Extracorporeal circulation thrombus identification method and system based on hyperspectral image

Abstract

The invention provides an extracorporeal circulation thrombus recognition method and system based on a hyperspectral image, and the method comprises the steps: collecting a current region image of a user body surface in an associated region of extracorporeal circulation treatment through hyperspectral image collection software, and carrying out the recognition of the extracorporeal circulation thrombus through an image recognition network based on the current region image; identifying each initial thrombus image region and thrombus image data in the region image; based on the pressure detection information of each region, generating region pressure distribution information of each initial thrombus image region; and on the basis of the regional pressure distribution information and the thrombus image data of each initial thrombus image region, through a thrombus prediction strategy, identifying the thrombus probability and the thrombus range of each initial thrombus image region, thereby generating regional thrombus prediction information of the associated region. By adopting the scheme, advanced early warning and real-time monitoring of extracorporeal circulation blood coagulation reaction can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of hyperspectral and equipment-assisted guidance, and in particular to an extracorporeal circulation thrombus recognition method and system based on hyperspectral images. Background Art

[0002] During continuous renal renal cell therapy (CRRT), the patient's blood needs to be drawn out of the body. When the blood is separated from the vascular environment and loses the protection of vascular endothelial cells, the stimulation caused by extracorporeal circulation will cause multiple biochemical pathways in blood cells and plasma to be activated, leading to blood coagulation, shortened filter life, and decreased toxin clearance rate, resulting in unplanned machine removal, which is still a major challenge in current clinical treatment. Therefore, how to accurately monitor the state of coagulation reaction and achieve advanced warning is the current research focus.

[0003] The traditional way to monitor the coagulation status of extracorporeal circulation is to use the pressure sensor of the blood purification treatment machine to display the arterial pressure, venous pressure, and transmembrane pressure sensing values. Medical staff judge the area where thrombosis may occur based on the dynamic changes of the values. The pressure value of the fluid is related to the blood flow, the minimum diameter of the extracorporeal circulation pipeline, blood viscosity, etc. Under the same blood flow, the smaller the diameter, the greater the pressure value. If the thrombus blocks the blood outflow outlet of the venous pot, arterial pot, or filter, causing coagulation in the extracorporeal circulation system to block the pipeline, the corresponding pressure sensing value will rise. However, since the diameter of the blood outflow outlet has not yet been blocked, the change in the pressure sensing value is significantly delayed, resulting in poor monitoring accuracy of the coagulation response. Summary of the invention

[0004] The main purpose of the present invention is to provide a method and system for identifying extracorporeal circulation thrombus based on hyperspectral images, aiming to solve the problems of hysteresis and poor accuracy in monitoring the coagulation state of extracorporeal circulation through pressure sensing in the prior art.

[0005] To achieve the above object, the present invention provides a method for identifying extracorporeal circulation thrombus based on hyperspectral images, the method comprising:

[0006] The current regional image of the user's body surface in the associated area of ​​the extracorporeal circulation treatment is collected by means of the hyperspectral image acquisition software, and based on the current regional image, each initial thrombus image region in the regional image and the thrombus image data of each initial thrombus image region are identified by means of the image recognition network;

[0007] Acquiring regional pressure detection information of each initial thrombus image region, and generating regional pressure distribution information of each initial thrombus image region based on each regional pressure detection information;

[0008] Based on the regional pressure distribution information of each initial thrombus image area and the thrombus image data of each initial thrombus image area, a thrombus prediction strategy is used to identify the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area, and based on the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area, regional thrombus prediction information of the associated area is generated.

[0009] Optionally, the identifying each initial thrombus image region in the regional image and the thrombus image data of each initial thrombus image region through an image recognition network based on the current regional image includes:

[0010] Performing grid splitting processing on the current region image to obtain a plurality of grid images, and inputting each grid image into an image recognition network to identify each thrombus grid image with a probability of thrombus existence;

[0011] Using an image edge recognition algorithm, the thrombus image edge information in each thrombus grid image is respectively recognized, and the image range enclosed by the thrombus image edge information in each thrombus grid image is used as the sub-thrombus image data in each thrombus grid image;

[0012] Adjacent thrombus grid images are subjected to image aggregation processing to obtain initial thrombus image areas, and sub-thrombus image data of each thrombus grid image in each initial thrombus image area are subjected to data splicing processing to obtain thrombus image data of each initial thrombus image area.

[0013] Optionally, the collecting of regional pressure detection information of each initial thrombus image region includes:

[0014] Based on the thrombus image data of each initial thrombus image region, identifying the position range of each thrombus image in the associated region;

[0015] Based on the position range of each thrombus image, in the associated area, each target detection area range that needs to be subjected to regional pressure detection is screened, and the target detection area range is sent to the staff client, instructing the staff to collect target pressure detection information of each target detection area range through the blood pressure sensor;

[0016] According to the sub-region range of each initial thrombus image region, in each target pressure detection information, the sub-pressure detection information of each sub-region range is screened, and the sub-pressure detection information of each sub-region range is used as the regional pressure detection information of the initial thrombus image region to which each sub-region range belongs.

[0017] Optionally, generating regional pressure distribution information of each initial thrombus image region based on each regional pressure detection information includes:

[0018] For each initial thrombus image region, based on the sub-pressure detection information of the initial thrombus image region, identifying the pressure data of each position point in the sub-region range of the initial thrombus image region;

[0019] The pressure data of each of the position points is distributed and arranged to obtain regional pressure distribution information of the initial thrombus image region.

[0020] Optionally, the identifying the thrombus probability of each initial thrombus image region and the thrombus range of each initial thrombus image region through a thrombus prediction strategy based on the regional pressure distribution information of each initial thrombus image region and the thrombus image data of each initial thrombus image region includes:

[0021] For each initial thrombus image region, based on the regional pressure distribution information of the initial thrombus image region, identifying a pressure regional range in the initial thrombus image region, and based on the pressure regional range and an image range corresponding to the thrombus image data of the initial thrombus image region, identifying an overlapping image range of the initial thrombus image region;

[0022] The overlapping image range is used as the thrombus range of the initial thrombus image region, and based on the pressure data of each position point in the thrombus range, the probability of a sub-thrombus at each position point in the thrombus range is predicted by a thrombus prediction strategy;

[0023] The sub-thrombus probabilities of all the position points are used as the thrombus probability of the initial thrombus image region.

[0024] Optionally, the generating the regional thrombus prediction information of the associated region based on the thrombus probability of each initial thrombus image region and the thrombus range of each initial thrombus image region includes:

[0025] Projecting the thrombus range of each of the initial thrombus image regions onto the current regional image of the regional image of the associated region to obtain regional thrombus range information;

[0026] The probability of sub-thrombus at each position point in each thrombus range is subjected to probability marking processing in the regional thrombus range information to obtain the regional thrombus prediction information of the associated region.

[0027] In addition, to achieve the above-mentioned purpose, the present invention also provides an extracorporeal circulation thrombus identification system based on hyperspectral images, and the extracorporeal circulation thrombus identification system based on hyperspectral images comprises:

[0028] An acquisition module, used to acquire a current regional image of a region associated with the extracorporeal circulation treatment on the user's body surface through a hyperspectral image acquisition software, and based on the current regional image, identify each initial thrombus image region in the regional image and thrombus image data of each initial thrombus image region through an image recognition network;

[0029] A generating module, used for collecting regional pressure detection information of each initial thrombus image region, and generating regional pressure distribution information of each initial thrombus image region based on each regional pressure detection information;

[0030] An identification module is used to identify the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area based on the regional pressure distribution information of each initial thrombus image area and the thrombus image data of each initial thrombus image area through a thrombus prediction strategy, and generate regional thrombus prediction information of the associated area based on the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area.

[0031] Optionally, the acquisition module is specifically used to:

[0032] Performing grid splitting processing on the current region image to obtain a plurality of grid images, and inputting each grid image into an image recognition network to identify each thrombus grid image with a probability of thrombus existence;

[0033] Using an image edge recognition algorithm, the thrombus image edge information in each thrombus grid image is respectively recognized, and the image range enclosed by the thrombus image edge information in each thrombus grid image is used as the sub-thrombus image data in each thrombus grid image;

[0034] Adjacent thrombus grid images are subjected to image aggregation processing to obtain initial thrombus image areas, and sub-thrombus image data of each thrombus grid image in each initial thrombus image area are subjected to data splicing processing to obtain thrombus image data of each initial thrombus image area.

[0035] Optionally, the generating module is specifically used for:

[0036] Based on the thrombus image data of each initial thrombus image region, identifying the position range of each thrombus image in the associated region;

[0037] Based on the position range of each thrombus image, in the associated area, each target detection area range that needs to be subjected to regional pressure detection is screened, and the target detection area range is sent to the staff client, instructing the staff to collect target pressure detection information of each target detection area range through the blood pressure sensor;

[0038] According to the sub-region range of each initial thrombus image region, in each target pressure detection information, the sub-pressure detection information of each sub-region range is screened, and the sub-pressure detection information of each sub-region range is used as the regional pressure detection information of the initial thrombus image region to which each sub-region range belongs.

[0039] Optionally, the generating module is specifically used for:

[0040] For each initial thrombus image region, based on the sub-pressure detection information of the initial thrombus image region, identifying the pressure data of each position point in the sub-region range of the initial thrombus image region;

[0041] The pressure data of each of the position points is distributed and arranged to obtain regional pressure distribution information of the initial thrombus image region.

[0042] Optionally, the identification module is specifically used to:

[0043] For each initial thrombus image region, based on the regional pressure distribution information of the initial thrombus image region, a high-pressure regional range in the initial thrombus image region is identified, and based on the high-pressure regional range and an image range corresponding to the thrombus image data of the initial thrombus image region, an overlapping image range of the initial thrombus image region is identified;

[0044] The overlapping image range is used as the thrombus range of the initial thrombus image region, and based on the pressure data of each position point in the thrombus range, the probability of a sub-thrombus at each position point in the thrombus range is predicted by a thrombus prediction strategy;

[0045] The sub-thrombus probabilities of all the position points are used as the thrombus probability of the initial thrombus image region.

[0046] Optionally, the identification module is specifically used to:

[0047] Projecting the thrombus range of each of the initial thrombus image regions onto the current regional image of the regional image of the associated region to obtain regional thrombus range information;

[0048] The probability of sub-thrombus at each position point in each thrombus range is subjected to probability marking processing in the regional thrombus range information to obtain the regional thrombus prediction information of the associated region.

[0049] In a third aspect, the present application provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of any one of the methods in the first aspect are implemented.

[0050] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0051] In a fifth aspect, the present application provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0052] The present invention provides a method and system for identifying extracorporeal circulation thrombus based on hyperspectral images. The method comprises: using hyperspectral image acquisition software to acquire a current regional image of an associated region of a user's body surface during extracorporeal circulation treatment, and based on the current regional image, using an image recognition network to identify each initial thrombus image region in the regional image, and thrombus image data of each initial thrombus image region; acquiring regional pressure detection information of each initial thrombus image region, and generating regional pressure distribution information of each initial thrombus image region based on each regional pressure detection information; based on the regional pressure distribution information of each initial thrombus image region and the thrombus image data of each initial thrombus image region, using a thrombus prediction strategy to identify the thrombus probability of each initial thrombus image region and the thrombus range of each initial thrombus image region, and generating regional thrombus prediction information of the associated region based on the thrombus probability of each initial thrombus image region and the thrombus range of each initial thrombus image region. This solution uses hyperspectral image acquisition software to identify in real time the current regional image of the user's body surface in the associated area of ​​extracorporeal circulation treatment, thereby first identifying the thrombus image area where there may be a probability of thrombus, and then using the pressure sensor to detect the regional pressure detection information in each thrombus image area, thereby avoiding the problem that a single position setting cannot detect thrombi in other positions. Then, the terminal uses the regional pressure distribution information of each initial thrombus image area and the thrombus image data of each initial thrombus image area to identify the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area through a thrombus prediction strategy, thereby determining the regional thrombus prediction information of the associated area, thereby improving the advance warning and real-time monitoring of the extracorporeal circulation coagulation reaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1is a flow chart of a method for identifying extracorporeal circulation thrombus based on hyperspectral images provided by an embodiment of the present invention;

[0055] Figure 2 It is a structural schematic diagram of an extracorporeal circulation thrombus identification system based on hyperspectral images provided in an embodiment of the present invention;

[0056] Figure 3 An internal structure diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The extracorporeal circulation thrombus identification method based on hyperspectral images provided in an embodiment of the present invention is applied to an extracorporeal circulation thrombus identification system based on hyperspectral images. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" in the specification and claims of the present application and the above-mentioned drawings and any variations thereof are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0058] 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 application. The appearance of the phrase in various locations 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.

[0059] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0060] The method for identifying thrombus in extracorporeal circulation based on hyperspectral images provided in the embodiment of the present application can be applied to the application environment of adjusting the parameters of the ventilator. The method can be applied to a terminal, a server, or a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, etc. The terminal uses hyperspectral image acquisition software to identify the current regional image of the user's body surface in the associated area of ​​extracorporeal circulation treatment in real time, thereby first identifying the thrombus image area where the probability of thrombus may exist, and then detecting the regional pressure detection information in each thrombus image area through a pressure sensor, thereby avoiding the problem that a single position setting cannot detect thrombus in other positions. Then, the terminal uses the regional pressure distribution information of each initial thrombus image area and the thrombus image data of each initial thrombus image area, and through the thrombus prediction strategy, identifies the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area, thereby determining the regional thrombus prediction information of the associated area, and improving the prediction comprehensiveness and detection accuracy of the monitoring accuracy of the coagulation reaction.

[0061] In one embodiment, Figure 1 As shown, a method for identifying extracorporeal circulation thrombus based on hyperspectral images is provided, and the method is described by taking the application of the method to a terminal as an example, and includes the following steps:

[0062] Step S101, using hyperspectral image acquisition software, collects the current regional image of the user's body surface in the associated area of ​​extracorporeal circulation treatment, and based on the current regional image, identifies each initial thrombus image area in the regional image and the thrombus image data of each initial thrombus image area through an image recognition network.

[0063] In this embodiment, the terminal collects the current regional image of the user's body surface in the associated area of ​​extracorporeal circulation treatment through image acquisition software based on hyperspectral technology, wherein the associated area is the area preset by the staff in the terminal, and the area is the location area of ​​the body surface connected to the staff when performing extracorporeal circulation treatment on the user's body surface, and the location area connected to the location area. Then, based on the current regional image, the terminal identifies each initial thrombus image area in the regional image and the thrombus image data of each initial thrombus image area through an image recognition network. Among them, the initial thrombus image area is an image area where there may be a probability of thrombus, and the thrombus image data is the image range that may be a thrombus in the image area. The image recognition network is a convolutional neural network based on a self-attention mechanism.

[0064] Step S102 : collecting regional pressure detection information of each initial thrombus image region, and generating regional pressure distribution information of each initial thrombus image region based on each regional pressure detection information.

[0065] In this embodiment, the terminal collects regional pressure detection information of each initial thrombus image area through a pressure sensor, and generates regional pressure distribution information of each initial thrombus image area based on each regional pressure detection information. The regional pressure distribution information is the distribution information of pressure data of each position point in the thrombus range in the initial thrombus image area. Each position point is a position point between the intervals preset by the staff. The preset interval distances can be 0.5mm, 0.8mm, 1mm and other interval distances.

[0066] Step S103, based on the regional pressure distribution information of each initial thrombus image area and the thrombus image data of each initial thrombus image area, the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area are identified through a thrombus prediction strategy, and based on the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area, regional thrombus prediction information of the associated area is generated.

[0067] In this embodiment, the terminal identifies the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area through a thrombus prediction strategy based on the regional pressure distribution information of each initial thrombus image area and the thrombus image data of each initial thrombus image area, and generates regional thrombus prediction information of the associated area based on the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area. The thrombus prediction strategy includes thrombus probability values ​​corresponding to each pressure data range. The specific process of identifying the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area will be described in detail later.

[0068] Based on the above scheme, the current regional image of the associated area of ​​the user's body surface during extracorporeal circulation treatment is identified in real time through the hyperspectral image acquisition software, so as to first identify the thrombus image area where the probability of thrombus may exist, and then detect the regional pressure detection information in each thrombus image area through the pressure sensor, thereby avoiding the problem that a single position setting cannot detect thrombus in other positions. Then, the terminal uses the regional pressure distribution information of each initial thrombus image area and the thrombus image data of each initial thrombus image area to identify the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area through the thrombus prediction strategy, thereby determining the regional thrombus prediction information of the associated area, thereby improving the comprehensiveness of the prediction and the detection accuracy of the monitoring accuracy of the coagulation reaction.

[0069] Optionally, based on the current regional image, each initial thrombus image region in the regional image and the thrombus image data of each initial thrombus image region are identified through an image recognition network, including: grid splitting the current regional image to obtain multiple grid images, and inputting each grid image into the image recognition network to identify each thrombus grid image with a probability of thrombus existence; identifying the thrombus image edge information in each thrombus grid image through an image edge recognition algorithm, and using the image range enclosed by the thrombus image edge information in each thrombus grid image as the sub-thrombus image data in each thrombus grid image; performing image aggregation processing on adjacent thrombus grid images to obtain each initial thrombus image region, and performing data splicing processing on the sub-thrombus image data of each thrombus grid image in each initial thrombus image region to obtain the thrombus image data of each initial thrombus image region.

[0070] In this embodiment, the terminal performs grid splitting processing on the current area image to obtain multiple grid images, and inputs each grid image into the image recognition network to identify each thrombus grid image with a probability of thrombus existence. Among them, the grid size of each grid image is the grid size preset by the staff in the terminal. By performing thrombus identification after grid splitting, the accuracy and comprehensiveness of thrombus identification can be improved. Among them, each thrombus grid image with a probability of thrombus existence is a grid image that is recognized as a thrombus grid image by the terminal through the image recognition network. The image recognition network is an image recognition network that is trained in large quantities through sample images of various annotated thrombus ranges and thrombus samples. The image recognition network can identify ten images as thrombus images or non-thrombus images.

[0071] Then, the terminal recognizes the thrombus image edge information in each thrombus grid image respectively through an image edge recognition algorithm, and uses the image range enclosed by the thrombus image edge information in each thrombus grid image as the sub-thrombus image data in each thrombus grid image.

[0072] Then, the terminal performs image aggregation processing on the adjacent thrombus grid images to obtain initial thrombus image regions, and performs data splicing processing on the sub-thrombus image data of each thrombus grid image in each initial thrombus image region to obtain thrombus image data of each initial thrombus image region. The image aggregation processing is to splice the adjacent thrombus grid images according to the positions of each thrombus grid image in the current region image to obtain the initial thrombus image region.

[0073] Based on the above scheme, the thrombus grid image is identified by grid splitting, which improves the recognition accuracy of the thrombus grid image. Then, the edge recognition algorithm and aggregation processing are used to improve the accuracy of the identified initial thrombus image area and sub-thrombus image data.

[0074] Optionally, regional pressure detection information of each initial thrombus image area is collected, including: based on the thrombus image data of each initial thrombus image area, identifying each thrombus image position range in the associated area; based on each thrombus image position range, in the associated area, screening each target detection area range that requires regional pressure detection, and sending the target detection area range to the staff client, guiding the staff to collect target pressure detection information of each target detection area range through the blood pressure sensor; according to the sub-area range of each initial thrombus image area, in each target pressure detection information, screening the sub-pressure detection information of each sub-area range, and using the sub-pressure detection information of each sub-area range as the regional pressure detection information of the initial thrombus image area to which each sub-area range belongs.

[0075] In this embodiment, the terminal identifies the position range of each thrombus image in the associated area based on the thrombus image data of each initial thrombus image area.

[0076] Based on the position range of each thrombus image, in the associated area, the target detection area ranges that need to be tested for regional pressure are screened, and the target detection area ranges are sent to the staff client to guide the staff to collect the target pressure detection information of each target detection area range through the blood pressure sensor. Among them, the target detection area range is the area range obtained by extending the thrombus image position range outward by a preset width. The preset width is the width preset by the staff in the terminal, thereby improving the comprehensiveness of the collection of pressure detection information on possible thrombi in the thrombus image position range.

[0077] Finally, the terminal filters the sub-pressure detection information of each sub-region range in each target pressure detection information according to the sub-region range of each initial thrombus image region, and uses the sub-pressure detection information of each sub-region range as the regional pressure detection information of the initial thrombus image region to which each sub-region range belongs.

[0078] Based on the above scheme, by setting the target detection area range larger than the thrombus image position range, the comprehensiveness of pressure detection of areas where thrombus may exist is improved.

[0079] Optionally, based on the pressure detection information of each region, regional pressure distribution information of each initial thrombus image region is generated, including: for each initial thrombus image region, based on the sub-pressure detection information of the initial thrombus image region, identifying the pressure data of each position point in the sub-region range of the initial thrombus image region; distributing and arranging the pressure data of each position point to obtain the regional pressure distribution information of the initial thrombus image region.

[0080] In this embodiment, the terminal identifies the pressure data of each position point in the sub-region range of the initial thrombus image region based on the sub-pressure detection information of the initial thrombus image region for each initial thrombus image region. Finally, the terminal distributes and arranges the pressure data of each position point through a position distribution arrangement strategy to obtain the regional pressure distribution information of the initial thrombus image region. The distribution arrangement strategy is to generate the pressure distribution information in the sub-region range of the initial blood vessel image region through a distribution fitting algorithm based on the pressure data of each position point.

[0081] Based on the above solution, the comprehensiveness of the recognition of pressure distribution information is improved by sorting the pressure data by distribution.

[0082] Optionally, based on the regional pressure distribution information of each initial thrombus image area and the thrombus image data of each initial thrombus image area, a thrombus prediction strategy is used to identify the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area, including: for each initial thrombus image area, based on the regional pressure distribution information of the initial thrombus image area, identifying the high-pressure area range in the initial thrombus image area, and based on the high-pressure area range and the image range corresponding to the thrombus image data of the initial thrombus image area, identifying the overlapping image range of the initial thrombus image area; using the overlapping image range as the thrombus range of the initial thrombus image area, and based on the pressure data of each position point in the thrombus range, predicting the sub-thrombus probability of each position point in the thrombus range through the thrombus prediction strategy; using the sub-thrombus probabilities of all position points as the thrombus probability of the initial thrombus image area.

[0083] In this embodiment, the terminal identifies the high pressure area range in the initial thrombus image area based on the regional pressure distribution information of the initial thrombus image area for each initial thrombus image area, and identifies the overlapping image range of the initial thrombus image area based on the high pressure area range and the image range corresponding to the thrombus image data of the initial thrombus image area. The high pressure area range is a range where the pressure data is greater than a preset pressure data threshold.

[0084] Finally, the terminal uses the overlapping image range as the thrombus range of the initial thrombus image area, and based on the pressure data of each position point in the thrombus range, predicts the sub-thrombus probability of each position point in the thrombus range through the thrombus prediction strategy. The sub-thrombus probabilities of all position points are used as the thrombus probability of the initial thrombus image area. The thrombus prediction strategy includes thrombus probability values ​​corresponding to each pressure data range. The terminal identifies the sub-thrombus probability of each position point by identifying the pressure data of each position point, and then generates the thrombus probability distribution information of the overlapping image range of the initial thrombus image area through the distribution fitting algorithm, and uses the thrombus probability distribution information as the thrombus probability of the initial thrombus image area.

[0085] Based on the above scheme, by combining the range of image recognition and the range overlapping with the range of sensor recognition, the thrombus range and thrombus probability of the initial thrombus image area are identified, thereby improving the accuracy and comprehensiveness of identifying the thrombus range and thrombus probability.

[0086] Optionally, based on the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area, regional thrombus prediction information of the associated area is generated, including: projecting the thrombus range of each initial thrombus image area to the current regional image of the regional image of the associated area to obtain the regional thrombus range information; and performing probability marking processing on the sub-thrombus probability of each position point in each thrombus range in the regional thrombus range information to obtain the regional thrombus prediction information of the associated area.

[0087] In this embodiment, the terminal projects the thrombus range of each initial thrombus image area into the current regional image of the associated area to obtain regional thrombus range information. Finally, the terminal performs probability marking processing on the sub-thrombus probability of each position point in each thrombus range in the regional thrombus range information to obtain regional thrombus prediction information of the associated area.

[0088] Based on the above scheme, by marking the identified thrombus ranges and thrombus probabilities with images, it is convenient for staff to identify the thrombus prediction information of related areas, and to guide the thrombus prediction information of related areas efficiently, so that staff can promptly detect coagulation reactions and promptly deal with the thrombi caused by coagulation reactions, thereby improving the accuracy and efficiency of monitoring coagulation reactions.

[0089] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0090] Based on the same inventive concept, the embodiment of the present application also provides an extracorporeal circulation thrombus identification system based on hyperspectral images for implementing the extracorporeal circulation thrombus identification method based on hyperspectral images involved above. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more embodiments of the extracorporeal circulation thrombus identification system based on hyperspectral images provided below can refer to the limitations of the extracorporeal circulation thrombus identification method based on hyperspectral images above, and will not be repeated here.

[0091] Further references Figure 2 , as a response to the above Figure 1 The implementation of the method shown in the figure, the present application provides an embodiment of an extracorporeal circulation thrombus identification system 200 based on hyperspectral images, the extracorporeal circulation thrombus identification system based on hyperspectral images comprises an acquisition module 210, a generation module 220 and an identification module 230, wherein:

[0092] The acquisition module 210 is used to acquire a current regional image of a region associated with the extracorporeal circulation treatment on the user's body surface through a hyperspectral image acquisition software, and based on the current regional image, identify each initial thrombus image region in the regional image and thrombus image data of each initial thrombus image region through an image recognition network;

[0093] A generating module 220, configured to collect regional pressure detection information of each initial thrombus image region, and generate regional pressure distribution information of each initial thrombus image region based on each regional pressure detection information;

[0094] The identification module 230 is used to identify the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area based on the regional pressure distribution information of each initial thrombus image area and the thrombus image data of each initial thrombus image area through a thrombus prediction strategy, and generate regional thrombus prediction information of the associated area based on the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area.

[0095] Optionally, the acquisition module 210 is specifically used for:

[0096] Performing grid splitting processing on the current region image to obtain a plurality of grid images, and inputting each grid image into an image recognition network to identify each thrombus grid image with a probability of thrombus existence;

[0097] Using an image edge recognition algorithm, the thrombus image edge information in each thrombus grid image is respectively recognized, and the image range enclosed by the thrombus image edge information in each thrombus grid image is used as the sub-thrombus image data in each thrombus grid image;

[0098] Adjacent thrombus grid images are subjected to image aggregation processing to obtain initial thrombus image areas, and sub-thrombus image data of each thrombus grid image in each initial thrombus image area are subjected to data splicing processing to obtain thrombus image data of each initial thrombus image area.

[0099] Optionally, the generating module 220 is specifically configured to:

[0100] Based on the thrombus image data of each initial thrombus image region, identifying the position range of each thrombus image in the associated region;

[0101] Based on the position range of each thrombus image, in the associated area, each target detection area range that needs to be subjected to regional pressure detection is screened, and the target detection area range is sent to the staff client, instructing the staff to collect target pressure detection information of each target detection area range through the blood pressure sensor;

[0102] According to the sub-region range of each initial thrombus image region, in each target pressure detection information, the sub-pressure detection information of each sub-region range is screened, and the sub-pressure detection information of each sub-region range is used as the regional pressure detection information of the initial thrombus image region to which each sub-region range belongs.

[0103] Optionally, the generating module 220 is specifically configured to:

[0104] For each initial thrombus image region, based on the sub-pressure detection information of the initial thrombus image region, identifying the pressure data of each position point in the sub-region range of the initial thrombus image region;

[0105] The pressure data of each of the position points is distributed and arranged to obtain regional pressure distribution information of the initial thrombus image region.

[0106] Optionally, the identification module 230 is specifically configured to:

[0107] For each initial thrombus image region, based on the regional pressure distribution information of the initial thrombus image region, a high-pressure regional range in the initial thrombus image region is identified, and based on the high-pressure regional range and an image range corresponding to the thrombus image data of the initial thrombus image region, an overlapping image range of the initial thrombus image region is identified;

[0108] The overlapping image range is used as the thrombus range of the initial thrombus image region, and based on the pressure data of each position point in the thrombus range, the probability of a sub-thrombus at each position point in the thrombus range is predicted by a thrombus prediction strategy;

[0109] The sub-thrombus probabilities of all the position points are used as the thrombus probability of the initial thrombus image region.

[0110] Optionally, the identification module 230 is specifically configured to:

[0111] Projecting the thrombus range of each of the initial thrombus image regions onto the current regional image of the regional image of the associated region to obtain regional thrombus range information;

[0112] The probability of sub-thrombus at each position point in each thrombus range is subjected to probability marking processing in the regional thrombus range information to obtain the regional thrombus prediction information of the associated region.

[0113] Each module in the above-mentioned hyperspectral image-based extracorporeal circulation thrombus identification system can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each of the above modules.

[0114] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input system connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for identifying extracorporeal circulation thrombus based on hyperspectral images is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input system of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0115] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0116] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of any one of the methods in the first aspect are implemented.

[0117] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0118] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the steps of any one of the methods in the first aspect.

[0119] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0120] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0121] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0122] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for identifying extracorporeal thrombus based on hyperspectral images, characterized in that: The method comprises: The current regional image of the user's body surface in the associated area of ​​the extracorporeal circulation treatment is collected by means of the hyperspectral image acquisition software, and based on the current regional image, each initial thrombus image region in the regional image and the thrombus image data of each initial thrombus image region are identified by means of the image recognition network; Acquiring regional pressure detection information of each initial thrombus image region, and generating regional pressure distribution information of each initial thrombus image region based on each regional pressure detection information; Based on the regional pressure distribution information of each initial thrombus image area and the thrombus image data of each initial thrombus image area, a thrombus prediction strategy is used to identify the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area, and based on the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area, regional thrombus prediction information of the associated area is generated.

2. The method according to claim 1, characterized in that The identifying, based on the current regional image, each initial thrombus image region in the regional image and the thrombus image data of each initial thrombus image region through an image recognition network includes: Performing grid splitting processing on the current region image to obtain a plurality of grid images, and inputting each grid image into an image recognition network to identify each thrombus grid image with a probability of thrombus existence; Using an image edge recognition algorithm, the thrombus image edge information in each thrombus grid image is respectively recognized, and the image range enclosed by the thrombus image edge information in each thrombus grid image is used as the sub-thrombus image data in each thrombus grid image; Adjacent thrombus grid images are subjected to image aggregation processing to obtain initial thrombus image areas, and sub-thrombus image data of each thrombus grid image in each initial thrombus image area are subjected to data splicing processing to obtain thrombus image data of each initial thrombus image area.

3. The method according to claim 2, characterized in that The collecting of regional pressure detection information of each initial thrombus image region includes: Based on the thrombus image data of each initial thrombus image region, identifying the position range of each thrombus image in the associated region; Based on the position range of each thrombus image, in the associated area, each target detection area range that needs to be subjected to regional pressure detection is screened, and the target detection area range is sent to the staff client, instructing the staff to collect target pressure detection information of each target detection area range through the blood pressure sensor; According to the sub-region range of each initial thrombus image region, in each target pressure detection information, the sub-pressure detection information of each sub-region range is screened, and the sub-pressure detection information of each sub-region range is used as the regional pressure detection information of the initial thrombus image region to which each sub-region range belongs.

4. The method according to claim 3, characterized in that The generating of regional pressure distribution information of each initial thrombus image region based on each regional pressure detection information includes: For each initial thrombus image region, based on the sub-pressure detection information of the initial thrombus image region, identifying the pressure data of each position point in the sub-region range of the initial thrombus image region; The pressure data of each of the position points is distributed and arranged to obtain regional pressure distribution information of the initial thrombus image region.

5. The method according to claim 4, characterized in that The method of identifying the thrombus probability of each initial thrombus image region and the thrombus range of each initial thrombus image region through a thrombus prediction strategy based on the regional pressure distribution information of each initial thrombus image region and the thrombus image data of each initial thrombus image region includes: For each initial thrombus image region, based on the regional pressure distribution information of the initial thrombus image region, identifying a pressure regional range in the initial thrombus image region, and based on the pressure regional range and an image range corresponding to the thrombus image data of the initial thrombus image region, identifying an overlapping image range of the initial thrombus image region; The overlapping image range is used as the thrombus range of the initial thrombus image region, and based on the pressure data of each position point in the thrombus range, the probability of a sub-thrombus at each position point in the thrombus range is predicted by a thrombus prediction strategy; The sub-thrombus probabilities of all the position points are used as the thrombus probability of the initial thrombus image region.

6. The method according to claim 4, characterized in that The generating of regional thrombus prediction information of the associated region based on the thrombus probability of each initial thrombus image region and the thrombus range of each initial thrombus image region comprises: Projecting the thrombus range of each of the initial thrombus image regions onto the current regional image of the regional image of the associated region to obtain regional thrombus range information; The probability of sub-thrombus at each position point in each thrombus range is subjected to probability marking processing in the regional thrombus range information to obtain the regional thrombus prediction information of the associated region.

7. An extracorporeal circulation thrombus identification system based on hyperspectral images, characterized in that: The system comprises: An acquisition module, used to acquire a current regional image of a region associated with the extracorporeal circulation treatment on the user's body surface through a hyperspectral image acquisition software, and based on the current regional image, identify each initial thrombus image region in the regional image and thrombus image data of each initial thrombus image region through an image recognition network; A generating module, used for collecting regional pressure detection information of each initial thrombus image region, and generating regional pressure distribution information of each initial thrombus image region based on each regional pressure detection information; An identification module is used to identify the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area based on the regional pressure distribution information of each initial thrombus image area and the thrombus image data of each initial thrombus image area through a thrombus prediction strategy, and generate regional thrombus prediction information of the associated area based on the thrombus probability of each initial thrombus image area and the thrombus range of each initial thrombus image area.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images