CN118412107B - Surgical medicine cost management and control method, system, storage medium and electronic equipment - Google Patents

Abstract

The present invention provides a method, system, storage medium and electronic device for controlling the cost of surgical medicine. The method obtains historical medical data of patient cases, surgical records, medication records, costs and treatment effects, performs feature analysis, retains target features, and then trains a neural network model with the target features, wherein at least the patient's age, gender, disease type, surgical complexity level and treatment effect level are used as inputs of the neural network model, and at least the drug type, dosage, medication duration and cost are used as outputs of the neural network model. Finally, the current patient's age, gender, disease type, surgical complexity level and treatment effect level are obtained and input into the trained neural network model, and the target drug type, target dosage, target medication duration and target cost are output. Specifically, through the analysis of big data, the surgical medicine cost is intelligently controlled, and manpower consumption is reduced while improving the control of surgical medicine cost.

Description

A method, system, storage medium and electronic device for controlling surgical drug cost

Technical Field

The present invention belongs to the technical field of surgical drug cost control, and in particular relates to a surgical drug cost control method, system, storage medium and electronic equipment.

Background Art

The control of drug costs for surgical patients is an important part of hospital cost control, and is of great significance for reducing the burden on patients and improving the efficiency of medical resource utilization.

The commonly used surgical drug cost control methods are as follows: for example, centralized procurement is implemented to reduce drug prices through economies of scale; bidding procurement is adopted to ensure drug quality while reducing costs through competition; clinical pathway management is implemented to standardize doctors' drug use behavior and ensure the rational use of drugs; pharmaceutical consultations are conducted, with pharmacists participating in drug decisions before and after surgery to ensure the necessity and effectiveness of drugs; and national and local medical insurance departments are cooperated to implement medical insurance payment method reforms, such as DIP payment, to promote the rational allocation of medical resources.

Although the above measures can control the drug costs of surgical patients to a certain extent while ensuring the safety and efficacy of medication, a large amount of manpower is still required in the review and decision-making process. At the same time, due to a certain degree of subjectivity in doctors' medication behavior, it is impossible to optimize the control of surgical drug costs.

Summary of the invention

Based on this, the embodiments of the present invention provide a method, system, storage medium and electronic device for controlling surgical drug costs, aiming to intelligently control surgical drug costs, thereby improving the control of surgical drug costs and reducing manpower consumption.

A first aspect of an embodiment of the present invention provides a method for controlling surgical drug cost, the method comprising:

Acquire historical medical data, the historical medical data including patient cases, surgical records, medication records, costs and treatment effects, and perform feature analysis based on the patient cases, surgical records, medication records, costs and treatment effects, and retain target features, wherein the target features at least include patient age, gender, disease type, surgical complexity level, drug type, dosage, medication duration, costs and treatment effect level;

The target features are subjected to neural network model training to construct a target neural network model, wherein at least the patient's age, the gender, the disease type, the surgery complexity level and the treatment effect level are used as inputs of the neural network model, and at least the drug type, the dosage, the medication duration and the cost are used as outputs of the neural network model;

Obtain the current patient's age, gender, disease type, surgical complexity level, and treatment effect level, input them into the target neural network model, and output the target drug type, target dose, target medication duration, and target cost;

Get the target cost and determine whether the quantity of the target cost is unique;

If the number of target costs is determined to be unique, the corresponding target drug type, target dosage, and target medication duration are determined;

If it is determined that the number of target costs is not unique, then the cost with the least cost among the target costs is determined, and it is determined whether the number of the cost with the least cost is unique;

If the number of the least expensive cost is unique, then the target drug type, target dosage, and target medication duration corresponding to the least expensive cost are determined;

If the number of drugs that cost the least is not unique, the target regimen shall be determined in the order of priority of target medication duration, target dosage, and target drug type.

Further, the step of acquiring historical medical data, which includes patient cases, surgical records, medication records, costs, and treatment effects, and performing feature analysis based on the patient cases, surgical records, medication records, costs, and treatment effects, and retaining target features includes:

According to the patient's medical records, surgical records and treatment effects, the complexity of the surgery and the treatment effect are graded to obtain the corresponding complexity level of the surgery and the treatment effect level;

Cleaning and integrating the patient cases, surgical records, medication records, costs, surgical complexity levels, and treatment effect levels in the historical medical data in turn to complete medical data preprocessing;

The pre-processed medical data is subjected to feature conversion, wherein the pre-processed patient records, surgical records, medication records and treatment effects are encoded, and the pre-processed costs are standardized or normalized;

The feature dimension of the data after feature conversion is reduced to obtain the target feature so as to retain important features.

Furthermore, in the step of reducing the feature dimension of the data after feature conversion to obtain the target feature and retaining important features, one of correlation analysis, principal component analysis and model-based feature selection is used to reduce the feature dimension of the data after feature conversion.

Furthermore, in the step of determining the target regimen according to the priority order of the target medication duration, target dose, and target drug type, when the target medication duration, target dose, and target drug type are the same, the drug names in each regimen are obtained, and it is determined whether there are generic drugs in the drug names;

If it is determined that there are generic drugs in the drug names, the number of generic drugs in each scheme is counted, and the scheme with the largest number of generic drugs is determined.

Furthermore, the neural network model includes an input layer, a hidden layer and an output layer, wherein:

The input layer includes n first nodes, and the number of the first nodes is consistent with the number of input types of the neural network model;

The hidden layer is composed of a first hidden sublayer, a second hidden sublayer, a third hidden sublayer, a fourth hidden sublayer and a fifth hidden sublayer, wherein the first hidden sublayer includes 64 nodes, which are used to extract features of input data and extract feature information related to the output result, and the first hidden sublayer uses the Leaky ReLU function; the second hidden sublayer and the third hidden sublayer each include 128 nodes, which are used to extract features from the output data of the first hidden sublayer and capture the intrinsic connection between the upper-layer nodes, and the second hidden sublayer and the third hidden sublayer use the Sigmoid function; the fourth hidden sublayer and the fifth hidden sublayer each include 64 nodes, which are used to process the data features output by the third hidden sublayer again, and the fourth hidden sublayer and the fifth hidden sublayer use the ReLU activation function;

The output layer has N second nodes, the number of the second nodes is consistent with the number of output types of the neural network model, wherein the output layer uses a linear activation function.

A second aspect of an embodiment of the present invention provides a surgical drug cost control system, the system comprising:

A feature analysis module is used to obtain historical medical data, wherein the historical medical data includes patient cases, operation records, medication records, costs and treatment effects, and perform feature analysis based on the patient cases, operation records, medication records, costs and treatment effects to retain target features, wherein the target features include at least patient age, gender, disease type, operation complexity level, drug type, dosage, medication duration, costs and treatment effect level;

A training module, used for training the target features with a neural network model to construct a target neural network model, wherein at least the patient's age, gender, disease type, surgical complexity level and treatment effect level are used as inputs of the neural network model, and at least the drug type, dosage, medication duration and cost are used as outputs of the neural network model;

The input module is used to obtain the current patient's age, gender, disease type, surgical complexity level and treatment effect level, input them into the target neural network model, and output the target drug type, target dosage, target medication duration and target cost.

A third aspect of an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the surgical drug cost control method provided in the first aspect.

A fourth aspect of an embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the surgical drug cost control method provided in the first aspect is implemented.

A method, system, storage medium and electronic device for controlling surgical drug costs are provided in an embodiment of the present invention. The method obtains historical medical data of patient cases, surgical records, medication records, costs and treatment effects, performs feature analysis, retains target features, and then trains a neural network model with the target features to construct a target neural network model, wherein at least the patient's age, gender, disease type, surgical complexity level and treatment effect level are used as inputs of the neural network model, and at least the drug type, dosage, medication duration and cost are used as outputs of the neural network model. Finally, the current patient's age, gender, disease type, surgical complexity level and treatment effect level are obtained and input into the target neural network model, and the target drug type, target dosage, target medication duration and target cost are output. Specifically, through the analysis of big data, the surgical drug costs are intelligently controlled, while improving the control of surgical drug costs and reducing manpower consumption.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a flowchart of a method for controlling surgical drug cost provided by Embodiment 1 of the present invention;

FIG2 is a structural block diagram of a surgical drug cost control system provided by Embodiment 2 of the present invention;

FIG3 is a structural block diagram of an electronic device provided in Embodiment 3 of the present invention.

DETAILED DESCRIPTION

In order to facilitate understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and/or" used herein includes any and all combinations of one or more of the related listed items.

Embodiment 1

According to an embodiment of the present invention, an embodiment of a method for controlling surgical drug costs is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

In the first embodiment of the present invention, a method for controlling the cost of surgical medicine expenses is provided, which can be used in electronic devices such as computers, mobile phones, tablet computers, etc. Please refer to Figure 1, which shows a flow chart of the implementation of a method for controlling the cost of surgical medicine expenses provided in the first embodiment of the present invention, specifically including steps S01 to S03.

Step S01, obtaining historical medical data, wherein the historical medical data includes patient cases, surgical records, medication records, costs and treatment effects, and performing feature analysis based on the patient cases, surgical records, medication records, costs and treatment effects, retaining target features, wherein the target features include at least patient age, gender, disease type, surgical complexity level, drug type, dosage, medication duration, cost and treatment effect level.

Specifically, relevant data can be collected from sources such as hospital information systems, financial systems, medical records, and drug databases. Specifically, first, according to patient cases, surgical records, and treatment effects, the complexity of the surgery and the treatment effect are graded to obtain the corresponding surgical complexity level and treatment effect level. It can be understood that the information strongly related to the condition in the patient cases, surgical records, and treatment effects is obtained by manual or machine semantic recognition. At the same time, the surgical complexity level and treatment effect level are pre-established, where different types of conditions have corresponding different surgical complexity levels and treatment effect levels. For example, for orthopedic surgery, it can be divided into four levels: first-level surgery: This is the simplest surgical level, mainly including some basic operations, such as debridement, bone traction, external fixation, joint cavity incision and drainage, simple fracture and dislocation reduction, etc.; second-level surgery: This level of surgical technology is generally difficult, the surgical process is not too complicated, and the risk is medium. Including internal fixation of single limb long tube bone fractures, treatment of open fractures, tendon displacement, simple joint dislocation reduction and internal fixation, bone transplantation, etc.; third-level surgery: This level of surgical technology is more difficult, the process is more complicated, and the risk is also higher. It includes internal fixation of complex limb shaft fractures, surgery for intra-articular fractures, treatment of complex hand injuries, lesion removal of large joints, arthroscopic meniscus removal, etc. Level 4 surgery: This is the highest level of surgery, with the greatest technical difficulty, complex surgical process and the highest risk. It includes cervical spine surgery, total joint replacement, revision surgery, scoliosis correction surgery, etc.

The treatment effect level can also be divided into three levels: level one means poor treatment effect, level two means average treatment effect, and level three means good treatment effect. The different treatment effect levels are mapped to the patient's recovery time and degree after surgery.

Furthermore, the patient cases, surgical records, medication records, costs, surgical complexity levels, and treatment effect levels in the historical medical data are cleaned and integrated in turn to complete the medical data preprocessing. Specifically, the data cleaning is mainly to deal with missing values, outliers, duplicate records, etc. to ensure data quality. In addition, based on the premise of treatment effect, the data corresponding to some drugs with abnormally increased usage and drugs with obviously too expensive prices are eliminated. The purpose of data integration is to integrate data from different sources to form a unified data set.

The pre-processed medical data is feature-converted, wherein the pre-processed patient cases, surgical records, medication records and treatment effects are coded, and the pre-processed costs are standardized or normalized, wherein the coding is mainly for classified data, which can be understood, that is, the pre-processed patient cases, surgical records, medication records and treatment effects need to be manually classified in advance, exemplary, according to the information in the patient case, the patient age is classified, for example, 0-5 years old, 6-17 years old, 18-24 years old, 25-44 years old, 45-64 years old, 65 years old and above, etc.; gender classification, for example, male and female; disease classification, for example, classification by disease type, which can be tumor resection surgery, organ transplantation surgery, congenital disease surgery, trauma surgery, infectious disease surgery, etc. In this embodiment, in the normalization step, the Z-score method is used to transform the original data to obtain data that obeys a normal distribution with a mean of 0 and a standard deviation of 1, wherein the formula for converting the original data using the Z-score method is:

;

in, Represents the normalized data. Represented as raw data, is expressed as the mean of all original data, Expressed as the standard deviation of all raw data.

Finally, the feature dimension of the data after feature conversion is reduced to obtain target features in order to retain important features. Specifically, one of correlation analysis, principal component analysis, and model-based feature selection is used to reduce the feature dimension of the data after feature conversion. It should be noted that the target features mentioned above are not limitations of this solution and can be set accordingly based on actual needs or actual feature analysis results.

Step S02, training the target features with a neural network model to construct a target neural network model, wherein at least the patient's age, gender, disease type, surgical complexity level and treatment effect level are used as inputs of the neural network model, and at least the drug type, dosage, medication duration and cost are used as outputs of the neural network model.

In this embodiment, the neural network model includes an input layer, a hidden layer and an output layer. Specifically, 80% of the randomly shuffled data set samples are used as training set samples, and the other 20% are used as test set samples. The neuron formulas of the hidden layer and the output layer can be expressed as:

First hidden sublayer:

;

Second hidden sublayer:

;

The third hidden sublayer:

;

Fourth hidden sublayer:

;

Fifth hidden sublayer:

;

Output layer:

;

Among them, _xj and _yi represent the jth node of the input layer and the ith node of the output layer respectively. Indicates the connection The weights of the jth node in the layer and the ith node in the output layer, Indicates The bias term of the i-th node in the layer, for example, is represented as the i-th neuron in the first hidden sublayer, is represented as the i-th neuron in the second hidden sublayer, is represented as the i-th neuron in the third hidden sublayer, is represented as the i-th neuron in the fourth hidden sublayer, Represented as the i-th neuron in the fifth hidden sublayer. Similarly, is represented as the jth neuron in the first hidden sublayer, is represented as the jth neuron in the second hidden sublayer, is represented as the jth neuron in the third hidden sublayer, is represented as the jth neuron in the fourth hidden sublayer, It represents the jth neuron in the fifth hidden sublayer. LeakyReLU, Sigmoid and ReLU are activation functions.

It should be noted that the input layer includes n first nodes, and the number of the first nodes is consistent with the number of input types of the neural network model;

The hidden layer consists of a first hidden sublayer, a second hidden sublayer, a third hidden sublayer, a fourth hidden sublayer and a fifth hidden sublayer, wherein the first hidden sublayer includes 64 nodes, which is used to extract features of the input data and extract feature information related to the output result, and the first hidden sublayer uses the Leaky ReLU function; the second hidden sublayer and the third hidden sublayer each include 128 nodes, which are used to extract features from the output data of the first hidden sublayer and capture the intrinsic connection between the upper-layer nodes, and the second hidden sublayer and the third hidden sublayer use the Sigmoid function; the fourth hidden sublayer and the fifth hidden sublayer each include 64 nodes, which are used to process the data features output by the third hidden sublayer again, and the fourth hidden sublayer and the fifth hidden sublayer use the ReLU activation function;

The output layer has N second nodes, and the number of the second nodes is consistent with the number of output types of the neural network model, wherein the output layer uses a linear activation function.

Step S03, obtaining the current patient's age, gender, disease type, surgical complexity level and treatment effect level, inputting them into the target neural network model, and outputting the target drug type, target dosage, target medication duration and target cost.

Among them, there may be multiple target drug types, target doses, target duration of medication, and target costs output by the target neural network model, that is, multiple parallel plans are included. At this time, the target cost is obtained to determine whether the number of target costs is unique;

If the number of target costs is determined to be unique, the corresponding target drug type, target dosage, and target medication duration are determined;

If it is determined that the number of target costs is not unique, then the cost with the least cost among the target costs is determined, and it is determined whether the number of the cost with the least cost is unique;

If the number of the least expensive cost is unique, then the target drug type, target dosage, and target medication duration corresponding to the least expensive cost are determined;

If the number of drugs with the least cost is not unique, the target plan is determined in the order of priority of target medication duration, target dosage, and target drug type. It can be understood that the shortest duration among the target medication durations is determined first. If there are multiple shortest durations at the same time, the dose with the least use among the target doses is determined. If there are multiple doses with the least use, the drug type with the least use among the target drug types is determined. If there are multiple drug types with the least use, the drug names in each plan are obtained, and it is determined whether there are generic drugs in the drug names; if it is determined that there are generic drugs in the drug names, the number of generic drugs in each plan is counted, and the plan with the largest number of generic drugs is determined. It should be noted that doctors are encouraged to give priority to generic drugs to reduce drug costs.

In summary, the surgical drug cost control method in the above-mentioned embodiment of the present invention obtains historical medical data of patient cases, surgical records, medication records, costs and treatment effects, performs feature analysis, retains target features, and then trains the target features for a neural network model to construct a target neural network model, wherein at least the patient's age, gender, disease type, surgical complexity level and treatment effect level are used as inputs of the neural network model, and at least the drug type, dosage, medication duration and cost are used as outputs of the neural network model. Finally, the current patient's age, gender, disease type, surgical complexity level and treatment effect level are obtained and input into the target neural network model, and the target drug type, target dosage, target medication duration and target cost are output. Specifically, through the analysis of big data, the surgical drug cost is intelligently controlled, thereby reducing manpower consumption while improving the control of surgical drug cost.

Embodiment 2

Please refer to Figure 2, which is a structural block diagram of a surgical drug cost control system provided in Example 2 of the present invention. The surgical drug cost control system 200 is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated here. As used below, the term "module" can implement a combination of software and/or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

Specifically, the surgical drug cost control system 200 includes: a feature analysis module 21, a training module 22 and an input module 23, wherein:

The feature analysis module 21 is used to obtain historical medical data, the historical medical data includes patient cases, operation records, medication records, costs and treatment effects, and perform feature analysis based on the patient cases, operation records, medication records, costs and treatment effects, and retain target features, wherein the target features at least include patient age, gender, disease type, operation complexity level, drug type, dosage, medication duration, cost and treatment effect level;

The training module 22 is used to train the target features into a neural network model to construct a target neural network model, wherein at least the patient's age, gender, disease type, surgical complexity level and treatment effect level are used as inputs of the neural network model, and at least the drug type, dosage, medication duration and cost are used as outputs of the neural network model. The neural network model includes an input layer, a hidden layer and an output layer, wherein:

The input layer includes n first nodes, and the number of the first nodes is consistent with the number of input types of the neural network model;

The hidden layer is composed of a first hidden sublayer, a second hidden sublayer, a third hidden sublayer, a fourth hidden sublayer and a fifth hidden sublayer, wherein the first hidden sublayer includes 64 nodes, which are used to extract features of input data and extract feature information related to the output result, and the first hidden sublayer uses the Leaky ReLU function; the second hidden sublayer and the third hidden sublayer each include 128 nodes, which are used to extract features from the output data of the first hidden sublayer and capture the intrinsic connection between the upper-layer nodes, and the second hidden sublayer and the third hidden sublayer use the Sigmoid function; the fourth hidden sublayer and the fifth hidden sublayer each include 64 nodes, which are used to process the data features output by the third hidden sublayer again, and the fourth hidden sublayer and the fifth hidden sublayer use the ReLU activation function;

The output layer has N second nodes, the number of the second nodes is consistent with the number of output types of the neural network model, wherein the output layer uses a linear activation function;

The input module 23 is used to obtain the current patient's age, gender, disease type, surgical complexity level and treatment effect level, input them into the target neural network model, and output the target drug type, target dosage, target medication duration and target cost.

Furthermore, in some optional embodiments of the present invention, the feature analysis module 21 includes:

A grading unit is used to grade the complexity of the surgery and the treatment effect according to the patient's case, surgery record and treatment effect, and obtain the corresponding surgery complexity level and treatment effect level;

A preprocessing unit, for sequentially cleaning and integrating the data of patient cases, surgical records, medication records, costs, surgical complexity levels, and treatment effect levels in the historical medical data, to complete the preprocessing of the medical data;

A feature conversion unit is used to perform feature conversion on the pre-processed medical data, wherein the pre-processed patient cases, operation records, medication records and treatment effects are encoded, and the pre-processed costs are standardized or normalized;

The target feature acquisition unit is used to reduce the feature dimension of the data after feature conversion to obtain the target feature so as to retain important features, wherein the feature dimension of the data after feature conversion is reduced by one of correlation analysis, principal component analysis and model-based feature selection.

Furthermore, in some optional embodiments of the present invention, the surgical drug cost control system 200 further includes:

The first judgment module is used to obtain the target cost and judge whether the quantity of the target cost is unique;

The first determination module is used to determine the corresponding target drug type, target dosage, and target medication duration if the number of target costs is determined to be unique;

The second judgment module is used for determining the least expensive cost among the target costs if it is determined that the number of target costs is not unique, and judging whether the number of the least expensive cost is unique;

The second determination module is used to determine the target drug type, target dosage, and target medication duration corresponding to the least expensive cost if the least expensive cost is unique;

The third determination module is used to determine the target plan according to the priority order of target medication duration, target dosage, and target drug type if the number of the least expensive expenses is not unique.

Further, in some optional embodiments of the present invention, the third determining module includes:

A judgment unit, used for obtaining the names of the drugs in each scheme when the target medication duration, target dosage and target drug type are the same, and judging whether there is a generic drug in the drug name;

The determination unit is used to count the number of generic drugs in each scheme if it is determined that there are generic drugs in the drug name, and determine the scheme with the largest number of generic drugs in the drug name.

Embodiment 3

On the other hand, the present invention further proposes an electronic device, please refer to Figure 3, which is shown as an electronic device in Embodiment 3 of the present invention, including a memory 20, a processor 10, and a computer program 30 stored in the memory and executable on the processor, and when the processor 10 executes the computer program 30, the surgical drug cost control method as described above is implemented.

In some embodiments, the processor 10 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 20, such as executing access restriction programs.

The memory 20 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a hard disk, a multimedia card, a card-type memory (for example, an SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 20 may be an internal storage unit of an electronic device, such as a hard disk of the electronic device. In other embodiments, the memory 20 may also be an external storage device of an electronic device, such as a plug-in hard disk equipped on the electronic device, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. Further, the memory 20 may also include both an internal storage unit and an external storage device of the electronic device. The memory 20 may be used not only to store application software and various types of data of the electronic device, but also to temporarily store data that has been output or is to be output.

It should be noted that the structure shown in FIG. 3 does not constitute a limitation on the electronic device. In other embodiments, the electronic device may include fewer or more components than shown in the figure, or a combination of certain components, or a different arrangement of components.

The embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the surgical drug cost control method as described above.

Those skilled in the art will appreciate that the logic and/or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable instructions for implementing logical functions, and may be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For purposes of this specification, "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or a combination thereof: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention patent shall be subject to the attached claims.

Claims (6)

1. A method for controlling cost of surgical charges, the method comprising:

Acquiring historical medical data, wherein the historical medical data comprises patient cases, operation records, medication records, fees and treatment effects, and carrying out feature analysis according to the patient cases, the operation records, the medication records, the fees and the treatment effects, and retaining target features, wherein the target features at least comprise patient ages, sexes, disease types, operation complexity levels, drug types, doses, medication durations, fees and treatment effect levels;

training the target feature by using a neural network model to construct a target neural network model, wherein at least the age of the patient, the sex, the disease type, the surgical complexity level and the treatment effect level are used as inputs of the neural network model, and at least the drug type, the dose, the drug duration and the cost are used as outputs of the neural network model;

Obtaining the current age, sex, disease type, operation complexity level and treatment effect level of a patient, inputting the current age, sex, disease type, operation complexity level and treatment effect level of the patient into the target neural network model, and outputting a target drug type, target dosage, target drug duration and target expense;

acquiring target fees and judging whether the number of the target fees is unique;

if the number of the target fees is judged to be unique, determining the corresponding target drug type, target dosage and target drug duration;

If the number of the target fees is judged to be not unique, determining the least cost in the target fees, and judging whether the number of the least cost is unique;

if the quantity of the least cost is judged to be unique, determining the target medicine type, the target dosage and the target medicine duration corresponding to the least cost;

If the quantity of the least cost is not unique, determining a target scheme according to the priority sequence of the target medication time length, the target dosage and the target medicine type, acquiring medicine names in the schemes when the target medication time length, the target dosage and the target medicine type are the same, and judging whether a common medicine exists in the medicine names;

if judging that the universal medicines exist in the medicine names, counting the number of the medicines with the universal names in the schemes, and determining the scheme with the maximum number of the medicines with the universal names;

the neural network model includes an input layer, a hidden layer, and an output layer, wherein:

the input layer comprises n first nodes, and the number of the first nodes is consistent with the input variety number of the neural network model;

The hidden layer consists of a first hidden sub-layer, a second hidden sub-layer, a third hidden sub-layer, a fourth hidden sub-layer and a fifth hidden sub-layer, wherein the first hidden sub-layer comprises 64 nodes and is used for extracting characteristics of input data and extracting characteristic information related to an output result, and the first hidden sub-layer uses a leak ReLU function; the second hidden sub-layer and the third hidden sub-layer comprise 128 nodes, are used for extracting characteristics in output data of the first hidden sub-layer, capture internal relations between upper nodes and use Sigmoid functions; the fourth hidden sub-layer and the fifth hidden sub-layer each comprise 64 nodes, and are used for processing the data characteristics output by the third hidden sub-layer again, and the fourth hidden sub-layer and the fifth hidden sub-layer use a ReLU activation function;

The output layer is provided with N second nodes, the number of the second nodes is consistent with the output variety number of the neural network model, and the output layer uses a linear activation function.

2. The method of claim 1, wherein the step of obtaining historical medical data, the historical medical data including patient cases, surgical records, medication records, fees, and treatment effects, and performing a feature analysis based on the patient cases, the surgical records, the medication records, the fees, and the treatment effects, the step of preserving target features comprises:

According to the patient cases, the operation records and the treatment effects, grading the operation complexity and the treatment effects to obtain corresponding operation complexity grade and treatment effect grade;

Sequentially cleaning and integrating data in patient cases, operation records, medication records, cost, operation complexity level and treatment effect level in the historical medical data to complete medical data preprocessing;

performing feature conversion on the preprocessed medical data, wherein the preprocessed patient cases, the preprocessed operation records, the preprocessed medication records and the preprocessed treatment effects are coded, and the preprocessed fees are standardized or normalized;

and reducing feature dimensions of the data after feature conversion to obtain the target features so as to retain important features.

3. The method according to claim 2, wherein the step of reducing feature dimensions of the feature-converted data to obtain the target feature to preserve important features includes reducing feature dimensions of the feature-converted data using one of correlation analysis, principal component analysis, and model-based feature selection.

4. A surgical pharmacy cost management system for implementing a surgical pharmacy cost management method according to any one of claims 1-3, the system comprising:

the characteristic analysis module is used for acquiring historical medical data, wherein the historical medical data comprises patient cases, operation records, medication records, fees and treatment effects, and carrying out characteristic analysis according to the patient cases, the operation records, the medication records, the fees and the treatment effects, and retaining target characteristics, and the target characteristics at least comprise patient ages, sexes, disease types, operation complexity levels, drug types, doses, medication duration, fees and treatment effect levels;

The training module is used for training the target characteristics into a neural network model so as to construct a target neural network model, wherein at least the age of the patient, the sex, the disease type, the surgical complexity level and the treatment effect level are taken as inputs of the neural network model, and at least the drug type, the dose, the drug duration and the cost are taken as outputs of the neural network model;

The input module is used for acquiring the current age, sex, disease type, operation complexity level and treatment effect level of the patient, inputting the current age, sex, disease type, operation complexity level and treatment effect level of the patient into the target neural network model, and outputting the target drug type, target dosage, target drug duration and target expense.

5. A computer readable storage medium having stored thereon a computer program, which when executed by a processor implements a surgical pharmacy cost management method according to any one of claims 1 to 3.

6. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the surgical pharmacy cost management method of any one of claims 1-3 when the program is executed.