (CIMT), carotid artery plaque area, and carotid artery diameter) of several samples.

[0072] (4) Based on the vascular parameters and carotid ultrasound measurement indexes of all samples, SPSS 21.0 statistical software was called to perform t-test and partial correlation analysis of the influence of carotid ultrasound measurement indexes on vascular parameters according to the sample grouping. According to the results of the partial correlation analysis, a regression model of the vascular parameters and the carotid ultrasound measurement indexes was constructed. Specifically, if there was a significant correlation between the vascular parameters and the carotid ultrasound measurement indexes, linear regression analysis was applied to obtain the relationship between all vascular parameters and each carotid ultrasound measurement index after controlling for age, blood pressure, and gender. All data were represented by mean ± standard deviation, frequency was represented by percentage, and statistical significance was represented by P<0.05.

[0073] As shown in Table 1, in the superficial capillary network of the retina, the average VD was reduced in the intima-media thickening group (group 1) compared with the normal intima-media group (group 2) (0.45±0.06 vs. 0.42±0.09, P=0.03). Similarly, the FD of SRL and DRL in group 1 was significantly reduced compared with group 2 (SRL layer: 1.69±0.02 vs. 1.68±0.15, P=0.01; DRL layer: 1.72±0.04 vs. 1.68±0.04, P=0.03). After excluding the factors of age, blood pressure, and gender, linear regression analysis showed that the VD and SD of the DRL in the macular region were significantly negatively correlated with the CIMT value (VD: β=-0.18, P=0.03; SD: β=-0.20, P=0.01). The FD in the superficial and deep capillary networks in the macular region was negatively linearly correlated with the CIMT (SRL layer: β=-0.18, P=0.03; DRL layer: β=-0.25, P<0.01).

[0074] Table 1, Correlation of vascular parameters with CIMT

[0075]

[0076] As shown in Table 2, the VD and SD of the superficial macular capillary network were significantly lower in the plaque group (Group 3) than in the non-plaque group (Group 4) (VD: 0.38 ± 0.01 vs. 0.44 ± 0.07, SD: 0.18 ± 0.01 vs. 0.21 ± 0.01; P < 0.01), as were the VD and SD of the deep retinal capillary network (VD: 0.40 ± 0.15 vs. 0.46 ± 0.12, SD: 0.19 ± 0.01 vs. 0.23 ± 0.01; P < 0.01). The FD of the macular area of the superficial and deep retinal capillary networks was also significantly lower in Group 3 than in Group 4 (SRL: 1.67 ± 0.01 vs. 1.71 ± 0.01, DRL: 1.67 ± 0.01 vs. 1.71 ± 0.01; P < 0.01).

[0077] Table 2, Vascular parameters in patients with and without carotid artery plaque

[0078]

[0079] As shown in Table 3, linear regression analysis showed that, after excluding the factors of age, blood pressure, and gender, there was a significant negative correlation between the VD and SD of the superficial and deep retinal capillary networks and the CCA diameter (SRL: β = -0.26, β = -0.27; DRL: β = -0.24, β = -0.25; all P < 0.05). As the CCA diameter narrowed, the FD values of the SRL and DRL also significantly decreased (SRL: β = -0.18, P = 0.02; DRL: β = -0.20, P = 0.01).

[0080] Table 3, Correlation of vascular parameters with carotid artery diameter

[0081]

[0082]

[0083] The system uses the OCTA high-resolution retinal imaging mode to quantitatively evaluate the retinal microvascular parameters, and proves that the retinal deep VD and SD are significantly negatively correlated with CIMT, which indicates that the quantitative detection of retinal microvessels in the deep capillary plexus of the retina can better reflect the CAS-related microvascular changes than the superficial capillary layer. The patients with carotid artery plaque can see a significant decrease in VD and SD of the macular superficial and deep layers. However, in the eyes with a narrow CCA diameter, the VD and SD of the macular capillary network are higher to maintain stable ocular perfusion. More importantly, in the superficial and deep capillary plexus of the retina, the intimal thickening and the presence of plaques are significantly negatively correlated with the macular FD. The decrease of the FD value measured by the OCTA image can become a more sensitive and useful indicator reflecting the microvascular damage caused by CAS.

[0084] The commonly used ultrasound parameters of carotid atherosclerosis are closely related to the quantitative indicators of retinal microvessels in the OCTA image. By using the novel marker of microcirculatory disturbance caused by carotid stenosis of the application, the progress of carotid atherosclerosis and the related microcirculatory changes can be detected with high sensitivity. The significant decrease of microvessel density and the impairment of capillary perfusion complexity can be used as an early indicator of carotid atherosclerotic lesions. The screening of VD, SD and FD of retinal vascular images by OCTA can be used as one of the non-invasive indicators to monitor the presence of early asymptomatic CAS, and to evaluate the progress of microcirculatory disturbance caused by carotid stenosis and treatment effect, etc.

[0085] Embodiment Two

[0086] The embodiment also provides an electronic device, comprising: one or more processors, one or more memories, and one or more computer programs; wherein the processor is connected with the memory, the one or more computer programs are stored in the memory, and when the electronic device is running, the processor executes the one or more computer programs stored in the memory, so that the electronic device executes the following steps:

[0087] obtaining a superficial retinal layer image and a deep retinal layer image of a to-be-detected person;

[0088] extracting a blood vessel parameter after sequentially performing binarization and skeletonization on the superficial retinal layer image and the deep retinal layer image, respectively;

[0089] based on the blood vessel parameter, using a regression model of the blood vessel parameter and a carotid ultrasound measurement index to obtain the carotid ultrasound measurement index, and determining a microcirculatory disturbance to which the to-be-detected person belongs.

[0090] It should be understood that in the embodiment, the processor can be a central processing unit (CPU), and the processor can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0091] The memory can include read-only memory and random access memory, and provide instructions and data to the processor, and a portion of the memory can also include non-volatile random access memory. For example, the memory can also store device type information.

[0092] In the implementation process, each step of the above method can be completed by integrated logic circuits of hardware in the processor or instructions in the form of software.

[0093] The method in the embodiment one can be directly embodied as hardware processor execution completion, or executed by hardware and software modules in the processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, registers, and other mature storage media in the art. The storage medium is located in the memory, and the processor reads the information in the memory and combines the hardware to complete the steps of the above method. To avoid repetition, it will not be described in detail here.

[0094] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are executed in hardware or software mode depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of the present application.

[0095] Embodiment three

[0096] The embodiment also provides a computer readable storage medium for storing computer instructions, which, when executed by a processor, performs the following steps:

[0097] Obtaining a superficial retinal layer image and a deep retinal layer image of a to-be-detected person;

[0098] After binarization and skeletonization of the superficial retinal layer image and the deep retinal layer image respectively, extracting a blood vessel parameter;

[0099] Based on the blood vessel parameter, using a regression model of the blood vessel parameter and a carotid ultrasound measurement index, obtaining the carotid ultrasound measurement index, and determining the microcirculation disorder to which the to-be-detected person belongs.

[0100] The above description is only the preferred embodiment of the present application, and is not intended to limit the present application. The present application can have various changes and modifications for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A carotid stenosis-induced microcirculation disorder detection system, characterized by, The method comprises the following steps: a data acquisition module is configured to acquire a superficial retinal layer image and a deep retinal layer image of a subject to be detected; an image analysis module is configured to extract blood vessel parameters by sequentially performing binarization and skeletonization on the superficial retinal layer image and the deep retinal layer image respectively; a detection module is configured to obtain a carotid ultrasonic measurement index based on the blood vessel parameters and a regression model of the blood vessel parameters and the carotid ultrasonic measurement index, and determine a microcirculation disorder category to which the subject to be detected belongs. The blood vessel parameters comprise an average blood vessel density, a skeleton density and a fractal dimension. The superficial retinal layer image and the deep retinal layer image are sequentially subjected to standardization, cropping and binarization to obtain a binarized image; based on each binarized image, the blood vessel parameters are calculated as follows: a skeleton graph of the binarized image is acquired, and the blood vessel skeleton density is calculated based on the skeleton graph. A model construction module is configured to construct a regression model of the blood vessel parameters and the carotid ultrasonic measurement index, and is specifically configured to: acquire blood vessel parameters and carotid ultrasonic measurement indexes of a plurality of samples, and group the samples; based on the blood vessel parameters and the carotid ultrasonic measurement indexes of all samples, perform a partial correlation analysis of the carotid ultrasonic measurement index and the blood vessel parameters according to the sample grouping, and construct a regression model of the blood vessel parameters and the carotid ultrasonic measurement index according to the partial correlation analysis result.

2. The system for detecting microcirculatory disturbance caused by carotid stenosis according to claim 1, wherein The image analysis module is further configured to sequentially perform standardization and cropping on the superficial retinal layer image and the deep retinal layer image respectively before binarization.

3. The system for detecting microcirculatory disturbance caused by carotid stenosis according to claim 1, wherein The fractal dimension is obtained by using a box counting method based on the binarized image.

4. The system for detecting microcirculatory disturbance caused by carotid stenosis according to claim 1, wherein The carotid ultrasonic measurement index comprises a carotid intima-media thickness, a common carotid artery plaque area and a common carotid artery diameter.

5. The carotid stenosis-induced microcirculatory disturbance detection system of claim 1, wherein The regression model is a linear regression model.

6. The system for detecting microcirculatory disturbance caused by carotid stenosis according to claim 1, wherein The superficial retinal layer image and the deep retinal layer image are obtained by scanning the subject's macular fovea as the center by an optical coherence tomography blood vessel imaging technology.

7. An electronic device, comprising: a memory for non-transiently storing computer readable instructions; and a processor for running the computer readable instructions, wherein the computer readable instructions, when run by the processor, perform the following steps: acquiring a superficial retinal layer image and a deep retinal layer image of a subject to be detected; extracting blood vessel parameters by sequentially performing binarization and skeletonization on the superficial retinal layer image and the deep retinal layer image respectively; based on the blood vessel parameters, obtaining a carotid ultrasonic measurement index by using a regression model of the blood vessel parameters and the carotid ultrasonic measurement index, and determining a microcirculation disorder to which the subject to be detected belongs. The computer readable instructions are stored in the memory, and when the computer readable instructions are executed by a computer, the following steps are performed:

8. A storage medium, characterized by a non-transitory acquiring a superficial retinal layer image and a deep retinal layer image of a subject to be detected; extracting blood vessel parameters by sequentially performing binarization and skeletonization on the superficial retinal layer image and the deep retinal layer image respectively; based on the blood vessel parameters, obtaining a carotid ultrasonic measurement index by using a regression model of the blood vessel parameters and the carotid ultrasonic measurement index, and determining a microcirculation disorder to which the subject to be detected belongs. ​

Images