CN109615204B

CN109615204B - Quality evaluation method, device and equipment of medical data and readable storage medium

Info

Publication number: CN109615204B
Application number: CN201811462376.9A
Authority: CN
Inventors: 黄越; 陈明东
Original assignee: Ping An Medical and Healthcare Management Co Ltd
Current assignee: Ping An Medical and Healthcare Management Co Ltd
Priority date: 2018-11-30
Filing date: 2018-11-30
Publication date: 2023-02-03
Anticipated expiration: 2038-11-30
Also published as: CN109615204A

Abstract

The invention provides a quality evaluation method, a device, equipment and a readable storage medium of medical data based on the idea of big data analysis and processing, wherein the method comprises the following steps: reading the basic data sheet, the medical insurance data sheet and the treatment data sheet of each patient to be treated, and setting the basic data sheet, the medical insurance data sheet and the treatment data sheet as group elements to form a medical data group; performing data reasonability detection, data correspondence detection and audit relation detection on each group of elements in the medical data group respectively to generate a detection result; and determining a target quality grade corresponding to the generated detection result according to the corresponding relation between the preset detection result and the quality grade, and performing quality evaluation on the medical data set according to the target quality grade. According to the scheme, the detection results of data reasonability, correspondence and checking relation of all groups of elements in the medical data group are used as the quality evaluation basis of the medical data, so that the evaluation is more accurate, and the efficiency and the degree of automation of the evaluation are improved.

Description

Quality evaluation method, device and equipment of medical data and readable storage medium

Technical Field

The invention mainly relates to the technical field of medical systems, in particular to a method, a device and equipment for evaluating the quality of medical data and a readable storage medium.

Background

With the development of social security system, more and more people have medical insurance and use the medical insurance to see a doctor, each person can generate various medical data when each medical institution uses the medical insurance to see a doctor, the various medical data are stored in different data tables for classified management, and the medical insurance fund is subsequently calculated and planned according to the medical data.

However, as the medical data of each medical staff is dispersed in each data table, once a certain data table lacks a certain piece of medical data of the medical staff, or the corresponding relationship of the medical data among the medical staff is incorrect, the relationship of each data in the data table and each data among the data tables is easily disordered, so that a large error exists in the calculation and planning of the medical insurance fund; therefore, the quality evaluation of the integrity and the corresponding relation of each data in the data table is particularly important. However, the existing evaluation scheme for the quality of the data mainly depends on manual operation, and the data in each data table is checked and evaluated in a manual mode, so that errors are easy to occur in checking, and inaccurate evaluation and low efficiency are caused.

Disclosure of Invention

The invention mainly aims to provide a method, a device and equipment for evaluating the quality of medical data and a readable storage medium, and aims to solve the problems of inaccurate quality evaluation and low efficiency of the medical data in the prior art.

In order to achieve the above object, the present invention provides a method for evaluating the quality of medical data, comprising the steps of:

reading a basic data sheet, a medical insurance data sheet and a treatment data sheet of each patient to be treated, and setting the basic data sheet, the medical insurance data sheet and the treatment data sheet as group elements to form a medical data group;

performing data rationality detection, data correspondence detection and audit relationship detection on each group of elements in the medical data group respectively to generate detection results;

and determining a target quality grade corresponding to the generated detection result according to a preset corresponding relation between the detection result and the quality grade, and performing quality evaluation on the medical data set according to the target quality grade.

Preferably, the detection result comprises a first detection result, a second detection result and a third detection result;

the steps of respectively carrying out data rationality detection, data correspondence detection and audit relationship detection on each group of elements in the medical data group and generating detection results comprise:

performing data rationality detection on each group of elements by taking data as a unit respectively to generate a first detection result;

performing data correspondence detection on each group of elements by using a data behavior unit respectively to generate a second detection result;

checking the cells with the operation data in each group of elements to generate a third detection result.

Preferably, the data reasonability detection includes null data detection and numerical range detection, the data reasonability detection is performed on each group of elements by using data columns as units, and the step of generating the first detection result includes:

respectively traversing the data columns in the group elements one by one, reading the column identifiers of the data columns, and determining the numerical value type data columns in the data columns according to the column identifiers;

detecting whether a cell with null data exists in each data column, and if the cell with null data exists, adding a first abnormal identifier to each cell;

detecting whether target data with values exceeding a preset range exists in each numerical value type data column or not, and if the target data with the values exceeding the preset range exists, adding a second abnormal identifier to the target data;

and generating a first detection result according to the number of the first abnormal identifiers and the second abnormal identifiers.

Preferably, the step of performing data correspondence detection on each group of elements by using a data line unit respectively and generating a second detection result includes:

judging whether a first abnormal rate in the first detection result is greater than a preset value or not, and if the first abnormal rate is greater than the preset value, stopping the corresponding detection of each group of elements;

if the first abnormal rate is not greater than the preset value, determining first corresponding data corresponding to each first abnormal identifier and second corresponding data corresponding to each second abnormal identifier according to the corresponding relation between the data in each group of elements;

and removing the first corresponding data and the second corresponding data from each group element to update each group element, and performing data correspondence detection on each group element by using a data behavior unit to generate a second detection result.

selecting any one group element from each group element as a target group element, and setting other group elements except the target group element in each group element as group elements to be compared;

reading a first row identifier of a data row in the target group element and a second row identifier of each data row in each group element to be compared, and judging whether a target identifier corresponding to the first row identifier exists in each second row identifier of each group element to be compared or not;

if the target identifiers corresponding to the first row identifiers exist, finishing data correspondence detection of each group element;

if there are no object identifiers each corresponding to the first line identifier, a third anomaly identifier is added to a data line having the first line identifier corresponding to the object identifier, and a second detection result is generated from the number of the third anomaly identifiers.

Preferably, the step of checking cells having operation data in each of the group elements to generate a third detection result includes:

reading cell identifiers of cells in each group of elements, and determining target cells with operational data according to the cell identifiers;

detecting whether target cell data with abnormal collusion relation detection exists in the cell data of each target cell;

and if the target cell data with abnormal collusion relation detection exists, adding a fourth abnormal identifier to the target cell data, and generating a third detection result according to the number of the fourth abnormal identifiers.

Preferably, the step of determining a target quality level corresponding to the generated detection result according to a preset correspondence between the detection result and the quality level includes:

reading a weighting coefficient corresponding to the first abnormality rate, a second abnormality rate in the second detection result, and a third abnormality rate in the third detection result;

respectively carrying out weighted calculation on the first abnormality rate, the second abnormality rate and the third abnormality rate by using the weighting coefficients to generate a target detection result;

and comparing the target detection result with a preset corresponding relation between the detection result and the quality grade, and determining the target quality grade corresponding to the target detection result.

In addition, in order to achieve the above object, the present invention also provides a quality evaluation apparatus for medical data, including:

the reading module is used for reading the basic data sheet, the medical insurance data sheet and the treatment data sheet of each patient to be treated, and setting the basic data sheet, the medical insurance data sheet and the treatment data sheet as group elements to form a medical data group;

the detection module is used for respectively carrying out data rationality detection, data correspondence detection and audit relationship detection on each group of elements in the medical data group to generate a detection result;

and the evaluation module is used for determining a target quality grade corresponding to the generated detection result according to the corresponding relation between the preset detection result and the quality grade, and evaluating the quality of the medical data group according to the target quality grade.

Further, to achieve the above object, the present invention also proposes a quality evaluation apparatus for medical data, comprising: a memory, a processor, a communication bus, and a quality assessment program of medical data stored on the memory;

the communication bus is used for realizing connection communication between the processor and the memory;

the processor is configured to execute a quality assessment procedure of the medical data to implement the steps of:

performing data reasonability detection, data correspondence detection and audit relationship detection on each group of elements in the medical data group respectively to generate a detection result;

Further, to achieve the above object, the present invention also provides a readable storage medium storing one or more programs, the one or more programs being executable by one or more processors for:

In the quality evaluation method of medical data of this embodiment, a medical data group of a patient to be treated is formed by taking the read basic data sheet, medical insurance data sheet and treatment data sheet of each patient to be treated as group elements; respectively carrying out data reasonability detection, data correspondence detection and audit relation detection on each group element in the medical data group to generate a detection result; and presetting a corresponding relation between the detection result and the quality grade, determining a corresponding target quality grade according to the corresponding relation after the detection result is generated, and further performing quality evaluation on the medical data group according to the target quality grade. The internal data of each group of elements and the integrity of the data among the elements and the correctness of the corresponding relation are reflected by a generated detection result by detecting the data reasonability, the correspondence and the checking relation of each group of elements in the medical data group; taking the target quality grade corresponding to the detection result as a basis to carry out quality evaluation on the medical data; the checking and evaluation in a manual mode are avoided, so that the evaluation of the medical data is more accurate, and the evaluation efficiency and the automation degree are improved.

Drawings

FIG. 1 is a schematic flow chart diagram of a first embodiment of a method for quality assessment of medical data according to the present invention;

FIG. 2 is a functional block diagram of a first embodiment of the medical data quality assessment apparatus of the present invention;

fig. 3 is a schematic device structure diagram of a hardware operating environment related to the method according to the embodiment of the present invention.

The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.

Detailed Description

It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The invention provides a quality evaluation method of medical data.

Referring to fig. 1, fig. 1 is a schematic flow chart of a medical data quality evaluation method according to a first embodiment of the present invention. In this embodiment, the method for evaluating the quality of medical data includes:

step S10, reading a basic data sheet, a medical insurance data sheet and a medical treatment data sheet of each patient to be treated, and setting the basic data sheet, the medical insurance data sheet and the medical treatment data sheet as group elements to form a medical data group;

the quality evaluation method of the medical data is applied to the server and is suitable for evaluating the quality of the medical data of the medical institution through the server; the medical data is data recorded by medical institutions on the basic information and the treatment process of patients, such as the name, age, sex, identification number, treatment place, treatment time, disease information, medication information, self-payment amount of patients, medical insurance reimbursement amount and the like of the patients. The medical institution records the data for each patient to be treated, classifies the data according to the basic data, the medical insurance data and the treatment data in the recording process, and respectively generates a basic data table, a medical insurance data table and a treatment data table so as to classify and manage the medical data of each patient to be treated; meanwhile, the same identifier is allocated to the data belonging to the same patient in each table so as to establish the association relationship of the patient in each table. For example, for the patient A, the identifier a is distributed to personal information such as the name, age, sex, identification number and the like of the patient A, and the personal information is recorded in a basic data table; the identifier a is also distributed to the medication information, the patient self-payment amount and the medical insurance reimbursement amount and is recorded in the medical insurance data sheet; and (4) assigning an identifier a to the visit place, the visit time and the disease information, and recording the identifier a into the visit data table. In order to ensure the uniqueness of the identifier of the patient, a medical record is established for the patient who initially visits the medical institution, and the uniqueness of the patient is characterized by the identification number and the identifier of the patient in the medical record. When a patient goes to a medical institution for a doctor, whether a medical file corresponding to the identity card number of the patient exists is judged, if yes, the identifier of the medical file is read and used as the identifier of the doctor data generated by the patient in the doctor process, if not, the medical file is established for the patient, and the identifier is distributed, so that the subsequent doctor data of the patient can be recorded.

In order to evaluate the quality of the recorded treatment data of each patient treated by the medical institution, a detection mechanism with interval time is set, such as detection once a month or detection once a quarter, and the like, and the quality evaluation is carried out according to the detection result. Specifically, when the inter-arrival time is detected, the server reads a basic data sheet, a medical insurance data sheet and a treatment data sheet of each patient recorded by the medical institution; because the data with the same identifier in each data table are different types of data of the same patient, the data tables have an association relation; and the read basic data sheet, medical insurance data sheet and treatment data sheet form a medical data group. Each data sheet is used as a group element in the medical data group, and the medical data of each patient is evaluated in quality by taking the medical data group as a whole.

Step S20, performing data rationality detection, data correspondence detection and audit relationship detection on each group of elements in the medical data group respectively to generate detection results;

further, after the basic data sheet, the medical insurance data sheet and the medical treatment data sheet are used as group elements to form a medical data group, the data in each group element is subjected to reasonability detection, correspondence detection and checking relation detection, and a detection result is generated. The reasonability detection is to determine whether each data in the group elements is null or not, whether the numerical range of each data is reasonable or not and the like so as to ensure the completeness and reasonability of the recorded medical data; the correspondence detection is to determine whether corresponding data exists in each group of elements of the patient to be treated so as to ensure the correctness of the data correspondence relationship among each group of elements; the checking relationship is used for detecting whether the operation result of the data needing to be operated in the data inside the group elements or the data among the group elements is correct or not so as to ensure the correctness of the operation of each data in the medical data group. For the different types of detection, the detection modes have differences, and the generated detection results are different; specifically, the detection results include a first detection result, a second detection result and a third detection result, the data rationality detection, the data correspondence detection and the checking relationship detection are respectively performed on each group of elements in the medical data group, and the step of generating the detection results includes:

step S21, performing data rationality detection on each group of elements by taking data as a unit respectively to generate a first detection result;

understandably, the elements of each group are in the form of a data table, and the data in the data table are in the form of "rows and columns", wherein "row" corresponds to each type of data of the same patient and "column" corresponds to each data of the same type. Considering that different types of data have different rationality requirements, such as for data of numerical type, firstly requiring that the data is not null, and simultaneously requiring that the data is within a reasonable range, and for data of different numerical types, the reasonable range is also different; for non-numeric data, the data is not required to be null. Each column in the data table represents the type of each data, so that when the rationality of the data in each group of elements is detected, the data is detected as a unit, and the rationality is detected according to each data type; namely, the reasonability detection of the data is carried out on each group element by taking the data column as a unit, and a first detection result is generated.

Step S22, performing data correspondence detection on each group of elements by using a data behavior unit respectively to generate a second detection result;

furthermore, the correspondence detection is used for detecting whether the patients to be diagnosed have corresponding data in each group of elements, and each row in the data table represents each data type of the patients to be diagnosed, so that when the data in each group of elements are subjected to the correspondence detection, the data are detected by a data row unit, and the correspondence detection is conveniently carried out according to each patient to be diagnosed; namely, the data correspondence detection is carried out on each group element in a data line unit, and a second detection result is generated.

Step S23, checking the cells with operation data in each group of elements to generate a third detection result.

Understandably, the data of the patient in the treatment process has the condition of operation; if the sum of the medication amounts in the medication information is equal to the total medication amount, the sum of the self-payment amount of the patient and the medical insurance reimbursement amount is equal to the total treatment amount, and the like; the data needing to be operated are taken as operation data, and different operation data exist in different cells of each group of elements in the form of a data table. In order to ensure the correctness of such operations, checking the cells with operation data is required; the checking relation detection is used for judging whether the operation between the operation data is correct or not and generating a third detection result.

Furthermore, performing reasonability detection, correspondence detection and construction check relation detection on the data in the group elements, wherein the generated first detection result, the second detection result and the third detection result are all components of the detection results; and generating a detection result by performing weighted integration on the first detection result, the second detection result and the third detection result, and further evaluating the quality of the medical data according to the quality grade corresponding to the generated detection result.

And S30, determining a target quality grade corresponding to the generated detection result according to a preset corresponding relation between the detection result and the quality grade, and performing quality evaluation on the medical data set according to the target quality grade.

Further, the embodiment is preset with a corresponding relationship between the detection result and the quality grade, the detection result indicates that the group element has an abnormal data quantity which is unreasonable, not corresponding or incorrect in operation, and the corresponding relationship between the detection result and the quality grade indicates that the abnormal data quantity in the group element has an influence on the quality of the medical data group. When the quantity of abnormal data in the detection result is more, the corresponding quality grade is lower, and the quality of the medical data recorded by the representation is worse; and the smaller the number of abnormal data, the higher the corresponding quality grade, and the better the quality of the medical data recorded by the representation. Comparing the generated detection result with a preset corresponding relation, and determining the quality grade corresponding to the generated detection result; the corresponding quality grade is taken as a target quality grade to represent each group element, namely the quality of each medical data recorded in the medical data group.

In the quality evaluation method of medical data of this embodiment, a medical data group of a patient to be treated is formed by taking the read basic data sheet, medical insurance data sheet and treatment data sheet of each patient to be treated as group elements; respectively carrying out data reasonability detection, data correspondence detection and audit relation detection on each group element in the medical data group to generate a detection result; the corresponding relation between the detection result and the quality grade is preset, after the detection result is generated, the corresponding target quality grade can be determined according to the corresponding relation, and then the quality of the medical data set is evaluated according to the target quality grade. The quality of the medical data group is evaluated substantially by the medical data formed by various types of treatment data in the treatment process; the internal data of each group of elements and the integrity of the data among the elements and the correctness of the corresponding relation are reflected by a generated detection result by detecting the data reasonability, the correspondence and the checking relation of each group of elements in the medical data group; taking the target quality grade corresponding to the detection result as a basis to evaluate the quality of the medical data; the checking and evaluation in a manual mode are avoided, so that the evaluation of the medical data is more accurate, and the evaluation efficiency and the automation degree are improved.

Further, in another embodiment of the method for evaluating the quality of medical data according to the present invention, the data reasonableness test includes null data test and numerical range test, the data reasonableness test is performed on each of the group elements by using data columns as units, and the step of generating the first test result includes:

step S211, respectively traversing the data columns in each group element one by one, reading the column identifier of each data column, and determining the numerical value type data column in each data column according to each column identifier;

understandably, when the data reasonability detection is carried out on each group element, the reasonability detection of the numerical range is required besides the null value data detection aiming at the numerical type data column in the group element; in order to distinguish the numerical value type data column and the non-numerical value type data column in the group element, a column identifier is preset for each data column, and the numerical value type and the non-numerical value type are characterized by different column identifiers. Before the rationality detection is carried out, data columns in each group of elements are traversed one by one, column identifiers in each data column are read, the read column identifiers are compared with preset column identifiers, column identifiers representing numerical value types are searched, and column data with the identifiers are determined as numerical value type data columns.

Step S212, detecting whether each data column has a cell with null data, and if yes, adding a first abnormal identifier to each cell;

further, both the numeric type data column and the non-numeric type data column in the group element need to perform null detection, such as detecting whether cells which are not recorded exist in the non-numeric type data column for recording gender and the numeric type data column for recording age; that is, each data column in the group element is detected, and it is determined whether or not a cell in which the data is null exists. If the cell with the null data exists, the cell does not record data, and the data in the data column is incomplete; thus, a first exception identifier is added to each cell where no data is recorded to characterize the exception data present in each set of elements.

Step S213, detecting whether target data with a numerical value exceeding a preset range exists in each numerical value type data column, and if the target data with the numerical value exceeding the preset range exists, adding a second abnormal identifier to the target data;

after null value detection is carried out on each data column in the group elements, the numerical value range detection is further carried out on the numerical value type data column; the detection is carried out according to the numerical range corresponding to the numerical type data because different numerical types have different numerical ranges, such as the numerical range of the age is 0-150 years, the expense amount of a certain type of medicine is 30-90 yuan, and the like. When the numerical value type data column is determined according to the read column identifier, the identifier representing the numerical value type data column correspondingly carries a numerical value range corresponding to the data column; taking the corresponding numerical range as a preset range of the data column, comparing the numerical value of each data in the data column with the preset range, judging whether the numerical value of each data is in the preset range, and if so, indicating that the numerical range of each data in the data column is reasonable; if the numerical value of some data in the numerical values of the data exceeds the preset range, the numerical range of the data is unreasonable, the data is used as target data in the numerical type data column, and a second abnormal identifier is added to the target data of which the numerical value exceeds the preset range to represent abnormal data in each group of elements.

Step S214, generating a first detection result according to the number of the first anomaly identifiers and the second anomaly identifiers.

Further, after null detection and numerical range detection are carried out on data of all data columns in each group of elements, and a first abnormal identifier and a second abnormal identifier are added to the cells and the target data which are detected to be null; through the first anomaly identifier and the second anomaly identifier, a first detection result can be generated, and the first detection result represents the proportion of the number of unreasonable anomalous data in the group elements. Specifically, respectively counting each first quantity of the first abnormal identifiers and the second quantity of the second abnormal identifiers, and then adding the first quantity and the second quantity, wherein the result obtained by adding is the first abnormal total quantity of the unreasonable abnormal data in the group elements; then, counting the total data amount of the data in the group elements, and taking the first abnormal total amount and the total data amount as a ratio, wherein the obtained ratio result is the ratio of the number of unreasonable abnormal data in the group elements; and taking the ratio result as a first abnormal rate, and forming a set by using data reflecting abnormality, such as the first abnormal total amount, the first abnormal rate, the first abnormal identifier number, the second abnormal identifier number and the like, as a first detection result to show the magnitude of the data with unreasonable abnormality in the group elements.

Further, in another embodiment of the method for evaluating the quality of medical data according to the present invention, the step of performing data correspondence detection on each group of elements by using data line units respectively to generate the second detection result includes:

step S24, judging whether a first abnormal rate in the first detection result is greater than a preset value, and if the first abnormal rate is greater than the preset value, stopping performing corresponding detection on each group of elements;

understandably, after the data of each data column in the group elements are subjected to rationality detection and the first detection result is generated, the number of the abnormal data in the group elements reflected by the first detection result is possibly higher, that is, the group elements contain more unreasonable data. If the medical data is continuously subjected to the correspondence detection and the checking relation detection, the reference value is low; therefore, in order to represent the proportion of the quantity of the abnormal data reflected in the first detection result, a preset numerical value is preset. Comparing the first abnormal rate in the first detection result with the preset value, and judging whether the first abnormal rate is greater than the preset value; if the first abnormal rate is larger than the preset value, unreasonable abnormal data in the group elements are more, and the detection of the correspondence of each group element is stopped.

Step S25, if the first abnormal rate is not larger than the preset value, determining first corresponding data corresponding to each first abnormal identifier and second corresponding data corresponding to each second abnormal identifier according to the corresponding relation between the data in each group of elements;

and when the first abnormal rate is judged to be not more than the preset value, unreasonable abnormal data in the group elements are less, the number ratio is within an acceptable reasonable range, and the group elements are subjected to corresponding detection. Specifically, before the correspondence detection is performed, unreasonable abnormal data in the group elements are removed to avoid affecting the correspondence detection. Because there are patients with identical data rows among the elements of each group, when abnormal data in one group of elements is removed, data of patients with identical abnormal data in other groups of elements needs to be correspondingly removed. And assigning the same identifier to the data of the same patient in each group of elements, taking the same identifier as the corresponding relation among the data in each group of elements, and determining other data of the patient to which the abnormal data belongs according to the corresponding relation. For example, if the data B in the data sheet for medical treatment is not in the preset range and the added identifier is B, all other data carrying the identifier B in the data sheet for medical treatment are taken as data to be eliminated; and meanwhile, searching data carrying the identifier b in the basic data table and the medical insurance data table to be used as data needing to be eliminated. The abnormal data comprises null value data and data with unreasonable numerical range, so that the determined data needing to be rejected are divided into first corresponding data corresponding to a first abnormal identifier and second corresponding data corresponding to a second abnormal identifier according to the two types, namely other data of the patient to be diagnosed to which the null value data belongs and other data of the patient to be diagnosed with unreasonable numerical range.

Step S26, removing the first corresponding data and the second corresponding data from each group element to update each group element, and performing data correspondence detection on each group element by using a data line unit to generate a second detection result.

Further, after determining the first corresponding data and the second corresponding data in each group of elements, removing the first corresponding data and the second corresponding data from each group of elements, and updating each group of elements after the removing operation into a new group of elements; and on the basis of the new group element, performing data correspondence detection by using a data line unit to generate a second detection result. Specifically, step S22 includes:

step S221, selecting any one of the group elements from each of the group elements as a target group element, and setting other group elements except the target group element in each of the group elements as group elements to be compared;

specifically, when the correspondence detection is performed on each group of elements, one group of elements is arbitrarily selected from the medical data set as a target group of elements, and the other group of elements are used as group elements to be compared; if the basic data table is used as a target group element, and the medical insurance data table and the clinic data table are used as group elements to be compared, the target group element and the group elements to be compared are compared, and whether corresponding data exist in each group element of each clinic patient is judged.

Step S222, reading a first row identifier of a data row in the target group element and a second row identifier of each data row in each group element to be compared, and determining whether a target identifier corresponding to the first row identifier exists in each second row identifier of each group element to be compared;

the data rows in the factor group elements correspond to the patients to be treated, and all data in the same data row are owned by one patient to be treated and carry the identifier corresponding to the patient to be treated; during comparison, reading any data in the target group element line by line, wherein the number of the read data corresponds to the line number of the data line in the target group element; and reading identifiers carried in the data, representing the patients to be treated corresponding to each row of data in the target group element, and taking the carried identifiers as first row identifiers. Simultaneously, randomly reading one data from each data line of the group element to be compared, wherein the number of the read data corresponds to the number of lines of the data line in the group element to be compared; and reading the identifier carried in each data, representing the patient to be treated corresponding to each row of data in the element group to be compared, and taking the carried identifier as a second row identifier. And comparing the first row identifiers with the second row identifiers one by one, and judging whether the second row identifiers all have target row identifiers corresponding to the first row identifiers. It should be noted that, because the group element to be compared includes a plurality of group elements, a second line identifier of data of each data line needs to be read for each group element, so that the formed second line identifier includes a plurality of groups; the number of the second row identifiers is consistent with the number of the group elements in the group elements to be compared, for example, for the medical insurance data sheet and the medical treatment data sheet serving as the group elements to be compared, the corresponding second identifiers are the second identifiers in the medical insurance data sheet and the second identifiers in the medical treatment data sheet respectively; when the first row identifier and the second row identifier are compared, the first row identifier and each group of second row identifiers are required to be compared, and whether a target identifier corresponding to the first row identifier exists in each group of second row identifiers is judged, so that the patient to be treated has corresponding data in each group of elements.

Step S223, if there are all target identifiers corresponding to the first row identifier, completing data correspondence detection of each group element;

further, in the comparison process, if it is determined that the target identifier corresponding to the first row identifier exists in each second row identifier, the patient to be diagnosed corresponding to the first row identifier is indicated, corresponding data exists in other elements of the group to be compared, and the data correspondence detection of the patient to be diagnosed is normal; and comparing other first row identifiers in the target group element with each second row identifier until all first row identifiers are compared. And if all the first row identifiers have target row identifiers corresponding to the first row identifiers in the second row identifiers, the data correspondence detection of each patient in the medical data group is normal, and the data correspondence detection of each group of elements is completed.

In step S224, if there are no object identifiers each corresponding to the first line identifier, a third anomaly identifier is added to the data line having the first line identifier corresponding to the object identifier, and a second detection result is generated from the number of the third anomaly identifiers.

When some target line identifiers which do not correspond to the first line identifiers exist in the first line identifiers in the second line identifiers, namely the target identifiers which do not correspond to the first line identifiers do not exist in the second line identifiers, the corresponding relation of the data of the patient corresponding to the first line identifiers in each group element is abnormal, the first line identifiers which do not correspond to the target identifiers are derived from the data lines, third abnormal identifiers are added to the data lines from which the first line identifiers are derived, and second detection results are generated by the third abnormal identifiers so as to represent the number of the abnormal data which do not correspond to the group elements. Specifically, the number of the third anomaly identifiers is counted as a second anomaly total number, the data total number of the data in the group element is counted, the counted second anomaly total number and the data total number are used as a ratio, and an obtained ratio result is the number proportion of the abnormal data which are not corresponding in the group element; and taking the ratio result as a second abnormal rate, and forming a set by using data reflecting abnormality, such as the second abnormal total amount, the second abnormal rate, the third abnormal identifier amount and the like, as a second detection result to represent the height of the amount of data with the different abnormal values in the group elements.

Further, in another embodiment of the method for evaluating the quality of medical data according to the present invention, the step of checking the cells having operation data in each of the group elements to generate a third detection result includes:

step S231, reading cell identifiers of cells in each group of elements, and determining target cells with operational data according to the cell identifiers;

in the embodiment, cell identifiers are arranged on the cells with operation data in the group elements, and the cell identifiers of the cells in the group elements are read when checking the checking relationship; when the cell identification is read, the cell is indicated to have operation data and needs to be operated; when the cell identification is not read in the cell, the cell to be detected does not have operation data, and operation is not needed; thereby determining the target cell in which the operation data exists according to the cell identification. Or different cell identifiers can be set for the cells with the operational data and the cells without the operational data, the cell identifiers of the cells are read, and then the cell identifiers are compared with the set cell identifiers representing the cells with the operational data, so that the target cells with the operational data are determined.

Step S232, detecting whether target cell data with abnormal collusion relation detection exists in the cell data of each target cell;

understandably, the operation modes and operation objects of different operation data are different, for example, some may be an addition operation with a certain data C of the patient with the operation data, and others may be a subtraction operation with another data D of the patient with the operation data; wherein the addition or subtraction operation characterizes the operation regime, while the data C and the data D characterize the operation object. Before checking the checking relationship of each operation data, the operation mode and the operation object of each operation data need to be determined. Specifically, when a target cell with operational data is determined according to the read cell identifier, representing an operational mode and an operational object corresponding to the operational data in the cell identifier with the operational data; the operation object comprises an operated object and an operation result, the operated object represents a cell where data operated by the operation data are located, and the operation result represents a cell where results obtained by operating the data in the operated object and the operation data are located. When checking the checking relationship, reading data in the representation cells of the operated object and cell data in the target cells, and operating the data in the representation cells and the cell data according to an operation mode to obtain an operation data result; and reading result data in the cells represented by the operation result, comparing the operation data result obtained by operation with the result data, and judging the consistency of the operation data result and the result data. Detecting whether the target cell data with abnormal checking relation exists in the cell data of each target cell or not through the consistency of the two target cell data; when the two are consistent, the operation of the cell data in the target cell is correct, the checking relation detection is normal, and the cell data in the next target cell is continuously read for detection until the cell data in all the target cells are detected; and when the result data of the operation data of the cell data in the target cell is detected to be inconsistent with the result data, the operation of the cell data in the target cell is incorrect, the checking relation detection is abnormal, and the target cell data with the abnormal checking relation detection exists in the cell data of each target cell.

In step S233, if there is target cell data with abnormal collusion relation detection, a fourth abnormal identifier is added to the target cell data, and a third detection result is generated according to the number of the fourth abnormal identifiers.

Furthermore, when the target cell data with abnormal checking relation does not exist in the cell data of each target cell, it indicates that the checking relation of each target cell is normal, and the operation of the cell data in each target cell in the group element is correct. And when the target cell data with the check relation detection abnormity exists in the cell data of each target cell, the target cell data with the operation errors exists in each target cell, a fourth abnormity identifier is added to the target cell data, and a third detection result is generated by the fourth abnormity identifier so as to represent the number proportion of the abnormal data with the check relation errors in the group elements. Specifically, the number of the fourth abnormal identifiers is counted as a fourth abnormal total number, the data total number of the data in the group element is counted, the counted fourth abnormal total number and the data total number are used as a ratio, and an obtained ratio result is the number of the abnormal data with incorrect collusion relation in the group element; and taking the ratio result as a third anomaly rate, and forming a set by using data reflecting anomalies, such as a third anomaly total amount, a third anomaly rate, a fourth anomaly identifier amount and the like, as a third detection result so as to reflect the quantity of the anomalous data with mismatching relation in the group elements.

Further, in another embodiment of the method for evaluating quality of medical data according to the present invention, the step of determining a target quality level corresponding to the generated detection result according to a preset correspondence between the detection result and the quality level includes:

step S31 of reading a weighting coefficient corresponding to the first abnormality rate, the second abnormality rate in the second detection result, and the third abnormality rate in the third detection result;

furthermore, after the first detection result, the second detection result and the third detection result are generated, the first abnormal rate in the first detection result represents the number proportion of unreasonable abnormal data in the group element, the second abnormal rate in the second detection result represents the number proportion of the abnormal data which is not corresponding in the group element, and the third abnormal rate in the third detection result represents the number proportion of the abnormal data which is incorrect in collusion relationship in the group element; by integrating the first abnormality rate, the second abnormality rate and the third abnormality rate which characterize the proportion of the number of various types of abnormalities, a detection result which reflects the overall quality of medical data can be generated. Considering that different abnormal data types have different influence degrees on the quality of the medical data, if the abnormality of the colluding relationship type has a large influence on the medical data, the influence of the abnormality of the unreasonable type on the medical data is relatively weak; therefore, in order to reflect the influence of various abnormalities on the quality of medical data more accurately, corresponding weighting coefficients are set for the various abnormalities, and the weighting coefficients corresponding to the various abnormalities, namely the first abnormality rate, the second abnormality rate and the third abnormality rate, are read and represented so as to integrate the first abnormality rate, the second abnormality rate and the third abnormality rate through the weighting coefficients.

Step S32, performing weighted calculation on the first abnormality rate, the second abnormality rate and the third abnormality rate by using each weighting coefficient to generate a target detection result;

after the weighting coefficients corresponding to the first abnormality rate, the second abnormality rate and the third abnormality rate are read, the weighting coefficients and the corresponding first abnormality rate, the second abnormality rate and the third abnormality rate are transmitted to a preset formula, the first abnormality rate, the second abnormality rate and the third abnormality rate are respectively weighted and calculated through the preset formula by using the weighting coefficients, and the calculated result is a target detection result for detecting the medical data group. The preset formula is as follows:

y＝(k ₁ *x ₁ +k ₂ *x ₂ +k ₃ *x ₃ )/3

where y denotes a target detection result, x1, x2, and x3 denote a first abnormality rate, a second abnormality rate, and a third abnormality rate, respectively, and k1, k2, and k3 denote weighting coefficients corresponding to x1, x2, and x3, respectively.

And transmitting the generated first abnormal rate, second abnormal rate and third abnormal rate and the read weighting coefficients into the preset formula, replacing x1, x2, x3, k1, k2 and k3 in the preset formula, and calculating to generate a target detection result.

And S33, comparing the target detection result with a preset corresponding relation between the detection result and the quality grade, and determining the target quality grade corresponding to the target detection result.

Furthermore, the generated target detection result is compared with a preset corresponding relationship between the detection result and the quality grade, and the detection result corresponding to the target detection result is determined, and the quality grade corresponding to the corresponding detection result in the corresponding relationship is the target quality grade corresponding to the target detection result. In the preset corresponding relation, the detection result is substantially a section of numerical range, and the target detection result is compared with the numerical range to determine the corresponding detection result. In the preset corresponding relation, the quality grade corresponding to the detection result between 0.05 and 0.1 is a first grade, the quality grade corresponding to the detection result between 0.11 and 0.2 is a second grade, and the quality grade corresponding to the detection result between 0.21 and 0.3 is a third grade; when the target detection result obtained by calculation of the preset formula is 0.15, the target detection result is compared with the detection result in the corresponding relation, and the corresponding target detection result is in a second stage. Further, the quality of the medical data set is evaluated according to the target quality grade; the corresponding relation between the quality grade and the quality can be evaluated, if the quality grade corresponds to high quality, the medical data with the quality grade of one grade is evaluated into the medical data with high quality, and the quality of the medical data is represented by the quality grade.

In addition, referring to fig. 2, the present invention provides a quality evaluation apparatus for medical data, in a first embodiment of the quality evaluation apparatus for medical data of the present invention, the quality evaluation apparatus for medical data includes:

the reading module 10 is used for reading the basic data sheet, the medical insurance data sheet and the treatment data sheet of each patient to be treated, and setting the basic data sheet, the medical insurance data sheet and the treatment data sheet as group elements to form a medical data group;

the detection module 20 is used for respectively performing data rationality detection, data correspondence detection and audit relationship detection on each group of elements in the medical data group to generate a detection result;

the evaluation module 30 is configured to determine a target quality level corresponding to the generated detection result according to a preset correspondence between the detection result and the quality level, and perform quality evaluation on the medical data set according to the target quality level.

In the quality evaluation device for medical data of the embodiment, the read basic data sheet, medical insurance data sheet and medical treatment data sheet of each patient are used as group elements by the reading module 10 to form a medical data group of the patient; the detection module 20 respectively performs data reasonability detection, data correspondence detection and audit relationship detection on each group element in the medical data group to generate a detection result; the corresponding relationship between the detection result and the quality grade is preset, and after the detection result is generated, the evaluation module 30 can determine the corresponding target quality grade according to the corresponding relationship, and further perform quality evaluation on the medical data set according to the target quality grade. The quality of the medical data group is evaluated substantially by the medical data formed by various types of treatment data in the treatment process; the internal data of each group of elements and the integrity of the data among the elements and the correctness of the corresponding relation are reflected by a generated detection result by detecting the data reasonability, the correspondence and the checking relation of each group of elements in the medical data group; taking the target quality grade corresponding to the detection result as a basis to evaluate the quality of the medical data; the checking and evaluation in a manual mode are avoided, so that the evaluation of the medical data is more accurate, and the evaluation efficiency and the automation degree are improved.

Further, in another embodiment of the apparatus for evaluating quality of medical data according to the present invention, the detection result includes a first detection result, a second detection result, and a third detection result, and the detection module is configured to:

performing data reasonability detection on each group of elements by taking data as a unit to generate a first detection result;

Further, in another embodiment of the apparatus for quality assessment of medical data of the present invention, the detection module is further configured to:

detecting whether target data with the numerical values exceeding a preset range exists in each numerical value type data column, and if the target data with the numerical values exceeding the preset range exists, adding a second abnormal identifier to the target data;

judging whether a first abnormal rate in the first detection result is greater than a preset value or not, and stopping performing corresponding detection on each group of elements if the first abnormal rate is greater than the preset value;

if there are no destination identifiers each corresponding to the first row identifier, a third anomaly identifier is added to the data row having the first row identifier corresponding to the destination identifier, and a second detection result is generated from the number of the third anomaly identifiers.

Further, in another embodiment of the apparatus for evaluating quality of medical data according to the present invention, the evaluation module further includes:

a reading unit configured to read a weighting coefficient corresponding to the first abnormality rate, a second abnormality rate in the second detection result, and a third abnormality rate in the third detection result;

a generating unit, configured to perform weighted calculation on the first abnormality rate, the second abnormality rate, and the third abnormality rate using each of the weighting coefficients, respectively, to generate a target detection result;

and the determining unit is used for comparing the target detection result with a preset corresponding relation between the detection result and the quality grade, and determining the target quality grade corresponding to the target detection result.

Here, each virtual function module of the above-described medical data quality evaluation apparatus is stored in the memory 1005 of the medical data quality evaluation device shown in fig. 3, and when the processor 1001 executes a medical data quality evaluation program, the function of each module in the embodiment shown in fig. 2 is realized.

Referring to fig. 3, fig. 3 is a schematic device structure diagram of a hardware operating environment related to the method according to the embodiment of the present invention.

The quality evaluation device of the medical data in the embodiment of the invention can be a Personal Computer (PC), or a terminal device such as a smart phone, a tablet computer, an electronic book reader, a portable computer, etc.

As shown in fig. 3, the quality evaluation apparatus of medical data may include: a processor 1001, such as a CPU (Central Processing Unit), a memory 1005, and a communication bus 1002. The communication bus 1002 is used for realizing connection communication between the processor 1001 and the memory 1005. The memory 1005 may be a Random Access Memory (RAM) or a non-volatile memory (disk memory). The memory 1005 may alternatively be a storage device separate from the processor 1001.

Optionally, the apparatus for evaluating quality of medical data may further include a user interface, a network interface, a camera, an RF (Radio Frequency) circuit, a sensor, an audio circuit, a WiFi (Wireless broadband) module, and the like. The user interface may comprise a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface may also comprise a standard wired interface, a wireless interface. The network interface may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface).

Those skilled in the art will appreciate that the medical data quality assessment device configuration shown in fig. 3 does not constitute a limitation of a medical data quality assessment device, and may include more or fewer components than those shown, or some components in combination, or a different arrangement of components.

As shown in fig. 3, a memory 1005, which is a kind of readable storage medium, may include therein an operating system, a network communication module, and a quality evaluation program of medical data. The operating system is a program that manages and controls the hardware and software resources of the quality assessment device for medical data, supporting the execution of the quality assessment program for medical data as well as other software and/or programs. The network communication module is used to implement communication between the components inside the memory 1005 and with other hardware and software in the quality assessment device of medical data.

In the quality evaluation apparatus for medical data shown in fig. 3, the processor 1001 is configured to execute a quality evaluation program for medical data stored in the memory 1005, and implement the steps in each embodiment of the quality evaluation method for medical data described above.

The present invention provides a readable storage medium storing one or more programs, which are further executable by one or more processors for implementing the steps in the various embodiments of the method for quality assessment of medical data described above.

It should also be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a component of' 8230; \8230;" does not exclude the presence of another like element in a process, method, article, or apparatus that comprises the element.

The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.

Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a readable storage medium (such as ROM/RAM, magnetic disk, optical disk) as described above, and includes several instructions for enabling a terminal device (which may be a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.

The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention, and all equivalent structural changes made by using the contents of the specification and drawings, or any other related technical fields, which are directly or indirectly applied to the present invention, are included in the scope of the present invention.

Claims

1. A method for quality assessment of medical data, characterized in that the method for quality assessment of medical data comprises the steps of:

determining a target quality grade corresponding to the generated detection result according to a corresponding relation between a preset detection result and the quality grade, and performing quality evaluation on the medical data set according to the target quality grade;

performing data reasonability detection on each group of elements by taking data as a unit respectively to generate a first detection result;

the data rationality detection comprises null data detection and numerical range detection, the data rationality detection is carried out on each group of elements by taking data as a unit, and the step of generating a first detection result comprises the following steps:

2. The method for quality assessment of medical data according to claim 1,

the step of performing data reasonability detection, data correspondence detection and audit relationship detection on each group of elements in the medical data group respectively and generating a detection result further comprises the following steps:

3. The method for evaluating the quality of medical data according to claim 2, wherein the step of performing data correspondence detection on each of the group elements in a data-line unit and generating the second detection result comprises:

4. The method for evaluating the quality of medical data according to claim 3, wherein the step of performing data correspondence detection on each of the group elements in a data-line unit and generating a second detection result comprises:

5. The method of claim 4, wherein the step of cross-checking the cells having the operational data in each of the plurality of groups of elements to generate a third test result comprises:

detecting whether the cell data of each target cell has target cell data with abnormal checking relation;

and if the target cell data with abnormal checking relation exists, adding a fourth abnormal identifier to the target cell data, and generating a third detection result according to the number of the fourth abnormal identifiers.

6. The method for quality assessment of medical data according to claim 5, wherein the step of determining a target quality level corresponding to the generated detection result according to a preset correspondence between detection results and quality levels comprises:

7. A quality evaluation apparatus for medical data, characterized by comprising:

the detection module is used for respectively carrying out data rationality detection, data correspondence detection and audit relationship detection on each group of elements in the medical data group to generate detection results, wherein the detection results comprise a first detection result, a second detection result and a third detection result;

the evaluation module is used for determining a target quality grade corresponding to the generated detection result according to a corresponding relation between a preset detection result and the quality grade, and evaluating the quality of the medical data set according to the target quality grade;

the detection module comprises a first detection unit, a second detection unit and a third detection unit, wherein the first detection unit is used for performing data reasonability detection on each group of elements by taking data as a unit respectively to generate a first detection result, and the data reasonability detection comprises null data detection and numerical range detection;

the first detection unit comprises a reading unit, a first adding unit, a second adding unit and a generating unit;

the reading unit is used for respectively traversing the data columns in the group elements one by one, reading the column identifiers of the data columns, and determining the numerical value type data columns in the data columns according to the column identifiers;

the first adding unit is used for detecting whether a cell with null data exists in each data column, and if the cell with null data exists, a first abnormal identifier is added to each cell;

the second adding unit is used for detecting whether target data with values exceeding a preset range exists in each numerical value type data column or not, and if the target data with the values exceeding the preset range exists, adding a second abnormal identifier to the target data;

the generation unit is used for generating a first detection result according to the number of the first abnormality identifiers and the second abnormality identifiers.

8. A quality evaluation apparatus for medical data, characterized by comprising: a memory, a processor, a communication bus, and a quality assessment program of medical data stored on the memory;

the processor is adapted to execute a quality assessment procedure of the medical data to implement the steps of the method of quality assessment of medical data according to any of claims 1-6.

9. A readable storage medium, characterized in that the readable storage medium has stored thereon a quality assessment program of medical data, which when executed by a processor implements the steps of the method of quality assessment of medical data according to any one of claims 1-6.