An information association method, device, electronic device and storage medium

Abstract

Embodiments of the present application disclose an information association method, apparatus, electronic device, and storage medium, belonging to the field of computers; embodiments of the present application obtain medical information in multiple different dimensions; perform lesion recognition on the medical information in each dimension to obtain lesion description information corresponding to the medical information in each dimension; determine the association relationship between the medical information in different dimensions based on the lesion description information; and generate association description information between the medical information in different dimensions according to the association relationship, which can improve the judgment of the association between the medical information in different dimensions, thereby improving the diagnosis efficiency.

Description

Technical Field

[0001] This application relates to the field of computer technology, and particularly to an information association method, apparatus, electronic device, and storage medium. Background Art

[0002] With the development of science and technology, diagnosing patients by relying on scientific and technological means has become a common medical method. For example, various scientific and technological means such as CT angiography (CTA), electrocardiogram equipment, and magnetic resonance examination (MR) can be used to diagnose patients, thereby obtaining medical information of patients in different dimensions.

[0003] When doctors diagnose the condition of patients, they usually retrieve the medical information of patients in different dimensions and comprehensively observe and diagnose the physiological structure of patients to confirm the true degree of the patient's condition and the location of the disease onset. However, the medical information of patients in different dimensions is generally large. If only relying on manual observation to find the association between the medical information in different dimensions, it will lead to a decrease in the diagnosis efficiency. Summary of the Invention

[0004] Embodiments of this application propose an information association method, apparatus, electronic device, and storage medium, which can improve the judgment of the association between medical information in different dimensions, thereby improving the diagnosis efficiency.

[0005] Embodiments of this application provide an information association method, including:

[0006] Obtain multiple pieces of medical information in different dimensions;

[0007] Perform lesion recognition on the medical information of each dimension to obtain lesion description information corresponding to the medical information of each dimension;

[0008] Based on the lesion description information, determine the association relationship between the medical information in different dimensions;

[0009] Generate association description information between the medical information in different dimensions according to the association relationship.

[0010] Correspondingly, embodiments of this application also provide an information association apparatus, including:

[0011] An obtaining unit, configured to obtain multiple pieces of medical information in different dimensions;

[0012] A lesion recognition unit, configured to perform lesion recognition on the medical information of each dimension to obtain lesion description information corresponding to the medical information of each dimension;

[0013] A determination unit, configured to determine the association relationship between the medical information of different dimensions based on the lesion description information;

[0014] A generation unit, configured to generate association description information between the medical information of different dimensions according to the association relationship.

[0015] In one embodiment, the lesion recognition unit may include:

[0016] A method determination subunit, configured to determine the corresponding recognition method for the medical information of each dimension according to the dimension of the medical information;

[0017] A recognition subunit, configured to recognize the medical information of the corresponding dimension by using the recognition method to obtain the lesion description information corresponding to the medical information of the dimension.

[0018] In one embodiment, the recognition subunit may include:

[0019] A first parsing module, configured to parse the recognition method to obtain the recognition logic corresponding to the medical information;

[0020] A parameter recognition module, configured to recognize the medical information according to the recognition logic to obtain at least one lesion parameter;

[0021] A description mapping module, configured to perform description mapping on the lesion parameter to obtain the lesion description information.

[0022] In one embodiment, the recognition subunit may include:

[0023] A second parsing module, configured to parse the recognition method to obtain the lesion recognition model corresponding to the medical information;

[0024] A model recognition module, configured to recognize the medical information by using the lesion recognition model to obtain the lesion description information corresponding to the medical information.

[0025] In one embodiment, the determination unit may include:

[0026] A division subunit, configured to divide the reference medical information and the candidate medical information from the medical information of multiple different dimensions based on the lesion description information;

[0027] A search subunit, configured to search for the associated medical information associated with the reference medical information in the candidate medical information based on the lesion description information of the reference medical information;

[0028] An association inference subunit, configured to perform association inference processing on the lesion description information of the reference medical information and the lesion description information of the associated medical information, so as to obtain the association relationship between the reference medical information and the associated medical information.

[0029] In one embodiment, the association inference subunit may include:

[0030] A comparison subunit, configured to use a preset association relationship table to perform comparison processing on the lesion description information of the reference medical information and the lesion description information of the associated medical information, so as to obtain a comparison result;

[0031] A generation subunit, configured to generate the association relationship based on the comparison result.

[0032] In one embodiment, the information association device further includes:

[0033] An integration unit, configured to perform integration processing on the association description information and the medical information to obtain a medical integration report;

[0034] A display unit, configured to display the medical integration report when receiving a report display instruction.

[0035] Correspondingly, an embodiment of the present application further provides an electronic device, where the electronic device includes a memory and a processor; the memory stores a computer program, and the processor is configured to run the computer program in the memory to execute the information association method provided in any one of the embodiments of the present application.

[0036] Correspondingly, an embodiment of the present application further provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the information association method provided in any one of the embodiments of the present application is implemented.

[0037] Embodiments of the present application can obtain medical information in multiple different dimensions; perform lesion recognition on the medical information in each dimension to obtain lesion description information corresponding to the medical information in each dimension; determine the association relationship between the medical information in different dimensions based on the lesion description information; and generate association description information between the medical information in different dimensions according to the association relationship, which can improve the judgment of the association between the medical information in different dimensions, thereby improving the diagnosis efficiency. Description of the Drawings

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative efforts.

[0039] Figure 1 It is a schematic diagram of the scenario of the information association method provided by the embodiment of the present application;

[0040] Figure 2 It is a schematic flowchart of the information association method provided by the embodiment of the present application;

[0041] Figure 3 It is a schematic diagram of the scenario of lesion recognition provided by the embodiment of the present application;

[0042] Figure 4 It is another schematic diagram of the scenario of lesion recognition provided by the embodiment of the present application;

[0043] Figure 5 It is another schematic diagram of the scenario of lesion recognition provided by the embodiment of the present application;

[0044] Figure 6 It is another schematic diagram of the scenario of lesion recognition provided by the embodiment of the present application;

[0045] Figure 7 It is another schematic flowchart of the information association method provided by the embodiment of the present application;

[0046] Figure 8 It is a schematic structural diagram of the information association device provided by the embodiment of the present application;

[0047] Figure 9 It is a schematic structural diagram of the electronic device provided by the embodiment of the present application. Detailed implementation manners

[0048] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. However, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present application.

[0049] The embodiment of the present application proposes an information association method. This information association method can be executed by an information association device, and this information association device can be integrated in an electronic device. Among them, the electronic device can include at least one of a terminal and a server, etc. That is, this information association method can be executed by the terminal or by the server.

[0050] Among them, the terminal can include a smart TV, a smart phone, a smart home, a wearable electronic device, a VR / AR product, an in-vehicle computer, a smart computer, and so on.

[0051] Among them, the server can be an interoperability server between multiple heterogeneous systems or a background server of a product verification and testing system, and can also be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, and big data and artificial intelligence platforms, and so on.

[0052] In one embodiment, as Figure 1 shown, the information association device can be integrated on an electronic device such as a terminal or a server to implement the information association method proposed in the embodiments of the present application. Specifically, the electronic device can obtain medical information in multiple different dimensions; perform lesion recognition on the medical information in each dimension to obtain lesion description information corresponding to the medical information in each dimension; determine the association relationship between the medical information in different dimensions based on the lesion description information; and generate association description information between the medical information in different dimensions according to the association relationship.

[0053] The following will be described in detail respectively. It should be noted that the description order of the following embodiments does not limit the preferred order of the embodiments.

[0054] The embodiments of the present application will be described from the perspective of the information association device. The information association device can be integrated in an electronic device, and the electronic device can include a terminal and can also include a server, and so on.

[0055] As Figure 2 shown, an information association method is provided. The specific process includes:

[0056] 101. Obtain medical information in multiple different dimensions.

[0057] Among them, the medical information can include information related to medical diagnosis. For example, the medical information can include medical images and / or medical chart data, and so on.

[0058] Among them, the medical images can include images related to medical diagnosis. For example, the medical images can include Computed Tomography (CT) images, Magnetic Resonance (MR) images, and ultrasonic images, and so on. For another example, the medical image can be a physiological tissue image, and so on. Among them, the physiological tissue can include the heart, brain, lungs, liver, and so on. For example, the medical image can be a CT image of a patient's heart. For another example, the medical image can be an MR image of a patient's brain, and so on.

[0059] Among them, medical chart data may include charts related to medical diagnoses. For example, medical chart data can be obtained by detecting the physiological structure or region of the human body target through electrocardiogram equipment, electroencephalogram equipment, Digital Radiography (DR) equipment, endoscope equipment, etc.

[0060] In one embodiment, when a patient's body shows abnormalities, scientific and technological means are often used to diagnose the physical condition. For example, scientific and technological means such as magnetic resonance imaging and B-ultrasound examination can be used to diagnose the physical condition.

[0061] Among them, when patients often use a variety of scientific and technological means for physical diagnosis, each scientific and technological means will correspond to a medical information.

[0062] For example, when a patient undergoes magnetic resonance imaging, an MR image will be obtained. When a patient uses electrocardiogram equipment for physical diagnosis, electrocardiogram chart data will be obtained.

[0063] In one embodiment, when a patient uses a variety of scientific and technological means for physical diagnosis, medical information of the patient in multiple different dimensions will be obtained. Among them, the medical information in one dimension can correspond to a scientific and technological means. For example, the patient's MR image can be medical information in one dimension, and the patient's electrocardiogram chart data can be medical information in another dimension.

[0064] In one embodiment, the medical information in one dimension can also correspond to a medical use.

[0065] In one embodiment, when a doctor diagnoses a patient's condition, the doctor usually retrieves the medical information of the patient's target physiological tissue in multiple different dimensions (or uses), so as to comprehensively observe and diagnose the situation of the target physiological tissue. For example, when diagnosing conditions such as cerebral infarction or cerebral ischemia, the doctor will retrieve CTA medical images and CTP (CT perfusion scan) medical images. Usually, CTP medical images can provide information such as infarction and ischemia of the patient, and CTA medical images can provide information on vascular lesions. By comprehensively observing the medical information in different dimensions, the true degree of the patient's condition and the location of the disease onset can be confirmed.

[0066] 102. Identify lesions for the medical information in each dimension to obtain the lesion description information corresponding to the medical information in each dimension.

[0067] In one embodiment, when a doctor confirms the true degree and location of a patient's illness by paying attention to medical information in different dimensions, if there are multiple medical images in different dimensions, it will take the doctor a lot of time and energy to determine the true degree and location of the patient's illness, which will lead to low diagnostic efficiency. Therefore, the embodiment of the present application proposes an information association method, which can automatically identify the association relationship between medical information in different dimensions, thereby saving the doctor's time and energy and improving the diagnostic efficiency.

[0068] In one embodiment, after obtaining multiple medical information in different dimensions, the information association device can perform lesion recognition on the medical information in each dimension to obtain the lesion description information corresponding to the medical information in each dimension.

[0069] Among them, performing lesion recognition on medical information may include identifying the area with lesions in the medical information and the description information corresponding to the lesion area.

[0070] Among them, the lesion description information may include information for describing the lesion area. For example, the lesion description information may include the type, benign or malignant nature, lesion size, and lesion range of the lesion, and so on.

[0071] In one embodiment, since the content of medical information in different dimensions is not the same, different methods can be used to perform lesion recognition on the medical information in different dimensions to obtain the lesion description information corresponding to the medical information in each dimension. Specifically, the step "perform lesion recognition on the medical information in each dimension to obtain the lesion description information corresponding to the medical information in each dimension" may include:

[0072] Determine the corresponding recognition method for the medical information in each dimension according to the dimension of the medical information;

[0073] Use the recognition method to recognize the medical information in the corresponding dimension to obtain the lesion description information corresponding to the medical information in the dimension.

[0074] Among them, the recognition method may include a set of logics that need to be followed when recognizing medical information, and so on. For example, the recognition method may be an interface for the logics that need to be followed when recognizing medical information. Through this interface, the corresponding logics can be directly called to recognize the medical information. By aggregating the logics that need to be followed for some information into an interface, the process of recognizing medical information can be simplified.

[0075] For example, the information association device obtains medical information in two dimensions, where the medical information in these two dimensions is medical image information and medical chart data respectively. Among them, the recognition method corresponding to the medical image information is to use a first lesion recognition model for images for recognition, while the recognition method corresponding to the medical chart information is to use a second lesion recognition model for chart data for recognition. Then, the information association device can use the first lesion recognition model for images to recognize the medical image information and obtain the lesion description information corresponding to the medical image information. In addition, the information association device can also use the second lesion recognition model for chart data to recognize the medical chart data and obtain the lesion description information corresponding to the medical chart data.

[0076] Among them, the lesion recognition model can be a deep learning model. For example, the lesion recognition model can include any one of Convolutional Neural Networks (CNN), De-Convolutional Networks (DN), Deep Neural Networks (DNN), Deep Convolutional Inverse Graphics Networks (DCIGN), Region-based Convolutional Networks (RCNN), Faster Region-based Convolutional Networks (Faster RCNN), and Bidirectional Encoder Representations from Transformers (BERT) model, etc.

[0077] For another example, different recognition methods can also be used to recognize medical information in the same dimension to obtain lesion description information of the medical information in the same dimension in different dimensions. For example, the information association device obtains a cardiac coronary artery vessel image. Then, the position, scope, degree, and other lesion description information of a blood vessel plaque in the cardiac coronary artery vessel image can be recognized using a traditional algorithm, and these lesion description information can be constructed into lesion description information. In addition, a fat recognition model can also be used to recognize the coronary artery peripheral fat index of the cardiac coronary artery in the cardiac coronary artery vessel image.

[0078] Therefore, when performing lesion recognition on medical information in each dimension, the recognition method corresponding to the medical information in each dimension can be determined according to the dimension of the medical information. Then, the recognition method is used to recognize the medical information in the corresponding dimension to obtain the lesion description information corresponding to the medical information in that dimension.

[0079] In one embodiment, since the recognition method includes the logic that needs to be followed when recognizing medical information, the recognition method can be parsed to obtain the recognition logic corresponding to the recognition method. Then, the medical information is recognized according to the recognition logic. Specifically, the step of "using the recognition method to recognize the medical information of the corresponding dimension and obtaining the lesion description information corresponding to the medical information of the dimension" may include:

[0080] Parse the recognition method to obtain the recognition logic corresponding to the medical information;

[0081] Recognize the medical information according to the recognition logic to obtain at least one lesion parameter;

[0082] Perform a description mapping on the lesion parameter to obtain the lesion description information.

[0083] Among them, the recognition logic includes the steps that need to be followed when performing lesion recognition on medical information.

[0084] Among them, the lesion parameter may include the original parameter obtained when recognizing medical information. For example, the lesion parameter may include the size of the lesion, the range of the lesion, the type of the lesion, and so on.

[0085] In one embodiment, since the recognition method may be an interface for the logic that needs to be followed when recognizing medical information, when using the recognition method to recognize the medical information of the corresponding dimension, the recognition method can be parsed to obtain the recognition logic corresponding to the medical information. For example, by traversing the recognition method, the storage unit corresponding to the recognition logic can be identified. Then, the recognition logic is called from the storage unit corresponding to the recognition logic to obtain the recognition logic corresponding to the medical information. Then, the medical information can be recognized according to the recognition logic to obtain at least one lesion parameter.

[0086] In one embodiment, after obtaining the lesion parameter, a description mapping can also be performed on the lesion parameter to obtain the lesion description information.

[0087] Among them, the lesion description information may include the information obtained by performing a description mapping on the lesion parameter.

[0088] Among them, the description mapping may include an operation of mapping the lesion parameter to the lesion description information. For example, the description mapping may include a calculation operation and / or a mapping operation, and so on.

[0089] For example, the identified lesion parameters include the size, scope, and type of the lesion. Then, the size and scope of the lesion can be calculated in combination with the type of the lesion to obtain the benign or malignant nature of the lesion. Next, the benign or malignant nature, size, scope, and type of the lesion can be integrated to obtain lesion description information.

[0090] In one embodiment, the identification method can be parsed to obtain the identification logic corresponding to the medical information; according to the identification logic, the medical information is identified to obtain at least one lesion parameter; the lesion parameters are calculated, and the calculated information and the lesion parameters are integrated to obtain lesion description information.

[0091] For example, the medical information includes medical information in a first dimension and medical information in a second dimension. By parsing the identification method, it is obtained that a first type of pre-trained identification model or a traditional algorithm is used to identify the medical information in the first dimension, and lesion parameters such as the size, scope, and type of the lesion of the medical information in the first dimension are obtained. Then, the size and scope of the lesion can be calculated in combination with the type of the lesion to obtain the benign or malignant nature of the lesion. Next, the benign or malignant nature, size, scope, and type of the lesion can be integrated to obtain the lesion description information of the medical information in the first dimension.

[0092] Similarly, by parsing the identification method, a second type of pre-trained identification model or a traditional algorithm can be used to identify the medical information in the second dimension, and lesion parameters such as the size, scope, and type of the lesion of the medical information in the second dimension are obtained. Then, the size and scope of the lesion can be calculated in combination with the type of the lesion to obtain the benign or malignant nature of the lesion. Next, the benign or malignant nature, size, scope, and type of the lesion can be integrated to obtain the lesion description information of the medical information in the second dimension.

[0093] In one embodiment, a model for lesion identification of medical information can be pre-trained, and each lesion identification model can be used as an identification method. Therefore, the identification method can be parsed to obtain the lesion identification model corresponding to the medical information, and then the lesion identification model is used to identify the medical information to obtain the lesion description information corresponding to the medical information. Specifically, the step of "using the identification method to identify the medical information in the corresponding dimension to obtain the lesion description information corresponding to the medical information in the dimension" may include:

[0094] Parse the identification method to obtain the lesion identification model corresponding to the medical information;

[0095] Use the lesion identification model to identify the medical information to obtain the lesion description information corresponding to the medical information.

[0096] For example, medical information may include medical information in a first dimension and medical information in a second dimension. Among them, the medical information in the first dimension can use a first type of pre-trained lesion recognition model to recognize the medical information, and obtain lesion description information corresponding to the medical information in the first dimension. Among them, the medical information in the second dimension can use a second type of pre-trained lesion recognition model to recognize the medical information, and obtain lesion description information corresponding to the medical information in the second dimension.

[0097] 103. Based on the lesion description information, determine the association relationship between medical information in different dimensions.

[0098] In one embodiment, after identifying the lesion description information corresponding to the medical information in each dimension, the association relationship between the medical information in different dimensions can be determined based on the lesion description information.

[0099] Among them, the association relationship may refer to the relationship between medical information in different dimensions. For example, the association relationship may include causal relationship, parallel relationship, inclusion relationship, etc. between medical information in different dimensions.

[0100] For example, taking the presence of myocardial abnormalities as an example, the medical information includes MR medical images and electrocardiograms. Among them, the lesion description information of the MR medical image is the area of myocardial abnormalities and the size of the area. The lesion description information of the electrocardiogram is the area where abnormal fluctuations exist in the electrocardiogram. Through the lesion description relationship of the MR medical image and the lesion description relationship of the electrocardiogram, the relationship between the MR medical image and the electrocardiogram can be obtained as a parallel relationship, that is, both describe the area of myocardial abnormalities from different perspectives.

[0101] In one embodiment, the step of "based on the lesion description information, determine the association relationship between medical information in different dimensions" may include:

[0102] Based on the lesion description information, divide the reference medical information and candidate medical information among multiple medical information in different dimensions;

[0103] Based on the lesion description information of the reference medical information, search for associated medical information in the candidate medical information that is associated with the reference medical information;

[0104] Perform an association inference process on the lesion description information of the reference medical information and the lesion description information of the associated medical information to obtain the association relationship between the reference medical information and the associated medical information.

[0105] Among them, the reference medical information may refer to the medical information used as the standard for judging the association relationship. For example, the reference medical information may include the medical information with the most lesion description information. Another example is that the reference medical information may include the medical information with the highest degree of importance, etc.

[0106] Among them, the candidate medical information may include medical information on whether there is an association relationship between the judgment and the reference medical information and what the association relationship is.

[0107] For example, there is medical information in three dimensions. Among them, if the medical information in the first dimension is divided into reference medical information, then the medical information in the second dimension and the third dimension can be divided into candidate medical information.

[0108] In one embodiment, after dividing the reference medical information and the candidate medical information, based on the lesion description information of the reference medical information, an associated medical information associated with the reference medical information can be searched for among the candidate medical information.

[0109] For example, the lesion description information of the reference medical information and the lesion description information of the candidate medical information can be matched, and the associated medical information associated with the reference medical information can be searched through the matching result.

[0110] For example, it can be determined whether there are similar or identical keywords between the lesion description information of the reference medical information and the lesion description information of the candidate medical information. When there are identical or similar keywords, it indicates that the reference medical information and the candidate medical information are matched, and then the candidate medical information can be determined as the associated medical information.

[0111] For example, the lesion description information of the reference medical information includes "lesion type is A, lesion area is B, and lesion benign / malignant is C". The lesion description information of the first candidate medical information includes "lesion type is A, lesion area is D". The lesion description information of the second candidate medical information includes "lesion type is B, lesion area is K". The first candidate medical information and the second candidate medical information are respectively matched with the reference medical information, and it is obtained that the first candidate medical information and the reference medical information have the same keyword lesion description information "lesion type is A". Therefore, the first candidate medical information can be determined as the associated medical information.

[0112] In one embodiment, after searching for the associated medical information, the lesion description information of the reference medical information and the lesion description information of the associated medical information can be subjected to associated inference processing to obtain the association relationship between the reference medical information and the associated medical information.

[0113] For example, the lesion description information of the reference medical information and the lesion description information of the associated medical information can be compared according to a preset association relationship table, so as to determine the association relationship between the reference medical information and the associated medical information. Specifically, the step "performing associated inference processing on the lesion description information of the reference medical information and the lesion description information of the associated medical information to obtain the association relationship between the reference medical information and the associated medical information" may include:

[0114] Using a preset association relationship table, compare and process the lesion description information of the reference medical information and the lesion description information of the associated medical information to obtain a comparison result;

[0115] Generate an association relationship based on the comparison result.

[0116] Among them, the preset association relationship table can illustrate what kind of association relationships may exist between what kinds of lesion description information.

[0117] For example, if the preset association relationship table records the content of "coronary artery; fat index; high; blood vessel; plaque vulnerability; causal relationship", when the information association device compares the lesion description information of the reference medical information and the lesion description information of the associated medical information according to the preset association relationship table, if it is detected that the lesion description information of the reference medical information and the lesion description information of the associated medical information conform to the content in the above preset association relationship table, it can be determined that the association relationship between the reference medical information and the associated medical information is a causal relationship.

[0118] In an embodiment, when there is medical information in multiple dimensions, different medical information can be used as the reference medical information respectively, so as to judge the association relationships between different medical information.

[0119] For example, for medical information including three dimensions, the medical information of the first dimension can be used as the reference medical information first, and the medical information of the second dimension and the medical information of the third dimension can be used as candidate medical information, so as to obtain the association relationships between the medical information of the first dimension and the medical information of the second dimension and the medical information of the third dimension. Then, the medical information of the second dimension can also be used as the reference medical information, and the medical information of the third dimension can be used as candidate medical information, so as to judge the association relationship between the medical information of the second dimension and the medical information of the third dimension.

[0120] 104. Generate association description information between medical information in different dimensions according to the association relationship.

[0121] In an embodiment, after obtaining the association relationship, association description information between medical information in different dimensions can be generated.

[0122] Among them, the association description information can include information describing the lesion description information of medical information in different dimensions based on the association relationships between medical information in different dimensions.

[0123] For example, as Figure 3As shown, taking cerebral infarction ischemia as an example. Using traditional algorithms, information such as the infarction range and infarction degree of the infarction location in cerebral artery CTP images of the brain is identified. These information are constructed into a lesion attribute set (which can be equivalent to lesion description information). In addition, as Figure 4 shown, using a lesion recognition model, lesion description information such as the location, range, and degree of arterial occlusion in coronary artery CTA images is identified. Then, according to the CPT image lesion attribute set, the coronary artery CTA image correlated with the cerebral artery CTP image containing cerebral infarction is found by comparison. Therefore, the associated description information between the CTP image and the CTA image can be described as: occlusion of the middle artery causes severe infarction in the left cerebrum.

[0124] For another example, taking a vascular plaque lesion as an example. Using traditional algorithms, information such as the location, range, and degree of a vascular plaque at a certain location in cardiac coronary artery images is identified. These information are constructed into a lesion attribute set. Using a fat recognition model, the fat around the coronary artery index (FAI) of the cardiac coronary artery is identified. Then, according to the lesion attribute set, the FAI index correlated with the vascular plaque in the FAI index is found by comparison. Then the associated description information between the cardiac coronary artery image and the FAI can be described as: an excessive FAI index in a certain area of the coronary artery leads to a risk of vulnerable plaques in the coronary artery (vulnerable: the plaque is likely to break off and is likely to cause vascular occlusion with blood flow, leading to death).

[0125] For another example, as Figure 5 shown, taking the presence of myocardial abnormalities as an example, a pre-trained lesion recognition model is used to identify the lesion description information of the lesions in MR images. As Figure 6 shown, a pre-trained abnormal area recognition model is used to identify the area with abnormal waveforms in an electrocardiogram. According to the lesion description information of the MR image, the electrocardiogram where the abnormal waveform is located and the lesion description information of the abnormal waveform are found correspondingly. Based on the lesion description information of the MR image and the lesion description information of the abnormal waveform, the associated description information of "lesions (ischemia, inflammation, etc.) of the myocardium can be seen in the XXX area of the MR image, and the waveform in the XXX area of the electrocardiogram shows an ascending area (descending / shortening, etc.)" is generated.

[0126] For another example, taking the presence of a lesion in the cardiovascular system as an example. The cardiac region is detected using CT, 4D ultrasound, and ECG respectively to obtain CT images, ultrasound images, and electrocardiograms. The pre-trained lesion recognition model or abnormal region recognition model corresponding to the CT image, ultrasound image, and electrocardiogram is used to identify the lesion region or abnormal fluctuation region respectively. According to the lesion information in the CT image, the corresponding information in the ultrasound image and the region with fluctuations in the electrocardiogram are found. According to the associations between the information identified in the above images or electrocardiogram, the associated description information can be described as: There is a thrombus in the blood vessels of the XX region of the heart, the imaging myocardium is working normally, resulting in myocardial ischemia. In the electrocardiogram, it can be seen that the wave in the XX region shows an upward trend, causing the patient to have an arrhythmia problem.

[0127] In one embodiment, after generating the associated description information, the associated description information and medical information can be integrated to obtain a medical integration report. When the information association device receives a report display instruction, the medical integration report can be displayed. Specifically, the method proposed in the embodiments of the present application may further include:

[0128] Integrate the associated description information and medical information to obtain a medical integration report;

[0129] When receiving a report display instruction, display the medical integration report.

[0130] For example, the medical information includes medical information in the first dimension, medical information in the second dimension, and medical information in the third dimension. Among them, the medical information in the first dimension and the medical information in the second dimension are related. Therefore, the medical information in the first dimension, the medical information in the second dimension, and their associated description information can be integrated to obtain a medical integration report.

[0131] Then, when the doctor wants to view the report, the information association device can receive a report display instruction, and thus display the medical integration report.

[0132] The information association method proposed in the embodiments of the present application can obtain multiple different dimensions of medical information; perform lesion recognition on the medical information in each dimension to obtain the lesion description information corresponding to the medical information in each dimension; determine the association relationship between the medical information in different dimensions based on the lesion description information; generate the associated description information between the medical information in different dimensions according to the association relationship. Through the information association method proposed in the embodiments of the present application, the association relationship between the medical information in different dimensions can be automatically identified, thus saving the doctor's time and effort and improving the diagnosis efficiency.

[0133] According to the method described in the above embodiments, the following will be further illustrated with examples in detail.

[0134] This embodiment of the application will take the integration of the information association method on an electronic device as an example to introduce the method of this embodiment of the application. For example, as Figure 7 shown, the information association method proposed in this embodiment of the application may include:

[0135] 201. The electronic device acquires medical information in multiple different dimensions.

[0136] For example, the physiological structure of the human body can be scanned by means of Computed Tomography (CT) images, Magnetic Resonance (MR) images, 4D ultrasound images, etc. to obtain medical images.

[0137] For another example, the physiological structure or region of the human body can be detected by an electrocardiogram device, an electroencephalogram device, a DR device (such as a fluoroscopy device), a DSA device, an endoscope device, etc. to obtain medical images or medical chart data.

[0138] 202. The electronic device performs lesion recognition on the medical information in each dimension to obtain lesion description information corresponding to the medical information in each dimension.

[0139] For example, a first pre-trained lesion recognition model or a traditional algorithm can be used to recognize the medical information in the first dimension, and lesion parameters such as the type and benign / malignant nature of the lesion are recognized. Information such as the size and scope of the lesion is calculated using the lesion parameters. Information such as the type, benign / malignant nature, lesion size, and lesion scope of the lesion is constructed into a lesion attribute set.

[0140] For example, a second pre-trained lesion recognition model or a traditional algorithm can be used to recognize the medical information in the second dimension, and lesion description information such as the lesion location and lesion scope is recognized.

[0141] 203. The electronic device determines the association relationship between the medical information in different dimensions based on the lesion description information.

[0142] For example, based on the lesion attribute set, candidate medical information is searched for associated medical information that is associated with the reference medical information. That is, based on medical experience, information with a causal relationship is found and integrated together.

[0143] Among them, medical information with clinical indication can be used to find medical information without clinical indication.

[0144] 204. The electronic device generates association description information between the medical information in different dimensions according to the association relationship.

[0145] For example, the causal relationship of a patient's lesion can be described based on reference medical information and associated medical information associated therewith.

[0146] In the information association method provided by the embodiments of the present application, an electronic device obtains medical information in multiple different dimensions; performs lesion recognition on the medical information in each dimension to obtain lesion description information corresponding to the medical information in each dimension; determines the association relationship between the medical information in different dimensions based on the lesion description information; and generates association description information between the medical information in different dimensions according to the association relationship. Through the information association method provided by the embodiments of the present application, the association relationship between the medical information in different dimensions can be automatically recognized, thereby saving the time and effort of doctors and improving the diagnosis efficiency.

[0147] To better implement the information association method provided by the embodiments of the present application, in one embodiment, an information association device is further provided. This information association device can be integrated into an electronic device. The meanings of the nouns are the same as those in the above information association method, and the specific implementation details can refer to the description in the method embodiments.

[0148] In one embodiment, an information association device is provided. This information association device can be specifically integrated in an electronic device, such as Figure 8 shown. This information association device includes: an acquisition unit 301, a lesion recognition unit 302, a determination unit 303, and a generation unit 304, specifically as follows:

[0149] The acquisition unit 301 is configured to acquire medical information in multiple different dimensions;

[0150] The lesion recognition unit 302 is configured to perform lesion recognition on the medical information in each dimension to obtain lesion description information corresponding to the medical information in each dimension;

[0151] The determination unit 303 is configured to determine the association relationship between the medical information in different dimensions based on the lesion description information;

[0152] The generation unit 304 is configured to generate association description information between the medical information in different dimensions according to the association relationship.

[0153] In one embodiment, the lesion recognition unit 302 may include:

[0154] A method determination subunit is configured to determine the corresponding recognition method for the medical information in each dimension according to the dimension of the medical information;

[0155] An identification subunit is configured to use the recognition method to identify the medical information in the corresponding dimension to obtain lesion description information corresponding to the medical information in the dimension.

[0156] In one embodiment, the recognition subunit may include:

[0157] A first parsing module, configured to parse the recognition method to obtain the recognition logic corresponding to the medical information;

[0158] A parameter recognition module, configured to recognize the medical information according to the recognition logic to obtain at least one lesion parameter;

[0159] A description mapping module, configured to perform description mapping on the lesion parameter to obtain the lesion description information.

[0160] In one embodiment, the recognition subunit may include:

[0161] A second parsing module, configured to parse the recognition method to obtain the lesion recognition model corresponding to the medical information;

[0162] A model recognition module, configured to recognize the medical information by using the lesion recognition model to obtain the lesion description information corresponding to the medical information.

[0163] In one embodiment, the determination unit 303 may include:

[0164] A division subunit, configured to divide the reference medical information and the candidate medical information from the medical information in multiple different dimensions based on the lesion description information;

[0165] A search subunit, configured to search for the associated medical information related to the reference medical information in the candidate medical information based on the lesion description information of the reference medical information;

[0166] An association inference subunit, configured to perform association inference processing on the lesion description information of the reference medical information and the lesion description information of the associated medical information to obtain the association relationship between the reference medical information and the associated medical information.

[0167] In one embodiment, the association inference subunit may include:

[0168] A comparison subunit, configured to use a preset association relationship table to perform comparison processing on the lesion description information of the reference medical information and the lesion description information of the associated medical information to obtain a comparison result;

[0169] A generation subunit, configured to generate the association relationship based on the comparison result.

[0170] In one embodiment, the information association device further includes:

[0171] An integration unit, configured to integrate the associated description information and the medical information to obtain a medical integration report;

[0172] A display unit, configured to display the medical integration report when receiving a report display instruction.

[0173] In specific implementation, each of the above units may be implemented as an independent entity, or may be combined arbitrarily and implemented as the same or several entities. For the specific implementation of each of the above units, reference may be made to the foregoing method embodiments, which will not be elaborated herein.

[0174] Through the above information association device, the association relationship between medical information in different dimensions can be automatically identified, thereby saving the time and energy of doctors and improving the diagnosis efficiency.

[0175] The embodiment of the present application further provides an electronic device, which may include a terminal or a server. For example, the electronic device may be used as an information association terminal, and the information association terminal may be a smart TV, etc.; for another example, the computer product may be a server, such as an information association server, etc. As Figure 9 shown, it shows a schematic structural diagram of the terminal involved in the embodiment of the present application. Specifically:

[0176] The electronic device may include a processor 401 with one or more processing cores, a memory 402 with one or more computer-readable storage media, a power supply 403, an input unit 404 and other components. Those skilled in the art can understand that Figure 9 the structural diagram of the electronic device shown in does not constitute a limitation on the electronic device, and may include more or fewer components than shown, or combine certain components, or arrange different components. Among them:

[0177] The processor 401 is the control center of the electronic device, connecting various parts of the entire electronic device through various interfaces and lines. By running or executing software programs and / or modules stored in the memory 402, and calling data stored in the memory 402, it executes various functions of the electronic device and processes data, thereby monitoring the entire electronic device. Optionally, the processor 401 may include one or more processing cores; preferably, the processor 401 may integrate an application processor and a modulation and demodulation processor. Among them, the application processor mainly processes the operating system, user interfaces and application programs, etc., and the modulation and demodulation processor mainly processes wireless communications. It can be understood that the above modulation and demodulation processor may not be integrated into the processor 401.

[0178] The memory 402 can be used to store software programs and modules. The processor 401 executes various functional applications and data processing by running the software programs and modules stored in the memory 402. The memory 402 mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the computer product. In addition, the memory 402 can include high-speed random access memory, and can also include non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 402 can also include a memory controller to provide the processor 401 with access to the memory 402.

[0179] The electronic device further includes a power supply 403 for powering each component. Preferably, the power supply 403 can be logically connected to the processor 401 through a power management system, so as to implement functions such as management of charging, discharging, and power consumption management through the power management system. The power supply 403 can also include any components such as one or more DC or AC power supplies, a recharge system, a power failure detection circuit, a power converter or an inverter, and a power status indicator.

[0180] The electronic device may further include an input unit 404, which can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function controls.

[0181] Although not shown, the electronic device may further include a typesetting unit, etc., which will not be elaborated here. Specifically, in this embodiment, the processor 401 in the electronic device will load the executable files corresponding to the processes of one or more application programs into the memory 402 according to the following instructions, and the processor 401 will run the application programs stored in the memory 402 to implement various functions as follows:

[0182] Obtain medical information in multiple different dimensions;

[0183] Perform lesion recognition on the medical information in each dimension to obtain lesion description information corresponding to the medical information in each dimension;

[0184] Based on the lesion description information, determine the association relationship between the medical information in different dimensions;

[0185] Generate association description information between the medical information in different dimensions according to the association relationship.

[0186] For the specific implementation of each of the above operations, reference can be made to the previous embodiments, which will not be elaborated here.

[0187] According to one aspect of the present application, there is provided a computer program product or a computer program. The computer program product or the computer program includes computer instructions stored in a computer-readable storage medium. A processor of the computer product reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer product executes the methods provided in various alternative implementations in the above embodiments.

[0188] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by a computer program or by controlling related hardware through a computer program. The computer program can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0189] For this reason, an embodiment of the present application further provides a storage medium, in which a computer program is stored. The computer program can be loaded by a processor to execute the steps in any of the information association methods provided by the embodiments of the present application. For example, the computer program can execute the following steps:

[0190] Obtain medical information in multiple different dimensions;

[0191] Perform lesion recognition on the medical information in each dimension to obtain lesion description information corresponding to the medical information in each dimension;

[0192] Based on the lesion description information, determine the association relationship between the medical information in different dimensions;

[0193] Generate association description information between the medical information in different dimensions according to the association relationship.

[0194] For the specific implementation of each of the above operations, reference may be made to the previous embodiments, which will not be elaborated here.

[0195] Since the computer program stored in the storage medium can execute the steps in any of the information association methods provided by the embodiments of the present application, the beneficial effects that can be achieved by any of the information association methods provided by the embodiments of the present application can be realized. For details, see the previous embodiments, which will not be elaborated here.

[0196] The above has introduced in detail an information association method, device, electronic device, and storage medium provided by the embodiments of the present application. Specific examples are used in this article to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those skilled in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. An information association method, characterized in that, Including: Obtain medical information in multiple different dimensions; Perform lesion recognition on the medical information in each dimension to obtain lesion description information corresponding to the medical information in each dimension; Based on the lesion description information, divide the medical information in the multiple different dimensions into reference medical information and candidate medical information; wherein, the reference medical information refers to the medical information used as the standard for judging the association relationship, and different medical information is respectively used as the reference medical information to judge the association relationship between different medical information; Based on the lesion description information of the reference medical information, search for associated medical information in the candidate medical information that is associated with the reference medical information; Use a preset association relationship table to compare and process the lesion description information of the reference medical information and the lesion description information of the associated medical information to obtain a comparison result; wherein, the preset association relationship table is the preset association relationship between different lesion description information; Generate the association relationship based on the comparison result; Generate association description information between the medical information in the different dimensions according to the association relationship.

2. The method according to claim 1, wherein The performing lesion recognition on the medical information in each dimension to obtain lesion description information corresponding to the medical information in each dimension includes: Determine the corresponding recognition method for the medical information in each dimension according to the dimension of the medical information; Use the recognition method to recognize the medical information in the corresponding dimension to obtain the lesion description information corresponding to the medical information in the dimension.

3. The method according to claim 2, wherein The using the recognition method to recognize the medical information in the corresponding dimension to obtain the lesion description information corresponding to the medical information in the dimension includes: Analyze the recognition method to obtain the recognition logic corresponding to the medical information; According to the recognition logic, recognize the medical information to obtain at least one lesion parameter; Perform description mapping on the lesion parameter to obtain the lesion description information.

4. The method according to claim 2, wherein The using the recognition method to recognize the medical information in the corresponding dimension to obtain the lesion description information corresponding to the medical information in the dimension includes: Analyze the recognition method to obtain the lesion recognition model corresponding to the medical information; Use the lesion recognition model to recognize the medical information to obtain the lesion description information corresponding to the medical information.

5. The method according to claim 1, characterized in that, The method further includes: Integrate the association description information and the medical information to obtain a medical integration report; When a report display instruction is received, display the medical integration report.

6. An information association device, characterized in that Including: An acquisition unit for obtaining medical information in multiple different dimensions; A lesion recognition unit for performing lesion recognition on the medical information in each dimension to obtain lesion description information corresponding to the medical information in each dimension; A determination unit for determining the association relationship between the medical information in the different dimensions based on the lesion description information; A generation unit for generating association description information between the medical information in the different dimensions according to the association relationship; The determination unit is further used for: Based on the lesion description information, divide the reference medical information and candidate medical information from the medical information in the multiple different dimensions; wherein, the reference medical information refers to the medical information used as the standard for judging the association relationship. Different medical information is used as the reference medical information respectively to judge the association relationship between different medical information. Based on the lesion description information of the reference medical information, search for the associated medical information that is associated with the reference medical information in the candidate medical information. Use the preset association relationship table to compare the lesion description information of the reference medical information and the lesion description information of the associated medical information to obtain a comparison result; wherein, the preset association relationship table is the pre-set association relationship between different lesion description information. Generate the association relationship based on the comparison result.

7. An electronic device, characterized in that, It includes a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to execute the steps in the information association method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores multiple computer programs, and the computer programs are suitable for being loaded by the processor to execute the steps in the information association method according to any one of claims 1 to 5.

Images