An operating room scheduling method, device, equipment and storage medium

Abstract

Embodiments of the present application disclose a kind of operating room scheduling method, device, equipment and storage medium.The method comprises: obtaining target information under each information dimension related to the operating room to be scheduled, wherein the target information under each information dimension includes at least one of doctor information, nurse information, patient information, surgical equipment information, surgery information and operating room information;For the target information under each information dimension, information extraction is carried out on the target information under each information dimension, and the first information extraction result is packaged as a first resource description framework multiset;Each first resource description framework multiset is input into the knowledge graph model for predicting operating room scheduling information which has been trained, and according to the output result of knowledge graph model, the operating room scheduling information of the operating room to be scheduled is obtained.The technical scheme of the embodiments of the present application can effectively guarantee the accuracy and timeliness in operating room scheduling.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the field of information processing, and in particular to a method and device for operating room scheduling, and a storage medium. BACKGROUND

[0002] Operating rooms are very important and scarce in medical treatment, so reasonable scheduling of operating rooms is very important. At present, manual scheduling is mainly used for operating rooms, but it is difficult to ensure the accuracy and timeliness of scheduling. SUMMARY

[0003] Embodiments of the present application provide a method and device for operating room scheduling, and a storage medium, which solve the problem that the accuracy and timeliness of operating room scheduling cannot be ensured.

[0004] In a first aspect, embodiments of the present application provide a method for operating room scheduling, which can include:

[0005] Obtaining target information in each information dimension related to the operating room to be scheduled, wherein the target information in each information dimension includes at least one of doctor information, nurse information, patient information, surgical equipment information, surgery information and operating room information;

[0006] For the target information in each information dimension, information extraction is performed on the target information in each information dimension, and the first information extraction result is packaged into a first resource description framework tuple;

[0007] Each first resource description framework tuple is input into a knowledge graph model for predicting operating room scheduling information which is pre-trained, and the operating room scheduling information of the operating room to be scheduled is obtained according to the output result of the knowledge graph model.

[0008] In a second aspect, embodiments of the present application also provide a device for operating room scheduling, which can include:

[0009] A target information obtaining module is configured to obtain target information in each information dimension related to the operating room to be scheduled, wherein the target information in each information dimension includes at least one of doctor information, nurse information, patient information, surgical equipment information, surgery information and operating room information;

[0010] A first resource description framework tuple obtaining module is configured to, for the target information in each information dimension, perform information extraction on the target information in each information dimension, and package the first information extraction result into a first resource description framework tuple;

[0011] The operating room scheduling information obtaining module is configured to input the first resource description framework multiple tuple into a knowledge graph model trained in advance for predicting operating room scheduling information, and obtain the operating room scheduling information of the operating room to be scheduled according to an output result of the knowledge graph model.

[0012] In a third aspect, an embodiment of the present application further provides an operating room scheduling device, which can include:

[0013] one or more processors;

[0014] a memory for storing one or more programs;

[0015] When the one or more programs are executed by the one or more processors, the one or more processors implement the operating room scheduling method provided by any embodiment of the present application.

[0016] In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the operating room scheduling method provided by any embodiment of the present application.

[0017] The technical solution of the embodiment of the present application obtains target information in each information dimension related to the operating room to be scheduled, the target information in each information dimension can include at least one of doctor information, nurse information, patient information, surgical equipment information, surgery information and operating room information, then, for the target information in each information dimension, information extraction is performed on the target information in each information dimension, and then the first information extraction result obtained thereby is packaged as a first RDF multiple tuple, thereby, the first RDF multiple tuple is input into a knowledge graph model trained in advance for predicting operating room scheduling information, and the operating room scheduling information of the operating room to be scheduled is obtained according to an output result of the knowledge graph model. The above technical solution automatically fully analyzes the target information in each information dimension based on the knowledge graph model, thereby solving the problem that the accuracy and timeliness cannot be guaranteed in operating room scheduling, and meeting the demand for effective operating room scheduling in various complex situations. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 is a flowchart of an operating room scheduling method in the embodiment one of the present application;

[0019] Figure 2 is a schematic diagram of a first optional example of the operating room scheduling method in the embodiment one of the present application;

[0020] Figure 3 is a schematic diagram of a second optional example of the operating room scheduling method in the embodiment one of the present application;

[0021] Figure 4 is a flow chart of a method for operating room scheduling in Embodiment Two of the present application;

[0022] Figure 5 is a flow chart of a method for operating room scheduling in Embodiment Three of the present application;

[0023] Figure 6 is a structural block diagram of an operating room scheduling device in Embodiment Four of the present application;

[0024] Figure 7 is a structural schematic diagram of an operating room scheduling device in Embodiment Five of the present application. DETAILED DESCRIPTION

[0025] The present application will be further described below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely intended for the purpose of interpretation of the present application and are not limiting of the present application. In addition, it should be noted that only the parts related to the present application are shown in the accompanying drawings for the purpose of description.

[0026] Embodiment One

[0027] Figure 1 is a flow chart of a method for operating room scheduling provided in Embodiment One of the present application. The present embodiment can be applicable to the case of predicting operating room scheduling information, especially the case of predicting operating room scheduling information based on multi-dimensional information. The method can be executed by an operating room scheduling device provided in the present application, which can be realized by software and / or hardware, and can be integrated on an operating room scheduling device, which can be various user terminals or servers.

[0028] Referring to Figure 1 , the method of the present embodiment specifically includes the following steps:

[0029] S110, obtaining target information under each information dimension related to the operating room to be scheduled, wherein the target information under each information dimension includes at least one of doctor information, nurse information, patient information, surgical equipment information, surgery information and operating room information.

[0030] The to-be-scheduled operating room can be an operating room whose operating room scheduling information is to be predicted, and the number thereof can be one, two or more, which is not specifically limited herein. The operating room scheduling information can be information that can reflect when the to-be-scheduled operating room will be occupied, and specifically can be information that will be occupied when and for what surgery. The target information can be information in a certain information dimension related to the to-be-scheduled operating room, such as doctor information, nurse information, patient information, surgical equipment information, operating room information and surgery information, i.e., the doctor information is target information in an information dimension, and the nurse information is target information in another information dimension, and similarly, the patient information, the surgical equipment information, the surgery information and the operating room information are similar, which will not be described herein.

[0031] In order to better understand the target information in each information dimension, the following exemplary description is made in combination with the application scenarios that the embodiments of the present application can involve. Exemplarily, the doctor information can be information related to a doctor, such as basic information of a certain doctor or doctors, surgery record information, shift information, etc.; the patient information can be information related to a patient, such as basic information of a certain patient or patients, surgery information, postoperative information (i.e., patient postoperative information), etc.; the nurse information can be information related to a nurse, such as basic information of a certain nurse or nurses, surgery record information, etc.; the surgical equipment information can be information related to surgical equipment, such as preoperative, intraoperative and / or postoperative inspection information, use record and duration of a certain surgical equipment or surgical equipment; the surgery information can be information related to surgery, such as basic information of a certain surgery or surgeries, sudden situation record information, doctor-patient information, etc., which are mostly relation representations and can be used as a connection point of surgical personnel (such as doctors, nurses, patients, etc.) and surgical equipment; the operating room information can be information related to the to-be-scheduled operating room, such as basic information of a certain to-be-scheduled operating room or to-be-scheduled operating rooms, use record and duration, sudden situation record information, etc. Comprehensive consideration of the target information in each information dimension helps to improve the determination accuracy of subsequent operating room scheduling information.

[0032] In S120, for the target information in each information dimension, information extraction is performed on the target information in each information dimension, and a first information extraction result is packaged as a first resource description framework multi-tuple.

[0033] The target information under each information dimension is subjected to information extraction and packaging operations in sequence. Specifically, for the target information under a certain information dimension, information extraction is performed, such as information cleaning (e.g., cleaning incomplete and / or redundant target information), semantic segmentation, and the like, to obtain a first information extraction result. In actual applications, the semantic segmentation operation can be implemented by the following steps: obtaining a semantic segmentation dictionary, wherein the semantic segmentation dictionary is a dictionary constructed in advance according to the annotation results when the knowledge graph model is trained; based on the semantic segmentation dictionary, the target information under each information dimension is subjected to semantic segmentation, and the target information under each information dimension is updated according to the semantic segmentation result. The annotation results given by the annotators when the knowledge graph model is trained can be the relationships between sample information under each information dimension. Based on such annotation results, a semantic segmentation dictionary for semantic segmentation can be constructed in advance. Further, for the target information under a certain information dimension, the target information under the information dimension is subjected to semantic segmentation based on the semantic segmentation dictionary, so that the effective information in the target information under the information dimension that can play a role in predicting the operating room scheduling information can be obtained.

[0034] The first information extraction result is packaged into a first Resource Description Framework (RDF) tuple, thereby achieving the effect of packaging the fragmented and discrete first information extraction result into an organic RDF tuple, which can represent the relationship between entities and entities (i.e., nodes and nodes). The tuple can be a triple, a quadruple, a quintuple, etc., which is not specifically limited herein. In actual applications, optionally, after obtaining the first RDF tuple, graph encoding can be performed on it, i.e., encapsulating it according to the requirements of the knowledge graph model, thereby achieving the effect of converting natural language into a data language that can be understood by a computer, and updating the first RDF tuple according to the graph encoding result. Further optionally, in order to enable the computer to better process the first RDF tuple, graph embedding can be performed on the first RDF tuple by Struc2Vec to obtain a vectorized first RDF tuple.

[0035] S130, input each first Resource Description Framework (RDF) tuple into a pre-trained knowledge graph model for predicting operating room scheduling information, and obtain the operating room scheduling information of the to-be-scheduled operating room according to the output result of the knowledge graph model.

[0036] Wherein, since each target information under each information dimension corresponds to a respective first RDF tuple, each first RDF tuple can be input into a pre-trained knowledge graph model, so that the knowledge graph model comprehensively analyzes each first RDF tuple, and thus obtains the operating room scheduling information according to the comprehensive analysis result. In actual application, optionally, the above prediction process of the operating room scheduling information can be understood as a process of converting each first RDF tuple into a target RDF tuple, and then analyzing the target RDF tuple to obtain the operating room scheduling information. The above conversion process can be performed by the knowledge graph model; the above analysis process can be performed by the knowledge graph model (i.e. the output result of the knowledge graph model is the operating room scheduling information), or can be performed by other modules / units (i.e. the output result is the target RDF tuple), etc., which is not limited here.

[0037] Alternatively, the above knowledge graph model can be a Graph Neural Networks (GNN) model, a Graph Convolutional Networks (GCN) model, a Graph Attention Networks (GAT) model, a GraphSAGE model, an MPNN model, etc., which is not limited here. In combination with the application scenarios that may be involved in the embodiments of the present application, since the operating personnel information (such as doctor information, nurse information, patient information, etc.), operating room information and / or operating equipment information, etc. can all change, the above knowledge graph model can be constructed based on the GAT model. For example, as shown in Figure 2 , it is assumed that the operating personnel information, the operating room information and the operating equipment information are represented by nodes u, v and w respectively. , wherein N is the number of nodes in the GAT model, F is the dimension of node features, and in Figure 2 , the F of node u is 5, and the feature vector of the target RDF tuple output after the node u passes through the Graph Attention Layer can be represented by , wherein F' can be any value, and in Figure 2 , the 5 dimensions around the node v.

[0038] The technical solution of the embodiment of the present invention obtains target information under each information dimension related to the operating room to be scheduled, and the target information under each information dimension may include at least one of doctor information, nurse information, patient information, surgical equipment information, surgery information and operating room information; then, for the target information under each information dimension, information extraction is performed on the target information under each information dimension, and then the first information extraction result obtained is packaged into a first RDF tuple; thereby, each first RDF tuple is input into a pre-trained knowledge graph model for predicting operating room scheduling information, and the operating room scheduling information of the operating room to be scheduled is obtained according to the output result of the knowledge graph model. The above technical solution automatically performs a full analysis of the target information under each information dimension based on the knowledge graph model, thereby solving the problem of difficulty in ensuring accuracy and timeliness in operating room scheduling, and meeting the demand for effective operating room scheduling under various complex and changeable situations.

[0039] An optional technical solution, the above-mentioned operating room scheduling method may further include: obtaining medical information related to the operating room scheduling information based on the output results, and the medical information may include at least one of the following: operation sequence information, doctor scheduling information, nurse scheduling information, postoperative patient information and risk warning information. Among them, the operation sequence information may be information related to the operation sequence, such as operation duration information, advance warning time information in case of emergencies, operation extension time information, etc.; the doctor scheduling information may be information related to the doctor scheduling, such as information on when a certain doctor or certain doctors will perform what operation; the situation of the nurse scheduling information is similar and will not be repeated here; the postoperative patient information may be relevant information about the patient after the operation, the key points of the work of the medical staff, etc.; the risk warning information may be information used to warn of risk items related to the operation, information for pushing risk handling plans, etc. It should be noted that what kind of medical information the knowledge graph model can predict is related to the labeling results in the model training phase, that is, what medical information is labeled in the labeling results, and then these medical information can be predicted in the model application phase. For example, if Figure 3 As shown, information extraction is performed on the target information under each information dimension on the left, and then the first information extraction results obtained are packaged into first RDF tuples, and each first RDF tuple is input into the knowledge graph model, and the medical information and operating room scheduling information on the right are obtained according to the output results of the knowledge graph model. In actual applications, optionally, when making predictions, the knowledge graph model can first predict the surgical timing information, and then predict the remaining information based on the surgical timing information to ensure the smooth progress of the patient's surgery. The above technical solution achieves the effect of reasonable scheduling of manpower and material resources related to the operating room to be scheduled.

[0040] On this basis, optionally, after obtaining the target information in each information dimension related to the to-be-scheduled operating room, the above operating room scheduling method can further include: extracting label information from the doctor information, and updating the doctor information according to the label information; accordingly, the medical information includes doctor scheduling information, and the medical information related to the operating room scheduling information obtained according to the output result can include: obtaining label scheduling information according to the output result, and determining the doctor scheduling information according to the label scheduling information and the attribute information of each candidate doctor. It should be noted that the above optional scheme is described by taking the doctor information in the operating personnel information as an example. It should be noted that the nurse information and / or patient information can also be processed by using the above optional scheme, which is not limited here. Since the doctors who can perform operations may change (such as a doctor being transferred to other departments, leaving, being hired, etc.), that is, the doctor information may change, in order to ensure that the doctor scheduling information predicted by the knowledge graph model conforms to the current situation, the doctor information can be fuzzified and processed using a fuzzy mapping relationship, that is, the concrete information is converted into fuzzy information, and then converted into concrete information after fuzzy prediction. Specifically, the label information is extracted from the doctor information. The label information can be information that can reflect the personal attributes of one or more doctors. Thus, the effect of converting the concrete information belonging to one or more doctors into fuzzy information that no longer belongs to one or more doctors but can reflect personal attributes is achieved. The doctor information is updated according to the label information, that is, the doctor information at this time is the label information. Since the first RDF tuple input into the knowledge graph model is related to the label information, the information related to the doctor scheduling obtained according to the output result is label scheduling information. The label scheduling information can be the scheduling information of the doctor represented by the label information, such as arranging the chief physician of the anorectal department to perform XX operation at XX time. The attribute information of each candidate doctor at the current time is determined, wherein the candidate doctor can be a doctor who can currently perform an operation. Then, the doctor scheduling information can be determined according to the label scheduling information and the attribute information of each candidate doctor, such as arranging the doctor whose department is anorectal department and whose title is chief physician to perform XX operation at XX time. Thus, the effect of still being able to predict effective doctor scheduling information when the doctor information changes is achieved.

[0041] Embodiment Two

[0042] Figure 4is a flowchart of a surgical room scheduling method provided in Embodiment Two of the present application. The present embodiment is optimized on the basis of the above technical solutions. In the present embodiment, after obtaining the surgical room scheduling information of the to-be-scheduled surgical room, the above surgical room scheduling method can further include: when a surgical emergency event is detected, obtaining target information in each information dimension including surgical emergency information corresponding to the surgical emergency event, and updating the target information in each information dimension according to the information acquisition result; for the target information in each information dimension, performing information extraction on the target information in each information dimension, and packing the second information extraction result into a second resource description framework tuple; comparing and matching the second resource description framework tuple with each node in the knowledge graph model to obtain a third resource description framework tuple; inputting the third resource description framework tuple into the knowledge graph model, and updating the surgical room scheduling information according to the output result of the knowledge graph model. Wherein, the same or corresponding terms as in the above embodiments are not repeated here.

[0043] Referring to Figure 4 The method of the present embodiment can specifically include the following steps:

[0044] S210, obtaining target information in each information dimension related to the to-be-scheduled surgical room, wherein the target information in each information dimension includes at least one of doctor information, nurse information, patient information, surgical equipment information, surgical information and surgical room information.

[0045] S220, for the target information in each information dimension, performing information extraction on the target information in each information dimension, and packing the first information extraction result into a first resource description framework tuple.

[0046] S230, inputting each first resource description framework tuple into a pre-trained knowledge graph model for predicting surgical room scheduling information, and obtaining the surgical room scheduling information of the to-be-scheduled surgical room according to the output result of the knowledge graph model.

[0047] S240, when a surgical emergency event is detected, obtaining target information in each information dimension including surgical emergency information corresponding to the surgical emergency event, and updating the target information in each information dimension according to the information acquisition result.

[0048] The surgical emergency can be triggered by an unplanned situation (i.e., an emergency situation) occurring before, during, and / or after surgery, where the emergency situation can be a patient emergency, a doctor on leave, a medical device failure, etc., which is not specifically limited herein. Upon detecting a surgical emergency, the target information under each information dimension including the surgical emergency information can be obtained, in other words, the target information under each information dimension at this time includes the surgical emergency information, which can be information of the emergency situation related to surgery, such as specific information of the emergency situation, treatment plan information, etc.

[0049] S250, for the target information under each information dimension, performing information extraction on the target information under each information dimension, and packing the second information extraction result as a second resource description framework (RDF) tuple.

[0050] In order to minimize the degree of influence on the predicted operating room scheduling information, after updating the target information under each information dimension based on the target information under each information dimension including the surgical emergency information, the target information under each information dimension can be subjected to information extraction and packing operations in sequence to obtain a second RDF tuple.

[0051] S260, comparing and matching the second RDF tuple with each node in the knowledge graph model to obtain a third RDF tuple.

[0052] The second RDF tuple is compared and matched with each node in the knowledge graph model to obtain a third RDF tuple, which can be an RDF tuple adjusted based on each node from the second RDF tuple. Thus, data integration is performed at the data level first, and then the third RDF tuple is input into the knowledge graph model for prediction of the operating room scheduling information, thereby achieving the effect of minimizing the degree of influence on the predicted operating room scheduling information.

[0053] S270, inputting the third RDF tuple into the knowledge graph model, and updating the operating room scheduling information according to the output result of the knowledge graph model.

[0054] The technical scheme of the embodiment of the present application can obtain target information in each information dimension including surgical emergency information corresponding to a surgical emergency when a surgical emergency occurs, and update the information; for the target information in each information dimension, information extraction is performed on the target information in each information dimension, and the second information extraction result is packaged as a second RDF tuple; then, the second RDF tuple is compared and matched with each node in the knowledge graph model to obtain a third RDF tuple, thereby realizing the effect of data integration at the data level to reduce the influence degree on the predicted operating room scheduling information; finally, the third RDF tuple is input into the knowledge graph model to re-predict the operating room scheduling information, thereby achieving the effect of real-time adjustment of the operating room scheduling information when a surgical emergency occurs.

[0055] Embodiment three

[0056] Figure 5 is a flowchart of an operating room scheduling method provided in Embodiment Three of the present application. The present embodiment is optimized on the basis of the above technical schemes. In the present embodiment, optionally, the knowledge graph model is obtained by the following steps: obtaining sample information in each information dimension related to the operating room to be scheduled, and annotation results corresponding to the sample information in each information dimension, wherein the annotation results include the relationships between the sample information in each information dimension; for the sample information in each information dimension, information extraction is performed on the sample information in each information dimension, and the sample information extraction result is packaged as a sample resource description framework tuple; the expected resource description framework tuple is determined according to the annotation results, and the expected resource description framework tuple and each sample resource description framework tuple are used as a group of training samples; the knowledge graph model to be trained is trained based on multiple groups of training samples to obtain the knowledge graph model. Wherein, the explanations of the same or corresponding terms as in the above embodiments are not repeated here.

[0057] Referring to Figure 5 , the method of the present embodiment can specifically include the following steps:

[0058] S310, obtaining sample information in each information dimension related to the operating room to be scheduled, and annotation results corresponding to the sample information in each information dimension, wherein the sample information in each information dimension includes at least one of doctor information, nurse information, patient information, surgical equipment information, surgical information and operating room information, and the annotation results include the relationships between the sample information in each information dimension.

[0059] The sample information and the target information are essentially the same, and are information related to the to-be-scheduled operating room in a certain information dimension. Different names are used here only to distinguish the model training stage and the model application stage, and not to specifically limit the substance. The annotation result can be a result given by an annotator according to the sample information in each information dimension to represent the relationship between the sample information in each information dimension. The relationship can be understood as the relationship between nodes (which can also be called entities). In actual application, if only the knowledge graph model is required to predict the operating room scheduling information, the annotation result can include relationships related to the operating room scheduling information. If the knowledge graph model is also required to predict certain medical information, the annotation result can also include relationships related to the medical information. Further optionally, if the knowledge graph model is required to handle surgical emergencies, the sample information in each information dimension can also include surgical emergency information.

[0060] S320, for the sample information in each information dimension, performing information extraction on the sample information in each information dimension, and packing the sample information extraction result as a sample resource description framework tuple.

[0061] Similar to the relationship between the sample information and the target information, the sample information extraction result and the first information extraction result, and the sample RDF tuple and the first RDF tuple are similar, and will not be described here.

[0062] S330, determining an expected resource description framework tuple according to the annotation result, and taking the expected resource description framework tuple and each sample resource description framework tuple as a set of training samples.

[0063] The expected RDF tuple can be an RDF tuple determined according to the annotation result, which can be used as an expected output in the model training process.

[0064] S340, training the knowledge graph model to be trained based on a plurality of sets of training samples, to obtain a trained knowledge graph model for predicting operating room scheduling information.

[0065] S350, obtaining target information in each information dimension related to the to-be-scheduled operating room.

[0066] S360, for the target information in each information dimension, performing information extraction on the target information in each information dimension, and packing the first information extraction result as a first resource description framework tuple.

[0067] S370, inputting each first resource description framework tuple into the knowledge graph model, and obtaining operating room scheduling information of the to-be-scheduled operating room according to an output result of the knowledge graph model.

[0068] The technical scheme of the embodiment of the present application is that sample information under each information dimension is subjected to information extraction and packaging, and then the sample RDF tuples obtained therefrom and the expected RDF tuples determined according to the labeling result are taken as a group of training samples for model training, learning and induction of multiple factors (i.e. information in each dimension) affecting operating room scheduling information, thereby obtaining a knowledge graph model for predicting operating room scheduling information.

[0069] Embodiment Four

[0070] Figure 6 A structural block diagram of an operating room scheduling device is provided for the fourth embodiment of the present application, which is used to execute the operating room scheduling method provided by any of the above embodiments. The device and the operating room scheduling method of each embodiment belong to the same inventive concept, and the details not described in the embodiment of the operating room scheduling device can be referred to the embodiment of the operating room scheduling method. Referring to Figure 6 , the device can specifically include a target information acquisition module 410, a first resource description framework tuple obtaining module 420 and an operating room scheduling information obtaining module 430. Among them,

[0071] The target information acquisition module 410 is configured to acquire target information in each information dimension related to the operating room to be scheduled, wherein the target information in each information dimension includes at least one of doctor information, nurse information, patient information, surgical equipment information, surgery information and operating room information;

[0072] The first resource description framework tuple obtaining module 420 is configured to, for the target information in each information dimension, perform information extraction on the target information in each information dimension, and package the first information extraction result into a first resource description framework tuple;

[0073] The operating room scheduling information obtaining module 430 is configured to input each first resource description framework tuple into the knowledge graph model for predicting operating room scheduling information which has been pre-trained, and obtain the operating room scheduling information of the operating room to be scheduled according to the output result of the knowledge graph model.

[0074] Optionally, the operating room scheduling device can further include:

[0075] The target information updating module is configured to, after obtaining the operating room scheduling information of the operating room to be scheduled, acquire target information in each information dimension corresponding to the surgical emergency including surgical emergency information when a surgical emergency is detected, and update the target information in each information dimension according to the information acquisition result;

[0076] The second resource description framework tuple obtaining module is configured to perform information extraction on the target information in each information dimension, and pack the second information extraction result as a second resource description framework tuple.

[0077] The third resource description framework tuple obtaining module is configured to compare and match the second resource description framework tuple with each node in the knowledge graph model, and obtain a third resource description framework tuple.

[0078] The operating room scheduling information updating module is configured to input the third resource description framework tuple into the knowledge graph model, and update the operating room scheduling information according to an output result of the knowledge graph model.

[0079] Optionally, the first resource description framework tuple obtaining module 420 can include:

[0080] The semantic segmentation dictionary obtaining unit is configured to obtain a semantic segmentation dictionary, wherein the semantic segmentation dictionary is a dictionary obtained in advance according to a labeling result when the knowledge graph model is trained.

[0081] The semantic segmentation unit is configured to perform semantic segmentation on the target information in each information dimension based on the semantic segmentation dictionary, and update the target information in each information dimension according to a semantic segmentation result.

[0082] And / or,

[0083] The operating room scheduling device can further include:

[0084] The graph coding module is configured to perform graph coding on the first resource description framework tuple after the first information extraction result is packed as the first resource description framework tuple, and update the first resource description framework tuple according to a graph coding result.

[0085] Optionally, the operating room scheduling device can further include:

[0086] The medical information obtaining module is configured to obtain medical information related to the operating room scheduling information according to the output result, wherein the medical information includes at least one of operation timing information, doctor scheduling information, nurse scheduling information, postoperative patient information, and risk warning information.

[0087] Optionally, the device can further include:

[0088] The label information extracting module is configured to extract label information from the doctor information after obtaining the target information in each information dimension related to the operating room to be scheduled, and update the doctor information according to the label information.

[0089] The medical information includes doctor scheduling information, and the medical information obtaining module can include:

[0090] The doctor scheduling information determining unit is configured to obtain label scheduling information according to the output result, and determine the doctor scheduling information according to the label scheduling information and attribute information of each candidate doctor.

[0091] Optionally, the knowledge graph model is obtained by pre-training through the following modules:

[0092] The annotation result obtaining module is configured to obtain sample information under each information dimension related to the to-be-scheduled operating room and annotation results corresponding to the sample information under each information dimension, wherein the annotation results include relationships between the sample information under each information dimension.

[0093] The sample resource description framework multi-tuple obtaining module is configured to perform information extraction on the sample information under each information dimension, and pack the sample information extraction result as a sample resource description framework multi-tuple.

[0094] The training sample obtaining module is configured to determine an expected resource description framework multi-tuple according to the annotation result, and use the expected resource description framework multi-tuple and each sample resource description framework multi-tuple as a group of training samples.

[0095] The knowledge graph model obtaining module is configured to train the knowledge graph model to be trained based on the multiple groups of training samples, and obtain the knowledge graph model.

[0096] Optionally, the knowledge graph model can include a graph attention model.

[0097] The operating room scheduling device provided in Embodiment Four of the present application obtains target information under each information dimension related to the to-be-scheduled operating room through the target information obtaining module, and the target information under each information dimension can include at least one of doctor information, nurse information, patient information, surgical equipment information, surgery information and operating room information. Then, the first resource description framework multi-tuple obtaining module performs information extraction on the target information under each information dimension, and packs the first information extraction result obtained thereby into a first RDF multi-tuple. Thus, the operating room scheduling information obtaining module inputs each first RDF multi-tuple into a pre-trained knowledge graph model used for predicting operating room scheduling information, and obtains operating room scheduling information of the to-be-scheduled operating room according to an output result of the knowledge graph model. The above device automatically fully analyzes the target information under each information dimension based on the knowledge graph model, thereby solving the problem that precision and timeliness cannot be guaranteed in operating room scheduling, and meeting the demand for effective operating room scheduling under various complex and changeable conditions.

[0098] The operating room scheduling device provided in the embodiment of the present invention can execute the operating room scheduling method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0099] It is worth noting that in the embodiment of the above-mentioned operating room scheduling device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0100] Example 5

[0101] Figure 7 This is a structural diagram of an operating room scheduling device provided in Example 5 of the present invention, such as Figure 7 As shown, the device includes a memory 510, a processor 520, an input device 530, and an output device 540. The number of processors 520 in the device can be one or more. Figure 7 In the embodiment, a processor 520 is used as an example; the memory 510, the processor 520, the input device 530 and the output device 540 in the device can be connected via a bus or other means. Figure 7 The connection via bus 550 is taken as an example.

[0102] The memory 510, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the operating room scheduling method in the embodiments of the present invention (e.g., the target information acquisition module 410, the first resource description frame tuple acquisition module 420, and the operating room scheduling information acquisition module 430 in the operating room scheduling device). The processor 520 executes the software programs, instructions, and modules stored in the memory 510 to execute various functional applications and data processing of the device, thereby implementing the above-mentioned operating room scheduling method.

[0103] The memory 510 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created based on the use of the device, etc. In addition, the memory 510 may include a high-speed random access memory and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 510 may further include a memory remotely located relative to the processor 520, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0104] The input device 530 can be used to receive inputted digital or character information, and to generate key signal input related to user settings and function control of the device. The output device 540 can include a display device such as a display screen.

[0105] Embodiment six

[0106] Embodiment six of the present application provides a storage medium containing computer executable instructions, which when executed by a computer processor, are used to perform an operating room scheduling method, which can include:

[0107] Obtaining target information in each information dimension related to the operating room to be scheduled, wherein the target information in each information dimension includes at least one of doctor information, nurse information, patient information, surgical equipment information, surgery information and operating room information;

[0108] For the target information in each information dimension, information extraction is performed on the target information in each information dimension, and the first information extraction result is packaged into a first resource description framework tuple;

[0109] Inputting each first resource description framework tuple into a pre-trained knowledge graph model for predicting operating room scheduling information, and obtaining operating room scheduling information of the operating room to be scheduled according to the output result of the knowledge graph model.

[0110] Of course, the computer executable instructions of the storage medium provided by the embodiment of the present application are not limited to the method operations as described above, but can also perform related operations in the operating room scheduling method provided by any embodiment of the present application.

[0111] From the above description of the embodiments, those skilled in the art can clearly understand that the present application can be realized by means of software and necessary general hardware, and of course can also be realized by hardware, but in many cases the former is a better embodiment. According to such understanding, the technical solutions of the present application can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH memory, a hard disk or an optical disk, etc., including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in various embodiments of the present application.

[0112] Note that the above merely describes preferred embodiments of the present application and the principles of the technology applied. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and that various obvious changes, modifications and substitutions can be made without departing from the scope of the present application. Therefore, although the present application has been described in detail through the above embodiments, the present application is not limited to the above embodiments, and can include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the claims.

Claims

1. An operating room scheduling method, characterized by, The method comprises the following steps: acquiring target information in each information dimension related to a to-be-scheduled operating room, wherein the target information in each information dimension comprises at least one of doctor information, nurse information, patient information, surgical equipment information, surgery information and operating room information; for each target information in each information dimension, performing information extraction on the target information in each information dimension, and packing a first information extraction result as a first resource description framework tuple; inputting each first resource description framework tuple into a pre-trained knowledge graph model for predicting operating room scheduling information, and obtaining the operating room scheduling information of the to-be-scheduled operating room according to an output result of the knowledge graph model; wherein the operating room scheduling information is information about when and for what surgery the to-be-scheduled operating room will be occupied, and the knowledge graph model comprises a graph attention model; when a surgery emergency event is detected, acquiring target information in each information dimension corresponding to the surgery emergency event, wherein the target information in each information dimension comprises surgery emergency information, and updating the target information in each information dimension according to the information acquisition result; wherein the surgery emergency event is an event triggered by a situation other than the plan before, during and / or after surgery; for each target information in each information dimension, performing information extraction on the target information in each information dimension, and packing a second information extraction result as a second resource description framework tuple; comparing and matching each node in the knowledge graph model with the second resource description framework tuple to obtain a third resource description framework tuple; inputting the third resource description framework tuple into the knowledge graph model, and updating the operating room scheduling information according to an output result of the knowledge graph model; wherein the knowledge graph model is pre-trained by the following steps: acquiring sample information in each information dimension related to the to-be-scheduled operating room, and a label result corresponding to the sample information in each information dimension, wherein the label result comprises a relationship between the sample information in each information dimension; for each sample information in each information dimension, performing information extraction on the sample information in each information dimension, and packing a sample information extraction result as a sample resource description framework tuple; determining an expected resource description framework tuple according to the label result, and taking the expected resource description framework tuple and each sample resource description framework tuple as a group of training samples; training the knowledge graph model to be trained based on multiple groups of training samples to obtain the knowledge graph model.

2. The method of claim 1, wherein, The information extraction on each target information in each information dimension comprises: acquiring a semantic segmentation dictionary, wherein the semantic segmentation dictionary is a dictionary pre-constructed according to a label result when the knowledge graph model is trained; based on the semantic segmentation dictionary, performing semantic segmentation on the target information in each information dimension, and updating the target information in each information dimension according to the semantic segmentation result; and / or, After the first information extraction result is packed as the first resource description framework tuple, the method further comprises: performing graph coding on the first resource description framework tuple, and updating the first resource description framework tuple according to the graph coding result.

3. The method of claim 1, wherein, Further comprising: obtaining medical information related to the operating room scheduling information according to the output result, wherein the medical information comprises at least one of operation timing information, doctor scheduling information, nurse scheduling information, postoperative patient information and risk warning information.

4. The method of claim 3, wherein, After the target information in each information dimension related to the to-be-scheduled operating room is obtained, further comprising: extracting label information from the doctor information, and updating the doctor information according to the label information; The medical information comprises the doctor scheduling information, and the medical information related to the operating room scheduling information is obtained according to the output result, comprising: obtaining label scheduling information according to the output result, and determining the doctor scheduling information according to the label scheduling information and attribute information of each candidate doctor.

5. An operating room scheduling apparatus, characterized by, Comprising: a target information acquisition module configured to acquire target information in each information dimension related to a to-be-scheduled operating room, wherein the target information in each information dimension comprises at least one of doctor information, nurse information, patient information, operation equipment information, operation information and operating room information; a first resource description framework tuple obtaining module configured to, for each target information in the information dimension, perform information extraction on the target information in each information dimension, and pack a first information extraction result as a first resource description framework tuple; an operating room scheduling information obtaining module configured to input each first resource description framework tuple into a knowledge graph model for predicting operating room scheduling information which is pre-trained, and obtain the operating room scheduling information of the to-be-scheduled operating room according to an output result of the knowledge graph model, wherein the operating room scheduling information is information about when and for what operation the to-be-scheduled operating room is occupied, and the knowledge graph model comprises a graph attention model; a target information updating module configured to, after the operating room scheduling information of the to-be-scheduled operating room is obtained, acquire target information in each information dimension corresponding to a surgical emergency event and comprising surgical emergency information when a surgical emergency event is detected, and update the target information in each information dimension according to the information acquisition result, wherein the surgical emergency event is an event triggered by a situation other than the plan before, during and / or after an operation; a second resource description framework tuple obtaining module configured to, for each target information in the information dimension, perform information extraction on the target information in each information dimension, and pack a second information extraction result as a second resource description framework tuple; a third resource description framework tuple obtaining module configured to compare and match the second resource description framework tuple with each node in the knowledge graph model, and obtain a third resource description framework tuple; The operating room scheduling information updating module is configured to input the third resource description framework multi-tuple into the knowledge graph model, and update the operating room scheduling information according to an output result of the knowledge graph model. The knowledge graph model is obtained by pre-training through the following modules: The annotation result obtaining module is configured to obtain sample information under each information dimension related to the operating room to be scheduled, and an annotation result corresponding to the sample information under each information dimension, wherein the annotation result includes relationships between the sample information under each information dimension. The sample resource description framework multi-tuple obtaining module is configured to, for each sample information under each information dimension, perform information extraction on each sample information under each information dimension, and pack the sample information extraction result as a sample resource description framework multi-tuple. The training sample obtaining module is configured to determine an expected resource description framework multi-tuple according to the annotation result, and use the expected resource description framework multi-tuple and each sample resource description framework multi-tuple as a group of training samples. The knowledge graph model obtaining module is configured to train the knowledge graph model to be trained based on multiple groups of training samples, and obtain the knowledge graph model.

6. An operating room scheduling apparatus, characterized by One or more processors; Memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the operating room scheduling method of any one of claims 1-4. The computer program is executed by the processor to implement the operating room scheduling method of any one of claims 1-4.

7. A computer readable storage medium having stored thereon a computer program, characterized in that, ​

Images