Artificial Intelligence-Based Design Method and System for Static Infusion Tubing

Abstract

The present invention provides an artificial intelligence-based static therapy pipeline design method and system. Among them, the method includes: obtaining patient information of a patient to be statically treated to determine a set of trunks to be statically treated, the set of trunks to be statically treated including at least one static therapy trunk area; obtaining the historical static therapy records of the patient to be statically treated within a preset time before the current time point, and performing area selection based on the historical static therapy records and the set of trunks to be statically treated to obtain the target trunk area of the patient to be statically treated; obtaining a corresponding area image based on the target trunk area, and creating a virtual trunk model having a mapping relationship with the target trunk area based on the area image; obtaining each virtual blood vessel in the virtual trunk model, and determining the virtual blood vessels meeting the medical ecological requirements as the blood vessels to be injected based on the ecological attributes of each virtual blood vessel; obtaining the static therapy drug information of the patient to be statically treated, and generating a drug pipeline model connected to the blood vessels to be injected based on the static therapy drug information. The present invention at least improves the efficiency of static therapy assistance.

Description

Technical Field

[0001] The present invention relates to data processing technologies, and in particular, to an artificial intelligence-based intravenous therapy pipeline design method and system. Background Art

[0002] Intravenous injection is a medical method of directly injecting liquid substances such as blood, medicine solution, and nutrient solution into a vein. Intravenous injection can be divided into transient and continuous types. Transient intravenous injection is mostly performed by directly injecting into a vein with a syringe, which is the generally common injection method; continuous intravenous injection is implemented by intravenous drip, commonly known as "drip".

[0003] The inventors found in their research that before performing an intravenous injection on a patient, it is generally necessary for medical staff to personally check the relevant medical records of the patient to determine the specific location of the current intravenous injection, which is inefficient and not easy to determine an appropriate medical plan. Summary of the Invention

[0004] Based on the above problems, the present invention is proposed to provide an artificial intelligence-based intravenous therapy pipeline design method and system that overcomes the above problems or at least partially solves the above problems, and has the advantages of high efficiency and an appropriate medical plan.

[0005] According to one aspect of the present invention, there is provided an artificial intelligence-based intravenous therapy pipeline design method, including the following steps:

[0006] Obtain patient information corresponding to the patient to be intravenously treated, and determine a set of trunks to be intravenously treated based on the patient information, where the set of trunks to be intravenously treated includes at least one intravenous therapy trunk region;

[0007] Obtain the historical intravenous therapy records of the patient to be intravenously treated within a preset time before the current time point, and perform region selection based on the historical intravenous therapy records and the set of trunks to be intravenously treated to obtain the target trunk region corresponding to the patient to be intravenously treated;

[0008] Obtain a corresponding regional image based on the target trunk region corresponding to the patient to be intravenously treated, and create a virtual trunk model having a mapping relationship with the target trunk region based on the regional image;

[0009] Obtain each virtual blood vessel located in the virtual trunk model, and determine the virtual blood vessels that meet the medical ecological requirements as the blood vessels to be injected based on the ecological attributes of each virtual blood vessel;

[0010] Obtain the intravenous therapy drug information corresponding to the patient to be intravenously treated, and generate a drug pipeline model connected to the blood vessels to be injected based on the intravenous therapy drug information.

[0011] Optionally, in the method according to the present invention, determining the set of trunks to be intravenously treated based on the patient information includes:

[0012] Based on the patient information, determine the patient age and the patient's medical record corresponding to the patient to be statically treated, and retrieve a preset static treatment trunk distribution table, wherein the preset static treatment trunk distribution table includes a plurality of static treatment trunk regions, and each static treatment trunk region includes a corresponding age-appropriate interval and a disease-exclusion interval;

[0013] Based on the preset static treatment trunk distribution table, perform a first screening to determine all the static treatment trunk regions corresponding to the age-appropriate intervals including the patient age, and determine them as the first screening group;

[0014] Based on the first screening group, perform a second screening, and determine the set of trunks to be statically treated based on the result of the second screening.

[0015] Optionally, in the method according to the present invention, determining the set of trunks to be statically treated based on the result of the second screening includes:

[0016] When at least one of the static treatment trunk regions corresponding to the disease-exclusion intervals not including the patient's medical record exists in the first screening group as the result of the second screening, determine the at least one static treatment trunk region as the set of static treatment trunks;

[0017] When no static treatment trunk region corresponding to the disease-exclusion intervals not including the patient's medical record exists in the first screening group as the result of the second screening, determine the first screening group as the set of static treatment trunks.

[0018] Optionally, in the method according to the present invention, performing region selection based on the historical static treatment record and the set of static treatment trunks to obtain the target trunk region corresponding to the patient to be statically treated includes:

[0019] Based on the historical static treatment record, determine the historical trunk region, and compare the historical trunk region with all the static treatment trunk regions in the set of static treatment trunks;

[0020] When there is one static treatment trunk region in the set of static treatment trunks and this static treatment trunk region is the same as the historical trunk region, determine this static treatment trunk region as the target trunk region corresponding to the patient to be statically treated;

[0021] When there is one static treatment trunk region in the set of static treatment trunks and it is not the same as the historical trunk region, determine the trunk connection relationship between the static treatment trunk region and the historical trunk region, and when the trunk connection relationship is upper connection, determine the static treatment trunk region as the target trunk region corresponding to the patient to be statically treated, and when the trunk connection relationship is lower connection, determine the historical trunk region as the target trunk region corresponding to the patient to be statically treated;

[0022] Or

[0023] When multiple static therapy torso regions are included in the static therapy torso set, determine the torso connection relationships between the multiple static therapy torso regions and the historical torso region respectively;

[0024] When the torso connection relationship between any static therapy torso region and the historical torso region is an upper connection, determine this static therapy torso region as the target torso region corresponding to the patient to be statically treated;

[0025] When the torso connection relationships between all static therapy torso regions and the historical torso region are lower connections, determine the historical torso region as the target torso region corresponding to the patient to be statically treated.

[0026] Optionally, in the method according to the present invention, obtaining each virtual blood vessel located in the virtual torso model and determining the virtual blood vessels meeting the medical ecological requirements as the blood vessels to be injected based on the ecological attributes of each virtual blood vessel includes:

[0027] Determine each virtual blood vessel located in the virtual torso model based on a preset image recognition model, and establish a model coordinate system corresponding to the virtual torso model;

[0028] Obtain each contour pixel point that respectively forms the blood vessel contour of each virtual blood vessel, and obtain each contour coordinate point corresponding to each of the contour pixel points based on the model coordinate points, and determine the contour coordinate points corresponding to the same blood vessel contour as a contour coordinate set;

[0029] Perform difference calculations on the contour coordinate points with the same abscissa in each contour coordinate set respectively to obtain each horizontal difference, and determine the horizontal difference corresponding to the maximum value as the first target difference;

[0030] Perform difference calculations on the contour coordinate points with the same ordinate in each contour coordinate set respectively to obtain each vertical difference, and determine the vertical difference corresponding to the maximum value as the second target difference;

[0031] Determine the larger one of the first target difference and the second target difference as the contour size corresponding to the blood vessel contour, obtain each contour size corresponding to each blood vessel contour respectively, and determine each contour size as the size attribute corresponding to each virtual blood vessel;

[0032] Obtain each contour pixel value corresponding to each contour pixel point respectively located in each contour coordinate set;

[0033] Calculate the mean value of the contour pixel values corresponding to the same set of contour coordinates, obtain the pixel means corresponding to the respective blood vessel contours, and determine the pixel means as the pixel attributes corresponding to the respective virtual blood vessels;

[0034] Based on a preset determination strategy, comprehensively evaluate the size attribute and pixel attribute corresponding to the same virtual blood vessel, obtain the respective attribute evaluation values corresponding to the respective virtual blood vessels, and determine the virtual blood vessel with the largest corresponding numerical attribute evaluation value as the blood vessel to be injected.

[0035] Optionally, in the method according to the present invention, comprehensively evaluating the size attribute and pixel attribute corresponding to the same virtual blood vessel based on a preset determination strategy to obtain the respective attribute evaluation values corresponding to the respective virtual blood vessels includes:

[0036] Retrieve the preset size evaluation weight and the preset pixel evaluation weight;

[0037] Perform a multiplication calculation on the contour size and pixel mean corresponding to the same virtual blood vessel with the preset size evaluation weight and the preset pixel evaluation weight respectively to obtain a first product value and a second product value;

[0038] Perform a normalization process on the first product value and the second product value to obtain a first evaluation value and a second evaluation value, and perform a summation calculation on the first evaluation value and the second evaluation value to obtain the respective attribute evaluation values corresponding to the respective virtual blood vessels.

[0039] Optionally, in the method according to the present invention, obtaining the intravenous drug information corresponding to the patient to be intravenously treated and generating a drug pipeline model connected to the blood vessel to be injected based on the intravenous drug information includes:

[0040] Determine the patient drug number based on the patient information corresponding to the patient to be intravenously treated, and retrieve the drug record table, where the drug record table includes multiple columns of drug sequences, and each column of drug sequences respectively includes different stored drug numbers and the stored drug information corresponding to the different stored drug numbers;

[0041] Traverse the drug record table to find the stored drug number that is the same as the patient drug number,

[0042] Delete the drug sequence including the intravenous drug information from the drug record table, and determine the stored drug information corresponding to the stored drug number as the intravenous drug information corresponding to the patient drug number;

[0043] Determine each injection drug corresponding to different drug types based on the intravenous drug information, and determine the characteristics of each injection drug according to the injection sequence determination strategy and the mixed injection determination strategy. Determine the sequence characteristics and connection characteristics corresponding to each injection drug respectively, and generate a drug pipeline model connected to the blood vessel to be injected based on the sequence characteristics and connection characteristics.

[0044] Optionally, in the method according to the present invention, determining the characteristics of each injection drug according to the injection sequence determination strategy and determining the sequence characteristics corresponding to each injection drug respectively includes:

[0045] Determine the components of each injection drug respectively to obtain an initial chemical component group corresponding to each injection drug. Among them, the initial chemical component group includes different component types and the type ratios corresponding to different component types respectively;

[0046] Retrieve a preset ratio threshold, and screen out the component types with corresponding type ratios less than or equal to the preset ratio threshold from each of the initial chemical component groups to obtain each current chemical component group;

[0047] Determine the content of each component type located in each current chemical component group respectively, and determine the content coefficient based on each component content respectively;

[0048] Determine the type coefficient of each component type located in each current chemical component group respectively based on a preset chemical injection sorting table, and sum the content coefficient and the type coefficient corresponding to the same component type to obtain the type priority value corresponding to this component type;

[0049] Sum up all the type priority values corresponding to the same current chemical component group to obtain the injection priority value corresponding to each current chemical component group respectively, sort each current chemical component group based on the injection priority value, and determine the injection sequences corresponding to each injection drug obtained as the sequence characteristics.

[0050] Optionally, in the method according to the present invention, determining the characteristics of each injection drug according to the mixed injection determination strategy and determining the connection characteristics corresponding to each injection drug respectively includes:

[0051] Retrieve a mixing record table, where the mixing record table includes multiple columns of mixing sequences, and each column of mixing sequences includes at least two different injection drugs that can be mixed for injection;

[0052] Traverse the mixing record table, and determine whether there are at least two of them located in the same mixing sequence among each injection drug, and determine all the injection drugs located in the same mixing sequence as a mixing group with mixing characteristics and the remaining all injection drugs as having independent characteristics.

[0053] Optionally, in the method according to the present invention, generating a drug pipeline model connected to the blood vessel to be injected based on the sequential characteristics and connection characteristics includes:

[0054] Determine the injection sequence corresponding to each injection drug located in the same mixing group, and determine the injection sequence with the largest value as the one corresponding to the mixing group;

[0055] Based on all injection drugs in each mixing group, generate a parallel pipeline model presenting a parallel structure, and based on all the remaining injection drugs except the mixing group, generate an independent pipeline model presenting an independent structure;

[0056] Based on the injection sequence corresponding to the mixing group and the injection sequences corresponding to all the remaining injection drugs except the mixing group, connect the parallel pipeline model and the independent pipeline model in sequence to obtain a drug pipeline model connected to the blood vessel to be injected and presenting a series structure.

[0057] Optionally, in the method according to the present invention, generating a drug pipeline model connected to the blood vessel to be injected based on the sequential characteristics and connection characteristics includes:

[0058] Extract the sequential characteristics and connection characteristics and compare them with the trained training feature groups respectively. If the features in the training feature groups correspond one by one to the sequential characteristics and connection characteristics, then retrieve the trained drug pipeline model and drug injection cycle corresponding to the training feature groups;

[0059] Traverse the trained drug pipeline model based on the position of each injection drug and add information to obtain a drug pipeline model;

[0060] If the features in the training feature groups do not completely correspond to the sequential characteristics and connection characteristics, then compare the features in the training feature groups with the sequential characteristics and connection characteristics to obtain a feature similarity;

[0061] If it is determined that there is a feature similarity greater than the selection threshold, then use the training feature group with the largest feature similarity as the feature group to be decomposed, and decompose and reassemble the feature group to be decomposed based on the sequential characteristics and connection characteristics to obtain a drug pipeline model;

[0062] If it is determined that there is no feature similarity greater than the selection threshold, then generate parallel pipeline models and independent pipeline models corresponding to all drugs respectively.

[0063] According to another aspect of the present invention, there is provided an artificial intelligence-based intravenous therapy pipeline design system, including:

[0064] An acquisition module, configured to acquire patient information corresponding to the patient to be statically treated, and determine a set of static treatment trunks based on the patient information, where the set of static treatment trunks includes at least one static treatment trunk area;

[0065] A region selection module, configured to acquire historical static treatment records of the patient to be statically treated within a preset time before the current time point, and perform region selection based on the historical static treatment records and the set of static treatment trunks to obtain a target trunk region corresponding to the patient to be statically treated;

[0066] A creation module, configured to acquire a corresponding region image based on the target trunk region corresponding to the patient to be statically treated, and create a virtual trunk model having a mapping relationship with the target trunk region based on the region image;

[0067] A determination module, configured to acquire each virtual blood vessel located in the virtual trunk model, and determine the virtual blood vessels that meet the medical ecological requirements as the blood vessels to be injected based on the ecological attributes of each virtual blood vessel;

[0068] A model generation module, configured to acquire static treatment drug information corresponding to the patient to be statically treated, and generate a drug pipeline model connected to the blood vessels to be injected based on the static treatment drug information.

[0069] According to the solution of the present invention, the server first acquires the patient information of the patient to be statically treated, and determines a set of static treatment trunks based on the patient information. The set of static treatment trunks includes at least one static treatment trunk area; then, the server acquires the historical static treatment records of the patient to be statically treated within a preset time before the current time point, and performs region selection based on the historical static treatment records and the set of static treatment trunks, so as to obtain the target trunk region corresponding to the patient to be statically treated; then, the server acquires the corresponding region image based on the target trunk region corresponding to the patient to be statically treated, and creates a virtual trunk model having a mapping relationship with the target trunk region based on the region image; then, the server acquires each virtual blood vessel located in the virtual trunk model, and determines the virtual blood vessels that meet the medical ecological requirements as the blood vessels to be injected based on the ecological attributes of each virtual blood vessel; finally, the server acquires the static treatment drug information corresponding to the patient to be statically treated, and generates a drug pipeline model connected to the blood vessels to be injected based on the static treatment drug information. The present invention can largely provide reference for medical staff on the injection area and injection drugs during intravenous injection, can reduce their working time to a certain extent, and thus improve the efficiency of intravenous injection. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 Shows a flowchart of a method for designing an intelligent static treatment pipeline based on artificial intelligence according to an embodiment of the present invention;

[0071] Figure 2 The figure shows a schematic diagram of the drug pipeline model in this embodiment;

[0072] Figure 3 The figure shows a schematic diagram of the structure of the terminal interconnection in this embodiment;

[0073] Figure 4 The figure shows a block diagram of the structure of an artificial intelligence-based intravenous therapy pipeline design system according to another embodiment of the present invention. Detailed implementation manners

[0074] Hereinafter, exemplary embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be completely conveyed to those skilled in the art.

[0075] Intravenous injection is a medical method, that is, injecting liquid substances such as blood, medicine solution, and nutrient solution directly into a vein. Intravenous injection can be divided into transient and continuous. Transient intravenous injection is mostly directly injected into the vein with a syringe, that is, the generally common injection method; continuous intravenous injection is implemented by intravenous drip, commonly known as "drip".

[0076] The inventors found in the research that before performing an intravenous injection on a patient, it is generally necessary for medical staff to personally check the relevant medical records of the patient to determine the specific location of the current intravenous injection, which will waste a certain amount of time.

[0077] To solve the problems existing in the above-mentioned prior art, the inventors proposed the solution of the present invention. An embodiment of the present invention provides an artificial intelligence-based intravenous therapy pipeline design method, which can be executed in a computing device.

[0078] Figure 1 The figure shows a flowchart of an artificial intelligence-based intravenous therapy pipeline design method according to an embodiment of the present invention, and this method is suitable for execution in a computing device (wherein, the computing device can be understood as a terminal device with processing functions such as a computer and a mobile phone).

[0079] As Figure 1 shown, the purpose of this embodiment is to implement an artificial intelligence-based intravenous therapy pipeline design method, starting from step S102, and in step S102, it includes the following content:

[0080] The medical end acquires the patient information corresponding to the patient to be intravenously treated and sends it to the server, and the server determines the set of trunks to be intravenously treated based on the patient information, wherein the set of trunks to be intravenously treated includes at least one trunk region to be intravenously treated.

[0081] For example, it can be illustrated that in an actual scenario, due to the different ages and diseases of patients awaiting intravenous therapy, medical staff may use age and disease as references to select appropriate areas for intravenous injection. For example, the elderly may be more suitable for intravenous injection in the lower limbs, while the youth may choose the upper and lower limbs for intravenous injection. Therefore, in order to better fit the actual scenario and ensure that the proposed solution in this embodiment has certain medical reference value, in this embodiment, when it is necessary to perform intravenous injection on a patient awaiting intravenous therapy, the server will first obtain the patient information corresponding to the patient awaiting intravenous therapy. The patient information may include the patient's age, medical records, etc. Furthermore, the server can preliminarily determine all the areas where the patient awaiting intravenous therapy may be suitable for intravenous injection, that is, the intravenous therapy trunk areas, and determine them as the corresponding set of trunks awaiting intravenous therapy.

[0082] Further, the above "determining the set of trunks awaiting intravenous therapy based on the patient information" further includes the following steps:

[0083] Based on the patient information, determine the patient's age and medical records corresponding to the patient awaiting intravenous therapy, and retrieve the preset intravenous therapy trunk distribution table. The preset intravenous therapy trunk distribution table includes multiple intravenous therapy trunk areas, and each intravenous therapy trunk area includes a corresponding age-appropriate interval and disease-excluded interval.

[0084] Perform a first screening based on the preset intravenous therapy trunk distribution table, determine all the intravenous therapy trunk areas corresponding to the age-appropriate intervals containing the patient's age, and determine them as the first screening group.

[0085] Perform a second screening based on the first screening group, and determine the set of trunks awaiting intravenous therapy based on the result of the second screening.

[0086] For example, in this embodiment, after the server obtains the patient information of the patient awaiting intravenous therapy, it can determine the patient's age and medical records of the patient awaiting intravenous therapy. The medical records include the current and previous medical record contents. At the same time, a preset intravenous therapy trunk distribution table is pre-set in the server. There are multiple intravenous therapy trunk areas in the preset intravenous therapy trunk distribution table, such as scalp, upper limbs, lower limbs and other intravenous therapy trunk areas. And each intravenous therapy trunk area includes a corresponding age-appropriate interval and disease-excluded interval. It should be noted that the age-appropriate interval can be understood as the age interval suitable for injection in this area, and the disease-excluded interval can be understood as the disease interval not suitable for injection in this area. For example, the age-appropriate interval for the lower limbs as an intravenous therapy trunk area is for the elderly over 70 years old, and the disease-excluded interval for the scalp as an intravenous therapy trunk area is for acute fever diseases, etc.

[0087] Next, the server retrieves the preset intravenous therapy trunk distribution table and conducts a first screening based on the preset intravenous therapy trunk distribution table. The specific steps of the first screening are as follows: The server first determines all the intravenous therapy trunk regions within the age-appropriate range that includes the patient's age, and then designates these intravenous therapy trunk regions as the first screening group. Next, the server conducts a second screening on the intravenous therapy trunk regions in the first screening group according to the disease exclusion range, and determines the set of trunk regions to be intravenously treated based on the results of the second screening. This embodiment can screen and determine the set of trunk regions to be intravenously treated according to the age-appropriate range and disease exclusion range of each intravenous therapy trunk region, which is not only rapid but also relatively objective, and can provide certain reference for medical staff.

[0088] Furthermore, the above-mentioned "determining the set of trunk regions to be intravenously treated based on the results of the second screening" further includes the following steps:

[0089] When the result of the second screening is that there is at least one intravenous therapy trunk region in the first screening group corresponding to the disease exclusion range that does not include the patient's medical record, the at least one intravenous therapy trunk region is determined as the set of trunk regions for intravenous therapy;

[0090] When the result of the second screening is that there is no intravenous therapy trunk region in the first screening group corresponding to the disease exclusion range that does not include the patient's medical record, the first screening group is determined as the set of trunk regions for intravenous therapy.

[0091] For example, in this embodiment, when the result of the server's second screening of the intravenous therapy trunk regions in the first screening group is that there is one or more intravenous therapy trunk regions in the first screening group that do not include the disease exclusion range of the patient's medical record, it means that these intravenous therapy trunk regions are suitable for the patient to be intravenously treated. Therefore, the server will determine this one or more intravenous therapy trunk regions as the set of trunk regions for intravenous therapy; when the result of the second screening of the intravenous therapy trunk regions in the first screening group is that there is no intravenous therapy trunk region in the first screening group corresponding to the disease exclusion range that does not include the patient's medical record, it means that all the intravenous therapy trunk regions in the first screening group are suitable for the patient to be intravenously treated. Therefore, the server will determine the first screening group as the set of trunk regions for intravenous therapy.

[0092] In step S104, the following content is included:

[0093] The server obtains the historical intravenous therapy records of the patient to be intravenously treated within a preset time before the current time point, and based on the historical intravenous therapy records and the set of trunk regions for intravenous therapy, conducts region selection to obtain the target trunk region corresponding to the patient to be intravenously treated.

[0094] For example, in this embodiment, when the patient to receive intravenous therapy is about to start intravenous injection, that is, at the current time point, the server will obtain the historical intravenous therapy records of the patient to receive intravenous therapy within a preset time before the current time point. The historical intravenous therapy records include the specific areas where the patient to receive intravenous therapy has received intravenous injection before. The preset time can be set by medical staff according to the actual situation. For example, it can be six months. The server can select the intravenous therapy trunk areas in the intravenous therapy trunk set according to the historical intravenous therapy records, and determine the area suitable for intravenous injection during the current intravenous therapy for the patient to receive intravenous therapy, that is, the target trunk area.

[0095] Further, the above "performing area selection based on the historical intravenous therapy record and the intravenous therapy trunk set to obtain the target trunk area corresponding to the patient to receive intravenous therapy" further includes the following steps:

[0096] Determine the historical trunk area based on the historical intravenous therapy record, and compare the historical trunk area with all the intravenous therapy trunk areas in the intravenous therapy trunk set;

[0097] When there is one intravenous therapy trunk area in the intravenous therapy trunk set and this intravenous therapy trunk area is the same as the historical trunk area, determine this intravenous therapy trunk area as the target trunk area corresponding to the patient to receive intravenous therapy;

[0098] When there is one intravenous therapy trunk area in the intravenous therapy trunk set and it is not the same as the historical trunk area, determine the trunk connection relationship between the intravenous therapy trunk area and the historical trunk area, and when the trunk connection relationship is upper connection, determine the intravenous therapy trunk area as the target trunk area corresponding to the patient to receive intravenous therapy, and when the trunk connection relationship is lower connection, determine the historical trunk area as the target trunk area corresponding to the patient to receive intravenous therapy;

[0099] Or

[0100] When there are multiple intravenous therapy trunk areas in the intravenous therapy trunk set, determine the trunk connection relationships between the multiple intravenous therapy trunk areas and the historical trunk area respectively;

[0101] When the trunk connection relationship between any intravenous therapy trunk area and the historical trunk area is upper connection, determine this intravenous therapy trunk area as the target trunk area corresponding to the patient to receive intravenous therapy;

[0102] When the trunk connection relationships between all the intravenous therapy trunk areas and the historical trunk area are lower connection, determine the historical trunk area as the target trunk area corresponding to the patient to receive intravenous therapy.

[0103] For example, in this embodiment, after the server obtains the historical intravenous therapy records of the patient to receive intravenous therapy within a preset time before the current time point, it will determine the specific area where the patient previously received intravenous injection based on the historical intravenous therapy records, that is, the historical torso area. Since during intravenous injection, the area for this intravenous injection needs to be above the historical torso area to prevent inflammation or leakage. For example, if the historical torso area is the back of the hand, the area for this intravenous injection can be the area above the back of the hand such as the arm. Therefore, the server will compare the historical torso area with all the intravenous therapy torso areas in the intravenous therapy torso set;

[0104] When there is one intravenous therapy torso area in the intravenous therapy torso set and this intravenous therapy torso area is the same as the historical torso area, it means that the intravenous therapy torso area coincides with the historical torso area. At this time, the server will determine this intravenous therapy torso area as the target torso area of the patient to receive intravenous therapy; when there is one intravenous therapy torso area in the intravenous therapy torso set and it is not the same as the historical torso area, the server will further determine the torso connection relationship between this intravenous therapy torso area and the historical torso area. When the intravenous therapy torso area is above the historical torso area, the torso connection relationship is upper connection; when the intravenous therapy torso area is below the historical torso area, the torso connection relationship is lower connection. Furthermore, the server will determine this intravenous therapy torso area as the target torso area of the patient to receive intravenous therapy when the torso connection relationship is upper connection, and determine the historical torso area as the target torso area of the patient to receive intravenous therapy when the torso connection relationship is lower connection;

[0105] Or

[0106] When there are multiple intravenous therapy torso areas in the intravenous therapy torso set, the server will determine the torso connection relationships between the multiple intravenous therapy torso areas and the historical torso area respectively. When the torso connection relationship between any one intravenous therapy torso area and the historical torso area is upper connection, the server will determine this intravenous therapy torso area as the target torso area of the patient to receive intravenous therapy; when the torso connection relationships between all the intravenous therapy torso areas and the historical torso area are lower connections, the server will determine the historical torso area as the target torso area corresponding to the patient to receive intravenous therapy.

[0107] It can be explained that this embodiment can compare the historical torso area with all the intravenous therapy torso areas in the intravenous therapy torso set, so as to determine the torso connection relationship between the intravenous therapy torso area and the historical torso area. When the torso connection relationships between all the intravenous therapy torso areas in the intravenous therapy torso set and the historical torso area are lower connections, the server will determine the historical torso area as the target torso area corresponding to the patient to receive intravenous therapy to avoid inflammation or leakage during the intravenous injection of the patient to receive intravenous therapy, and can largely provide a reference for the medical staff on the injection area during intravenous injection.

[0108] In step S106, the following is included:

[0109] The server obtains a corresponding regional image based on the target torso area corresponding to the patient to be statically treated, and creates a virtual torso model having a mapping relationship with the target torso area based on the regional image.

[0110] For example, in this embodiment, when the server completes the determination of the target torso area of the patient to be statically treated, the server will obtain a regional image corresponding to the target torso area, and then create a virtual torso model having a mapping relationship with the target torso area based on the obtained regional image; here, the establishment of the virtual torso model can be realized based on digital twin technology, that is, corresponding sensing data is collected based on the target torso area, and a digital twin space corresponding to the target torso area, that is, a virtual torso model, is generated based on the collected sensing data.

[0111] In step S108, the following is included:

[0112] The server obtains each virtual blood vessel located in the virtual torso model, and determines the virtual blood vessels that meet the medical ecological requirements as the blood vessels to be injected based on the ecological attributes of each virtual blood vessel.

[0113] For example, in this embodiment, after the server creates a virtual torso model having a mapping relationship with the target torso area, it can obtain each virtual blood vessel in the virtual torso model, and determine the virtual blood vessels that meet the medical ecological requirements as the blood vessels to be injected according to the ecological attributes of each virtual blood vessel.

[0114] Further, the above "obtaining each virtual blood vessel located in the virtual torso model, and determining the virtual blood vessels that meet the medical ecological requirements as the blood vessels to be injected" further includes the following steps:

[0115] Determine each virtual blood vessel located in the virtual torso model based on a preset image recognition model, and establish a model coordinate system corresponding to the virtual torso model;

[0116] Obtain each contour pixel point that composes the blood vessel contour of each virtual blood vessel, and obtain each contour coordinate point corresponding to each of the contour pixel points based on the model coordinate points, and determine the contour coordinate points corresponding to the same blood vessel contour as a contour coordinate set;

[0117] Perform difference calculations on the contour coordinate points with the same abscissa in each contour coordinate set respectively to obtain each horizontal difference, and determine the horizontal difference corresponding to the maximum value as the first target difference;

[0118] In each set of contour coordinates, perform difference calculations on the contour coordinate points with the same ordinate respectively to obtain respective longitudinal differences, and determine the longitudinal difference with the corresponding maximum value as the second target difference;

[0119] Determine the larger one of the first target difference and the second target difference as the contour size corresponding to the blood vessel contour, obtain respective contour sizes corresponding to each blood vessel contour, and determine each contour size as the size attribute corresponding to each virtual blood vessel;

[0120] Obtain respective contour pixel values corresponding to each contour pixel point located in each set of contour coordinates;

[0121] Perform mean value calculations on the contour pixel values corresponding to the same set of contour coordinates to obtain respective pixel means corresponding to each blood vessel contour, and determine each pixel mean as the pixel attribute corresponding to each virtual blood vessel;

[0122] Based on a preset determination strategy, perform comprehensive attribute evaluation on the size attribute and pixel attribute corresponding to the same virtual blood vessel to obtain respective attribute evaluation values corresponding to each virtual blood vessel, and determine the virtual blood vessel with the largest corresponding numerical attribute evaluation value as the blood vessel to be injected.

[0123] For example, in this embodiment, after the server creates a virtual torso model having a mapping relationship with the target torso area, it will determine each virtual blood vessel in the virtual torso model based on a preset image recognition model. The preset image recognition model can be a neural network recognition method. Then, the server will establish a coordinate system corresponding to the virtual torso model, that is, a model coordinate system. Next, the server will obtain all the pixel points of the blood vessel contours constituting each virtual blood vessel, that is, each contour pixel point, and obtain the coordinate points corresponding to each contour pixel point based on the model coordinate points, that is, each contour coordinate point. Furthermore, the contour coordinate points corresponding to the same blood vessel contour are determined as a set of contour coordinates.

[0124] The server will perform difference calculations on the contour coordinate points with the same abscissa in each set of contour coordinates respectively to obtain the respective horizontal differences, and determine the horizontal difference corresponding to the maximum value as the first target difference. For example, if the contour coordinate points with the same abscissa are (1, 2), (1, 3), and (1, 7) respectively, the calculation methods of the respective horizontal differences are 3 - 2 = 1, 7 - 2 = 5, and 7 - 3 = 4, so the horizontal difference with the maximum value of 5 is determined as the first target difference; then, the server will perform difference calculations on the contour coordinate points with the same ordinate in each set of contour coordinates respectively to obtain the respective vertical differences, and determine the vertical difference corresponding to the maximum value as the second target difference. At this time, the server will determine the larger one of the first target difference and the second target difference as the contour size of the blood vessel contour, so as to obtain the contour sizes of each blood vessel contour, and determine each contour size as the size attribute of each virtual blood vessel.

[0125] Since the distances of each blood vessel from the epithelial region are different, the colors of each blood vessel are different. Therefore, after the server determines the size attributes of each virtual blood vessel, it will obtain the pixel values corresponding to each contour pixel point in each set of contour coordinates, that is, each contour pixel value, and then perform a mean calculation on the contour pixel values in the same set of contour coordinates, so as to obtain the pixel means of each blood vessel contour, and determine each pixel mean as the pixel attribute of each virtual blood vessel.

[0126] Since thicker and closer blood vessels to the epithelial region are more suitable for intravenous injection, the server will perform a comprehensive attribute evaluation on the size attribute and pixel attribute corresponding to the same virtual blood vessel based on a preset determination strategy, so as to obtain the respective attribute evaluation values corresponding to each virtual blood vessel, and determine the virtual blood vessel with the largest corresponding attribute evaluation value as the blood vessel to be injected.

[0127] This embodiment can roughly obtain the contour sizes of each blood vessel contour based on the model coordinate system, which is a way to quickly obtain the sizes representing each virtual blood vessel, and can more objectively evaluate the comprehensive attributes of virtual blood vessels, saving a certain amount of working time.

[0128] Furthermore, the above "performing a comprehensive attribute evaluation on the size attribute and pixel attribute corresponding to the same virtual blood vessel based on a preset determination strategy to obtain the respective attribute evaluation values corresponding to each virtual blood vessel" further includes the following steps:

[0129] Retrieve the preset size evaluation weight and preset pixel evaluation weight;

[0130] Perform product calculations on the contour size and pixel mean corresponding to the same virtual blood vessel with the preset size evaluation weight and preset pixel evaluation weight respectively to obtain a first product value and a second product value;

[0131] Normalize the first product value and the second product value to obtain a first evaluation value and a second evaluation value, and perform a summation calculation on the first evaluation value and the second evaluation value to obtain the attribute evaluation values corresponding to each virtual blood vessel.

[0132] For example, in this embodiment, a preset size evaluation weight and a preset pixel evaluation weight are pre-set in the server. After the server determines the size attribute and pixel attribute of each virtual blood vessel, it will retrieve the preset size evaluation weight and the preset pixel evaluation weight, and then perform a product calculation on the contour size and pixel mean value corresponding to the same virtual blood vessel with the preset size evaluation weight and the preset pixel evaluation weight respectively to obtain a first product value and a second product value. Since the units of the contour size and the pixel mean value are different, the server will divide the first product value and the second product value by the preset planning value respectively, so as to normalize the first product value and the second product value to obtain a first evaluation value and a second evaluation value, and perform a summation calculation on the first evaluation value and the second evaluation value to obtain the attribute evaluation values of each virtual blood vessel.

[0133] In step S110, the following content is included:

[0134] The server obtains the intravenous therapy drug information corresponding to the patient to be intravenously treated, and generates a drug pipeline model connected to the blood vessel to be injected based on the intravenous therapy drug information and feeds it back to the medical end.

[0135] For example, in this embodiment, after the server determines the blood vessel to be injected, it will obtain the intravenous therapy drug information of the patient to be intravenously treated, and generate a drug pipeline model connected to the blood vessel to be injected according to the intravenous therapy drug information, so as to complete the corresponding intravenous pipeline design for the reference of medical staff, improving the corresponding user experience and medical assistance.

[0136] Further, in this embodiment, the above "obtain the intravenous therapy drug information corresponding to the patient to be intravenously treated, and generate a drug pipeline model connected to the blood vessel to be injected based on the intravenous therapy drug information" may further include the following steps:

[0137] Determine the patient drug number based on the patient information corresponding to the patient to be intravenously treated, and retrieve the drug record table, where the drug record table includes multiple columns of drug sequences, and each column of drug sequence includes different stored drug numbers and the stored drug information corresponding to different stored drug numbers respectively;

[0138] Traverse the drug record table to find the stored drug number that is the same as the patient drug number,

[0139] Delete the drug sequence including the intravenous drug information from the drug record form, and determine the stored drug information corresponding to the stored drug number as the intravenous drug information corresponding to the patient drug number;

[0140] Based on the intravenous drug information, determine each injection drug corresponding to different drug types, and determine the characteristics of each injection drug according to the injection order determination strategy and the mixed injection determination strategy, determine the sequence characteristics and connection characteristics corresponding to each injection drug respectively, and generate a drug pipeline model connected to the blood vessel to be injected based on the sequence characteristics and the connection characteristics.

[0141] For example, in this embodiment, since the intravenous drugs required for each patient to be intravenously treated may be different, therefore, when creating the corresponding drug pipeline model, the corresponding patient drug number can be determined first based on the patient information of the patient to be intravenously treated. It should be noted that each patient to be intravenously treated has a corresponding patient drug number, and the patient drug number is unique and sole; after obtaining the corresponding patient drug number, the drug record form pre-stored in the server can be retrieved. Here, the drug record form includes multiple columns of drug sequences arranged in a columnar manner, and each column of drug sequence includes a stored drug number and the stored drug information corresponding to the stored drug number. By checking the drug record form to find the stored drug number with the same number content as the patient drug number, and further determine the corresponding stored drug information as the intravenous drug information corresponding to the patient to be intravenously treated; at the same time, since the corresponding drug information in the drug record form has been retrieved, therefore, in order to reduce the corresponding data storage amount, the corresponding drug sequence can be deleted from the drug record form, thereby improving the corresponding data processing amount;

[0142] Finally, after obtaining the corresponding intravenous drug information, the respective injection drugs including different drug types can be determined based on the intravenous drug information. Since the components of different drugs are different during the process of intravenous injection, it is necessary to determine the corresponding injection order. For example, the injection order can be determined according to the following rules: 1. Salt first and then sugar. Supplementing salt first and then sugar helps to timely supplement the components missing in the patient's body and adjust the balance of the patient's experience; 2. Crystals first and then colloids. Crystals refer to the liquids usually infused in the human body, such as normal saline, while colloids refer to artificial or natural colloids, such as plasma albumin. Injecting crystals first can play a role in supplementing volume. In addition, during the process of fluid replacement, proportional fluid replacement should be carried out according to the lost fluid components. After determining the order characteristics of the respective injection drugs based on the above rules, it is also necessary to determine the connection characteristics of the respective injection drugs, that is, it is necessary to determine whether there is a possibility of mixed injection between the respective injection drugs. If mixed injection is possible, the connection characteristics of the corresponding injection drugs can be determined as mixed characteristics. If mixed injection is not possible, the connection characteristics of the corresponding injection items can be determined as independent characteristics. After completing the determination of the order characteristics and connection characteristics of each injection drug, a corresponding drug pipeline model is generated.

[0143] Further, in this embodiment, the above "determining the characteristics of each injection drug according to the injection order determination strategy and determining the order characteristics corresponding to each injection drug respectively" may further include the following steps:

[0144] Determine the components of each injection drug respectively to obtain the initial chemical component groups corresponding to each injection drug, where the initial chemical component groups include different component types and the type ratios corresponding to different component types respectively;

[0145] Retrieve the preset ratio threshold, and screen out the component types with the corresponding type ratios less than or equal to the preset ratio threshold from each of the initial chemical component groups to obtain each current chemical component group;

[0146] Determine the respective component contents corresponding to the respective component types in each current chemical component group, and determine the content coefficients based on the respective component contents;

[0147] Determine the type coefficients of the respective component types in each current chemical component group based on the preset chemical injection sorting table, and sum up the content coefficients and type coefficients corresponding to the same component type to obtain the type priority value corresponding to the component type;

[0148] Sum up all the species priority values corresponding to the same current chemical composition group to obtain the injection priority values corresponding to each current chemical composition group respectively. Then, sort each current chemical composition group based on the injection priority values, and determine the injection order corresponding to each injection drug obtained as the order characteristic.

[0149] For example, in this embodiment, it should be noted that from the above content, it can be seen that the injection order of each injection drug is mainly based on the difference in the chemical composition of each injection drug.

[0150] Therefore, based on this, the corresponding component determination can be performed on each injection drug first to obtain the initial chemical composition groups corresponding to each injection drug respectively. Among them, the initial chemical composition group includes different component types and the type ratios corresponding to different component types respectively.

[0151] Next, since the chemical composition in each injection drug may be relatively complex, and the chemical components with relatively small proportions may also have relatively small impacts on the injection order. Therefore, in order to reduce the corresponding data calculation amount, the chemical components with relatively small proportions can be removed. The removal method is to compare the component content of each chemical component with a preset ratio threshold, so as to screen out the component types with component contents less than or equal to the preset threshold ratio from the initial chemical composition group to obtain the corresponding current chemical composition group.

[0152] Next, the content of each component type in the current chemical composition group can be determined to obtain the corresponding component contents, and the corresponding content coefficients can be determined based on the component contents. It should be noted that the determination of the type coefficient can be set based on the principle that the larger the component content, the higher the coefficient value. For example, the corresponding coefficient values are set according to different content intervals, and the corresponding content coefficient is determined by comparing the component content with the content interval. The higher the content coefficient, the higher the corresponding component content, so as to obtain the content coefficients corresponding to all component types in the same current chemical composition group respectively.

[0153] Then, after the determination of the content coefficients is completed, the corresponding preset chemical injection sorting table can be retrieved to determine the type coefficients of each component type in each current chemical composition group. It should be noted that the preset chemical injection sorting table lists the type coefficients corresponding to different component types, that is, the type coefficients of each component type in each current chemical composition group can be found by checking the preset chemical injection sorting table.

[0154] Then, after obtaining the content coefficient and type coefficient corresponding to the same component type, through summation calculation, the species priority value corresponding to this component type can be obtained, and by summing up all the species priority values corresponding to the same current chemical composition group again, the injection priority values corresponding to each current chemical composition group can be obtained.

[0155] Finally, sort in descending order according to the injection priority value of each current chemical component group, so as to obtain the injection order corresponding to each injection drug, and the corresponding injection order can be determined as the order characteristic to be obtained.

[0156] Furthermore, in this embodiment, the above "determine the characteristics of each injection drug according to the mixed injection determination strategy and determine the connection characteristics corresponding to each injection drug" may further include the following steps:

[0157] Retrieve the mixing record table, where the mixing record table includes multiple columns of mixing sequences, and each column of mixing sequences respectively includes at least two different injection drugs that can be mixed for injection;

[0158] Traverse the mixing record table, and determine whether there are at least two injection drugs in the same mixing sequence among the injection drugs, and determine all the injection drugs in the same mixing sequence as a mixing group with mixing characteristics, and determine the remaining all injection drugs as having independent characteristics.

[0159] For example, in this embodiment, in order to be able to determine the connection characteristics of each injection drug, a corresponding mixing record table can be preset. Among them, the mixing record table includes multiple columns of mixing sequences arranged vertically, and each column of mixing sequences respectively includes at least two different injection drugs that can be mixed for injection; after the presetting of the mixing record table is completed, by traversing the mixing record table, it is determined whether there are injection drugs that can be mixed for injection among the injection drugs, and the corresponding injection drugs that can be mixed for injection are determined as having mixing characteristics, while the remaining injection drugs can be respectively determined as having independent characteristics.

[0160] Furthermore, in this embodiment, the above "generate a drug pipeline model connected to the blood vessel to be injected based on the order characteristic and the connection characteristic" may further include the following steps:

[0161] Determine the injection order corresponding to all the injection drugs in the same mixing group respectively, and determine the injection order with the largest value as the one corresponding to the mixing group;

[0162] Generate a parallel pipeline model presenting a parallel structure based on all the injection drugs in each mixing group, and generate an independent pipeline model presenting an independent structure based on the remaining all injection drugs except the mixing group;

[0163] Based on the injection sequence corresponding to the mixing group and the injection sequence corresponding to all the remaining injection drugs except the mixing group, the parallel pipeline model and the independent pipeline model are connected in sequence to obtain a drug pipeline model connected to the blood vessel to be injected and presenting a series structure.

[0164] For example, in this embodiment, after determining the sequence characteristics and connection characteristics of each injection drug through the above content, a corresponding drug pipeline model can be generated based on the corresponding characteristics;

[0165] First, since the injection sequences of all the injection drugs in the mixing group are different, corresponding unification is required. The unification method is to determine the injection sequence with the largest value as the one corresponding to the mixing group;

[0166] Next, after determining the injection sequence, a parallel pipeline model with a parallel structure can be generated corresponding to all the injection drugs in each mixing group, and an independent pipeline model presenting an independent structure can be generated based on the other injection drugs;

[0167] Finally, the corresponding pipeline models can be connected according to the injection sequence to obtain a drug pipeline model presenting a series structure.

[0168] For example, as Figure 2 shown, Figure 2 shows a schematic diagram of the corresponding drug pipeline model. In Figure 2 it can be seen that there is a parallel pipeline model presenting a parallel structure and an independent pipeline model presenting an independent structure. Among them, the parallel pipeline model and the independent pipeline model together form a series pipeline model presenting a series structure, and the series pipeline model is connected to the corresponding virtual torso model for corresponding intravenous injection.

[0169] In addition, in this embodiment, the above "generating a drug pipeline model connected to the blood vessel to be injected based on the sequence characteristics and connection characteristics" can also be implemented based on the following method steps:

[0170] Extract the sequence characteristics and connection characteristics and compare them with the trained training feature groups respectively. If the features in the training feature groups correspond one by one to the sequence characteristics and connection characteristics, then retrieve the trained drug pipeline model and drug injection cycle corresponding to the training feature groups.

[0171] For example, the present invention compares the sequential characteristics and connection characteristics with the trained training feature groups respectively, and the comparison also includes comparing the drug types. When the sequential characteristics and connection characteristics correspond to each other, the corresponding drug types are also corresponding. If the previous corresponding patients or other patients have the same needs as the currently injected drug, at this time, the characteristics in the training feature group correspond one by one to the sequential characteristics and mixing characteristics. Therefore, the present invention can retrieve the training drug pipeline model and drug injection cycle corresponding to the training feature group, quickly determine the training drug pipeline model to be generated and the relatively accurate drug injection cycle. The drug injection cycle can be the average injection cycle of all previous users;

[0172] Traverse the training drug pipeline model based on the position of each injected drug and add information to obtain the drug pipeline model. The present invention adds corresponding drug information according to the position of each injected drug in the training drug pipeline model to obtain a drug pipeline model with the information of the injected drug;

[0173] If the characteristics in the training feature group do not completely correspond to the sequential characteristics and connection characteristics, then compare the characteristics in the training feature group with the sequential characteristics and connection characteristics to obtain the feature similarity. At this time, there is no historical injection scenario corresponding to the current injection scenario. Therefore, the characteristics in the training feature group do not completely correspond to the sequential characteristics and connection characteristics. At this time, there may be some drug injection word searches. At this time, the present invention can calculate the comparison between the characteristics in the training feature group and the sequential characteristics and connection characteristics to obtain the feature similarity.

[0174] When the present invention calculates the feature similarity, it can be calculated through the following steps, including:

[0175] Obtain the historical drug corresponding to the historical sequential characteristics in the training feature group and the current drug corresponding to the current sequential characteristics, and calculate the first same quantity value and the second different quantity value of the historical drug and the current drug.

[0176] Obtain the historical connection characteristics and the current connection characteristics in the training feature group, calculate the third same quantity value and the fourth different quantity value of the historical connection characteristics and the current connection characteristics, and calculate the feature similarity through the following formula,

[0177] Wherein, is the feature similarity, is the first same quantity value, is the second different quantity value, is the third same quantity value, is the fourth different quantity value, is the first weight value, is the second weight value, at When it is larger, there is a tendency for the feature similarity to increase in the dimension corresponding to the sequential characteristic. In When it is larger, there is a tendency for the feature similarity to increase in the dimension corresponding to the connection characteristic. and are preset, where is preferably greater than .

[0178] If it is determined that there is a feature similarity greater than the selection threshold, the training feature group with the maximum feature similarity is used as the feature group to be decomposed. After decomposing the feature group to be decomposed based on the sequential characteristic and the connection characteristic and then re - splicing, a drug pipeline model is obtained. At this time, it can be considered that there is a drug pipeline model stored in the history that is similar to the current medication scenario. Modifying and re - splicing based on the premise of the drug pipeline model stored in the history to obtain the drug pipeline model is the most efficient. Therefore, if there is a feature similarity greater than the selection threshold in the present invention at this time, the training feature group with the maximum feature similarity is used as the feature group to be decomposed. The feature group to be decomposed is decomposed, all sub - models corresponding to different drugs are deleted, and sub - models of drugs that exist in the current drug but do not exist in the stored drug pipeline model are established. Automatic splicing settings are performed on the sub - models of the corresponding drugs based on the sequential characteristic and the connection characteristic, that is, the newly added sub - models are connected by pipelines according to their corresponding sequential characteristics and connection characteristics, or the final drug pipeline model is obtained through manual adjustment by medical staff. The present invention does not limit the splicing method. And the final drug pipeline model is stored corresponding to the corresponding new sequential characteristic and connection characteristic to obtain

[0179] If it is determined that there is no feature similarity greater than the selection threshold, parallel pipeline models and independent pipeline models corresponding to all drugs are generated. At this time, it proves that there is no relatively close stored drug pipeline model. Therefore, at this time, parallel pipeline models and independent pipeline models corresponding to each injected drug need to be established and connected in the corresponding order to obtain a training drug pipeline model, and the injection cycles of all injection times under the training drug pipeline model are calculated and averaged to obtain the drug injection cycle.

[0180] Furthermore, in this embodiment, after the creation of the corresponding drug pipeline model is completed, in order to enable the injected drug to complete the corresponding injection according to the corresponding injection order, therefore, a corresponding clamping control model can also be generated. Among them, the clamping control model is located on each pipeline model and is used to clamp the corresponding pipeline model. By performing clamping control on the clamping control model, it can be ensured that the corresponding pipeline model is in a clamped state or an injection state, thereby realizing the automatic control of each pipeline model and improving the corresponding medical assistance.

[0181] It should be noted that in this embodiment, as Figure 3 shown, the corresponding method can be implemented by interconnecting terminals based on a server and a medical terminal.

[0182] In summary, according to the solution of this embodiment, the server will first obtain the patient information of the patient to be statically treated, and determine the set of statically treated trunks based on the patient information. The set of statically treated trunks includes at least one statically treated trunk area; then, the server will obtain the historical static treatment records of the patient to be statically treated within a preset time before the current time point, and perform area selection based on the historical static treatment records and the set of statically treated trunks, so as to obtain the target trunk area corresponding to the patient to be statically treated; then, the server will obtain the corresponding area image based on the target trunk area of the patient to be statically treated, and create a virtual trunk model having a mapping relationship with the target trunk area based on the area image; then, the server will obtain each virtual blood vessel located in the virtual trunk model, and determine the virtual blood vessels that meet the medical ecological requirements as the blood vessels to be injected based on the ecological attributes of each virtual blood vessel; finally, the server will obtain the static treatment drug information corresponding to the patient to be statically treated, and generate a drug pipeline model connected to the blood vessels to be injected based on the static treatment drug information. The present invention can greatly provide reference for medical staff on the injection area and injection drugs during intravenous injection, can reduce their working time to a certain extent, and thus improve the efficiency of intravenous injection.

[0183] Another embodiment of the present invention provides an artificial intelligence-based static treatment pipeline design system, Figure 4 and the corresponding system block diagram is as follows. The system includes:

[0184] An acquisition module, configured to acquire the patient information corresponding to the patient to be statically treated, and determine a set of statically treated trunks based on the patient information, wherein the set of statically treated trunks includes at least one statically treated trunk area;

[0185] An area selection module, configured to acquire the historical static treatment records of the patient to be statically treated within a preset time before the current time point, and perform area selection based on the historical static treatment records and the set of statically treated trunks to obtain the target trunk area corresponding to the patient to be statically treated;

[0186] A creation module, configured to acquire the corresponding area image based on the target trunk area of the patient to be statically treated, and create a virtual trunk model having a mapping relationship with the target trunk area based on the area image;

[0187] A determination module, configured to acquire each virtual blood vessel located in the virtual trunk model, and determine the virtual blood vessels that meet the medical ecological requirements as the blood vessels to be injected based on the ecological attributes of each virtual blood vessel;

[0188] A model generation module, configured to obtain intravenous therapy drug information corresponding to the patient to receive intravenous therapy, and generate a drug pipeline model connected to the blood vessel to be injected based on the intravenous therapy drug information.

[0189] In the specification provided herein, the algorithms and displays are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems may also be used in conjunction with the examples of the present invention. The structure required to construct such systems will be apparent from the above description. Additionally, the present invention is not directed to any particular programming language. It should be understood that the content of the present invention described herein can be implemented using various programming languages, and the description of a particular language above is for the purpose of disclosing the preferred embodiments of the present invention.

[0190] In the specification provided herein, numerous specific details are set forth. However, it can be understood that embodiments of the present invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0191] Similarly, it should be understood that, in order to streamline this disclosure and assist in understanding one or more of the various inventive aspects, in the foregoing description of the exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together in a single embodiment, figure, or description thereof.

[0192] Those skilled in the art should understand that the modules or units or components of the devices in the examples disclosed herein may be arranged in the devices as described in the embodiments, or alternatively may be located in one or more devices different from the devices in the examples. The modules in the foregoing examples may be combined into one module or may be further divided into multiple sub-modules.

[0193] Those skilled in the art can understand that the modules in the devices in the embodiments can be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition, they can be divided into multiple sub-modules or sub-units or sub-components.

[0194] Furthermore, those skilled in the art will be able to understand that, although some of the embodiments described herein include certain features included in other embodiments but not others, the combination of the features of different embodiments is within the scope of the present invention and forms different embodiments.

[0195] In addition, some of the embodiments described herein are described as methods or combinations of method elements that can be implemented by a processor of a computer system or by other devices performing the functions. Therefore, a processor having the necessary instructions for implementing the method or method elements forms a device for implementing the method or method elements. In addition, the elements described herein in the device embodiments are examples of the following devices: the device is used to implement the functions performed by the elements for the purpose of implementing the invention.

[0196] As used herein, unless otherwise specified, the use of ordinal numbers "first", "second", "third", etc. to describe ordinary objects only indicates different instances of similar objects, and is not intended to imply that the objects so described must have a given order in time, space, order, or in any other way.

[0197] Although the invention has been described in terms of a limited number of embodiments, those skilled in the art in this technical field will appreciate that other embodiments can be contemplated within the scope of the invention as thus described. In addition, it should be noted that the language used in this specification has been principally selected for readability and instructional purposes, rather than to limit or define the subject matter of the invention.

Claims

1. A method for designing intravenous therapy circuits based on artificial intelligence, characterized in that: include: Determine the trunk set to be treated; Obtaining historical static therapy records, and determining historical torso regions based on the historical static therapy records; When the static treatment trunk set includes a static treatment trunk area and the static treatment trunk area is the same as the historical trunk area, it is determined as the target trunk area; When the static treatment trunk set includes a static treatment trunk region and it is not the same region as the historical trunk region, determining the trunk connection relationship, and determining it as the target trunk region in the case of an upper connection, and determining the historical trunk region as the target trunk region in the case of a lower connection; or When the static therapy trunk set includes multiple static therapy trunk regions, determining a trunk connection relationship; When any one of them is an upper connection, it is determined as the target trunk area; When all are lower connections, the historical trunk area is determined as the target trunk area; Acquiring a regional image based on the target torso region, and creating a virtual torso model based on the regional image; determining each virtual blood vessel located in the virtual torso model; Obtain each contour pixel point of the virtual blood vessel, and obtain each corresponding contour coordinate point, Performing difference calculation on the contour coordinate points having the same horizontal coordinate, and determining the obtained horizontal difference corresponding to the maximum value as the first target difference; Performing difference calculation on the contour coordinate points having the same longitudinal coordinate, and determining the longitudinal difference corresponding to the maximum value obtained as the second target difference; The larger of the first target difference and the second target difference is determined as the size attribute; The pixel mean obtained by calculating the mean of each contour pixel value corresponding to each contour pixel point is determined as the pixel attribute; Performing a comprehensive attribute evaluation on the size attribute and the pixel attribute, and determining the blood vessel with the largest attribute evaluation value as the blood vessel to be injected; A drug pipeline model is generated based on the acquired intravenous drug information and fed back to the medical end.

2. The artificial intelligence-based intravenous therapy circuit design method according to claim 1 is characterized in that: Identify the trunk set to be treated, including: Determine the patient age and medical history of the patient to be treated with intravenous therapy based on the patient information, and retrieve a preset intravenous therapy trunk distribution table, wherein the preset intravenous therapy trunk distribution table includes multiple intravenous therapy trunk areas, and each intravenous therapy trunk area includes a corresponding age-appropriate interval and a disease-prohibited interval; Performing a primary screening based on the preset static therapy trunk distribution table, determining all static therapy trunk areas corresponding to the appropriate age ranges including the patient's age, and determining them as a primary screening group; A secondary screening is performed based on the primary screening group, and the to-be-treated trunk set is determined based on the secondary screening result.

3. The artificial intelligence-based intravenous therapy circuit design method according to claim 2 is characterized in that: Determining the to-be-treated trunk set based on the secondary screening result includes: When the secondary screening result is that the primary screening group has at least one static treatment trunk area corresponding to a disease-free interval that does not include the patient's medical record, determining the at least one static treatment trunk area as a static treatment trunk set; When the secondary screening result is that the primary screening group does not have a static therapy trunk area corresponding to a forbidden disease interval that does not include the patient's medical record, the primary screening group is determined as a static therapy trunk set.

4. The method for designing intravenous therapy circuit based on artificial intelligence according to claim 1 is characterized in that: The size attributes and pixel attributes are evaluated comprehensively, including: Retrieve the preset size evaluation weight and the preset pixel evaluation weight; The contour size and pixel mean value corresponding to the same virtual blood vessel are respectively multiplied by the preset size evaluation weight and the preset pixel evaluation weight to obtain a first product value and a second product value; The first product value and the second product value are normalized to obtain a first evaluation value and a second evaluation value, and the first evaluation value and the second evaluation value are summed to obtain each attribute evaluation value corresponding to each virtual blood vessel.

5. The artificial intelligence-based intravenous therapy circuit design method according to claim 1 is characterized in that: Generate a drug pipeline model based on the acquired intravenous drug information and feed it back to the medical end, including: Determine the patient's drug number based on the patient information corresponding to the patient to be treated with intravenous therapy, and retrieve a drug record table, wherein the drug record table includes a plurality of drug sequences, each of which includes different recorded drug numbers and recorded drug information corresponding to the different recorded drug numbers; Traverse the medication record table to find the same medication number as the patient's medication number. Deleting the drug sequence including the intravenous therapy drug information from the drug record table, and determining the recorded drug information corresponding to the recorded drug number as the intravenous therapy drug information corresponding to the patient's drug number; Based on the intravenous therapy drug information, various injectable drugs corresponding to different drug types are determined, and the characteristics of each injectable drug are determined according to the injection sequence determination strategy and the mixed injection determination strategy, and the sequence characteristics and connection characteristics corresponding to each injectable drug are determined, and a drug pipeline model connected to the blood vessel to be injected is generated based on the sequence characteristics and the connection characteristics.

6. The artificial intelligence-based intravenous therapy circuit design method according to claim 5 is characterized in that: The characteristics of each injection drug are determined according to the injection sequence determination strategy, and the sequence characteristics corresponding to each injection drug are determined, including: Determine the components of each injectable medicine to obtain an initial chemical component group corresponding to each injectable medicine, wherein the initial chemical component group includes different component types and type ratios corresponding to the different component types; Retrieving a preset ratio threshold, and screening out component types whose corresponding type ratio is less than or equal to the preset ratio threshold from each of the initial chemical component groups to obtain each current chemical component group; Determine the content of each component corresponding to each component type in each current chemical component group, and determine the content coefficient based on the content of each component; Based on the preset chemical injection sorting table, the type coefficient of each component type in each current chemical component group is determined respectively, and the content coefficient and the type coefficient corresponding to the same component type are summed and calculated to obtain the type priority value corresponding to the component type; The priority values ​​of all types corresponding to the same current chemical component group are summed up to obtain the injection priority values ​​corresponding to each current chemical component group, and the current chemical component groups are sorted based on the injection priority values, and the injection sequences corresponding to the respective injection drugs are determined as the sequence characteristics.

7. The artificial intelligence-based intravenous therapy circuit design method according to claim 6 is characterized in that: The characteristics of each injection drug are determined according to the mixed injection determination strategy, and the connection characteristics corresponding to each injection drug are determined, including: Retrieving a mixing record table, wherein the mixing record table includes a plurality of columns of mixing sequences, each column of mixing sequence includes at least two different injection drugs that can be mixed and injected; The mixing record table is traversed, and it is determined whether there are at least two injection drugs in the same mixing sequence in each injection drug, and all injection drugs in the same mixing sequence are determined as a mixing group with mixing characteristics, and all remaining injection drugs are determined as having independent characteristics.

8. The artificial intelligence-based intravenous therapy circuit design method according to claim 7 is characterized in that: Generating a drug pipeline model connected to the blood vessel to be injected based on the sequence characteristics and the connection characteristics includes: Determine the injection order corresponding to all the injectable drugs in the same mixed group, and determine the injection order with the largest value as corresponding to the mixed group; Generate a parallel pipeline model with a parallel structure based on all the injection drugs in each mixed group, and generate an independent pipeline model with an independent structure based on all the remaining injection drugs except the mixed group; Based on the injection sequence corresponding to the mixed group and the injection sequence corresponding to all remaining injection drugs except the mixed group, the parallel pipeline model and the independent pipeline model are connected in sequence to obtain a drug pipeline model that is connected to the blood vessel to be injected and presents a series structure.

9. The artificial intelligence-based intravenous therapy circuit design method according to claim 8 is characterized in that: Generating a drug pipeline model connected to the blood vessel to be injected based on the sequence characteristics and the connection characteristics includes: Extracting the sequence characteristics and the connection characteristics and comparing them with the trained training feature group respectively, if the features in the training feature group correspond to the sequence characteristics and the connection characteristics one by one, then retrieving the training drug pipeline model and drug injection cycle corresponding to the training feature group; Based on the position of each injected drug, the trained drug pipeline model is traversed and information is added to obtain a drug pipeline model; If the features in the training feature group do not completely correspond to the sequence characteristics and the connection characteristics, then based on the comparison between the features in the training feature group and the sequence characteristics and the connection characteristics, the feature similarity is obtained; If it is determined that there is a feature similarity greater than the selected threshold, the training feature group with the largest feature similarity is used as the feature group to be decomposed, and the feature group to be decomposed is decomposed based on the sequence characteristics and the connection characteristics and then reassembled to obtain the drug pipeline model; If it is determined that there is no feature similarity greater than the selected threshold, a parallel pipeline model and an independent pipeline model corresponding to all drugs are generated.

10. An artificial intelligence-based intravenous therapy circuit design system, characterized in that: include: Acquisition module, determining the trunk set to be treated statically; A region selection module obtains historical static therapy records, and determines historical trunk regions based on the historical static therapy records; when the static therapy trunk set includes a static therapy trunk region and is the same region as the historical trunk region, determines it as the target trunk region; when the static therapy trunk set includes a static therapy trunk region and is not the same region as the historical trunk region, determines the trunk connection relationship, and determines it as the target trunk region when it is an upper connection, and determines the historical trunk region as the target trunk region when it is a lower connection; or when the static therapy trunk set includes multiple static therapy trunk regions, determines the trunk connection relationship; when any one is an upper connection, determines it as the target trunk region; when all are lower connections, determines the historical trunk region as the target trunk region; A creation module is used to obtain a regional image based on a target torso region and to create a virtual torso model based on the regional image; A determination module determines each virtual blood vessel located in the virtual torso model; obtains each contour pixel point of the virtual blood vessel, and obtains each corresponding contour coordinate point, performs difference calculation on the contour coordinate points with the same horizontal coordinate, and determines the obtained horizontal difference value corresponding to the maximum value as the first target difference value; performs difference calculation on the contour coordinate points with the same vertical coordinate, and determines the obtained vertical difference value corresponding to the maximum value as the second target difference value; determines the larger value between the first target difference value and the second target difference value as the size attribute; The pixel mean obtained by calculating the mean of each contour pixel value corresponding to each contour pixel point is determined as the pixel attribute; Performing a comprehensive attribute evaluation on the size attribute and the pixel attribute, and determining the blood vessel with the largest attribute evaluation value as the blood vessel to be injected; The model generation module generates a drug pipeline model based on the acquired intravenous drug information and feeds it back to the medical end.

Images