Oral mucosal vascular OCT image processing method and system

Abstract

The present invention discloses an oral mucosal vascular OCT image processing method and system, the method comprising: acquiring an OCT image of a target oral mucosal tissue; segmenting a plurality of vascular regions from the OCT image based on a trained image segmentation model; determining the disease probability corresponding to each of the vascular regions; performing image enhancement on the image region of the OCT image near each of the vascular regions according to the disease probability to obtain an enhanced OCT image result; the intensity and range of the image enhancement are proportional to the disease probability. It can be seen that the present invention can achieve more targeted and efficient image enhancement, provide a more reference data image basis for disease diagnosis or research of related tissues, and reduce image processing costs.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an oral mucosal blood vessel OCT image processing method and system. Background Art

[0002] Oral mucosal blood vessel OCT images can effectively characterize the condition of oral mucosal blood vessels and assist in the diagnosis of oral mucosal diseases, which are diseases of various types and with a high incidence rate, and can have a serious impact on the patient's life and production activities. Since the lesions of oral mucosal diseases are complex and have different forms, it is often difficult to make an accurate judgment on the lesions without the help of effective technical images. When conducting corresponding disease research or diagnosis through OCT images of oral mucosal blood vessels, the existing technology does not fully consider the rich blood vessels of the oral mucosa, and does not weight the images of the vascular position part in the OCT image to realize image processing based on the vascular veins, and further identify possible lesion locations. Therefore, it is impossible to provide researchers or medical staff with clearer and more targeted image results, and it is impossible to provide an accurate data basis for disease research or treatment. It can be seen that the existing technology has defects that need to be solved urgently. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide an oral mucosal vascular OCT image processing method and system, which can achieve more targeted and efficient image enhancement, provide a more reference data image basis for disease diagnosis or research of related tissues, and reduce image processing costs.

[0004] In order to solve the above technical problems, the first aspect of the present invention discloses an oral mucosal vascular OCT image processing method, the method comprising:

[0005] Obtaining OCT images of target oral mucosal tissues;

[0006] Segmenting a plurality of blood vessel regions from the OCT image based on a trained image segmentation model;

[0007] Determining a disease probability corresponding to each of the vascular regions;

[0008] According to the disease probability, image enhancement is performed on the image area of ​​the OCT image near each blood vessel area to obtain an enhanced OCT image result; the intensity and range of the image enhancement are proportional to the disease probability.

[0009] As an optional implementation, in the first aspect of the present invention, segmenting a plurality of vascular regions from the OCT image based on a trained image segmentation model comprises:

[0010] Segmenting the OCT image according to a preset image segmentation rule to obtain a plurality of image blocks;

[0011] Inputting each of the image blocks into a trained blood vessel image probability prediction neural network to obtain a blood vessel probability corresponding to the image block;

[0012] According to the blood vessel probability of each of the image blocks, any plurality of the image blocks belonging to the same blood vessel are combined to obtain a plurality of blood vessel regions.

[0013] As an optional implementation, in the first aspect of the present invention, combining any multiple image blocks belonging to the same blood vessel according to the blood vessel probability of each image block to obtain multiple blood vessel regions includes:

[0014] Screening out all the image blocks whose blood vessel probabilities are greater than a preset probability threshold to obtain a plurality of blood vessel image blocks;

[0015] Inputting each of the blood vessel image blocks into a trained blood vessel direction prediction neural network to obtain an angle parameter corresponding to each of the blood vessel image blocks;

[0016] For any two adjacent blood vessel image blocks, calculating a probability difference between the blood vessel probabilities of the two blood vessel image blocks;

[0017] Calculating the angle difference between the angle parameters of the two blood vessel image blocks;

[0018] Calculating a weighted average of the probability difference and the angle difference to obtain a correlation coefficient corresponding to the two blood vessel image blocks;

[0019] According to the dynamic programming algorithm and the correlation coefficient, all the blood vessel image blocks are iteratively combined to obtain a plurality of blood vessel regions; each of the blood vessel regions includes a plurality of blood vessel image blocks belonging to the same blood vessel.

[0020] As an optional implementation, in the first aspect of the present invention, the iterative combination calculation is performed on all the blood vessel image blocks according to the dynamic programming algorithm and the correlation coefficient to obtain multiple blood vessel regions, including:

[0021] Setting the objective function to minimize the number of the blood vessel image blocks in each blood vessel region;

[0022] The setting restriction conditions include that any one of the blood vessel image blocks in each blood vessel region is in a position adjacent to at least another of the blood vessel image blocks in the OCT image, the correlation coefficient corresponding to any two adjacent blood vessel image blocks in each blood vessel region is greater than a preset first coefficient threshold, and the correlation coefficient corresponding to any two blood vessel image blocks belonging to different blood vessel regions is less than a preset second coefficient threshold; and the second coefficient threshold is less than the first coefficient threshold;

[0023] According to the objective function and the restriction condition, all the blood vessel image blocks are iteratively combined and calculated based on a dynamic programming algorithm until convergence, so as to obtain a plurality of blood vessel regions.

[0024] As an optional implementation, in the first aspect of the present invention, determining the disease probability corresponding to each of the vascular regions includes:

[0025] Inputting each of the vascular regions into a trained disease recognition neural network to obtain a disease type and a predicted probability corresponding to each of the vascular regions; the disease recognition neural network is trained by a training data set including a plurality of training vascular region images and corresponding disease type annotations;

[0026] The disease probability corresponding to each of the vascular regions is determined according to the disease type and the predicted probability of each of the vascular regions and adjacent vascular regions.

[0027] As an optional implementation, in the first aspect of the present invention, determining the disease probability corresponding to each of the vascular regions according to the disease type and the predicted probability of each of the vascular regions and the adjacent vascular regions includes:

[0028] For each of the blood vessel regions, a plurality of other blood vessel regions whose center position distances from the blood vessel region are less than a preset distance threshold are determined; the center position distance is the position distance between the geometric center of the other blood vessel regions and the geometric center of the blood vessel region;

[0029] Determine the other vascular regions corresponding to the same disease type as the vascular region as relevant vascular regions;

[0030] Calculating a weighted sum average of the predicted probabilities corresponding to each of the related blood vessel regions to obtain a reference probability value corresponding to the blood vessel region; wherein the weighted calculation weight corresponding to the predicted probability corresponding to each of the related blood vessel regions is inversely proportional to the corresponding distance from the center position;

[0031] The average value of the predicted probability and the reference probability value corresponding to the blood vessel region is calculated to obtain the disease probability corresponding to the blood vessel region.

[0032] As an optional implementation, in the first aspect of the present invention, performing image enhancement on an image region of the OCT image near each of the blood vessel regions according to the disease probability to obtain an enhanced OCT image result includes:

[0033] For each of the blood vessel regions, determining a center line corresponding to the blood vessel region;

[0034] Determining enhancement parameters corresponding to the vascular region according to the disease probability corresponding to the vascular region and the corresponding relationship between the preset probability and enhancement parameters; the enhancement parameters include enhancement range, enhancement sharpness and enhancement clarity; the enhancement parameters are proportional to the disease probability;

[0035] At each point on the center line, a range area is generated with the enhancement range as a radius, and the sum of all the range areas is determined as the enhancement area;

[0036] Based on the enhanced sharpness and the enhanced clarity, according to an image enhancement algorithm, the OCT image in the enhanced region is enhanced to obtain an enhanced image corresponding to the blood vessel region;

[0037] The union of the enhanced images corresponding to all the vascular regions is determined as the enhanced OCT image result.

[0038] As an optional implementation, in the first aspect of the present invention, determining the center line corresponding to the blood vessel region includes:

[0039] segmenting the blood vessel region into a plurality of region portions;

[0040] Calculate the minimum enclosing rectangle corresponding to each of the area parts;

[0041] Determine a partial center line corresponding to each of the area portions according to a center line in the long side direction corresponding to the corresponding minimum enclosing rectangle;

[0042] For any two adjacent area parts, calculating the blood vessel probability parameters respectively corresponding to the two area parts; the blood vessel probability parameter is the average value of the blood vessel probabilities corresponding to all the image blocks in the corresponding area parts;

[0043] Calculate the weighted sum average of the extension direction angles of the partial center lines corresponding to the two area parts to obtain the adjusted direction angle; wherein the weight of the weighted calculation of each extension direction angle is proportional to the corresponding blood vessel probability parameter;

[0044] Determine the partial center lines corresponding to the two area parts as line segments at the adjusted direction angles, and obtain the adjusted center lines corresponding to the two area parts;

[0045] The set of the adjusted center lines corresponding to all the region parts is determined as the center line corresponding to the blood vessel region.

[0046] A second aspect of an embodiment of the present invention discloses an oral mucosal vascular OCT image processing system, the system comprising:

[0047] An acquisition module, used for acquiring an OCT image of a target oral mucosal tissue;

[0048] A segmentation module, used for segmenting a plurality of blood vessel regions from the OCT image based on a trained image segmentation model;

[0049] A determination module, used to determine the disease probability corresponding to each of the blood vessel regions;

[0050] The enhancement module is used to perform image enhancement on the image area of ​​the OCT image near each blood vessel area according to the disease probability to obtain an enhanced OCT image result; the intensity and range of the image enhancement are proportional to the disease probability.

[0051] As an optional implementation, in the second aspect of the present invention, the specific manner in which the segmentation module segments multiple vascular regions from the OCT image based on the trained image segmentation model includes:

[0052] Segmenting the OCT image according to a preset image segmentation rule to obtain a plurality of image blocks;

[0053] Inputting each of the image blocks into a trained blood vessel image probability prediction neural network to obtain a blood vessel probability corresponding to the image block;

[0054] According to the blood vessel probability of each of the image blocks, any plurality of the image blocks belonging to the same blood vessel are combined to obtain a plurality of blood vessel regions.

[0055] As an optional implementation, in the second aspect of the present invention, the segmentation module combines any multiple image blocks belonging to the same blood vessel according to the blood vessel probability of each image block to obtain a specific manner of multiple blood vessel regions, including:

[0056] Screening out all the image blocks whose blood vessel probabilities are greater than a preset probability threshold to obtain a plurality of blood vessel image blocks;

[0057] Inputting each of the blood vessel image blocks into a trained blood vessel direction prediction neural network to obtain an angle parameter corresponding to each of the blood vessel image blocks;

[0058] For any two adjacent blood vessel image blocks, calculating a probability difference between the blood vessel probabilities of the two blood vessel image blocks;

[0059] Calculating the angle difference between the angle parameters of the two blood vessel image blocks;

[0060] Calculating a weighted average of the probability difference and the angle difference to obtain a correlation coefficient corresponding to the two blood vessel image blocks;

[0061] According to the dynamic programming algorithm and the correlation coefficient, all the blood vessel image blocks are iteratively combined to obtain a plurality of blood vessel regions; each of the blood vessel regions includes a plurality of blood vessel image blocks belonging to the same blood vessel.

[0062] As an optional implementation, in the second aspect of the present invention, the segmentation module performs iterative combination calculation on all the blood vessel image blocks according to the dynamic programming algorithm and the correlation coefficient to obtain a specific method of multiple blood vessel regions, including:

[0063] Setting the objective function to minimize the number of the blood vessel image blocks in each blood vessel region;

[0064] The setting restriction conditions include that any one of the blood vessel image blocks in each blood vessel region is in a position adjacent to at least another of the blood vessel image blocks in the OCT image, the correlation coefficient corresponding to any two adjacent blood vessel image blocks in each blood vessel region is greater than a preset first coefficient threshold, and the correlation coefficient corresponding to any two blood vessel image blocks belonging to different blood vessel regions is less than a preset second coefficient threshold; and the second coefficient threshold is less than the first coefficient threshold;

[0065] According to the objective function and the restriction condition, all the blood vessel image blocks are iteratively combined and calculated based on a dynamic programming algorithm until convergence, so as to obtain a plurality of blood vessel regions.

[0066] As an optional implementation, in the second aspect of the present invention, the specific manner in which the determination module determines the disease probability corresponding to each of the vascular regions includes:

[0067] Inputting each of the vascular regions into a trained disease recognition neural network to obtain a disease type and a predicted probability corresponding to each of the vascular regions; the disease recognition neural network is trained by a training data set including a plurality of training vascular region images and corresponding disease type annotations;

[0068] The disease probability corresponding to each of the vascular regions is determined according to the disease type and the predicted probability of each of the vascular regions and adjacent vascular regions.

[0069] As an optional implementation, in the second aspect of the present invention, the specific manner in which the determination module determines the disease probability corresponding to each of the vascular regions according to the disease type and the predicted probability of each of the vascular regions and the adjacent vascular regions includes:

[0070] For each of the blood vessel regions, a plurality of other blood vessel regions whose center position distances from the blood vessel region are less than a preset distance threshold are determined; the center position distance is the position distance between the geometric center of the other blood vessel regions and the geometric center of the blood vessel region;

[0071] Determine the other vascular regions corresponding to the same disease type as the vascular region as relevant vascular regions;

[0072] Calculating a weighted sum average of the predicted probabilities corresponding to each of the related blood vessel regions to obtain a reference probability value corresponding to the blood vessel region; wherein the weighted calculation weight corresponding to the predicted probability corresponding to each of the related blood vessel regions is inversely proportional to the corresponding distance from the center position;

[0073] The average value of the predicted probability and the reference probability value corresponding to the blood vessel region is calculated to obtain the disease probability corresponding to the blood vessel region.

[0074] As an optional implementation, in the second aspect of the present invention, the enhancement module performs image enhancement on the image area of ​​the OCT image near each of the blood vessel regions according to the disease probability to obtain the enhanced OCT image result in a specific manner, including:

[0075] For each of the blood vessel regions, determining a center line corresponding to the blood vessel region;

[0076] Determining enhancement parameters corresponding to the vascular region according to the disease probability corresponding to the vascular region and the corresponding relationship between the preset probability and enhancement parameters; the enhancement parameters include enhancement range, enhancement sharpness and enhancement clarity; the enhancement parameters are proportional to the disease probability;

[0077] At each point on the center line, a range area is generated with the enhancement range as a radius, and the sum of all the range areas is determined as the enhancement area;

[0078] Based on the enhanced sharpness and the enhanced clarity, according to an image enhancement algorithm, the OCT image in the enhanced region is enhanced to obtain an enhanced image corresponding to the blood vessel region;

[0079] The union of the enhanced images corresponding to all the vascular regions is determined as the enhanced OCT image result.

[0080] As an optional implementation, in the second aspect of the present invention, the specific manner in which the enhancement module determines the center line corresponding to the blood vessel region includes:

[0081] segmenting the blood vessel region into a plurality of region portions;

[0082] Calculate the minimum enclosing rectangle corresponding to each of the area parts;

[0083] Determine a partial center line corresponding to each of the area portions according to a center line in the long side direction corresponding to the corresponding minimum enclosing rectangle;

[0084] For any two adjacent area parts, calculating the blood vessel probability parameters respectively corresponding to the two area parts; the blood vessel probability parameter is the average value of the blood vessel probabilities corresponding to all the image blocks in the corresponding area parts;

[0085] Calculate the weighted sum average of the extension direction angles of the partial center lines corresponding to the two area parts to obtain the adjusted direction angle; wherein the weight of the weighted calculation of each extension direction angle is proportional to the corresponding blood vessel probability parameter;

[0086] Determine the partial center lines corresponding to the two area parts as line segments at the adjusted direction angles, and obtain the adjusted center lines corresponding to the two area parts;

[0087] The set of the adjusted center lines corresponding to all the region parts is determined as the center line corresponding to the blood vessel region.

[0088] The third aspect of the present invention discloses another oral mucosal vascular OCT image processing system, the system comprising:

[0089] A memory storing executable program code;

[0090] a processor coupled to the memory;

[0091] The processor calls the executable program code stored in the memory to execute part or all of the steps in the oral mucosal vascular OCT image processing method disclosed in the first aspect of the present invention.

[0092] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute some or all of the steps in the oral mucosal vascular OCT image processing method disclosed in the first aspect of the present invention.

[0093] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0094] The present invention can segment multiple vascular regions according to a trained image segmentation model. After determining the disease probability of each vascular region, image enhancement is performed on other nearby image regions along each vascular region according to the disease probability to obtain enhanced OCT image results, thereby achieving more targeted and efficient image enhancement, providing a more reference data image basis for disease diagnosis or research of related tissues, and reducing image processing costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0096] Figure 1 It is a flow chart of an oral mucosal vascular OCT image processing method disclosed in an embodiment of the present invention.

[0097] Figure 2 It is a structural schematic diagram of an oral mucosal vascular OCT image processing system disclosed in an embodiment of the present invention.

[0098] Figure 3 It is a structural schematic diagram of another oral mucosal vascular OCT image processing system disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0099] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0100] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or equipment.

[0101] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0102] The present invention discloses an oral mucosal vascular OCT image processing method and system, which can segment multiple vascular regions according to a trained image segmentation model, and after determining the disease probability of each vascular region, perform image enhancement on other nearby image regions along each vascular region according to the disease probability to obtain enhanced OCT image results, thereby achieving more targeted and more efficient image enhancement, providing a more reference data image basis for disease diagnosis or research of related tissues, and reducing image processing costs. The following are detailed descriptions.

[0103] Embodiment 1

[0104] See also Figure 1 , Figure 1 : is a flow chart of an oral mucosal vascular OCT image processing method disclosed in an embodiment of the present invention. Figure 1 The described oral mucosal vascular OCT image processing method can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 1 As shown, the oral mucosal vascular OCT image processing method may include the following operations:

[0105] 101. Obtain an OCT image of the target oral mucosal tissue.

[0106] 102. Based on the trained image segmentation model, multiple vascular regions are segmented from the OCT image.

[0107] 103. Determine the disease probability corresponding to each vascular region.

[0108] 104. According to the disease probability, image enhancement is performed on the image region of the nearby OCT image along each blood vessel region to obtain an enhanced OCT image result.

[0109] Optionally, the strength and extent of image enhancement is proportional to the probability of disease.

[0110] It can be seen that the above-mentioned embodiment of the invention can segment multiple vascular regions according to the trained image segmentation model. After determining the disease probability of each vascular region, image enhancement is performed on other nearby image regions along each vascular region according to the disease probability to obtain enhanced OCT image results, thereby achieving more targeted and efficient image enhancement, providing a more reference data image basis for disease diagnosis or research of related tissues, and reducing image processing costs.

[0111] As an optional embodiment, in the above steps, segmenting multiple blood vessel regions from the OCT image based on the trained image segmentation model includes:

[0112] The OCT image is segmented according to a preset image segmentation rule to obtain a plurality of image blocks;

[0113] Input each image block into the trained vascular image probability prediction neural network to obtain the vascular probability corresponding to the image block;

[0114] According to the blood vessel probability of each image block, any multiple image blocks belonging to the same blood vessel are combined to obtain multiple blood vessel regions.

[0115] It can be seen that through the above optional embodiments, the OCT image can be segmented into multiple image blocks, and the probability of each image block belonging to a vascular image can be predicted based on the vascular image probability prediction neural network, and multiple vascular regions can be combined to facilitate subsequent targeted image enhancement, assist in achieving more targeted and efficient image enhancement, provide a more reference data image basis for disease diagnosis or research of related tissues, and reduce image processing costs.

[0116] As an optional embodiment, in the above steps, according to the blood vessel probability of each image block, any multiple image blocks belonging to the same blood vessel are combined to obtain multiple blood vessel regions, including:

[0117] Screening out all image blocks whose blood vessel probabilities are greater than a preset probability threshold to obtain a plurality of blood vessel image blocks;

[0118] Input each blood vessel image block into the trained blood vessel direction prediction neural network to obtain the angle parameter corresponding to each blood vessel image block;

[0119] For any two adjacent blood vessel image blocks, calculating the probability difference between the blood vessel probabilities of the two blood vessel image blocks;

[0120] Calculating the angle difference between the angle parameters of the two blood vessel image blocks;

[0121] Calculate the weighted average of the probability difference and the angle difference to obtain the correlation coefficient corresponding to the two blood vessel image blocks;

[0122] According to the dynamic programming algorithm and the correlation coefficient, all the blood vessel image blocks are iteratively combined to obtain a plurality of blood vessel regions; each blood vessel region includes a plurality of blood vessel image blocks belonging to the same blood vessel.

[0123] It can be seen that through the above optional embodiments, the correlation coefficient between two vascular image blocks can be calculated according to the difference in angle parameters corresponding to the vascular image blocks and the difference in predicted probabilities, and a dynamic programming calculation is performed based on this to obtain multiple vascular regions, so as to facilitate subsequent targeted image enhancement, assist in achieving more targeted and efficient image enhancement, provide a more reference data image basis for disease diagnosis or research of related tissues, and reduce image processing costs.

[0124] As an optional embodiment, in the above steps, all blood vessel image blocks are iteratively combined and calculated according to the dynamic programming algorithm and the correlation coefficient to obtain multiple blood vessel regions, including:

[0125] The objective function is set to minimize the number of blood vessel image blocks in each blood vessel region;

[0126] The setting restriction condition includes that any one blood vessel image block in each blood vessel region is in a position adjacent to at least one other blood vessel image block in the OCT image, the correlation coefficient corresponding to any two adjacent blood vessel image blocks in each blood vessel region is greater than a preset first coefficient threshold, and the correlation coefficient corresponding to any two blood vessel image blocks belonging to different blood vessel regions is less than a preset second coefficient threshold; optionally, the second coefficient threshold is less than the first coefficient threshold;

[0127] According to the objective function and constraints, all vascular image blocks are iteratively combined and calculated based on the dynamic programming algorithm until convergence, and multiple vascular regions are obtained.

[0128] It can be seen that through the above-mentioned optional embodiments, multiple precise vascular areas can be obtained according to the preset objective function and constraints and the iterative calculation of the dynamic programming algorithm, so as to facilitate the subsequent targeted image enhancement, assist in achieving more targeted and efficient image enhancement, and provide a more reference data image basis for disease diagnosis or research of related tissues, thereby reducing image processing costs.

[0129] As an optional embodiment, in the above step, determining the disease probability corresponding to each blood vessel region includes:

[0130] Inputting each vascular region into a trained disease recognition neural network to obtain the disease type and predicted probability corresponding to each vascular region; optionally, the disease recognition neural network is trained using a training data set including multiple training vascular region images and corresponding disease type annotations;

[0131] The disease probability corresponding to each vascular region is determined according to the disease type and predicted probability of each vascular region and adjacent vascular regions.

[0132] It can be seen that through the above-mentioned optional embodiments, the disease type and predicted probability corresponding to each vascular region can be identified according to the disease recognition neural network to comprehensively determine the disease probability corresponding to each vascular region, so as to facilitate subsequent targeted image enhancement, assist in achieving more targeted and efficient image enhancement, and provide a more reference data image basis for disease diagnosis or research of related tissues, thereby reducing image processing costs.

[0133] As an optional embodiment, in the above step, determining the disease probability corresponding to each vascular region according to the disease type and predicted probability of each vascular region and adjacent vascular regions includes:

[0134] For each blood vessel region, determine a plurality of other blood vessel regions whose center position distances from the blood vessel region are less than a preset distance threshold; optionally, the center position distance is the position distance between the geometric centers of the other blood vessel regions and the geometric center of the blood vessel region;

[0135] Determine other vascular regions with the same disease type as the vascular region as related vascular regions;

[0136] Calculate the weighted sum average of the predicted probabilities corresponding to each relevant blood vessel region to obtain a reference probability value corresponding to the blood vessel region; optionally, the weighted calculation weight corresponding to the predicted probability corresponding to each relevant blood vessel region is inversely proportional to the corresponding center position distance;

[0137] The average of the predicted probability and the reference probability value corresponding to the vascular region is calculated to obtain the disease probability corresponding to the vascular region.

[0138] It can be seen that through the above optional embodiments, it is possible to comprehensively and accurately determine the disease probability corresponding to each vascular region based on the weighted calculation of the predicted probabilities of each vascular region and multiple nearby related vascular regions, so as to facilitate subsequent targeted image enhancement, assist in achieving more targeted and efficient image enhancement, provide a more reference data image basis for disease diagnosis or research of related tissues, and reduce image processing costs.

[0139] As an optional embodiment, in the above steps, according to the disease probability, image enhancement is performed on the image region of the nearby OCT image along each blood vessel region to obtain an enhanced OCT image result, including:

[0140] For each blood vessel region, determine the center line corresponding to the blood vessel region;

[0141] Determine the enhancement parameter corresponding to the vascular region according to the disease probability corresponding to the vascular region and the corresponding relationship between the preset probability and the enhancement parameter; optionally, the enhancement parameter includes an enhancement range, an enhancement sharpness and an enhancement clarity; and the enhancement parameter is proportional to the disease probability;

[0142] At each point on the center line, a range area is generated with the enhancement range as the radius, and the sum of all range areas is determined as the enhancement area;

[0143] Based on the enhanced sharpness and enhanced clarity, the OCT image in the enhanced area is enhanced according to the image enhancement algorithm to obtain an enhanced image corresponding to the vascular area;

[0144] The union of the enhanced images corresponding to all vascular regions is determined as the enhanced OCT image result.

[0145] It can be seen that through the above-mentioned optional embodiments, image enhancement parameters can be determined based on the disease probability with the center line of each vascular area as a reference, and the enhancement range and intensity can be determined accordingly, so as to achieve more targeted and efficient image enhancement, provide a more reference data image basis for disease diagnosis or research of related tissues, and reduce image processing costs.

[0146] As an optional embodiment, in the above step, determining the center line corresponding to the blood vessel region includes:

[0147] segmenting the blood vessel region into a plurality of region portions;

[0148] Calculate the minimum enclosing rectangle corresponding to each area part;

[0149] Determine the partial center line corresponding to each area portion according to the center line in the long side direction corresponding to the corresponding minimum enclosing rectangle;

[0150] For any two adjacent area parts, calculate the blood vessel probability parameters corresponding to the two area parts respectively; optionally, the blood vessel probability parameter is an average value of the blood vessel probabilities corresponding to all image blocks in the corresponding area parts;

[0151] Calculate the weighted sum average of the extension direction angles of the partial center lines corresponding to the two area parts to obtain the adjusted direction angle; optionally, the weight of the weighted calculation of each extension direction angle is proportional to the corresponding blood vessel probability parameter;

[0152] Determine the partial center line corresponding to the two area parts as a line segment with an adjusted direction angle, and obtain the adjusted center line corresponding to the two area parts;

[0153] The set of adjusted center lines corresponding to all the area parts is determined as the center line corresponding to the blood vessel area.

[0154] It can be seen that through the above optional embodiments, the center line of each vascular area can be accurately determined through the minimum enclosing rectangle algorithm and the vascular probability comparison between adjacent area parts, so as to realize image enhancement based on this, realize more targeted and efficient image enhancement, provide a more reference data image basis for disease diagnosis or research of related tissues, and reduce image processing costs.

[0155] Embodiment 2

[0156] See also Figure 2 , Figure 2 : is a schematic diagram of the structure of an oral mucosal vascular OCT image processing system disclosed in an embodiment of the present invention. Figure 2 The described oral mucosal vascular OCT image processing system can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 2 As shown, the oral mucosal vascular OCT image processing system may include:

[0157] The acquisition module 201 is used to acquire an OCT image of the target oral mucosal tissue.

[0158] The segmentation module 202 is used to segment multiple blood vessel regions from the OCT image based on the trained image segmentation model.

[0159] The determination module 203 is used to determine the disease probability corresponding to each blood vessel region.

[0160] The enhancement module 204 is used to perform image enhancement on the image region of the nearby OCT image along each blood vessel region according to the disease probability, so as to obtain an enhanced OCT image result.

[0161] Optionally, the strength and extent of image enhancement is proportional to the probability of disease.

[0162] It can be seen that the above-mentioned embodiment of the invention can segment multiple vascular regions according to the trained image segmentation model. After determining the disease probability of each vascular region, image enhancement is performed on other nearby image regions along each vascular region according to the disease probability to obtain enhanced OCT image results, thereby achieving more targeted and efficient image enhancement, providing a more reference data image basis for disease diagnosis or research of related tissues, and reducing image processing costs.

[0163] As an optional embodiment, the specific manner in which the segmentation module segments multiple vascular regions from the OCT image based on the trained image segmentation model includes:

[0164] The OCT image is segmented according to a preset image segmentation rule to obtain a plurality of image blocks;

[0165] Input each image block into the trained vascular image probability prediction neural network to obtain the vascular probability corresponding to the image block;

[0166] According to the blood vessel probability of each image block, any multiple image blocks belonging to the same blood vessel are combined to obtain multiple blood vessel regions.

[0167] It can be seen that through the above optional embodiments, the OCT image can be segmented into multiple image blocks, and the probability of each image block belonging to a vascular image can be predicted based on the vascular image probability prediction neural network, and multiple vascular regions can be combined to facilitate subsequent targeted image enhancement, assist in achieving more targeted and efficient image enhancement, provide a more reference data image basis for disease diagnosis or research of related tissues, and reduce image processing costs.

[0168] As an optional embodiment, the segmentation module combines any multiple image blocks belonging to the same blood vessel according to the blood vessel probability of each image block to obtain a specific manner of multiple blood vessel regions, including:

[0169] Screening out all image blocks whose blood vessel probabilities are greater than a preset probability threshold to obtain a plurality of blood vessel image blocks;

[0170] Input each blood vessel image block into the trained blood vessel direction prediction neural network to obtain the angle parameter corresponding to each blood vessel image block;

[0171] For any two adjacent blood vessel image blocks, calculating the probability difference between the blood vessel probabilities of the two blood vessel image blocks;

[0172] Calculating the angle difference between the angle parameters of the two blood vessel image blocks;

[0173] Calculate the weighted average of the probability difference and the angle difference to obtain the correlation coefficient corresponding to the two blood vessel image blocks;

[0174] According to the dynamic programming algorithm and the correlation coefficient, all the blood vessel image blocks are iteratively combined to obtain a plurality of blood vessel regions; each blood vessel region includes a plurality of blood vessel image blocks belonging to the same blood vessel.

[0175] It can be seen that through the above optional embodiments, the correlation coefficient between two vascular image blocks can be calculated according to the difference in angle parameters corresponding to the vascular image blocks and the difference in predicted probabilities, and a dynamic programming calculation is performed based on this to obtain multiple vascular regions, so as to facilitate subsequent targeted image enhancement, assist in achieving more targeted and efficient image enhancement, provide a more reference data image basis for disease diagnosis or research of related tissues, and reduce image processing costs.

[0176] As an optional embodiment, the segmentation module performs iterative combination calculations on all blood vessel image blocks according to the dynamic programming algorithm and the correlation coefficient to obtain a specific method of multiple blood vessel regions, including:

[0177] The objective function is set to minimize the number of blood vessel image blocks in each blood vessel region;

[0178] The setting restriction condition includes that any one blood vessel image block in each blood vessel region is in a position adjacent to at least one other blood vessel image block in the OCT image, the correlation coefficient corresponding to any two adjacent blood vessel image blocks in each blood vessel region is greater than a preset first coefficient threshold, and the correlation coefficient corresponding to any two blood vessel image blocks belonging to different blood vessel regions is less than a preset second coefficient threshold; optionally, the second coefficient threshold is less than the first coefficient threshold;

[0179] According to the objective function and constraints, all vascular image blocks are iteratively combined and calculated based on the dynamic programming algorithm until convergence, and multiple vascular regions are obtained.

[0180] It can be seen that through the above-mentioned optional embodiments, multiple precise vascular areas can be obtained according to the preset objective function and constraints and the iterative calculation of the dynamic programming algorithm, so as to facilitate the subsequent targeted image enhancement, assist in achieving more targeted and efficient image enhancement, and provide a more reference data image basis for disease diagnosis or research of related tissues, thereby reducing image processing costs.

[0181] As an optional embodiment, the specific manner in which the determination module determines the disease probability corresponding to each blood vessel region includes:

[0182] Inputting each vascular region into a trained disease recognition neural network to obtain the disease type and predicted probability corresponding to each vascular region; optionally, the disease recognition neural network is trained using a training data set including multiple training vascular region images and corresponding disease type annotations;

[0183] The disease probability corresponding to each vascular region is determined according to the disease type and predicted probability of each vascular region and adjacent vascular regions.

[0184] It can be seen that through the above-mentioned optional embodiments, the disease type and predicted probability corresponding to each vascular region can be identified according to the disease recognition neural network to comprehensively determine the disease probability corresponding to each vascular region, so as to facilitate subsequent targeted image enhancement, assist in achieving more targeted and efficient image enhancement, and provide a more reference data image basis for disease diagnosis or research of related tissues, thereby reducing image processing costs.

[0185] As an optional embodiment, the specific manner in which the determination module determines the disease probability corresponding to each vascular region according to the disease type and predicted probability of each vascular region and adjacent vascular regions includes:

[0186] For each blood vessel region, determine a plurality of other blood vessel regions whose center position distances from the blood vessel region are less than a preset distance threshold; optionally, the center position distance is the position distance between the geometric centers of the other blood vessel regions and the geometric center of the blood vessel region;

[0187] Determine other vascular regions with the same disease type as the vascular region as related vascular regions;

[0188] Calculate the weighted sum average of the predicted probabilities corresponding to each relevant blood vessel region to obtain a reference probability value corresponding to the blood vessel region; optionally, the weighted calculation weight corresponding to the predicted probability corresponding to each relevant blood vessel region is inversely proportional to the corresponding center position distance;

[0189] The average of the predicted probability and the reference probability value corresponding to the vascular region is calculated to obtain the disease probability corresponding to the vascular region.

[0190] It can be seen that through the above optional embodiments, it is possible to comprehensively and accurately determine the disease probability corresponding to each vascular region based on the weighted calculation of the predicted probabilities of each vascular region and multiple nearby related vascular regions, so as to facilitate subsequent targeted image enhancement, assist in achieving more targeted and efficient image enhancement, provide a more reference data image basis for disease diagnosis or research of related tissues, and reduce image processing costs.

[0191] As an optional embodiment, the enhancement module performs image enhancement on the image region of the nearby OCT image along each blood vessel region according to the disease probability to obtain the enhanced OCT image result in a specific manner including:

[0192] For each blood vessel region, determine the center line corresponding to the blood vessel region;

[0193] Determine the enhancement parameter corresponding to the vascular region according to the disease probability corresponding to the vascular region and the corresponding relationship between the preset probability and the enhancement parameter; optionally, the enhancement parameter includes an enhancement range, an enhancement sharpness and an enhancement clarity; and the enhancement parameter is proportional to the disease probability;

[0194] At each point on the center line, a range area is generated with the enhancement range as the radius, and the sum of all range areas is determined as the enhancement area;

[0195] Based on the enhanced sharpness and enhanced clarity, the OCT image in the enhanced area is enhanced according to the image enhancement algorithm to obtain an enhanced image corresponding to the vascular area;

[0196] The union of the enhanced images corresponding to all vascular regions is determined as the enhanced OCT image result.

[0197] It can be seen that through the above-mentioned optional embodiments, image enhancement parameters can be determined based on the disease probability with the center line of each vascular area as a reference, and the enhancement range and intensity can be determined accordingly, so as to achieve more targeted and efficient image enhancement, provide a more reference data image basis for disease diagnosis or research of related tissues, and reduce image processing costs.

[0198] As an optional embodiment, the specific manner in which the enhancement module determines the center line corresponding to the blood vessel region includes:

[0199] segmenting the blood vessel region into a plurality of region portions;

[0200] Calculate the minimum enclosing rectangle corresponding to each area part;

[0201] Determine the partial center line corresponding to each area portion according to the center line in the long side direction corresponding to the corresponding minimum enclosing rectangle;

[0202] For any two adjacent area parts, calculate the blood vessel probability parameters corresponding to the two area parts respectively; optionally, the blood vessel probability parameter is an average value of the blood vessel probabilities corresponding to all image blocks in the corresponding area parts;

[0203] Calculate the weighted sum average of the extension direction angles of the partial center lines corresponding to the two area parts to obtain the adjusted direction angle; optionally, the weight of the weighted calculation of each extension direction angle is proportional to the corresponding blood vessel probability parameter;

[0204] Determine the partial center line corresponding to the two area parts as a line segment with an adjusted direction angle, and obtain the adjusted center line corresponding to the two area parts;

[0205] The set of adjusted center lines corresponding to all the area parts is determined as the center line corresponding to the blood vessel area.

[0206] It can be seen that through the above optional embodiments, the center line of each vascular area can be accurately determined through the minimum enclosing rectangle algorithm and the vascular probability comparison between adjacent area parts, so as to realize image enhancement based on this, realize more targeted and efficient image enhancement, provide a more reference data image basis for disease diagnosis or research of related tissues, and reduce image processing costs.

[0207] Embodiment 3

[0208] See also Figure 3 , Figure 3 It is another oral mucosal blood vessel OCT image processing system disclosed in an embodiment of the present invention. Figure 3 The described oral mucosal vascular OCT image processing system is applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 3 As shown, the oral mucosal vascular OCT image processing system may include:

[0209] A memory 301 storing executable program codes;

[0210] a processor 302 coupled to the memory 301;

[0211] The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the oral mucosal vascular OCT image processing method described in the first embodiment.

[0212] Embodiment 4

[0213] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the oral mucosal vascular OCT image processing method described in the first embodiment.

[0214] Embodiment 5

[0215] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps of the oral mucosal vascular OCT image processing method described in the first embodiment.

[0216] The above describes specific embodiments of the present specification, and other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily have to be performed in the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0217] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0218] For the convenience of description, the above device is described by dividing it into various units according to its functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0219] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may be in the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of this specification may be in the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0220] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0221] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0222] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0223] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0224] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0225] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0226] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0227] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0228] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0229] Finally, it should be noted that the oral mucosal vascular OCT image processing method and system disclosed in the embodiments of the present invention disclose only the preferred embodiments of the present invention, which are only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for processing oral mucosal blood vessels OCT images, characterized in that: The method comprises: Obtaining OCT images of target oral mucosal tissues; Based on the trained image segmentation model, multiple vascular regions are segmented from the OCT image, including: Segmenting the OCT image according to a preset image segmentation rule to obtain a plurality of image blocks; Inputting each of the image blocks into a trained blood vessel image probability prediction neural network to obtain a blood vessel probability corresponding to the image block; Screening out all the image blocks whose blood vessel probabilities are greater than a preset probability threshold to obtain a plurality of blood vessel image blocks; Inputting each of the blood vessel image blocks into a trained blood vessel direction prediction neural network to obtain an angle parameter corresponding to each of the blood vessel image blocks; For any two adjacent blood vessel image blocks, calculating a probability difference between the blood vessel probabilities of the two blood vessel image blocks; Calculating the angle difference between the angle parameters of the two blood vessel image blocks; Calculating a weighted average of the probability difference and the angle difference to obtain a correlation coefficient corresponding to the two blood vessel image blocks; According to the dynamic programming algorithm and the correlation coefficient, all the blood vessel image blocks are iteratively combined to obtain a plurality of blood vessel regions; each of the blood vessel regions includes a plurality of blood vessel image blocks belonging to the same blood vessel; Determining a disease probability corresponding to each of the vascular regions; According to the disease probability, image enhancement is performed on an image region of the OCT image near each of the blood vessel regions to obtain an enhanced OCT image result, including: For each of the blood vessel regions, determining a center line corresponding to the blood vessel region; Determining enhancement parameters corresponding to the vascular region according to the disease probability corresponding to the vascular region and the corresponding relationship between the preset probability and enhancement parameters; the enhancement parameters include enhancement range, enhancement sharpness and enhancement clarity; the enhancement parameters are proportional to the disease probability; At each point on the center line, a range area is generated with the enhancement range as a radius, and the sum of all the range areas is determined as the enhancement area; Based on the enhanced sharpness and the enhanced clarity, according to an image enhancement algorithm, the OCT image in the enhanced region is enhanced to obtain an enhanced image corresponding to the blood vessel region; The union of the enhanced images corresponding to all the vascular regions is determined as the enhanced OCT image result.

2. The oral mucosal vascular OCT image processing method according to claim 1, characterized in that: The iterative combination calculation is performed on all the blood vessel image blocks according to the dynamic programming algorithm and the correlation coefficient to obtain multiple blood vessel regions, including: Setting the objective function to minimize the number of the blood vessel image blocks in each blood vessel region; The setting restriction conditions include that any one of the blood vessel image blocks in each blood vessel region is in a position adjacent to at least another of the blood vessel image blocks in the OCT image, the correlation coefficient corresponding to any two adjacent blood vessel image blocks in each blood vessel region is greater than a preset first coefficient threshold, and the correlation coefficient corresponding to any two blood vessel image blocks belonging to different blood vessel regions is less than a preset second coefficient threshold; and the second coefficient threshold is less than the first coefficient threshold; According to the objective function and the restriction condition, all the blood vessel image blocks are iteratively combined and calculated based on a dynamic programming algorithm until convergence, so as to obtain a plurality of blood vessel regions.

3. The oral mucosal vascular OCT image processing method according to claim 1, characterized in that: The determining of the disease probability corresponding to each of the blood vessel regions comprises: Inputting each of the vascular regions into a trained disease recognition neural network to obtain a disease type and a predicted probability corresponding to each of the vascular regions; the disease recognition neural network is trained by a training data set including a plurality of training vascular region images and corresponding disease type annotations; The disease probability corresponding to each of the vascular regions is determined according to the disease type and the predicted probability of each of the vascular regions and adjacent vascular regions.

4. The oral mucosal vascular OCT image processing method according to claim 3, characterized in that: Determining the disease probability corresponding to each of the vascular regions according to the disease type and the predicted probability of each of the vascular regions and the adjacent vascular regions includes: For each of the blood vessel regions, a plurality of other blood vessel regions whose center position distances from the blood vessel region are less than a preset distance threshold are determined; the center position distance is the position distance between the geometric center of the other blood vessel regions and the geometric center of the blood vessel region; Determine the other vascular regions corresponding to the same disease type as the vascular region as relevant vascular regions; Calculating a weighted sum average of the predicted probabilities corresponding to each of the related blood vessel regions to obtain a reference probability value corresponding to the blood vessel region; wherein the weighted calculation weight corresponding to the predicted probability corresponding to each of the related blood vessel regions is inversely proportional to the corresponding distance from the center position; The average value of the predicted probability and the reference probability value corresponding to the blood vessel region is calculated to obtain the disease probability corresponding to the blood vessel region.

5. The oral mucosal vascular OCT image processing method according to claim 1, characterized in that: Determining the center line corresponding to the blood vessel region includes: segmenting the blood vessel region into a plurality of region portions; Calculate the minimum enclosing rectangle corresponding to each of the area parts; Determine a partial center line corresponding to each of the area portions according to a center line in the long side direction corresponding to the corresponding minimum enclosing rectangle; For any two adjacent area parts, calculating the blood vessel probability parameters respectively corresponding to the two area parts; the blood vessel probability parameter is the average value of the blood vessel probabilities corresponding to all the image blocks in the corresponding area parts; Calculate the weighted sum average of the extension direction angles of the partial center lines corresponding to the two area parts to obtain the adjusted direction angle; wherein the weight of the weighted calculation of each extension direction angle is proportional to the corresponding blood vessel probability parameter; Determine the partial center lines corresponding to the two area parts as line segments at the adjusted direction angles, and obtain the adjusted center lines corresponding to the two area parts; The set of the adjusted center lines corresponding to all the region parts is determined as the center line corresponding to the blood vessel region.

6. An oral mucosal vascular OCT image processing system, characterized in that: The system comprises: An acquisition module, used for acquiring an OCT image of a target oral mucosal tissue; A segmentation module, used to segment multiple blood vessel regions from the OCT image based on a trained image segmentation model, comprising: Segmenting the OCT image according to a preset image segmentation rule to obtain a plurality of image blocks; Inputting each of the image blocks into a trained blood vessel image probability prediction neural network to obtain a blood vessel probability corresponding to the image block; Screening out all the image blocks whose blood vessel probabilities are greater than a preset probability threshold to obtain a plurality of blood vessel image blocks; Inputting each of the blood vessel image blocks into a trained blood vessel direction prediction neural network to obtain an angle parameter corresponding to each of the blood vessel image blocks; For any two adjacent blood vessel image blocks, calculating a probability difference between the blood vessel probabilities of the two blood vessel image blocks; Calculating the angle difference between the angle parameters of the two blood vessel image blocks; Calculating a weighted average of the probability difference and the angle difference to obtain a correlation coefficient corresponding to the two blood vessel image blocks; According to the dynamic programming algorithm and the correlation coefficient, all the blood vessel image blocks are iteratively combined to obtain a plurality of blood vessel regions; each of the blood vessel regions includes a plurality of blood vessel image blocks belonging to the same blood vessel; A determination module, used to determine the disease probability corresponding to each of the blood vessel regions; An enhancement module, used to perform image enhancement on an image region of the OCT image near each of the blood vessel regions according to the disease probability to obtain an enhanced OCT image result, including: For each of the blood vessel regions, determining a center line corresponding to the blood vessel region; Determining enhancement parameters corresponding to the vascular region according to the disease probability corresponding to the vascular region and the corresponding relationship between the preset probability and enhancement parameters; the enhancement parameters include enhancement range, enhancement sharpness and enhancement clarity; the enhancement parameters are proportional to the disease probability; At each point on the center line, a range area is generated with the enhancement range as a radius, and the sum of all the range areas is determined as the enhancement area; Based on the enhanced sharpness and the enhanced clarity, according to an image enhancement algorithm, the OCT image in the enhanced region is enhanced to obtain an enhanced image corresponding to the blood vessel region; The union of the enhanced images corresponding to all the vascular regions is determined as the enhanced OCT image result.

7. An oral mucosal vascular OCT image processing system, characterized in that: The system comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the oral mucosal vascular OCT image processing method according to any one of claims 1 to 5.

Images