Method, device, equipment and medium for detecting in-stent restenosis
Through machine learning-based methods, the stent feature information is extracted from original medical images and the stenosis of the vascular segment is analyzed, which solves the problem of intrastent restenosis detection in the prior art, and achieves efficient and reliable diagnostic support.
Patent Information
- Application Number
- CN202210389542.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-04-14
- Publication Date
- 2025-04-04
- Estimated Expiration
- 2042-04-14
AI Technical Summary
The prior art is difficult to effectively detect and predict restenosis in the stent, resulting in high diagnostic pressure, inconsistent results and lack of reproducibility.
By obtaining the user's original medical images, the characteristic information of the scaffold is extracted based on the machine learning model, the blood vessel segment to be examined is determined, and feature extraction and analysis are performed to obtain stenosis information, including the location, degree and plaque type.
Effective analysis and prediction of restenosis in the stent is achieved, the diagnostic process is simplified, the reliability and reproducibility of the results are improved, and the diagnostic pressure of doctors is reduced.
Smart Images
Figure CN114881935B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of image processing technology, and in particular to a method, device, equipment and medium for detecting in-stent restenosis. Background Art
[0002] The growth, rupture, and detachment of plaques in blood vessels pose a fatal risk to patients. Manual diagnosis often requires very experienced doctors to complete, which will bring greater diagnostic pressure to doctors and take longer to diagnose. Secondly, it is susceptible to differences between and within observers. For the same blood vessel, the diagnosis results of different doctors or the same doctor at different times will be different, and manual diagnosis is difficult to reproduce. Summary of the invention
[0003] The main purpose of the present application is to provide a method, device, equipment and medium for detecting restenosis in a stent, so as to achieve effective analysis and prediction of restenosis in the stent.
[0004] To achieve the above-mentioned purpose, the present application provides a method for detecting in-stent restenosis, which is characterized by comprising:
[0005] Obtaining original medical images of the user to be detected;
[0006] Acquire characteristic information of the stent to be inspected based on the original medical image, and acquire a blood vessel segment to be inspected including the stent to be inspected according to the characteristic information of the stent to be inspected;
[0007] Extracting features of the blood vessel segment to be inspected to obtain feature information of the blood vessel segment to be inspected;
[0008] The characteristic information of the blood vessel segment to be inspected is analyzed to obtain stenosis information of the blood vessel segment to be inspected.
[0009] Preferably, the original medical image includes: at least one of a CT image, an intravascular image or an angiography image. Specifically, the intravascular image may be intravascular ultrasound or optical coherence tomography (OCT).
[0010] Preferably, the determining of the blood vessel segment to be inspected that contains the stent to be inspected comprises:
[0011] The stent information of the blood vessel segment to be inspected is predicted based on the stent information in the original medical image and in combination with a prediction method of a preset stent detection model based on machine learning.
[0012] Preferably, the prediction method of the preset stent detection model based on machine learning includes:
[0013] The machine learning model for stent positioning is trained in advance by using the original medical image of the first blood vessel segment containing the first stent as direct input, so that the candidate area of the first stent predicted by the machine learning model for stent positioning includes the first stent.
[0014] Preferably, the step of acquiring characteristic information of the stent to be inspected based on the original medical image, and dividing the blood vessel segment to be inspected containing the stent to be inspected according to the information of the stent to be inspected comprises:
[0015] The position information and size information of the blood vessel segment to be inspected are determined according to the position information and size information of the stent to be inspected.
[0016] Preferably, the acquiring the stenosis information of the to-be-detected blood vessel segment includes:
[0017] Obtain information such as stenosis information, stenosis location, stenosis degree, plaque type, etc. of the blood vessel segment to be inspected.
[0018] Preferably, the step of analyzing the characteristic information of the to-be-tested blood vessel segment to obtain the stenosis information of the to-be-tested blood vessel segment includes:
[0019] The original medical image of the first blood vessel segment containing the first stent is used as direct input in advance to train a machine learning model for analyzing stenosis information within the first stent, so that the information analyzed by the machine learning model for analyzing stenosis information within the first stent includes the stenosis information of the first blood vessel segment.
[0020] To achieve the above-mentioned object, the present application further provides an in-stent restenosis detection device, the in-stent restenosis detection device comprising:
[0021] A detection system, used to obtain an original medical image of a user's blood vessel, obtain the position information of the user's stent to be inspected according to the original medical image, and then analyze the stenosis information of the blood vessel segment to be inspected according to the position information of the stent to be inspected;
[0022] The analysis system is used to analyze at least one of the following: whether stenosis exists, the location of stenosis, the degree of stenosis, and the type of plaque in the blood vessel segment to be inspected, based on the stenosis information of the blood vessel segment to be inspected.
[0023] Furthermore, to achieve the above-mentioned purpose, the present application also provides an in-stent restenosis detection device, the in-stent restenosis detection device comprising a memory, a processor, and an in-stent restenosis detection program stored in the memory and executable on the processor, wherein the in-stent restenosis detection program, when executed by the processor, implements the steps of the in-stent restenosis detection method described above.
[0024] Furthermore, to achieve the above-mentioned purpose, the present application also provides a medium, wherein the computer-readable storage medium stores an in-stent restenosis detection program, and when the in-stent restenosis detection program is executed by a processor, the steps of the above-mentioned in-stent restenosis detection method are implemented.
[0025] Furthermore, to achieve the above-mentioned purpose, the present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned in-stent restenosis detection method when executed by a processor.
[0026] The embodiments of the present application provide a method, device, equipment and medium for detecting in-stent restenosis. The in-stent restenosis detection method of the present application includes obtaining the original medical image of the user to be detected; obtaining the characteristic information of the stent to be detected based on the original medical image, and obtaining the blood vessel segment to be detected containing the stent to be detected according to the characteristic information of the stent to be detected; extracting the characteristics of the blood vessel segment to be detected to obtain the characteristic information of the blood vessel segment to be detected; analyzing the characteristic information of the blood vessel segment to be detected to obtain the stenosis information of the blood vessel segment to be detected. The present application can obtain the characteristic information of the stent to be detected based on the original medical image of the user's blood vessel, and extract the characteristic information of the blood vessel segment to be detected based on the characteristic information of the stent to be detected, so as to obtain the characteristic information of the blood vessel segment to be detected of the user's blood vessel, so as to obtain the stenosis information of the user's blood vessel segment according to the characteristic information of the blood vessel segment to be detected. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 This is a schematic diagram of the structure of the hardware operating environment involved in the embodiment of the method for detecting in-stent restenosis of the present application;
[0028] Figure 2 This is a flow chart of the first embodiment of the in-stent restenosis detection method of the present application;
[0029] Figure 3 Schematic diagram of functional modules of a preferred embodiment of an in-stent restenosis detection device.
[0030] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0031] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0032] The main solution of the embodiment of the present invention is: obtaining the original medical image of the user's blood vessels, obtaining the characteristic information of the stent to be inspected based on the original medical image, and obtaining the blood vessel segment to be inspected containing the stent to be inspected according to the characteristic information of the stent to be inspected; extracting the characteristics of the blood vessel segment to be inspected to obtain the characteristic information of the blood vessel segment to be inspected; analyzing the characteristic information of the blood vessel segment to be inspected to obtain the stenosis information of the blood vessel segment to be inspected. The present application can obtain the characteristic information of the stent to be inspected based on the original medical image of the user's blood vessels, and extract the characteristic information of the blood vessel segment to be inspected based on the characteristic information of the stent to be inspected, so as to obtain the stenosis information of the user's blood vessel segment according to the characteristic information of the blood vessel segment to be inspected.
[0033] The professional terms involved in the embodiments of the present invention include:
[0034] Intra-observer difference: It is often used in clinical experimental observations. When the same observer observes the same set of data, different subjective and objective factors may lead to different observation results. For example, the observer's psychological state: nervous or calm, physiological state: good spirits or tired, etc.; external objective environment: the brightness of the light, the noisiness of the sound, etc. The above are all important factors that affect the intra-observer difference.
[0035] Inter-observer difference: Commonly used in clinical trial observations, different observers may observe the same set of data and have different results due to different subjective and objective factors. The experience level of the observers and factors that affect intra-observer differences.
[0036] CTA: Computed tomography angiography; refers to the use of multi-slice spiral CT (Computed Tomography) to scan the coronary arteries and obtain coronary tomography images to understand the condition of coronary artery lesions.
[0037] Plaque: Plaque refers to the diseased tissue that grows in the vascular dissection; plaque has many harmful effects on blood vessels, such as: compressing the vascular lumen, affecting blood supply; rupturing and falling off into the blood vessels to form blood clots.
[0038] Machine Learning: Machine Learning is a multidisciplinary discipline that covers knowledge of probability theory, statistics, approximate theory and complex algorithms. It uses computers as tools and is committed to simulating human learning methods in real time, and divides existing content into knowledge structures to effectively improve learning efficiency.
[0039] The present application embodiment takes into account that there are relatively few detection technologies for in-stent restenosis at present, which generally focus on the detection of stenosis in blood vessels, but there is almost no detection of in-stent restenosis. That is, there is currently no complete and systematic in-stent restenosis detection method to predict the stenosis in the stent, including the stenosis location, stenosis degree and plaque type prediction.
[0040] Therefore, an embodiment of the present invention proposes a solution that can evaluate and monitor the plaque growth risk and plaque rupture risk of a user's blood vessels.
[0041] like Figure 1 As shown, Figure 1 It is a schematic diagram of the structure of a vascular plaque risk prediction device in the hardware operating environment involved in the embodiment of the present application.
[0042] In the subsequent description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present application, and have no specific meanings. Therefore, "module", "component" or "unit" can be used in a mixed manner.
[0043] The in-stent restenosis detection device in the embodiment of the present application may be a PC, or a portable terminal device such as a tablet computer or a portable computer.
[0044] like Figure 1 As shown, the in-stent restenosis detection device may include: a processor, such as a CPU, a network interface, a user interface, a memory, and a communication bus. The communication bus is used to realize the connection and communication between these components. The user interface may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface may also include a standard wired interface and a wireless interface. The network interface may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory may be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a disk memory. The memory may also be a storage device independent of the aforementioned processor.
[0045] Those skilled in the art will understand that Figure 1 The structure of the vascular plaque risk prediction device shown in the figure does not constitute a limitation on the in-stent restenosis detection device, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.
[0046] like Figure 1 As shown, the memory as a storage medium may include an operating system, a network communication module, a user interface module, and an in-stent restenosis detection program.
[0047] exist Figure 1 In the device shown, the network interface is mainly used to connect to the backend server and perform data communication with the backend server; the user interface is mainly used to connect to the client (user end) and perform data communication with the client; and the processor can be used to call the in-stent restenosis detection program stored in the memory and perform the following operations:
[0048] Obtaining original medical images of the user to be detected;
[0049] Acquire characteristic information of the stent to be inspected based on the original medical image, and acquire a blood vessel segment to be inspected including the stent to be inspected according to the characteristic information of the stent to be inspected;
[0050] Extracting features of the blood vessel segment to be inspected to obtain feature information of the blood vessel segment to be inspected;
[0051] The characteristic information of the blood vessel segment to be inspected is analyzed to obtain stenosis information of the blood vessel segment to be inspected.
[0052] Furthermore, the original medical image includes: at least one of a CT image, an intravascular image or an angiography image. Specifically, the intravascular image may be intravascular ultrasound or optical coherence tomography (OCT).
[0053] Further, the determining of the blood vessel segment to be inspected that contains the stent to be inspected comprises:
[0054] The stent information of the blood vessel segment to be inspected is predicted based on the stent information in the original medical image and in combination with a prediction method of a preset stent detection model based on machine learning.
[0055] Furthermore, the prediction method of the preset stent detection model based on machine learning includes:
[0056] The machine learning model for stent positioning is trained in advance by using the original medical image of the first blood vessel segment containing the first stent as direct input, so that the candidate area of the first stent predicted by the machine learning model for stent positioning includes the first stent.
[0057] Furthermore, the step of acquiring characteristic information of the stent to be inspected based on the original medical image, and dividing the blood vessel segment to be inspected containing the stent to be inspected according to the information of the stent to be inspected includes:
[0058] The position information and size information of the blood vessel segment to be inspected are determined according to the position information and size information of the stent to be inspected.
[0059] Furthermore, the obtaining of the stenosis information of the to-be-detected blood vessel segment includes:
[0060] Obtain information such as stenosis information, stenosis location, stenosis degree, plaque type, etc. of the blood vessel segment to be inspected.
[0061] Furthermore, the step of analyzing the characteristic information of the to-be-detected blood vessel segment to obtain the stenosis information of the to-be-detected blood vessel segment includes:
[0062] The original medical image of the first blood vessel segment containing the first stent is used as direct input in advance to train a machine learning model for analyzing stenosis information within the first stent, so that the information analyzed by the machine learning model for analyzing stenosis information within the first stent includes the stenosis information of the first blood vessel segment.
[0063] In order to better understand the above technical solution, exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.
[0064] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0065] Reference Figure 2 , Figure 2 A schematic flow chart of a method for detecting in-stent restenosis provided in the first embodiment of the present application. In this embodiment, the method for detecting in-stent restenosis comprises the following steps:
[0066] Step S10, obtaining the original medical image of the user to be detected;
[0067] In the present embodiment, the in-stent restenosis detection method is applied to the in-stent restenosis detection system. The in-stent restenosis detection method of the present application is executed by the in-stent restenosis detection system. The blood vessel segment to be inspected can be obtained by obtaining the original medical image based on the user's blood vessel, and then the characteristic information of the blood vessel segment to be inspected can be analyzed to obtain the stenosis information of the blood vessel segment to be inspected, so as to effectively predict whether there is restenosis in the area where the stent has been installed.
[0068] Specifically, for users with in-stent restenosis detection, 2D or 3D original medical images of the user's blood vessels are obtained, and the blood vessels include but are not limited to coronary arteries, carotid arteries, and peripheral arteries. The original medical images can be specifically scanned by computer tomography equipment, X-ray angiography equipment, or ultrasound imaging equipment to obtain CTA (CT angiography) images or ICA (Invasive Coronary Angiography) images or IVUS (Intravenous Ultrasound) images or OCT (Optical Coherence Tomography) images as the original medical images of the user's blood vessels. Since the material of the stent is mostly metal, stainless steel, alloy, etc., nuclear magnetic resonance images cannot be used; in vitro ultrasound cannot effectively obtain information such as the position, size, material, etc. of the stent in the body, while IVUS images can directly obtain the stent image in the blood vessel, so only part of the effective original medical images can be obtained through ultrasound imaging.
[0069] Step S20, acquiring characteristic information of the stent to be inspected based on the original medical image, and acquiring a blood vessel segment to be inspected including the stent to be inspected according to the characteristic information of the stent to be inspected;
[0070] After obtaining the CTA image, ICA image, IVUS image or OCT image of the user's blood vessel as the original medical image, the position of the blood vessel segment to be inspected is determined by obtaining the characteristic information of the stent to be inspected, that is, the position information, size information, etc. of the stent to be inspected. Since this scheme is to detect the restenosis condition within the range of the stent, it is necessary to determine the information of the blood vessel segment to be inspected based on the characteristic information of the stent. In addition, by obtaining the characteristic information of the stent to be inspected, the incidence of restenosis can be summarized based on the different characteristic information collected of the stent to be inspected, and a preliminary analysis is performed to determine whether the incidence of restenosis is increased due to factors such as the material, size or specific structure of the stent to be inspected.
[0071] Further, the determining of the blood vessel segment to be inspected that contains the stent to be inspected comprises:
[0072] The stent information of the blood vessel segment to be inspected is predicted based on the stent information in the original medical image and in combination with a prediction method of a preset stent detection model based on machine learning.
[0073] After obtaining a CTA image, an ICA image, an IVUS image or an OCT image of the user's blood vessels as an original medical image, a vascular model of the user's blood vessels is generated according to a prediction algorithm of a stent detection model based on machine learning and the original medical image.
[0074] In this embodiment, the original medical image obtained can also be processed by the stent detection algorithm based on machine learning to generate feature information of the stent to be detected in the user's blood vessel. Specifically, the original medical image can be processed, and the processing method includes but is not limited to data normalization and data augmentation. Data normalization is to linearly map the grayscale value [0-255] of the original medical image to [0, 1]; data augmentation is to expand the data sample, i.e., the original medical image, by translation, rotation, mirroring, brightness change, noise addition and scaling, so as to improve the accuracy and robustness of the machine learning method. Further, before the original medical image obtained is processed by the prediction algorithm of the stent detection model based on machine learning, an initial neural network is constructed based on the current scene, and the medical image processed by data normalization and data augmentation is predicted by the neural network to obtain the predicted result; the predicted result is compared with the manual annotation and fed back to the neural network to update the neural network so that it evolves in the direction of reducing the prediction error; by using a large amount of data (medical images), the above iterative process is repeated thousands of times, and finally the prediction result is close to the manual standard, and a prediction model including stent detection is obtained. Therefore, when the original medical image is processed by the stent detection algorithm based on machine learning, the original medical image is input into the trained vascular prediction model, and the input original medical image is predicted by the stent detection algorithm contained therein, and the predicted region of interest, i.e., the foreground, is used as the vascular segment to be inspected. It should be noted that the stent detection model based on machine learning in this embodiment can be based on traditional machine learning methods, and traditional machine learning methods include but are not limited to random forest method, multi-layer perceptron, Bayesian method, support vector machine, etc.
[0075] When building a stent detection model based on machine learning, a large number of original vascular images will be directly input. We will also collect information such as the material, structure, and type of the stent, and form a statistical table to analyze whether different stents will cause changes in the stenosis rate of the vascular segment.
[0076] Step S30, extracting features of the blood vessel segment to be inspected to obtain feature information of the blood vessel segment to be inspected;
[0077] After obtaining the blood vessel segment to be inspected containing the stent to be inspected according to the characteristic information of the stent to be inspected, the information of the blood vessel segment to be inspected can be determined by a machine learning method, and a new machine learning model for analyzing the stenosis information in the stent to be inspected is established to determine the stenosis information of the blood vessel segment to be inspected of the user's blood vessel;
[0078] After generating the characteristic information of the stent to be inspected of the user's blood vessel according to the original medical image, the algorithm of the machine learning model that analyzes the stenosis information in the stent to be inspected can be used to predict the blood vessel segment to be inspected, and the size information, position information, etc. of the blood vessel segment to be inspected of the user can be predicted. It can be understood that before the prediction is made by the algorithm of the machine learning model that analyzes the stenosis information in the stent to be inspected, an initial neural network is constructed based on the current scene, and the blood vessel segment to be inspected in the original medical image acquired historically is predicted by the neural network to obtain the predicted result; the predicted result is compared with the manual annotation and fed back to the neural network to update the neural network so that it evolves in the direction of reducing the prediction error; by using a large amount of data, the above iterative process is repeated thousands of times, and finally the prediction result is close to the manual standard, and a machine learning model containing the stenosis information in the stent to be inspected is obtained. Therefore, when predicting the blood vessel segment to be inspected by the algorithm of the machine learning model that analyzes the stenosis information in the stent to be inspected, the original medical image is input into the trained machine learning model that analyzes the stenosis information in the stent to be inspected, and the input blood vessel segment to be inspected is predicted by the algorithm of the machine learning model that analyzes the stenosis information in the stent to be inspected, and the characteristic information of the blood vessel segment to be inspected is obtained. In order to facilitate the subsequent analysis and acquisition of the stenosis of the blood vessel segment to be inspected based on the above-mentioned characteristic information. It should be noted that the machine learning model for analyzing the stenosis information in the stent to be inspected in this embodiment can be a model obtained by modeling based on traditional machine learning methods, and traditional machine learning methods include but are not limited to random forest method, multi-layer perceptron, Bayesian method, support vector machine, etc.
[0079] Step S40: analyzing the characteristic information of the blood vessel segment to be inspected, and obtaining stenosis information of the blood vessel segment to be inspected.
[0080] It can be understood that after obtaining the stenosis information of the user's blood vessel segment to be inspected, in order to facilitate medical personnel or users (i.e., patients) to more intuitively understand the risk of stenosis of the user's blood vessel, this embodiment can also display the output stenosis information in the form of quantitative values in the original medical image to provide the user with visual results. For example, the corresponding areas of the blood vessels in the original image are marked with corresponding values, and the corresponding areas can be displayed in corresponding colors according to the values, so as to monitor the user's blood vessels according to the stenosis information and perform stenosis treatment when necessary, which is conducive to reducing the mortality rate of cardiovascular and cerebrovascular diseases.
[0081] The in-stent restenosis detection method of the embodiment of the present application includes obtaining the original medical image of the user to be tested; obtaining the characteristic information of the stent to be tested based on the original medical image, and obtaining the blood vessel segment to be tested containing the stent to be tested according to the characteristic information of the stent to be tested; extracting the characteristics of the blood vessel segment to be tested to obtain the characteristic information of the blood vessel segment to be tested; analyzing the characteristic information of the blood vessel segment to be tested to obtain the stenosis information of the blood vessel segment to be tested. The present application can obtain the characteristic information of the stent to be tested based on the original medical image of the user's blood vessel, and extract the characteristic information of the blood vessel segment to be tested based on the characteristic information of the stent to be tested, so as to obtain the characteristic information of the blood vessel segment to be tested of the user's blood vessel, so as to obtain the stenosis information of the user's blood vessel segment according to the characteristic information of the blood vessel segment to be tested.
[0082] In addition, this embodiment has the following advantages by executing the in-stent restenosis detection method of the present application: 1. It is simple, and only the original medical images of the patient need to be obtained for analysis; and the coronary artery images can be obtained non-invasively, such as through CT; 2. It has clinical reference value, and by analyzing and predicting the information of in-stent restenosis, it can assist doctors in determining the stenosis area, stenosis location and even stenosis type in the stent.
[0083] Furthermore, the present application also provides a device for predicting vascular plaque risk.
[0084] Reference Figure 3 , Figure 3 This is a schematic diagram of the functional modules of the first embodiment of the in-stent restenosis detection device of the present application.
[0085] The in-stent restenosis detection includes:
[0086] The detection system 10 is used to obtain the original medical image of the user's blood vessel, obtain the position information of the user's stent to be inspected according to the original medical image, and then analyze the stenosis information of the blood vessel segment to be inspected according to the position information of the stent to be inspected;
[0087] The analysis system 20 is used to analyze at least one of the following: whether stenosis exists in the blood vessel segment to be inspected, the location of stenosis, the degree of stenosis, and the type of plaque, etc., based on the stenosis information of the blood vessel segment to be inspected.
[0088] In addition, the present application also provides a medium, which is preferably a computer-readable storage medium, on which an in-stent restenosis detection program is stored. When the in-stent restenosis detection program is executed by a processor, the steps of each embodiment of the above-mentioned in-stent restenosis detection method are implemented.
[0089] In addition, the present application also provides a computer program product, including a computer program, which implements the steps of each embodiment of the above-mentioned in-stent restenosis detection method when executed by a processor.
[0090] In the embodiments of the in-stent restenosis detection device, computer-readable storage medium and computer program product of the present application, all technical features of the above-mentioned embodiments of the in-stent restenosis detection method are included, and the descriptions and explanations are basically the same as those of the above-mentioned embodiments of the in-stent restenosis detection method, and will not be repeated here.
[0091] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article 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, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.
[0092] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.
[0093] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application can essentially or in other words, the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD), including a number of instructions to enable a terminal device (which can be a fixed terminal, such as an IoT smart device, including smart air conditioners, smart lights, smart power supplies, smart routers and other smart homes; it can also be a mobile terminal, including smart phones, wearable networked AR / VR devices, smart speakers, self-driving cars and many other networked devices) to execute the methods described in the various embodiments of the present application.
[0094] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.
Claims
1. An in-stent restenosis detection device, characterized in that: include: Acquire an original medical image of the user to be detected, wherein the original medical image includes at least one of a CT image, an intravascular image, or an angiography image; Acquire characteristic information of the stent to be inspected based on the original medical image, and acquire a blood vessel segment to be inspected including the stent to be inspected according to the characteristic information of the stent to be inspected; Wherein, the step of obtaining the blood vessel segment to be inspected that includes the stent to be inspected comprises: Predicting the stent feature information of the blood vessel segment to be inspected according to the stent feature information in the original medical image and in combination with a prediction method of a preset stent detection model based on machine learning; Acquiring the position information and size information of the blood vessel segment to be inspected according to the position information and size information of the stent to be inspected; Extracting features of the blood vessel segment to be inspected to obtain feature information of the blood vessel segment to be inspected; The characteristic information of the blood vessel segment to be inspected is analyzed to obtain stenosis information of the blood vessel segment to be inspected.
2. The in-stent restenosis detection device according to claim 1, characterized in that: The prediction method of the preset stent detection model based on machine learning includes: The machine learning model for stent positioning is trained in advance by using the original medical image of the first blood vessel segment containing the first stent as direct input, so that the candidate area of the first stent predicted by the machine learning model for stent positioning includes the first stent.
3. The in-stent restenosis detection device according to claim 1, characterized in that: The step of obtaining the stenosis information of the to-be-detected blood vessel segment comprises: Obtain stenosis information, stenosis location, stenosis degree and plaque type information of the blood vessel segment to be inspected.
4. The in-stent restenosis detection device according to claim 2, characterized in that: The step of analyzing the characteristic information of the blood vessel segment to be inspected and obtaining the stenosis information of the blood vessel segment to be inspected comprises: The original medical image of the first blood vessel segment containing the first stent is used as direct input in advance to train a machine learning model for analyzing stenosis information within the first stent, so that the information analyzed by the machine learning model for analyzing stenosis information within the first stent includes the stenosis information of the first blood vessel segment.
5. An in-stent restenosis detection device, characterized in that: The in-stent restenosis detection device comprises a memory, a processor and an in-stent restenosis detection program stored in the memory and executable on the processor. When the in-stent restenosis detection program is executed by the processor, the steps of the in-stent restenosis detection apparatus as described in any one of claims 1 to 4 are implemented.
6. A medium, the medium being a computer-readable storage medium, characterized in that: The computer-readable storage medium stores an in-stent restenosis detection program, which, when executed by a processor, implements the steps of the in-stent restenosis detection device according to any one of claims 1 to 4.