Method, device and computer equipment for identifying chronic disease reimbursement drugs

Abstract

The present invention provides a method, device and computer equipment for identifying drugs that are reimbursed for chronic diseases, wherein the method comprises: obtaining text information on a prescription; obtaining the identity information and drug information of the corresponding patient based on the text information; extracting the patient's disease information based on the identity information, and obtaining the corresponding chronic disease qualification information based on the disease information; inputting the drug information and the chronic disease qualification information into a chronic disease qualification reimbursement model for calculation, and obtaining the target distance between each drug in the drug information and the chronic disease qualification information; the beneficial effects of the present invention are as follows: by obtaining the patient's identity information, the corresponding chronic disease qualification information is obtained in the database, and then the target drug is reimbursed based on the chronic disease qualification information. Compared with setting rule-based supervision, a certain standard is set for reimbursement of drugs, which reduces the abuse of supervision and saves a lot of time and manpower costs.

Description

Technical Field

[0001] The present invention relates to the field of digital medicine, and in particular to a method, a device and a computer device for identifying reimbursable drugs for chronic diseases. Background Art

[0002] At present, the population is aging rapidly, and the number of chronic disease patients is huge, among which the number of patients with hypertension and diabetes exceeds 300 million. 20% of chronic disease patients spend about 70% of the medical insurance funds, and the high expenditure on chronic diseases is also the main factor in the growth of medical insurance costs. When prescribing, doctors will consider other conditions of chronic disease patients and prescribe drugs that are not for treating chronic diseases. There are cases where drugs that are not for treating chronic diseases are reimbursed as chronic disease drugs. Therefore, a method for identifying drugs on prescriptions is needed. Summary of the invention

[0003] The main purpose of the present invention is to provide a method, device and computer equipment for identifying chronic disease reimbursement drugs, aiming to solve the problem of how to identify chronic disease drugs in prescriptions.

[0004] The present invention provides a method for identifying reimbursable drugs for chronic diseases, comprising:

[0005] Get the text information on the prescription;

[0006] Acquire the identity information and drug information of the corresponding patient according to the text information;

[0007] Extracting the patient's disease information according to the identity information, and acquiring corresponding chronic disease qualification information according to the disease information; wherein the chronic disease qualification information is the authority information for reimbursement of drugs corresponding to the disease information;

[0008] The drug information and the chronic disease qualification information are input into a chronic disease qualification reimbursement model for calculation to obtain a target distance between each drug in the drug information and the chronic disease qualification information; wherein the chronic disease qualification reimbursement model is trained based on various chronic disease types and the insured drugs corresponding to the various chronic disease types;

[0009] Determine whether there is a target drug among the drugs whose target distance is less than a first preset target distance;

[0010] If the target drug has a target distance that is smaller than the first preset target distance, the target drug is determined to be a reimbursed drug for a chronic disease of the patient.

[0011] Furthermore, before the step of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information, it also includes:

[0012] Obtain the medical information of various chronic diseases in the outpatient database and outpatient chronic disease database;

[0013] The medical information is input into the training model as training data, wherein different types of chronic disease information are trained through corresponding calculation formulas. After the training is completed, the chronic disease qualification reimbursement model is obtained. The calculation formula is: W k represents the target distance of the kth chronic disease, N is the size of the training data, X ij Represents word vector a j In the word vector a i The number of times the context appears, f(X ij ) is the weighting function, a i represents the i-th word vector, represents the independent j-th context word vector, b i , Indicates the parameters after training, log n X ij The n in represents the parameters after training.

[0014] Furthermore, the step of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information includes:

[0015] Based on the information of each drug, determine whether each drug is recorded in the chronic disease reimbursement drug database;

[0016] Extracting reimbursable drugs recorded in the chronic disease reimbursement drug database from among the drugs;

[0017] Obtaining corresponding drug category information according to the information of the reimbursed drugs;

[0018] The drug category information is input into the chronic disease qualification reimbursement model for calculation to obtain the target distance.

[0019] Furthermore, the step of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information includes:

[0020] When the disease information includes multiple chronic disease information, the drug information is respectively input into the calculation formulas corresponding to the multiple chronic disease information;

[0021] Multiple target distances corresponding to each drug in the drug information and the multiple chronic disease information are calculated.

[0022] Furthermore, the step of obtaining text information on the prescription includes:

[0023] Obtaining image information corresponding to the prescription;

[0024] Collecting the values ​​of the R color channel, the G color channel, and the B color channel in the RGB color model of the pixel points in the picture, and setting the RGB colors of the pixel points in the picture to (0,0,0), (255,255,255), or (P,P,P) according to a preset ternary method, where P is a preset value greater than 0 and less than 255, thereby obtaining a temporary picture composed of three colors;

[0025] Calculating the areas occupied by the three colors in the temporary image, and using a preset text segmentation method for the areas occupied by the two colors with smaller areas, thereby obtaining segmented single handwritten characters and segmented single printed characters;

[0026] The text features of the single handwritten text and the text features of the single printed text are extracted and input into a preset support vector machine for classification, thereby obtaining the handwritten text and the printed text in the image.

[0027] Furthermore, after the step of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information, the method further includes:

[0028] Determining whether there is a common medicine among the medicines whose target distance is greater than a second preset target distance;

[0029] If the above-mentioned common medicines are available, it is deemed that the above-mentioned common medicines cannot be reimbursed.

[0030] Furthermore, after the step of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information, the method further includes:

[0031] Determine whether there is a suspicious drug among the drugs whose target distance is less than the second preset target distance and greater than the first preset distance target;

[0032] If the suspicious drug is present, the suspicious drug is extracted and sent to the processing terminal for identification.

[0033] The present invention also provides a device for identifying chronic disease drugs, comprising:

[0034] A text information acquisition module is used to acquire text information on the prescription;

[0035] A patient information acquisition module, which acquires the identity information and drug information of the corresponding patient according to the text information;

[0036] A disease information extraction module is used to extract the disease information of the patient according to the identity information, and obtain the corresponding chronic disease qualification information according to the disease information; wherein the chronic disease qualification information is the authority information of the reimbursement of drugs corresponding to the disease information;

[0037] A target distance calculation module is used to input the drug information and the chronic disease qualification information into a chronic disease qualification reimbursement model for calculation to obtain a target distance between each drug in the drug information and the chronic disease qualification information; wherein the chronic disease qualification reimbursement model is trained based on various chronic disease types and insured drugs corresponding to various chronic disease types;

[0038] A target distance judgment module, used to judge whether there is a target drug whose target distance is less than a first preset target distance among the drugs;

[0039] The reimbursement module is used to determine that the target drug is a reimbursable drug for a patient's chronic disease if the target drug has a target distance that is less than the first preset target distance.

[0040] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0041] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0042] The beneficial effects of the present invention are as follows: by obtaining the patient's identity information, the corresponding chronic disease qualification information is obtained in the database, and then the drug information and chronic disease qualification information on the prescription are input into the chronic disease qualification reimbursement model for calculation, and the target drug related to the chronic disease qualification information is obtained, and then the target drug is reimbursed. Compared with setting rules for supervision, a certain standard is set for reimbursement of drugs, which reduces the abuse of supervision and saves a lot of time and manpower costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flow chart of a method for identifying chronic disease reimbursement drugs according to an embodiment of the present invention;

[0044] Figure 2 This is a schematic block diagram of the structure of a device for identifying reimbursable drugs for chronic diseases according to an embodiment of the present invention;

[0045] Figure 3A schematic block diagram of the structure of a computer device according to an embodiment of the present application.

[0046] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly. The connection can be a direct connection or an indirect connection.

[0049] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there may be three relationships. For example, A and B can represent three situations: A exists alone, A and B exist at the same time, and B exists alone.

[0050] In addition, in the present invention, descriptions such as "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0051] Reference Figure 1 The present invention proposes a method for identifying chronic disease reimbursement drugs, comprising:

[0052] S1: Get the text information on the prescription;

[0053] S2: Obtaining the identity information and drug information of the corresponding patient according to the text information;

[0054] S3: extracting the patient's disease information according to the identity information, and acquiring corresponding chronic disease qualification information according to the disease information; wherein the chronic disease qualification information is the authority information for reimbursement of drugs corresponding to the disease information;

[0055] S4: Input the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation, and obtain the target distance between each drug in the drug information and the chronic disease qualification information; wherein the chronic disease qualification reimbursement model is trained based on various chronic disease types and the insured drugs corresponding to various chronic disease types, and the chronic disease qualification information includes one or more chronic disease information, and different chronic disease information corresponds to different calculation formulas;

[0056] S5: Determine whether there is a target drug whose target distance is less than a first preset target distance among the drugs;

[0057] S6: If the target drug has the target distance that is smaller than the first preset target distance, the target drug is determined to be a reimbursed drug for a chronic disease of the patient.

[0058] As described in the above step S1, the text information on the prescription is obtained. The prescription can be an electronic prescription, a paper prescription, or a picture containing the prescription. For an electronic prescription, the text information therein can be directly obtained; for a paper prescription, its scanned copy can be obtained through a printer, or a photo can be taken through a camera to obtain its corresponding picture containing the prescription; for pictures and scanned copies, since they are both in picture format, the text information in the picture can be extracted through a text extraction method, thereby achieving the acquisition of the text information on the prescription.

[0059] As described in step S2 above, the identity information and drug information of the corresponding patient are obtained based on the text information, and the text information is extracted based on semantic recognition. Specifically, the text information is preprocessed first, and the preprocessing includes unifying the language of the text information, replacing synonyms, deleting unnecessary sentences, etc. The identity information is generally in the name column before the prescription, which can be directly identified and extracted, and then the corresponding drug information is extracted based on the drug vocabulary, where the drug vocabulary includes the name and / or alias of each drug.

[0060] As described in the above step S3, the patient's disease information is then extracted from the chronic disease database based on the identity information, wherein the chronic disease database records the information of all chronic diseases suffered by the patients, and then the corresponding chronic disease qualification information is obtained based on the disease information, wherein the chronic disease qualification information is the reimbursable drug authority information for the corresponding chronic disease.

[0061] As described in the above step S4, since chronic diseases can be treated with different medicines, even for the same chronic disease, there are many kinds of corresponding reimbursable medicines, so it is necessary to calculate the target distance between each medicine and the chronic disease qualification information. Since there are many kinds of medicines, directly comparing each medicine with the chronic disease qualification information will result in a large amount of calculation; and for the same type of medicines, they can generally treat the chronic disease, so the corresponding drug category can be directly detected, and then the target distance between the drug category and the chronic disease qualification information can be calculated based on the drug category. Of course, some drugs may have therapeutic effects, but they do not fall within the scope of reimbursement. Later, it can be determined whether the drug can be reimbursed based on the chronic disease reimbursement drug library.

[0062] As described in the above steps S5-S6, determine whether there is a target drug with a distance less than the first preset target distance, wherein the first preset target distance is a pre-set value. When it is less than this value, it can be considered that the drug is a chronic disease reimbursement drug corresponding to the chronic disease qualification information, so this part of the drug included in the prescription can be extracted for reimbursement.

[0063] In one embodiment, before step S4 of inputting the relevant word vector and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information, the method further includes:

[0064] S301: Obtaining the medical information of various chronic disease qualification information in the outpatient database and the outpatient chronic disease database;

[0065] S302: Input the medical information as training data into the training model, wherein different types of chronic disease information are trained using corresponding calculation formulas. After the training is completed, the chronic disease qualification reimbursement model is obtained. The calculation formula is: W k represents the target distance of the kth chronic disease, N is the size of the training data, X ij Represents word vector a j In the word vector a i The number of times the context appears, f(X ij ) is the weighting function, a i represents the i-th word vector, represents the independent j-th context word vector, b i , Indicates the parameters after training, log n X ij The n in represents the parameters after training.

[0066] As described in the above steps S301-S302, since outpatient and chronic diseases have certain similarities in diagnosing diseases and prescribing drugs, the medical information of outpatient and chronic diseases can be selected as training data to train the corresponding training model. It should be noted that the training model may include multiple calculation formulas, each of which corresponds to a chronic disease information. This embodiment takes the first chronic disease information as an example, and it may also include multiple chronic disease information, which correspond to different calculation formulas. After obtaining the chronic disease information of the patient, different calculation formulas in the trained chronic disease qualification reimbursement model are used to calculate the target distance between each drug and the chronic disease information. In the formula, you only need to input the extracted text information into the formula. During the calculation process, the formula In the formula, multiple corresponding word vectors are obtained according to the text information (i.e., a i ), is the word vector related to the first chronic disease information. After training, it can be regarded as a certain parameter value. i , And log n X ij can be regarded as specific parameter values, f(X ij ) is a weighting function, which can be regarded as i The relevant parameter values ​​can be considered as a i The value on which the function is calculated, which in some embodiments may be a constant of 1.

[0067] In one embodiment, the step S4 of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information includes:

[0068] S401: judging whether each drug is recorded in the chronic disease reimbursement drug database according to the information of each drug;

[0069] S402: extracting reimbursable drugs recorded in the chronic disease reimbursement drug database from among the drugs;

[0070] S403: Obtaining corresponding drug category information according to the information of the reimbursed drug;

[0071] S404: Input the drug category information into the chronic disease qualification reimbursement model for calculation to obtain the target distance.

[0072] As described in the above steps S401-S404, the drugs can be filtered first, that is, some drugs that do not belong to the chronic disease reimbursement drug database, such as ordinary cold medicines, but the prescription includes these drugs. Therefore, the drug information on the prescription can be directly judged. If it is not in the chronic disease reimbursement drug database, then this part of the drug information can be directly eliminated, and there is no need to input it into the chronic disease qualification reimbursement model for calculation. It is only necessary to input the reimbursement drugs recorded in the chronic disease reimbursement drug database into the chronic disease qualification reimbursement model for calculation, which can reduce the amount of calculation of the model and help to quickly obtain the final target drug.

[0073] In one embodiment, the step S4 of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information includes:

[0074] S411: When the disease information includes multiple types of chronic disease information, the drug information is respectively input into the calculation formulas corresponding to the multiple types of chronic disease information;

[0075] S412: Calculate and obtain multiple target distances between each drug in the drug information and the multiple chronic disease information.

[0076] As described in the above steps S411-S412, the patient may have multiple chronic diseases, and the corresponding disease information will also include multiple chronic disease information. The patient has reimbursement rights corresponding to multiple chronic disease information. Therefore, each drug can be input into a different calculation formula and calculated separately, that is, different target distances of each drug in different calculation formulas are obtained. It should be understood that among the target distances corresponding to the drug, as long as one target distance is greater than the first preset target distance, the drug can be considered as a target drug.

[0077] In one embodiment, the step S1 of obtaining text information on the prescription includes:

[0078] S101: Obtaining image information corresponding to the prescription;

[0079] S102: collecting values ​​of the R color channel, the G color channel, and the B color channel in the RGB color model of the pixel points in the image, and setting the RGB colors of the pixel points in the image to (0,0,0), (255,255,255), or (P,P,P) according to a preset ternary method, where P is a preset value greater than 0 and less than 255, thereby obtaining a temporary image composed of three colors;

[0080] S103: calculating the areas occupied by the three colors in the temporary image, and using a preset text segmentation method for the areas occupied by the two colors with smaller areas, so as to obtain segmented single handwritten characters and segmented single printed characters;

[0081] S104: extracting the text features of the single handwritten text and the text features of the single printed text, and inputting them into a preset support vector machine for classification, so as to obtain the handwritten text and the printed text in the image.

[0082] As described in the above steps S101-S104, the corresponding image information is obtained by a printer or a camera, and then the handwritten text and the printed text obtained by recognition are obtained by a ternary method. Since the prescription may be handwritten by the doctor, in order to more accurately distinguish handwritten text from printed text, the present application uses a ternary method, that is, according to the preset ternary method, the RGB color of the pixel points in the image is set to (0,0,0), (255,255,255) or (P,P,P), where P is a preset value greater than 0 and less than 255, thereby obtaining a temporary image composed of three colors, and calculating the area occupied by the three colors in the temporary image, and using the preset text segmentation method for the areas occupied by the two colors with smaller areas (since the area with the largest area is definitely the background, there is no need to analyze the area with the largest area), thereby obtaining a segmented single handwritten text and a segmented single printed text. The support vector machine is a generalized linear classifier that performs binary classification of data in a supervised learning manner, and is suitable for comparing the text to be recognized with the pre-stored text to output the most similar text. Based on this, the text features of the single handwritten text and the text features of the single printed text are extracted and input into the preset support vector machine for classification, thereby obtaining the handwritten text and printed text contained in the recognized image. The text features are, for example, special points in the pixel points corresponding to the text, such as extreme points, isolated points, etc.

[0083] In one embodiment, after the step S4 of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information, the step further includes:

[0084] S501: Determine whether there is a common medicine among the medicines whose target distance is greater than a second preset target distance;

[0085] S502: If the common medicine is available, it is determined that the common medicine cannot be reimbursed.

[0086] As described in the above steps S501-S502, when the calculated target distance is less than the second preset target distance, it can be considered as an ordinary drug, where the second preset target distance is a pre-set value. When it is greater than this value, it can be identified as an ordinary drug. Ordinary drugs are drugs that are not reimbursed for chronic diseases. In this article, drugs that are not reimbursed for chronic diseases only refer to drugs that cannot be reimbursed for chronic diseases, not drugs that cannot be reimbursed for other items. Then this part of ordinary drugs is identified as non-reimbursable.

[0087] In one embodiment, after the step S4 of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information, the step further includes:

[0088] S511: Determine whether there is a suspicious drug among the drugs whose target distance is less than the second preset target distance and greater than the first preset distance target;

[0089] S512: If the suspicious drug is present, the suspicious drug is extracted and sent to a processing terminal for identification.

[0090] As described in the above steps S511-S512, in order to improve the recognition of drugs, the first preset target distance will be set small enough to ensure that drugs smaller than the first preset target distance are target drugs, and the second preset target distance will be set large enough to ensure that drugs larger than the second preset target distance are ordinary drugs. Drugs between the first preset target distance and the second preset target distance are defined as suspicious drugs, which may be target drugs or ordinary drugs. Therefore, the suspicious drugs can be extracted and sent to relevant personnel for further identification. After identification, the relevant personnel will directly screen out the target drugs and then integrate all target drugs for reimbursement identification processing.

[0091] Reference Figure 2 The present invention proposes a device for identifying chronic disease reimbursement drugs, comprising:

[0092] A text information acquisition module 10 is used to acquire text information on a prescription;

[0093] A patient information acquisition module 20, which acquires the identity information and drug information of the corresponding patient according to the text information;

[0094] The disease information extraction module 30 is used to extract the disease information of the patient according to the identity information, and obtain the corresponding chronic disease qualification information according to the disease information; wherein the chronic disease qualification information is the authority information of the reimbursement of the medicine corresponding to the disease information;

[0095] The target distance calculation module 40 is used to input the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information; wherein the chronic disease qualification reimbursement model is trained based on various chronic disease types and the insured drugs corresponding to various chronic disease types;

[0096] A target distance determination module 50 is used to determine whether there is a target drug whose target distance is less than a first preset target distance among the drugs;

[0097] The reimbursement module 60 is used to determine that the target drug is a reimbursable drug for a patient's chronic disease if the target drug has a target distance that is less than the first preset target distance.

[0098] Obtain the text information on the prescription. The prescription can be an electronic prescription, a paper prescription, or a picture containing the prescription. For an electronic prescription, the text information therein can be directly obtained; for a paper prescription, its scanned copy can be obtained through a printer, or a photo can be taken through a camera to obtain its corresponding picture containing the prescription; for pictures and scanned copies, since they are both in picture format, the text information in the picture can be extracted through a text extraction method, thereby achieving the acquisition of the text information on the prescription.

[0099] The patient's identity information and drug information are obtained based on the text information, and the text information is extracted based on semantic recognition. Specifically, the text information is preprocessed first, including unifying the language of the text information, replacing synonyms, deleting unnecessary sentences, etc. The identity information is generally in the name column before the prescription, which can be directly identified and extracted, and then the corresponding drug information is extracted based on the drug vocabulary, where the drug vocabulary includes the name and / or alias of each drug.

[0100] Then, the patient's disease information is extracted from the chronic disease database based on the identity information, where the chronic disease database records the information of all chronic diseases suffered by the patients. Then, the corresponding chronic disease qualification information is obtained based on the disease information, where the chronic disease qualification information is the reimbursable drug authority information for the corresponding chronic disease.

[0101] Since chronic diseases can be treated with different medicines, even for the same chronic disease, there are many kinds of corresponding reimbursement medicines. Therefore, it is necessary to calculate the target distance between each medicine and the chronic disease qualification information. Since there are many kinds of medicines, directly comparing each medicine with the chronic disease qualification information will result in a large amount of calculation; and for the same type of medicines, they can generally treat the chronic disease. Therefore, the corresponding drug category can be directly detected, and then the target distance between the drug category and the chronic disease qualification information can be calculated based on the drug category. Of course, some drugs may have therapeutic effects, but they are not within the scope of reimbursement. Later, it can be determined whether the drug can be reimbursed based on the chronic disease reimbursement drug library.

[0102] Determine whether there is a target drug with a distance less than a first preset target distance, where the first preset target distance is a pre-set value. When it is less than this value, it can be considered that the drug is a chronic disease reimbursement drug corresponding to the chronic disease qualification information, so this part of the drug included in the prescription can be extracted for reimbursement.

[0103] In one embodiment, the device for identifying chronic disease drugs further includes:

[0104] The medical information acquisition module is used to obtain the medical information of various chronic disease qualification information in the outpatient database and the outpatient chronic disease database;

[0105] The medical information input module is used to input the medical information as training data into the training model, wherein different types of chronic disease information are trained through corresponding calculation formulas, and after the training is completed, the chronic disease qualification reimbursement model is obtained, and the calculation formula is: W k represents the target distance of the kth chronic disease, N is the size of the training data, X ij Represents word vector a j In the word vector a i The number of times the context appears, f(X ij ) is the weighting function, a i represents the i-th word vector, represents the independent j-th context word vector, b i , Indicates the parameters after training, log n X ij The n in represents the parameters after training.

[0106] Since outpatient and chronic disease diagnosis and prescription are similar to each other, the medical information of outpatient and chronic disease can be selected as training data to train the corresponding training model. It should be noted that the training model may include multiple calculation formulas, each of which corresponds to a chronic disease information. This embodiment takes the first chronic disease information as an example, and it may also include multiple chronic disease information, which correspond to different calculation formulas. After obtaining the chronic disease information of the patient, different calculation formulas in the trained chronic disease qualification reimbursement model are used to calculate the target distance between each drug and the chronic disease information. In the formula, you only need to input the extracted text information into the formula. During the calculation process, the formula In the formula, multiple corresponding word vectors are obtained according to the text information (i.e., a i ), is the word vector related to the first chronic disease information. After training, it can be regarded as a certain parameter value. i , And log n X ij can be regarded as specific parameter values, f(X ij ) is a weighting function, which can be regarded as i The relevant parameter values ​​can be considered as a i The value on which the function is calculated, which in some embodiments may be a constant of 1.

[0107] In one embodiment, the target distance calculation module 40 includes:

[0108] The drug judgment submodule is used to judge whether each drug is recorded in the chronic disease reimbursement drug database based on the information of each drug;

[0109] A reimbursement drug extraction submodule is used to extract reimbursement drugs recorded in the chronic disease reimbursement drug database from among the drugs;

[0110] A drug category information acquisition submodule is used to acquire the corresponding drug category information according to the information of the reimbursed drug;

[0111] The target distance calculation submodule is used to input the drug category information into the chronic disease qualification reimbursement model for calculation to obtain the target distance.

[0112] The drugs can be filtered first, that is, some drugs do not belong to the chronic disease reimbursement drug database, such as ordinary cold medicines, but these drugs are included in the prescription. Therefore, the drug information on the prescription can be directly judged. If it is not in the chronic disease reimbursement drug database, then this part of the drug information can be directly eliminated. There is no need to input it into the chronic disease qualification reimbursement model for calculation. It is only necessary to input the reimbursement drugs recorded in the chronic disease reimbursement drug database into the chronic disease qualification reimbursement model for calculation. This can reduce the amount of calculation of the model and help to quickly obtain the final target drug.

[0113] In one embodiment, the target distance calculation module 40 includes:

[0114] A drug information input submodule, for inputting the drug information into the calculation formulas corresponding to the various chronic disease information respectively when the disease information includes various chronic disease information;

[0115] Multiple target distance calculation submodules are used to calculate multiple target distances corresponding to each drug in the drug information and the multiple chronic disease information.

[0116] A patient may have multiple chronic diseases, and the corresponding disease information will also include multiple chronic disease information. The patient has reimbursement rights corresponding to multiple chronic disease information. Therefore, each drug can be input into different calculation formulas and calculated separately, that is, different target distances for each drug in different calculation formulas are obtained. It should be understood that among the target distances corresponding to the drug, as long as one target distance is greater than the first preset target distance, the drug can be considered as a target drug.

[0117] In one embodiment, the text information acquisition module 10 includes:

[0118] A picture information acquisition submodule, used to acquire picture information corresponding to the prescription;

[0119] a value acquisition submodule, used for acquiring the values ​​of the R color channel, the G color channel and the B color channel in the RGB color model of the pixel points in the picture, and setting the RGB color of the pixel points in the picture to (0,0,0), (255,255,255) or (P,P,P) according to a preset ternary method, where P is a preset value greater than 0 and less than 255, so as to obtain a temporary picture composed of three colors;

[0120] An area calculation submodule, used to calculate the areas occupied by the three colors in the temporary image, and to use a preset text segmentation method for the areas occupied by the two colors with smaller areas, so as to obtain segmented single handwritten characters and segmented single printed characters;

[0121] The text feature extraction submodule is used to extract the text features of the single handwritten text and the text features of the single printed text, and input them into a preset support vector machine for classification, so as to obtain the handwritten text and the printed text in the image.

[0122] The corresponding image information is obtained through a printer or a camera, and then the handwritten text and printed text obtained by recognition are obtained by a ternary method. Since the prescription may be handwritten by the doctor, in order to more accurately distinguish handwritten text from printed text, the present application uses a ternary method, that is, according to the preset ternary method, the RGB color of the pixel point in the picture is set to (0,0,0), (255,255,255) or (P,P,P), where P is a preset value greater than 0 and less than 255, thereby obtaining a temporary picture composed of three colors, and calculating the area occupied by the three colors in the temporary picture, and using a preset text segmentation method for the areas occupied by the two colors with smaller areas (since the area with the largest area is definitely the background, there is no need to analyze the area with the largest area), thereby obtaining a segmented single handwritten text and a segmented single printed text. The support vector machine is a generalized linear classifier that performs binary classification of data in a supervised learning manner, which is suitable for comparing the recognized text with the pre-stored text to output the most similar text. Based on this, the text features of the single handwritten text and the single printed text are extracted and input into a preset support vector machine for classification, thereby obtaining the handwritten text and printed text contained in the recognition image. The text features are, for example, special points such as extreme points, isolated points, etc. in the pixel points corresponding to the text.

[0123] In one embodiment, the device for identifying chronic disease reimbursement drugs further includes:

[0124] A common medicine judging module, used for judging whether there is a common medicine whose target distance is greater than a second preset target distance among the medicines;

[0125] The non-reimbursement identification module is used to identify the common medicine as non-reimbursable if the common medicine is available.

[0126] When the calculated target distance is less than the second preset target distance, it can be considered as an ordinary drug, where the second preset target distance is a pre-set value. When it is greater than this value, it can be considered as an ordinary drug. Ordinary drugs are drugs that are not reimbursed for chronic diseases. In this article, drugs that are not reimbursed for chronic diseases only refer to drugs that cannot be reimbursed for chronic diseases, not drugs that cannot be reimbursed for other items. Then this part of ordinary drugs is considered non-reimbursable.

[0127] In one embodiment, the device for identifying chronic disease reimbursement drugs further includes:

[0128] A suspicious drug judging module, used for judging whether there is a suspicious drug among the drugs whose target distance is less than the second preset target distance and greater than the first preset distance target;

[0129] The suspicious drug sending module is used to extract the suspicious drug if it is available and send it to the processing terminal for identification.

[0130] In order to improve the recognition of drugs, this article will set the first preset target distance small enough to ensure that drugs smaller than the first preset target distance are target drugs, and set the second preset target distance large enough to ensure that drugs larger than the second preset target distance are ordinary drugs. Drugs between the first preset target distance and the second preset target distance are defined as suspicious drugs, which may be target drugs or ordinary drugs. Therefore, the suspicious drugs can be extracted and sent to relevant personnel for further identification. After identification, the relevant personnel will directly screen out the target drugs and then integrate all target drugs for reimbursement identification processing.

[0131] The beneficial effects of the present invention are as follows: by obtaining the patient's identity information, the corresponding chronic disease qualification information is obtained in the database, and then the drug information and chronic disease qualification information on the prescription are input into the chronic disease qualification reimbursement model for calculation, and the target drug related to the chronic disease qualification information is obtained, and then the target drug is reimbursed. Compared with setting rules for supervision, a certain standard is set for reimbursement of drugs, which reduces the abuse of supervision and saves a lot of time and manpower costs.

[0132] Reference Figure 3 In an embodiment of the present application, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store various chronic disease reimbursement drug data, etc. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the method for identifying chronic disease reimbursement drugs described in any of the above embodiments can be implemented.

[0133] Those skilled in the art will understand that Figure 3The structure shown in is merely a block diagram of a portion of the structure related to the present application solution and does not constitute a limitation on the computer device to which the present application solution is applied.

[0134] An embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the method for identifying chronic disease reimbursement drugs described in any of the above embodiments can be implemented.

[0135] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0136] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, device, article or method including the element.

[0137] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention shall be included in the scope of the claims of the present invention.

Claims

1. A method for identifying reimbursement drugs for chronic diseases, characterized in that: include: Get the text information on the prescription; Acquire the identity information and drug information of the corresponding patient according to the text information; Extracting the patient's disease information according to the identity information, and acquiring corresponding chronic disease qualification information according to the disease information; wherein the chronic disease qualification information is the authority information for reimbursement of drugs corresponding to the disease information; The drug information and the chronic disease qualification information are input into a chronic disease qualification reimbursement model for calculation to obtain a target distance between each drug in the drug information and the chronic disease qualification information; wherein the chronic disease qualification reimbursement model is trained based on various chronic disease types and the insured drugs corresponding to the various chronic disease types; Determine whether there is a target drug among the drugs whose target distance is less than a first preset target distance; If the target drug has a target distance that is less than the first preset target distance, the target drug is determined to be a chronic disease reimbursement drug for the patient; Before the step of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information, the step further includes: Obtain the medical information of various chronic diseases in the outpatient database and outpatient chronic disease database; The medical information is input into the training model as training data, wherein different types of chronic disease information are trained through corresponding calculation formulas. After the training is completed, the chronic disease qualification reimbursement model is obtained. The calculation formula is: W k represents the target distance of the kth chronic disease, N is the size of the training data, X ij Represents word vector a j In the word vector a i The number of times the context appears, f(X ij ) is the weighting function, a i represents the i-th word vector, represents the independent j-th context word vector, b i , Indicates the parameters after training, log n X ij The n in represents the parameters after training.

2. The method for identifying reimbursable drugs for chronic diseases according to claim 1, characterized in that: The step of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information includes: Based on the information of each drug, determine whether each drug is recorded in the chronic disease reimbursement drug database; Extracting reimbursable drugs recorded in the chronic disease reimbursement drug database from among the drugs; Obtaining corresponding drug category information according to the information of the reimbursed drugs; The drug category information is input into the chronic disease qualification reimbursement model for calculation to obtain the target distance.

3. The method for identifying reimbursable drugs for chronic diseases according to claim 1, characterized in that: The step of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information includes: When the disease information includes multiple chronic disease information, the drug information is respectively input into the calculation formulas corresponding to the multiple chronic disease information; Multiple target distances corresponding to each drug in the drug information and the multiple chronic disease information are calculated.

4. The method for identifying reimbursable drugs for chronic diseases according to claim 1, characterized in that: The step of obtaining text information on the prescription includes: Obtaining image information corresponding to the prescription; Collecting the values ​​of the R color channel, the G color channel, and the B color channel in the RGB color model of the pixel points in the picture, and setting the RGB colors of the pixel points in the picture to (0,0,0), (255,255,255), or (P,P,P) according to a preset ternary method, where P is a preset value greater than 0 and less than 255, thereby obtaining a temporary picture composed of three colors; Calculating the areas occupied by the three colors in the temporary image, and using a preset text segmentation method for the areas occupied by the two colors with smaller areas, thereby obtaining segmented single handwritten characters and segmented single printed characters; The text features of the single handwritten text and the text features of the single printed text are extracted and input into a preset support vector machine for classification, thereby obtaining the handwritten text and the printed text in the image.

5. The method for identifying reimbursable drugs for chronic diseases according to claim 1, characterized in that: After the step of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information, the method further includes: Determining whether there is a common medicine among the medicines whose target distance is greater than a second preset target distance; If the above-mentioned common medicines are available, it is deemed that the above-mentioned common medicines cannot be reimbursed.

6. The method for identifying reimbursable drugs for chronic diseases according to claim 1, characterized in that: After the step of inputting the drug information and the chronic disease qualification information into the chronic disease qualification reimbursement model for calculation to obtain the target distance between each drug in the drug information and the chronic disease qualification information, the method further includes: Determine whether there is a suspicious drug among the drugs whose target distance is less than the second preset target distance and greater than the first preset distance target; If the suspicious drug is present, the suspicious drug is extracted and sent to the processing terminal for identification.

7. A device for identifying chronic disease drugs, used to implement the method according to any one of claims 1 to 6, characterized in that: include: A text information acquisition module is used to acquire text information on the prescription; A patient information acquisition module, which acquires the identity information and drug information of the corresponding patient according to the text information; A disease information extraction module is used to extract the disease information of the patient according to the identity information, and obtain the corresponding chronic disease qualification information according to the disease information; wherein the chronic disease qualification information is the authority information of the reimbursement of drugs corresponding to the disease information; A target distance calculation module is used to input the drug information and the chronic disease qualification information into a chronic disease qualification reimbursement model for calculation to obtain a target distance between each drug in the drug information and the chronic disease qualification information; wherein the chronic disease qualification reimbursement model is trained based on various chronic disease types and insured drugs corresponding to various chronic disease types; A target distance judgment module, used to judge whether there is a target drug whose target distance is less than a first preset target distance among the drugs; The reimbursement module is used to determine that the target drug is a reimbursable drug for a patient's chronic disease if the target drug has a target distance that is less than the first preset target distance.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images