Information query method and computer equipment applied to intravenous catheter information system

Abstract

Embodiments of this application relate to the field of symptom identification, specifically to an information query method and computer device for use in an intravenous catheter information system. A specific implementation of this method includes: a knowledge graph construction subsystem constructing a complication knowledge graph; a symptom identification subsystem training a symptom identification model; an information transmission subsystem sending patient intravenous catheter information to a symptom identification subsystem and a symptom retrieval subsystem; the symptom identification subsystem inputting the patient intravenous catheter information into the symptom identification model to obtain patient symptom type information; the symptom retrieval subsystem, in response to receiving the patient intravenous catheter information, searches a complication type database to generate complication retrieval information; and the information transmission subsystem, in response to receiving the patient symptom type information and complication retrieval information, sends the patient symptom type information and complication retrieval information to a management terminal. This implementation can reduce the waste of medical resources.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of disease identification, and more particularly to an information query method and computer equipment applied to an intravenous catheter information system. Background Art

[0002] Intravenous therapy is one of the most commonly used and direct treatments in clinical practice, playing an irreplaceable role in disease prevention, treatment, and life-saving efforts. However, intravenous therapy is often associated with a range of complications, including phlebitis, extravasation / extravasation, catheter occlusion, and catheter-related bloodstream infections. These complications not only impact the function and effectiveness of intravenous catheters but can even endanger patients' lives. Currently, clinical care often relies on experience-based judgment to determine whether to continue or remove the catheter after adjustment. Commonly used intravenous therapy methods include peripheral intravenous catheters and central venous catheters. While complications associated with peripheral intravenous catheters (such as indwelling catheters) can often be managed promptly through visual inspection, central venous catheters are more commonly used clinically due to their adaptability (accommodating all types of infusion therapy and medication). Furthermore, their unique location (the catheter tip is located in the lower third of the superior vena cava, at the entrance to the heart), high treatment costs, wide applicability to diverse patient needs, and the difficulty in assessing and managing complications present significant challenges for clinical care. Misjudgments can waste medical resources and impact patient progress. Summary of the Invention

[0003] The content of this application is used to briefly introduce concepts that will be described in detail in the detailed description section below. The content of this application is not intended to identify key features or essential features of the technical solution for which protection is sought, nor is it intended to limit the scope of the technical solution for which protection is sought.

[0004] Some embodiments of the present application provide an information query method, a computer device, and a computer-readable storage medium applied to an intravenous catheter information system to solve one or more of the technical problems mentioned in the above background technology section.

[0005] In a first aspect, some embodiments of the present application provide an information query method applied to an intravenous catheter information system, the intravenous catheter information system comprising: a data acquisition subsystem, a knowledge graph construction subsystem, a symptom identification subsystem, a symptom retrieval subsystem, and an information transmission subsystem, the method comprising: the knowledge graph construction subsystem constructing a complication knowledge graph based on an acquired complication type information set; the symptom identification subsystem training a symptom identification model based on the acquired complication type information set and the acquired complication knowledge graph; the information transmission subsystem sending the patient's intravenous catheter information to the symptom identification subsystem and the symptom detection subsystem in response to receiving the patient's intravenous catheter information sent by the management terminal; the symptom identification subsystem, in response to receiving the patient's venous catheter information, inputs the patient's venous catheter information into the symptom identification model to obtain the patient's symptom type information, and sends the patient's symptom type information to the information transmission subsystem; the symptom retrieval subsystem, in response to receiving the patient's venous catheter information, searches the complication type database based on the patient's venous catheter information to generate complication retrieval information, and sends the complication retrieval information to the information transmission subsystem; the information transmission subsystem, in response to receiving the patient's symptom type information and the complication retrieval information, sends the patient's symptom type information and the complication retrieval information to the management terminal.

[0006] In a second aspect, the present application also provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, it implements the method described in any implementation of the first aspect.

[0007] In a third aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, wherein when the computer program is executed by a processor, the method described in any implementation of the first aspect is implemented.

[0008] The above-described embodiments of the present application have the following beneficial effects: The information query method for an intravenous catheter information system, as described in some embodiments of the present application, can reduce the waste of medical resources. Specifically, the waste of medical resources is caused by the unique location of central venous catheters (the catheter tip is located in the lower third of the superior vena cava, at the entrance to the heart), high treatment costs, a wide range of applicability to different patient treatment needs, and the difficulty in identifying and managing complications. This makes clinical management more challenging for medical staff. Misjudgments can waste medical resources and affect the patient's treatment progress. Based on this, in some embodiments of the present application, the information query method for an intravenous catheter information system includes: first, the knowledge graph construction subsystem constructs a complication knowledge graph based on the acquired complication type information set. Thus, the knowledge graph construction subsystem can construct the complication knowledge graph, which can then be used to train a symptom recognition model. Second, the symptom recognition subsystem trains a symptom recognition model based on the acquired complication type information set and the acquired complication knowledge graph. Thus, the symptom recognition subsystem can train the symptom recognition model, which can then be used to identify the patient's symptom type information. Next, in response to receiving the patient's intravenous catheter information from the management terminal, the information transmission subsystem transmits the patient's intravenous catheter information to the symptom identification subsystem and the symptom retrieval subsystem. The symptom identification subsystem then inputs the patient's intravenous catheter information into the symptom identification model to obtain patient symptom type information, and transmits the patient's symptom type information to the information transmission subsystem. This allows the symptom identification subsystem to accurately identify the patient's symptom type based on the symptom identification model. Subsequently, in response to receiving the patient's intravenous catheter information, the symptom retrieval subsystem searches a complication type database based on the patient's intravenous catheter information to generate complication retrieval information, and transmits the complication retrieval information to the information transmission subsystem. This allows the complication retrieval subsystem to retrieve the complication retrieval information from the complication type database. Finally, in response to receiving the patient's symptom type information and the complication retrieval information, the information transmission subsystem transmits the patient's symptom type information and the complication retrieval information to the management terminal. This allows the patient's symptom type information identified by the symptom identification model and the complication retrieval information retrieved from the complication type database to be transmitted to the management terminal. Therefore, the management terminal can obtain more comprehensive and accurate complication information, thereby reducing the waste of medical resources and the impact on the patient's treatment process. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The above and other features, advantages, and aspects of the various embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that the elements and components are not necessarily drawn to scale.

[0010] Figure 1 is a flowchart of some embodiments of an information query method applied to an intravenous catheter information system according to the present application;

[0011] Figure 2 is a schematic diagram of the structure of a computer device suitable for implementing some embodiments of the present application;

[0012] Figure 3 Schematic diagram of application scenarios of some embodiments of the information query method applied to a venous catheter information system according to the present application. DETAILED DESCRIPTION

[0013] The following will describe embodiments of the present application in more detail with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are for illustrative purposes only and are not intended to limit the scope of protection of the present application.

[0014] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of this application can be combined with each other.

[0015] It should be noted that the concepts of "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0016] It should be noted that the modifications of "one" and "multiple" mentioned in this application are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly indicated in the context, they should be understood as "one or more".

[0017] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0018] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0019] Figure 1A process 100 of some embodiments of an information query method for an intravenous catheter information system according to the present application is shown. The information query method for an intravenous catheter information system, wherein the intravenous catheter information system includes: a data acquisition subsystem, a knowledge graph construction subsystem, a symptom identification subsystem, a symptom retrieval subsystem, and an information transmission subsystem, includes the following steps:

[0020] Step 101: The knowledge graph construction subsystem constructs a complication knowledge graph based on the acquired complication type information set.

[0021] In some embodiments, the knowledge graph construction subsystem constructs a complication knowledge graph based on the acquired complication type information set. The complication type information in the complication type information set may be information representing complication types stored in a complication type database. The complication type database may be a database storing complication type information. The complication type information in the complication type information set may include, but is not limited to, patient basic information and complication type. For example, patient basic information may include, but is not limited to, at least one of the following: drug extravasation, patient pain, location of pain, thermal imaging, and ultrasound. Complication types may include, but are not limited to, phlebitis, drug extravasation / extravasation, catheter blockage, catheter-related venous thrombosis, catheter-related bloodstream infection (CRBSI), central venous catheter misplacement / displacement, medical adhesive-related skin injury, and catheter clamp syndrome. Here, the data acquisition subsystem may be the system used to acquire the initial complication type information set. The knowledge graph construction subsystem may be the system used to construct the complication knowledge graph. The symptom identification subsystem may be the system used to identify patient symptom type information. The symptom retrieval subsystem may be the system used to search the complication type database. The information transmission subsystem may be the system used to transmit information.

[0022] In practice, the above-mentioned knowledge graph construction subsystem can construct a complication knowledge graph based on the acquired complication type information set through the following steps:

[0023] The first step is to perform data preprocessing on each complication type information in the complication type information set to generate complication preprocessing information, thereby obtaining the complication preprocessing information set. In practice, data preprocessing is first performed on each complication type information in the complication type information set. The knowledge graph construction subsystem can remove information representing empty spaces from the complication type information to obtain complication preprocessing information. The knowledge graph construction subsystem can then determine the obtained individual complication preprocessing information as complication preprocessing information.

[0024] The second step is to perform entity recognition processing on the complication pretreatment information set to generate a complication entity information set. In practice, the knowledge graph construction subsystem can perform entity recognition processing on the complication pretreatment information set using a preset entity recognition algorithm to generate a complication entity information set. The complication pretreatment information in the complication pretreatment information set corresponds one-to-one to the complication entity information in the complication entity information set. For example, the preset entity recognition algorithm can be a NER (Named Entity Recognition) algorithm (e.g., hidden Markov model, conditional random field, bidirectional long short-term memory network).

[0025] The third step is to perform relationship extraction on the complication preprocessing information set based on the complication entity information set to generate a complication entity relationship information set. In practice, the knowledge graph construction subsystem can perform relationship extraction on the complication preprocessing information set based on the complication entity information set using a preset relationship extraction algorithm to generate a complication entity relationship information set. For example, the preset relationship extraction algorithm can include, but is not limited to, a convolutional neural network, a long short-term memory network, or an attention model.

[0026] The fourth step is to perform attribute extraction on the above complication preprocessing information set based on the above complication entity information set to generate a complication attribute information set.

[0027] In practice, the above-mentioned knowledge graph construction subsystem is based on the above-mentioned complication entity information set, and can extract attributes from the above-mentioned complication preprocessing information set through the following sub-steps to generate a complication attribute information set:

[0028] In the first sub-step, for each complication pre-processing information in the complication pre-processing information set, perform the following attribute extraction steps:

[0029] In the first attribute extraction step, text recognition is performed on the complication preprocessing information to generate complication text information. In practice, the knowledge graph construction subsystem can use a preset text recognition algorithm to perform text recognition on the complication preprocessing information to generate complication text information. For example, the preset text recognition algorithm can be an OCR (Optical Character Recognition) algorithm.

[0030] The second attribute extraction step removes the complication text information based on the complication entity information corresponding to the complication preprocessing information to generate complication text-removed information. In practice, the knowledge graph construction subsystem can remove information from the complication text information that is identical to the corresponding complication entity information to generate complication text-removed information.

[0031] In the third attribute extraction step, information in the complication text-removed information that meets a first preset attribute condition is determined as first complication attribute information. The first preset attribute condition may be information representing the type of clinical manifestation. Here, the clinical manifestation type may include, but is not limited to, drug leakage, drug extravasation, catheter blockage, catheter malposition, and catheter displacement.

[0032] In the fourth attribute extraction step, information in the complication text-removed information that meets the second preset attribute condition is determined as second complication attribute information. The second preset attribute condition may be information representing the patient's experience type. Here, the patient experience type may include, but is not limited to, patient pain, patient chest pain, patient stomach pain, patient acute chest pain, and patient shortness of breath.

[0033] In the fifth attribute extraction step, information in the complication text removal information that satisfies a third preset attribute condition is determined as third complication attribute information. The third preset attribute condition may be information representing the type of tool evaluation. The tool evaluation type may include, but is not limited to, thermal imaging, ultrasound, and the like.

[0034] In the sixth attribute extraction step, the first complication attribute information, the second complication attribute information and the third complication attribute information are determined as complication attribute information.

[0035] The second sub-step is to determine the obtained attribute information of each complication as a complication attribute information set.

[0036] First, text recognition technology can be used to identify the text representing complication pretreatment information. Then, three types of attribute information can be identified to identify the attribute information corresponding to the complication from three different perspectives: clinical manifestations, patient perception, and tool evaluation. This facilitates the subsequent consideration of these three types of attribute information based on the patient's intravenous catheter information to more accurately identify the complication type.

[0037] The fifth step is to perform knowledge fusion processing on the above-mentioned complication entity information set, the above-mentioned complication entity relationship information set, and the above-mentioned complication attribute information set to generate a complication knowledge fusion information set. In practice, first, for each complication entity information in the complication entity information set, the above-mentioned knowledge graph construction subsystem can fuse the various complication entity relationship information corresponding to the above-mentioned complication entity information in the above-mentioned complication entity relationship information set, the complication attribute information corresponding to the above-mentioned complication entity information in the above-mentioned complication attribute information set, and the above-mentioned limb information to generate complication knowledge fusion information. Then, the above-mentioned knowledge graph construction subsystem can determine the generated various complication knowledge fusion information as the complication knowledge fusion information set.

[0038] The sixth step is to model the above-mentioned complication knowledge fusion information set to generate a complication knowledge graph. In practice, the above-mentioned knowledge graph construction subsystem can model the above-mentioned complication knowledge fusion information set through a preset modeling algorithm to generate a complication knowledge graph. For example, the preset modeling algorithm can be: using the various complication entity information included in the complication knowledge fusion information set as nodes, and using the various complication entity relationship information included in the complication knowledge fusion information set as the relationship between nodes to construct a complication knowledge graph. Here, the complication attribute information included in the complication knowledge fusion information set can serve as a supplementary description of the nodes in the complication knowledge graph.

[0039] Before step 101, the method further includes:

[0040] In the first step, the data acquisition subsystem obtains an initial complication type information set, and stores the initial complication type information set in a complication type database.

[0041] In some embodiments, the data acquisition subsystem may obtain an initial complication type information set from a terminal device via a wired or wireless connection, and store the initial complication type information set in a complication type database. The initial complication type information in the initial complication type information set may be information representing the complication type obtained from the terminal device. The initial complication type information in the initial complication type information set may include, but is not limited to, basic patient information and complication type.

[0042] It should be noted that the above wireless connection methods may include but are not limited to 3G / 4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0043] In the second step, the above-mentioned knowledge graph construction subsystem obtains the complication type information set from the above-mentioned complication type database.

[0044] In some embodiments, the knowledge graph construction subsystem obtains a complication type information set from the complication type database. In practice, the knowledge graph construction subsystem may obtain all complication type information included in the complication type database from the complication type database to obtain the complication type information set.

[0045] Step 102: The symptom identification subsystem trains a symptom identification model based on the acquired complication type information set and the acquired complication knowledge graph.

[0046] In some embodiments, the symptom identification subsystem trains a symptom identification model based on the acquired complication type information set and the acquired complication knowledge graph. The complication type information set is acquired from a complication type database, and the complication knowledge graph is acquired from the knowledge graph construction subsystem.

[0047] In practice, the above-mentioned symptom identification subsystem can train the symptom identification model based on the acquired complication type information set and the acquired complication knowledge graph through the following steps:

[0048] The first step is to obtain sample venous catheter information corresponding to each complication type information in the complication type information set to obtain a sample venous catheter information set. In practice, first, for each complication type information in the complication type information set, the symptom identification subsystem may determine the basic patient information included in the complication type information as sample venous catheter information. Then, the symptom identification subsystem may determine each piece of determined sample venous catheter information as a sample venous catheter information set.

[0049] The second step is to determine the initial disease recognition model, which includes: initial feature extraction model, initial vectorization model, initial graph embedding model, initial aggregation model, and initial matching model.

[0050] Here, the initial feature extraction model may be a neural network model that takes the target sample venous catheter information as input and outputs the initial venous catheter feature information. For example, the initial feature extraction model may be a SIFT (Scale-Invariant Feature Transform) algorithm.

[0051] The initial vectorization model may be a model that takes initial venous catheter feature information as input and outputs initial venous catheter vector information. For example, the initial vectorization model may be an IF-IDF (Term Frequency-Inverse Document Frequency) model.

[0052] The initial graph embedding model may be a model that takes the complication knowledge graph as input and outputs an initial knowledge graph embedding information set. For example, the initial graph embedding model may be a TransE (Translation Distance Model for Knowledge Graph Embedding) model.

[0053] The initial aggregation model can be a neural network model that takes the initial venous catheter vector information and the initial knowledge graph embedding information set as input and outputs the initial aggregation information. For example, the initial aggregation model can be a graph convolutional neural network model.

[0054] The initial matching model may be a model that takes the initial venous catheter vector information and the initial aggregation information as input and takes the initial disease type information as output. For example, the initial matching model may be, but is not limited to, a cosine similarity algorithm or a Euclidean distance algorithm.

[0055] The third step is to select target sample venous catheter information from the above sample venous catheter information set.

[0056] In the fourth step, the selected target sample venous catheter information is input into the initial feature extraction model to obtain initial venous catheter feature information.

[0057] In the fifth step, the initial venous catheter feature information is input into the initial vectorization model to obtain the initial venous catheter vector information.

[0058] In the sixth step, the complication knowledge graph is input into the initial graph embedding model to obtain the initial knowledge graph embedding information set.

[0059] In the seventh step, the initial venous catheter vector information and the initial knowledge graph embedding information set are input into the initial aggregation model to obtain the initial aggregation information.

[0060] In the eighth step, the initial venous catheter vector information and the initial aggregation information are input into the initial matching model to obtain the initial disease type information.

[0061] In the ninth step, the difference between the initial disease type information and the complication type information corresponding to the target sample intravenous catheter information is determined based on a preset loss function. For example, the preset loss function may include, but is not limited to, mean square error (MSE), hinge loss, cross entropy, 0-1 loss, absolute value loss, logarithmic loss, square loss, exponential loss, etc.

[0062] In step 10, in response to determining that the difference value is greater than or equal to a preset difference value, the network parameters of the initial disease recognition model are adjusted. For example, the difference value can be calculated by subtracting the preset difference value. Based on this, the parameters of the initial disease recognition model are adjusted using methods such as backpropagation and gradient descent. For example, the preset difference value can be 0.1.

[0063] Optionally, in response to determining that the difference value is less than a preset difference value, the above-mentioned disease recognition subsystem may also determine the initial disease recognition model as the disease recognition model.

[0064] First, the initial feature extraction model can be used to extract feature information from the target sample's venous catheter information to reduce the impact of redundant information. Second, the initial vectorization model can be used to vectorize the initial venous catheter feature information for subsequent identification with the knowledge graph. Next, the initial graph embedding model can be used to embed the knowledge graph for aggregation with the initial venous catheter feature information. Finally, the initial aggregation model can be used to aggregate features of entities in the knowledge graph that are adjacent to the initial venous catheter feature information. Finally, the initial matching model can be used to match relatively accurate initial disease type information from the knowledge graph. Finally, by training the initial disease recognition model, which includes the initial feature extraction model, initial vectorization model, initial graph embedding model, initial aggregation model, and initial matching model, a relatively accurate disease recognition model can be trained. This relatively accurate disease recognition model can then accurately identify the patient's disease type information. This can reduce the waste of medical resources and minimize the impact on patient treatment progress.

[0065] Step 103 : In response to receiving the patient's venous catheter information sent by the management terminal, the information transmission subsystem sends the patient's venous catheter information to the symptom identification subsystem and the symptom retrieval subsystem.

[0066] In some embodiments, the information transmission subsystem transmits the patient's intravenous catheter information to the symptom identification subsystem and the symptom retrieval subsystem in response to receiving the patient's intravenous catheter information from the management terminal. The management terminal may be the terminal that transmits the patient's intravenous catheter information to the information transmission subsystem.

[0067] In practice, in response to receiving the patient's intravenous catheter information sent by the management terminal, the information transmission subsystem may send the patient's intravenous catheter information to the symptom identification subsystem and the symptom retrieval subsystem through the following steps:

[0068] The first step is to store the patient's intravenous catheter information in an asynchronous query queue, wherein the asynchronous query queue can be a queue for performing asynchronous queries.

[0069] In a second step, in response to determining that the asynchronous query queue and the patient's venous catheter information meet a preset queue condition, the patient's venous catheter information is sent to the symptom identification subsystem and the symptom retrieval subsystem. The preset queue condition may be that the patient's venous catheter information is first in the asynchronous query queue.

[0070] Therefore, asynchronous communication can be achieved through the asynchronous query queue, which can optimize the utilization of computing resources, improve system performance, and reduce the waiting time of the management terminal due to a long wait for query results.

[0071] Step 104 : In response to receiving the patient's venous catheter information, the disease identification subsystem inputs the patient's venous catheter information into the disease identification model to obtain patient disease type information, and sends the patient disease type information to the information transmission subsystem.

[0072] In some embodiments, in response to receiving the patient's intravenous catheter information, the disease identification subsystem inputs the patient's intravenous catheter information into the disease identification model to obtain the patient's disease type information, and sends the patient's disease type information to the information transmission subsystem.

[0073] Step 105 , in response to receiving the patient's intravenous catheter information, the symptom retrieval subsystem searches the complication type database based on the patient's intravenous catheter information to generate complication retrieval information, and sends the complication retrieval information to the information transmission subsystem.

[0074] In some embodiments, the above-mentioned disease retrieval subsystem, in response to receiving the patient's intravenous catheter information, searches and processes the complication type database based on the above-mentioned patient's intravenous catheter information to generate complication retrieval information, and sends the above-mentioned complication retrieval information to the above-mentioned information transmission subsystem.

[0075] In practice, in response to receiving the patient's intravenous catheter information, the symptom retrieval subsystem may perform a search process on the complication type database based on the patient's intravenous catheter information through the following steps to generate complication retrieval information:

[0076] The first step is to perform data cleansing on the patient's intravenous catheter information to generate clean patient information. In practice, the symptom retrieval subsystem can perform data cleansing on the patient's intravenous catheter information using a preset data cleansing algorithm to generate clean patient information. For example, the preset data cleansing algorithm can include removing missing values.

[0077] The second step is to perform feature extraction on the patient cleansing information to generate a patient feature information set. In practice, the symptom retrieval subsystem can perform feature extraction on the patient cleansing information using a preset feature extraction algorithm to generate the patient feature information set. For example, the preset feature extraction algorithm may include, but is not limited to, a histogram feature extraction algorithm, a wavelet transform algorithm, or a SIFT (Scale-Invariant Feature Transform) algorithm.

[0078] In the third step, for each patient characteristic information in the above patient characteristic information set, perform the following retrieval sub-steps:

[0079] The first sub-step is to determine the complication similarity between the patient characteristic information and each complication type information included in the complication type database, thereby obtaining a complication similarity set. In practice, the symptom retrieval subsystem may determine the complication similarity between the patient characteristic information and each complication type information included in the complication type database using a preset similarity algorithm to obtain the complication similarity set. For example, the preset similarity algorithm may include, but is not limited to, a Euclidean distance algorithm, a cosine similarity algorithm, or a Pearson correlation coefficient algorithm.

[0080] In the second sub-step, the complication type information corresponding to each complication similarity that satisfies a preset similarity condition in the complication similarity set is determined as the target complication type information. The preset similarity condition may be that the complication similarity is greater than a similarity threshold. For example, the similarity threshold may be 0.8.

[0081] The fourth step is to determine the obtained target complication type information as a target complication type information set.

[0082] The fifth step is to classify the target complication type information set to generate a set of complication classification information groups. In practice, the above-mentioned symptom retrieval subsystem can classify the target complication type information set using a preset classification algorithm to generate a set of complication classification information groups. For example, the preset classification algorithm can be to group together each target complication type information that represents the same complication type.

[0083] In the sixth step, for each complication classification information group in the complication classification information group set, the sum of the similarities of each complication corresponding to the complication classification information group is determined as the target complication similarity.

[0084] In the seventh step, the maximum target complication similarity among the obtained target complication similarities and the corresponding complication classification information in the complication classification information group are determined as the complication retrieval information. In practice, the above-mentioned disease retrieval subsystem can randomly select a complication classification information from the maximum target complication similarity among the obtained target complication similarities and the corresponding complication classification information group as the complication retrieval information.

[0085] Thus, first, the patient's intravenous catheter information can be cleaned to remove missing values. Secondly, the patient's cleaned information can be subjected to feature extraction to extract the characteristic information of the patient's cleaned information so that more accurate complication retrieval information can be subsequently identified. Next, the similarity between each piece of information in the complication type database and the patient's characteristic information can be calculated, and each piece of information with a similarity greater than a similarity threshold can be selected. Afterwards, the obtained information can be classified to group the target complication type information of the same category. Next, the similarities corresponding to each target complication type information in each group can be summed. Finally, the complication classification information corresponding to the maximum target complication similarity can be determined as the complication retrieval information. Thus, the complication retrieval information with the highest similarity to the patient's intravenous catheter information can be obtained.

[0086] Step 106 : In response to receiving the patient's symptom type information and the complication search information, the information transmission subsystem sends the patient's symptom type information and the complication search information to the management terminal.

[0087] In some embodiments, in response to receiving the patient's symptom type information and the complication retrieval information, the information transmission subsystem sends the patient's symptom type information and the complication retrieval information to the management terminal.

[0088] In practice, the information transmission subsystem may perform the following sending steps in response to receiving a query request from the management terminal:

[0089] The first sending step is to send the received patient disease type information and complication search information to the management terminal in response to receiving the patient disease type information and complication search information. The query request may represent a request from the management terminal to receive the patient disease type information and complication search information.

[0090] The second sending step is to send the preset search information to the management terminal in response to not receiving the patient's disease type information or complication search information. The preset search information can be a message indicating "the search has not been completed yet, please wait".

[0091] Therefore, since asynchronous queries do not directly send query results to the management terminal after completion, the management terminal must send a query request to receive the query results. After receiving the query request from the management terminal, the information transmission subsystem sends the query results to the management terminal. However, if the information transmission subsystem receives the query request from the management terminal but has not yet received the patient's symptom type information or complication search information, the query operation is incomplete and the management terminal must wait. This completes the asynchronous query operation.

[0092] The above-described embodiments of the present application have the following beneficial effects: The information query method for an intravenous catheter information system, as described in some embodiments of the present application, can reduce the waste of medical resources. Specifically, the waste of medical resources is caused by the unique location of central venous catheters (the catheter tip is located in the lower third of the superior vena cava, at the entrance to the heart), high treatment costs, a wide range of applicability to different patient treatment needs, and the difficulty in identifying and managing complications. This makes clinical management more challenging for medical staff. Misjudgments can waste medical resources and affect the patient's treatment progress. Based on this, in some embodiments of the present application, the information query method for an intravenous catheter information system includes: first, the knowledge graph construction subsystem constructs a complication knowledge graph based on the acquired complication type information set. Thus, the knowledge graph construction subsystem can construct the complication knowledge graph, which can then be used to train a symptom recognition model. Second, the symptom recognition subsystem trains a symptom recognition model based on the acquired complication type information set and the acquired complication knowledge graph. Thus, the symptom recognition subsystem can train the symptom recognition model, which can then be used to identify the patient's symptom type information. Next, in response to receiving the patient's intravenous catheter information from the management terminal, the information transmission subsystem transmits the patient's intravenous catheter information to the symptom identification subsystem and the symptom retrieval subsystem. The symptom identification subsystem then inputs the patient's intravenous catheter information into the symptom identification model to obtain patient symptom type information, and transmits the patient's symptom type information to the information transmission subsystem. This allows the symptom identification subsystem to accurately identify the patient's symptom type based on the symptom identification model. Subsequently, in response to receiving the patient's intravenous catheter information, the symptom retrieval subsystem searches a complication type database based on the patient's intravenous catheter information to generate complication retrieval information, and transmits the complication retrieval information to the information transmission subsystem. This allows the complication retrieval subsystem to retrieve the complication retrieval information from the complication type database. Finally, in response to receiving the patient's symptom type information and the complication retrieval information, the information transmission subsystem transmits the patient's symptom type information and the complication retrieval information to the management terminal. This allows the patient's symptom type information identified by the symptom identification model and the complication retrieval information retrieved from the complication type database to be transmitted to the management terminal. Therefore, the management terminal can obtain more comprehensive and accurate complication information, thereby reducing the waste of medical resources and the impact on the patient's treatment process.

[0093] This application also provides a computer device 200. Figure 2As shown, computer device 200 includes a bus 201, a processor 202, a memory 203, and a communication interface 204. Processor 202, memory 203, and communication interface 204 communicate with each other via bus 201. Computer device 200 can be a server or a terminal device. It should be understood that this application does not limit the number of processors and memories in computer device 200.

[0094] The bus 201 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 The bus 201 may include a path for transmitting information between various components of the computer device 200 (eg, the memory 203, the processor 202, and the communication interface 204).

[0095] The processor 202 may include any one or more processors such as a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0096] The memory 203 may include a volatile memory, such as a random access memory (RAM). The memory 203 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0097] Memory 203 stores executable program code, which processor 202 executes to implement the functions of the acquisition module, sampling module, determination module, and mixing module, thereby implementing the aforementioned information query method for an intravenous catheter information system. Specifically, memory 203 stores instructions for executing the aforementioned information query method for an intravenous catheter information system.

[0098] The communication interface 204 uses a transceiver module such as, but not limited to, a network interface card or a transceiver to implement communication between the computer device 200 and other devices or a communication network.

[0099] An embodiment of the present application further provides a chip, which includes a processor and a data interface. The processor reads instructions stored in a memory through the data interface to execute the above-mentioned information query method applied to the intravenous catheter information system.

[0100] Embodiments of the present application also provide a computer-readable storage medium. This computer-readable storage medium can be any available medium capable of storing data on a computing device, or a data storage device such as a data center that contains one or more available media. This available medium can be a magnetic medium (e.g., a floppy disk, hard disk, or magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive). This computer-readable storage medium includes instructions that instruct a computing device to execute the aforementioned information query method for an intravenous catheter information system.

[0101] Figure 3 Schematic diagram of application scenarios of some embodiments of the information query method applied to a venous catheter information system according to the present application.

[0102] exist Figure 3In this application scenario, first, the knowledge graph construction subsystem constructs a complication knowledge graph 302 based on the acquired complication type information set 301. In this application scenario, the complication type information in complication type information set 301 may be "drug extravasation, patient pain-free, thermal imaging, drug extravasation." Complication knowledge graph 302 may be "Complication Knowledge Graph 1." Second, the symptom identification subsystem trains a symptom identification model 303 based on the acquired complication type information set 301 and complication knowledge graph 302. In this application scenario, symptom identification model 303 may be "symptom identification model 1." Next, in response to receiving patient intravenous catheter information 304 from the management terminal, the information transmission subsystem transmits this patient intravenous catheter information 304 to the symptom identification subsystem and the symptom retrieval subsystem. In this application scenario, patient intravenous catheter information 304 may be "drug extravasation, patient pain-free, thermal imaging 1." Then, in response to receiving the patient's intravenous catheter information 3044, the symptom identification subsystem inputs the patient's intravenous catheter information 304 into the symptom identification model 303 to obtain patient symptom type information 305, and transmits the patient's symptom type information 305 to the information transmission subsystem. In this application scenario, the patient's symptom type information 305 may be "drug extravasation." Subsequently, in response to receiving the patient's intravenous catheter information 304, the symptom retrieval subsystem searches the complication type database 306 based on the patient's intravenous catheter information 304 to generate complication retrieval information 307, and transmits the complication retrieval information 307 to the information transmission subsystem. In this application scenario, the complication type database 306 may be "Complication Type Database 1." The complication retrieval information 307 may be "drug extravasation." Finally, in response to receiving the patient's symptom type information 305 and complication retrieval information 307, the information transmission subsystem transmits the patient's symptom type information 305 and complication retrieval information 307 to the management terminal.

[0103] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0104] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the protection scope of the technical solutions of the embodiments of the present application.

Claims

1. An information query method applied to an intravenous catheter information system, the intravenous catheter information system comprising: Data collection subsystem, knowledge graph construction subsystem, symptom identification subsystem, symptom retrieval subsystem, and information transmission subsystem, including: The knowledge graph construction subsystem constructs a complication knowledge graph based on the acquired complication type information set; The symptom identification subsystem trains a symptom identification model based on the acquired complication type information set and the acquired complication knowledge graph, wherein the symptom identification model is obtained by training an initial symptom identification model using the complication type information set and the complication knowledge graph, wherein the initial symptom identification model includes: an initial feature extraction model, an initial vectorization model, an initial graph embedding model, an initial aggregation model, and an initial matching model, wherein the initial feature extraction model is a neural network model that takes the target sample venous catheter information as input and the initial venous catheter feature information as output, the initial vectorization model is a model that takes the initial venous catheter feature information as input and the initial venous catheter vector information as output, the initial graph embedding model is a model that takes the complication knowledge graph as input and the initial knowledge graph embedding information set as output, the initial aggregation model is a neural network model that takes the initial venous catheter vector information and the initial knowledge graph embedding information set as input and the initial aggregation information as output, and the initial matching model is a model that takes the initial venous catheter vector information and the initial aggregation information as input and the initial symptom type information as output; The information transmission subsystem, in response to receiving the patient's intravenous catheter information sent by the management terminal, sends the patient's intravenous catheter information to the symptom identification subsystem and the symptom retrieval subsystem; In response to receiving the patient's intravenous catheter information, the condition identification subsystem inputs the patient's intravenous catheter information into the condition identification model to obtain patient's condition type information, and sends the patient's condition type information to the information transmission subsystem; In response to receiving the patient's intravenous catheter information, the symptom retrieval subsystem searches the complication type database based on the patient's intravenous catheter information to generate complication retrieval information, and sends the complication retrieval information to the information transmission subsystem; The information transmission subsystem, in response to receiving the patient's symptom type information and the complication retrieval information, sends the patient's symptom type information and the complication retrieval information to the management terminal; The complication type database is searched and processed to generate complication search information, including: For each patient characteristic information, perform the following retrieval steps: Determining the complication similarity between the patient characteristic information and each complication type information included in the complication type database; Determine the complication type information corresponding to each complication similarity that meets the preset similarity condition in the complication similarity set as the target complication type information; Classify the target complication type information set to generate a complication classification information group set; For each complication classification information group, the sum of the similarities of each complication corresponding to the complication classification information group is determined as the target complication similarity; Determine the largest target complication similarity among the target complication similarities and the complication classification information in the corresponding complication classification information group as complication retrieval information; The step of sending the patient's venous catheter information to the symptom identification subsystem and the symptom retrieval subsystem includes: Storing the patient's intravenous catheter information in an asynchronous query queue; In response to determining that the asynchronous query queue and the patient's intravenous catheter information meet a preset queue condition, the patient's intravenous catheter information is sent to the symptom identification subsystem and the symptom retrieval subsystem.

2. The information query method applied to the intravenous catheter information system according to claim 1, wherein: Before the knowledge graph construction subsystem constructs the complication knowledge graph based on the acquired complication type information set, the method further includes: The data acquisition subsystem obtains an initial complication type information set, and stores the initial complication type information set in a complication type database; The knowledge graph construction subsystem obtains a complication type information set from the complication type database.

3. The information query method applied to the intravenous catheter information system according to claim 1, wherein: The complication knowledge graph is constructed based on the acquired complication type information set, including: performing data preprocessing on each complication type information in the complication type information set to generate complication preprocessing information, thereby obtaining a complication preprocessing information set; performing entity recognition processing on the complication preprocessing information set to generate a complication entity information set, wherein the complication preprocessing information in the complication preprocessing information set corresponds one-to-one to the complication entity information in the complication entity information set; Based on the complication entity information set, performing relationship extraction processing on the complication preprocessing information set to generate a complication entity relationship information set; Based on the complication entity information set, performing attribute extraction processing on the complication preprocessing information set to generate a complication attribute information set; performing knowledge fusion processing on the complication entity information set, the complication entity relationship information set, and the complication attribute information set to generate a complication knowledge fusion information set; The complication knowledge fusion information set is modeled to generate a complication knowledge graph.

4. The information query method applied to the intravenous catheter information system according to claim 3, wherein: The step of performing attribute extraction processing on the complication preprocessing information set based on the complication entity information set to generate a complication attribute information set includes: For each complication preprocessing information in the complication preprocessing information set, perform the following attribute extraction steps: performing text recognition processing on the complication pre-processing information to generate complication text information; Based on the complication entity information corresponding to the complication preprocessing information, the complication text information is removed to generate complication text removal information; Determining the information that satisfies a first preset attribute condition in the complication text removal information as first complication attribute information; Determining the information that satisfies a second preset attribute condition in the complication text removal information as second complication attribute information; Determining the information that satisfies the third preset attribute condition in the complication text removal information as third complication attribute information; determining the first complication attribute information, the second complication attribute information, and the third complication attribute information as complication attribute information; The obtained individual complication attribute information is determined as a complication attribute information set.

5. The information query method applied to the intravenous catheter information system according to claim 1, wherein: In response to receiving the patient's symptom type information and the complication search information, sending the patient's symptom type information and the complication search information to the management terminal includes: In response to receiving the query request from the management terminal, the following sending steps are performed: In response to receiving the patient's condition type information and complication search information, sending the received patient's condition type information and complication search information to the management terminal; In response to not receiving the patient's condition type information or complication search information, the preset search information is sent to the management terminal.

6. A computer device, wherein: The computer device comprises a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 5 are implemented.

7. A computer-readable storage medium, wherein: The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images