CN114944208B - Quality control method, quality control device, electronic equipment and storage medium - Google Patents
Quality control method, quality control device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN114944208B CN114944208B CN202210575556.8A CN202210575556A CN114944208B CN 114944208 B CN114944208 B CN 114944208B CN 202210575556 A CN202210575556 A CN 202210575556A CN 114944208 B CN114944208 B CN 114944208B
- Authority
- CN
- China
- Prior art keywords
- quality control
- data
- detection item
- control model
- permutation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/40—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for data related to laboratory analysis, e.g. patient specimen analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Abstract
The application provides a quality control method, a quality control device, electronic equipment and a storage medium, and relates to the technical field of inspection medicine. The method comprises the following steps: acquiring specimen data of a detection item; determining a quality control model corresponding to the detection item according to the variation coefficient of the detection item and an error scene where the detection item is located, wherein the quality control model comprises a floating average algorithm and a control limit, and the quality control models corresponding to the same detection item under different error scenes are different; determining whether the specimen data is a monitoring result of uncontrolled data or not through the quality control model; the quality control model is used for calculating the specimen data by utilizing the floating average algorithm to obtain a floating average value, and determining whether the specimen data is a monitoring result of out-of-control data or not by judging whether the floating average value exceeds the control limit. The mode improves the accuracy of specimen data inspection.
Description
Technical Field
The present disclosure relates to the field of inspection medicine, and in particular, to a quality control method, apparatus, electronic device, and storage medium.
Background
Clinical examination is an important auxiliary reference basis for a doctor to make scientific diagnosis on a patient, but the examination result often generates errors due to some objective or subjective reasons, so that the doctor makes an incorrect diagnosis, further delays treatment and even causes serious injury to the patient. Errors may occur in each link of the test process, such as sample collection, transportation or handling, instrumentation, personnel handling, interpretation of results, etc. In general, errors occurring during the inspection process are mostly instrument-related, and laboratories typically use internal quality control programs to monitor, guarantee and manage the quality of the inspection throughout the inspection process.
In order to ensure the reliability of the detection result, the mode of performing indoor quality control by using a quality control object is introduced earliest in a medical laboratory, and one of the important means for monitoring the detection performance is currently commonly used. Meanwhile, the quality control is also an instructive evaluation index in medical institution grade evaluation, and specifically, the national sanitary industry standard clinical examination quantitative determination indoor quality control definitely prescribes how to design the indoor quality control in the medical detection process, and the traditional quality control scheme is effectively implemented.
However, conventional quality control schemes are greatly affected by quality control products (i.e., the above quality control substances) in clinical applications, such as: the quality control product has limited concentration level, higher cost, lower detection frequency and the like. Studies have shown that quality control methods such as real-time-based real-time quality Control (PBRTQC) based on Patient data can detect variations in test result deviation relatively sensitively, and can detect error events before the next quality control action, thereby minimizing the number of unreliable Patient detection results in a clinical report, being not affected by the quality control matrix effect, and being capable of monitoring the whole process of analysis of the test. Once set up successfully in the medical laboratory, no additional detection and maintenance costs are required. In 2020, the international consortium of clinical chemistry and laboratory medicine (IFCC) analytical quality working group issued expert consensus on the establishment and testing of the PBRTQC quality control method. However, the quality control method of the PBRTQC still has the following defects: aiming at the same detection project, when coping with different error scenes, the technical defect of low accuracy of specimen data inspection caused by monitoring by adopting the same quality control model is overcome.
Disclosure of Invention
The application provides a quality control method, a quality control device, electronic equipment and a storage medium, which are used for monitoring by adopting different quality control models when coping with different error scenes, so as to ensure high accuracy of specimen data inspection.
According to a first aspect of the present application, there is provided a quality control method comprising:
acquiring specimen data of a detection item;
determining a quality control model corresponding to the detection item according to a variation coefficient of the detection item and an error scene where the detection item is located, wherein the quality control model comprises a floating average algorithm and a control limit, the variation coefficient is a numerical value formulated by the detection item according to biological variation and is used for reflecting the discrete degree of the detection item, and the quality control models corresponding to the same detection item under different error scenes are different;
determining whether the specimen data is a monitoring result of uncontrolled data or not through the quality control model; the quality control model is used for calculating the specimen data by utilizing the floating average algorithm to obtain a floating average value, and determining whether the specimen data is a monitoring result of out-of-control data or not by judging whether the floating average value exceeds the control limit.
According to a second aspect of the present application, there is provided a quality control apparatus comprising:
the data acquisition module is used for acquiring specimen data of the detection item;
the data analysis module is used for determining a quality control model corresponding to the detection item according to the variation coefficient of the detection item and an error scene where the detection item is located, the quality control model comprises a floating average algorithm and a control limit, the variation coefficient is a numerical value formulated by the detection item according to biological variation and is used for reflecting the discrete degree of the detection item, and the quality control models corresponding to the same detection item under different error scenes are different;
the abnormality identification module is used for determining whether the specimen data is a monitoring result of uncontrolled data or not through the quality control model; the quality control model is used for calculating the specimen data by utilizing the floating average algorithm to obtain a floating average value, and determining whether the specimen data is a monitoring result of out-of-control data or not by judging whether the floating average value exceeds the control limit.
According to a third aspect of the present application, there is provided an electronic device comprising: at least one processor and memory;
The memory stores computer-executable instructions;
the at least one processor executes computer-executable instructions stored in the memory, causing the at least one processor to perform the quality control method as described in the first aspect above.
According to a fourth aspect of the present application, there is provided a computer readable storage medium having stored therein computer executable instructions for implementing the quality control method as described in the first aspect above when executed by a processor.
According to the quality control method, the device, the electronic equipment and the storage medium, when the quality control model is determined, not only is the discrete degree of patient data (namely specimen data) taken as a model selection basis, but also the error scene of a detection item is considered, the monitoring result of the quality control model can be more accurate through the effective influence of the error scene on accurate modeling, and when aiming at the same detection item under different error scenes, the floating mean algorithm in the quality control model is not single any more, so that the accuracy and timeliness of specimen data inspection can be improved.
It should be understood that the description of this section is not intended to identify key or critical features of the embodiments of the application or to delineate the scope of the application. Other features of the present application will become apparent from the description that follows.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.
Fig. 1 is a schematic diagram of an application scenario according to an embodiment of the present application;
fig. 2 is a schematic flow chart of a quality control method according to an embodiment of the present application;
fig. 3 is a schematic flow chart of another quality control method according to an embodiment of the present application;
FIG. 4 is a schematic diagram of a floating-average algorithm;
FIG. 5 is a MNPED graph of each floating-mean algorithm for 5 detection items under proportional error;
FIG. 6 is a graph showing the effect of quality control under various inspection items;
fig. 7 is a schematic structural diagram of a quality control device according to an embodiment of the present application;
fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Specific embodiments thereof have been shown by way of example in the drawings and will herein be described in more detail. These drawings and the written description are not intended to limit the scope of the inventive concepts in any way, but to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present application.
For easy understanding, first, an application scenario of the embodiment of the present application will be described.
Fig. 1 is a schematic diagram of an application scenario according to an embodiment of the present application. As shown in fig. 1, the application scenario of the embodiment relates to an instrument and a monitoring device, wherein the instrument and the device are used for providing detection data of a patient, the monitoring device monitors the detection data through a PBRTQC quality control method, and when the detection data is out of control, an alarm is given.
The existing PBRTQC quality control method still has the defects, and the method is expressed as follows: 1) The quality control scheme is required to be formulated for each detection item object according to different crowds, specifically, the existing PBRTQC technology is a quality control method based on a floating average value, namely, the average value, the median value or other quantities of samples are monitored through rolling of a sliding window with a fixed length, and the samples are regarded as out of control when the average value, the median value or other quantities exceed a set range (control limit). The control limit is selected based on the floating amount statistics of the normal samples. The method is extremely sensitive to data performance, and different methods have different performances under different projects, so that the lack of establishment of a quality control system makes up the defect of a single algorithm; 2) In the early setting, the analysis method used by the characteristics and detection projects of the patient group in a laboratory is required to be known, that is, the current different PBRTQC quality control methods have different performance under the data with different characteristics, the data characteristics are closely related to the patient group and the type of the instrument, so that the data characteristics are required to be summarized according to the characteristics of the patient group and the working principle of the instrument, and the medicine is delivered for the symptoms, and finally, the special quality control method is matched; 3) The detection accuracy and detection speed of the method cannot completely meet clinical diagnosis and treatment requirements, and in practical application, different quality control methods are sensitive to data quality and lack of guiding standards in the selection process of the method and the setting modes of parameters such as window length, control limit and the like, corresponding to the monitored quantity. Once the identification range is unreasonable, only larger deviation can be effectively identified, and finally the identification range is reduced, so that the accuracy is low. However, too long a window results in hundreds of consecutive uncontrolled data being required to be identified, and alarm dullness affecting the speed of detection. In particular, for the project requirement of important clinical value, the detection efficiency of small errors is still to be improved, for example: prostate Specific Antigen (PSA) detection is critical for prostate cancer diagnosis, monitoring and treatment. In a continuous PSA assay, a change of 0.2 μg/L may be considered by the system as an indication of a poor prognosis. If the small deviation with clinical significance can be accurately identified, the occurrence of subsequent misdiagnosis can be effectively reduced. 4) The functional requirements for laboratory information systems are high. It is therefore necessary to provide quality control models containing different types of floating mean algorithm for the same detection item under different error scenarios to form a quality control system that can cope with different error scenarios.
To solve at least one of the above technical problems, the embodiments of the present application provide a quality control method, apparatus, electronic device and storage medium, which are intended to introduce PBRTQC technology into the field of inspection medicine, and establish a new real-time dynamic quality control framework capable of covering various error types (i.e. error scenarios), sizes, and based on patient inspection results (or referred to as inspection results, specimen data). The establishment and clinical application of the framework can further improve the accuracy and timeliness of the test result, better meet clinical demands and industry regulation requirements, further save medical resources, reduce medical cost, guarantee medical safety and bring benefit to both doctors and patients. In order to achieve the above objective, the present application provides a real-time quality control system based on patient test results, and the overall framework is divided into the following four layers: the system comprises a data acquisition layer, a data analysis layer, an anomaly identification layer and a reinforcement learning layer. On the basis of the real-time quality control system, the application provides a quality control method.
The following describes the technical solutions of the present application and how the technical solutions of the present application solve the above technical problems in detail with specific embodiments. The following embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.
Fig. 2 is a schematic flow chart of a quality control method according to an embodiment of the present application. As shown in fig. 2, the method of the present embodiment includes:
s201: acquiring specimen data of a detection item;
in the embodiments of the present application, the inspection item is alternatively referred to as a check item. Conventional inspection items include, but are not limited to: blood routine items, wherein the blood routine items are serum or protein, and biochemical items, including but not limited to: glucose (GLU), total Protein (TP), alanine Aminotransferase (ALT), aspartyl enzyme (AST), albumin (ALB), and the like. The specimen data may be understood as a patient sample, patient test data, patient test results or patient test results. The sample data may be raw sample data or may be data obtained by processing raw sample data, and the details are described in S301 to S303 below. Further, S201 may be understood as a specific operation of the data acquisition layer. While obtaining the specimen data, the present application may also record patient information corresponding to the specimen data, including but not limited to: age, gender, subject, time of acquisition, season, etc. It should be noted that only reported clinical results are included in the specimen data collection process.
S202: and determining a quality control model corresponding to the detection item according to the variation coefficient of the detection item and an error scene where the detection item is located, wherein the quality control model comprises a floating average algorithm and a control limit. The variation coefficient is a numerical value formulated by the detection item according to biological variation and is used for reflecting the discrete degree of the detection item, and quality control models corresponding to the same detection item under different error scenes are different.
In this embodiment, the variation coefficient is an index characteristic of the detection item, and an appropriate quality control model may be provided according to the index characteristic of the detection item, where the quality control model is used to recommend a corresponding floating average algorithm. S202 can be understood as an operation of the data analysis layer. At the data analysis layer, the embodiment of the application can judge whether the detection item is suitable for a floating average algorithm according to the biological variation coefficient (simply called variation coefficient) of the detection item aiming at different detection items. The coefficient of variation includes the intra-individual coefficient of variation CV i And coefficient of variation CV between individuals g . The embodiment can be based on CV i And CV (CV) g To determine the degree of discrete detection items, such as: if CV i And CV (CV) g Greater than 60% indicates that the probability of detecting data instability is great, and the detection item is not suitable for the floating average algorithm.
S203: determining whether the specimen data is a monitoring result of uncontrolled data or not through the quality control model; the quality control model is used for calculating the specimen data by utilizing the floating average algorithm to obtain a floating average value, and determining whether the specimen data is a monitoring result of out-of-control data or not by judging whether the floating average value exceeds the control limit.
In the present embodiment, S203 can be understood as the operation of the abnormality recognition layer. And at the abnormality identification layer, monitoring new patient samples in the same batch in real time by using a control limit calculated by using historical patient samples.
Because the existing quality control technology is based on quality control products, the technical problems of high cost and limited inspection exist, the embodiment of the application is based on patient samples, the cost problem of the quality control products is avoided, the accuracy and timeliness of detection results are further improved, and the clinical requirements and the industry regulation requirements can be better met.
In a possible implementation manner, since the quality control models in the present application are all pre-established before application, before the quality control model corresponding to the detection item is determined according to the coefficient of variation of the detection item and the error scene in which the detection item is located, the quality control method further includes the following steps S204 to S205, where:
S204: and obtaining the variation coefficient of the detection item, and determining a plurality of alternative floating average algorithms according to the variation coefficient of the detection item.
The step of determining a plurality of alternative floating-mean algorithms is a step of implementing algorithm selection. One or more of Moving weighted Average (Exponentially Weighted Moving Average, EWMA), moving Average (MA), moving Median (MM), moving statistical Median Hd50, moving standard deviation (Moving Standard deviation, movSD), moving division number MovSO may be provided as an alternative floating Average algorithm.
According to the statistical result of each floating mean algorithm under a normal sample, the formulas of 6 floating mean algorithm calculation are as follows:
MA：mean value of data within fixed window.
EWMA：EWMA t ＝(1-λ)×EWMA t-1 +λ×x t The weighted average is sliding on the basis of the MA algorithm.
MM：MM t ＝Median(x t-N ，x t-N+1 ，...，x t ) The median of the data within the window is fixed.
Hd50：Hd50＝(x t-N ，x t-N+1 ，...，x t ) The statistical median of the data within the window is fixed.
MovSD：MovSD t ＝SD(x t-N ，x t-N+1 ，...，x t ) The statistical standard deviation of the data within the window is fixed.
MovSO：The number of data within the fixed window that exceeds the clinical decision limit.
In each of the above floating-average algorithms, the step size may be in accordance with [10, 30, 50, 90, 110, 130, 150]Is searched by way of (a). It should be noted that different algorithms have different error resolutions for the detected items, and the magnitude of the error resolution is related to the numerical distribution of the detected items. The mean value and the discrete degree of the detection item influence the recognition of errors, and the selection of an alternative floating mean algorithm suitable for the detection item can effectively improve the unbalanced relation between sensitivity and specificity, so that the embodiment can be used for controlling the variation coefficient CV between individuals according to the detection item g And judging the discrete degree, and further selecting a proper moving average algorithm as an alternative floating average algorithm.
S205: and carrying out grid search experiments based on the multiple alternative floating mean algorithms and error scenes where the detection items are located, and establishing a quality control model corresponding to the detection items based on experimental results.
In the embodiment of the present application, S204 to S205 may be understood as a statistical flow including algorithm selection, grid search experiment and result optimization, where the grid search experiment in S205 includes a series of operations including data filtering, data dividing, data transformation (i.e. transformation into normal), window length setting, control limit selection calculation method, simulation test and statistical result selection. Specific analysis of data filtering, data dividing and data transferring are described in the following steps S301 to S303. Setting the window length refers to setting the data displayed in the movable window to be a fixed length. For example: the window length is 50, then the embodiment of the present application uses 50 units, the sliding window scrolls to monitor the mean, median or other amount of samples, the samples in the window going in and out, and the number is kept at 50. The control limit calculation method is a method adopted in the process of setting the left control limit and the right control limit after learning a normal sample and comprises a symmetrical method, an integral dividing method and a daily dividing method. After the control limit calculation method is selected, a control limit calculation process may be implemented. And the specific process details of the simulation test and the statistical result are shown in the following S1 to S3.
In a possible implementation manner, S205 is described above: performing a grid search experiment based on the plurality of alternative floating mean algorithms and an error scene where the detection item is located, and establishing a quality control model corresponding to the detection item based on an experiment result, wherein the method comprises the following steps: S1-S3, wherein:
s1: and providing a plurality of control limit calculation methods and a plurality of window lengths under the error scene of the detection item, and forming a plurality of permutation and combination based on the plurality of alternative floating average algorithms, the plurality of control limit calculation methods and the plurality of window lengths.
In the embodiment of the application, S1 is a step of a simulation test, and according to a clinical actual error scene, errors can be introduced according to three scenes of a proportional error, a constant error and a random error, and the error magnitude is changed according to an error factor. For example, the embodiment of the application may set a plurality of simulation days, introduce 1000 consecutive error data daily as test objects, and use the number of patient samples (Nped) affected by statistics as an important index for evaluating the recognition rate.
S2: and counting index analysis results of the permutation and combination under each preset evaluation index aiming at each permutation and combination.
In an embodiment of the present application, the preset evaluation index includes at least one of the following: true Positive Rate (TPR), false Positive Rate (FPR), true Negative Rate (TNR), false Negative Rate (FNR), cumulative ANPed (i.e., average of Nped), cumulative 95Nped (i.e., 95% quantile of Nped), and MNPed (i.e., median of Nped). In practical applications, the True Positive Rate (TPR), false Positive Rate (FPR), true Negative Rate (TNR) and False Negative Rate (FNR) are evaluation criteria commonly used under classification statistics, and since the quality control process essentially judges the two situations of control and out of control by identifying unbiased and biased samples, the scene in which the present application is located essentially belongs to the classification process. If the index is deleted, the realization of the whole scheme is not affected. Throughout the verification process, the embodiment of the application needs to divide the data set into multiple simulation days, and each day contains 1000 continuous deviation data. The number of samples passed from the start of the detection data from the offset insertion point until the offset is identified is referred to as NPed. At each deviation, the average, median and 95 quantiles (i.e., ANPed, MNPED and 95 NPed) of all simulation days NPed can be counted, which is necessary because it is an accepted index in the clinical profession.
S2 is a step of statistical results, and is used for summarizing index analysis results corresponding to all the permutation and combination related to the grid search experiment into a table.
S3: and selecting one target permutation and combination from the permutation and combination according to the priority order of all the preset evaluation indexes based on index analysis results of the permutation and combination under all the preset evaluation indexes as a quality control model corresponding to the detection item.
In the step of optimizing the result, the embodiment of the application synthesizes each index of each permutation and combination in the statistical result table, and selects the optimal permutation and combination as the optimal solution of the grid search experiment according to a certain priority order.
Before S1, sample data used for modeling may be processed, and the data processing process is consistent with the following steps S301 to S303, which are not described herein.
In a possible implementation manner, after the determining whether the specimen data is the monitoring result of the uncontrolled data by determining whether the floating average exceeds the control limit, the quality control method further includes S206:
s206: when the specimen data are accumulated to a preset quantity, updating a quality control model corresponding to the detection item; wherein the updated quality control model has a re-optimized control limit.
It should be appreciated that S206 is a specific operation of the reinforcement learning layer, and when patient samples accumulate to some extent, the model may be re-updated according to a statistical procedure, and control limits are preferred to improve accuracy. The updating process is similar to the model building process, and will not be described in detail here. That is, when the model is first used, the control limit is preset through the reinforcement learning step. The preset number can be set by user, and the calculation method or the numerical value of the control limit is changed when the model is updated.
Based on the above embodiments, the technical solutions of the present application will be described in more detail below in conjunction with specific embodiments.
Fig. 3 is a flow chart of another quality control method according to an embodiment of the present application. On the basis of the embodiment shown in fig. 2, this embodiment focuses on refining S201 in fig. 2. As shown in fig. 3, the method of the present embodiment includes:
s301: and obtaining original sample data, and filtering the original sample data to obtain filtered sample data. Acquisition in embodiments of the present application may be understood as collection. The model can be used as sample data for model training in the model establishing process, and can be used as input data of the model for testing in the clinical application process.
In a possible implementation manner, S301: the filtering processing is performed on the original sample data to obtain filtered sample data, which comprises the following steps: s401, when the filtering process is interception process, setting an upper cutoff value and a lower cutoff value; s402, according to the upper cut-off value and the lower cut-off value, intercepting abnormal data in the original sample data to obtain intercepted sample data.
In the embodiment of the application, abnormal data is intercepted, and an interception limit (namely a cut-off limit) is defined as a quantile t, t= [0,0.01,0.05,0.2,0.4 ] corresponding to a threshold boundary of normal data]The upper limit of the interception limit (i.e. the upper cut-off value) is: utl=quaterile (1-n/2), and the lower limit of the interception limit (lower cutoff value) is: ltl=quatile (n/2). The intercepting process is to intercept the abnormal point, and the mode of intercepting the abnormal point is divided into two modes of replacement and removal, wherein the formula of the replacement mode is as follows:the formula of the removal mode is as follows: />Wherein, NAN is a non-number.
In the embodiment of the application, the filtering processing in S301 is to clean the original specimen data, and can reject outliers (for example, values smaller than 25% of the whole quantiles or larger than 75% of the whole quantiles can be removed as outliers) and paradoxical values (for example, null values, character strings, negative numbers, etc.); or, the sample data in a non-digital format is removed, the structure does not accord with the conventional data, and adverse effects of abnormal points on the control limit calculation process in the follow-up process are avoided. After the data interception process, a transition may be made.
S302: performing data transformation on the filtered sample data to obtain transformed sample data;
in one possible implementation, S302: the step of performing data transformation on the filtered specimen data to obtain transformed specimen data comprises the following steps: and carrying out data transformation on the filtered sample data by using a Box-Cox transformation mode or a prototype transformation mode to obtain transformed sample data.
The prototype transformation mode or referred to as prototype transformation is referred to as the "unit transformation," and refers to processing according to the original level, i.e., directly using the original data without any processing. The Box-Cox transformation method or called Box-Cox transformation is a generalized power transformation method proposed in 1964, is a data transformation commonly used in statistical modeling, and is used for the situation that continuous response variables do not meet normal distribution. After Box-Cox transformation, the correlation of the non-observable errors and the predicted variables can be reduced to some extent. The Box-Cox transformation can significantly improve the normalization, symmetry, and variance equality properties of the data. The Box-Cox conversion formula isWherein λ is a box-cox transformationThe transformation parameters can influence the normal performance after transformation, and the parameters can directly obtain the optimal value by using a programming tool. The data transformation in S302 thus achieves a transition state. For the transformation, the deviation of the data distribution is different for different types of detection projects, and if the sample data is transformed into the normal state, the accuracy of the experimental result can be improved to a certain extent.
S303: determining the transformed sample data as sample data of the detection item;
in the embodiment of the present application, the above S301 to S303 implement cleaning and processing of the original specimen data, which can provide a data base for improving the accuracy of the inspection of the specimen data. Besides filtering and data transformation, the data processing may also include data division, wherein the data division is only aimed at movSD algorithm, a clinically specified level decision limit is used as a demarcation point of the data division, and the number of sample points at two sides is counted, wherein the division results corresponding to the decision limits at different positions are different.
In this embodiment, the data processing may also be classified by controlling age, department, and category, and selecting a laboratory data quality control period. It should be noted that the raw specimen data collected in the present application is data of a quality control controlled period.
S304: and determining a quality control model corresponding to the detection item according to the variation coefficient of the detection item and an error scene where the detection item is located, wherein the quality control model comprises a floating average algorithm and a control limit. The variation coefficient is a numerical value formulated by the detection item according to biological variation and is used for reflecting the discrete degree of the detection item, and quality control models corresponding to the same detection item under different error scenes are different.
It should be understood that the specific implementation of S304 is similar to S202 in fig. 2, and will not be described here.
S305: determining whether the specimen data is a monitoring result of uncontrolled data or not through the quality control model; the quality control model is used for calculating the specimen data by utilizing the floating average algorithm to obtain a floating average value, and determining whether the specimen data is a monitoring result of out-of-control data or not by judging whether the floating average value exceeds the control limit.
It should be understood that the specific implementation of S305 is similar to S203 in fig. 2, and will not be described here.
In connection with the above detailed step analysis, an example analysis of the model building process is given below for one embodiment:
step 11: alanine Aminotransferase (ALT), aspartyl enzyme (AST), glucose (GLU), total Protein (TP) and Albumin (ALB) were selected as detection items by the hospital laboratory information system. The reason for choosing these analytes as detection items is that these 5 analytes represent different data distribution segments in the clinical laboratory and there is no significant sex difference, since fluctuations in sex distribution will lead to unpredictable sex differences in the mean values other than the analytical performance, and thus the present embodiments can avoid such sex differences. For each detection item, data processing such as interception and transformation is performed to obtain sample data (hereinafter referred to as a sample).
Step 12: the samples were divided into 1150 samples per day for 200 days, the first 150 samples being normal samples and the last 1000 samples introducing errors as abnormal samples. The process of modeling 1000 error data was analyzed as follows: errors are defined as three types of proportional errors, constant errors, and random errors, where the formula of proportional errors is x' =x× (1+n×teas), where n×teas= [ -50%, -48%, -46%,..46%, 48%,50%]The method comprises the steps of carrying out a first treatment on the surface of the The formula of the constant error isWherein n= [ -3, -2.5, -2, -1.5, -1, -0.5,0.5,1,1.5,2,2.5,3]TEa is an industry quality specification for each test item in the clinic, and the corresponding values are different for each test item and are updated by the association each year; the formula of random error is +.>Wherein n= [0.5,1,1.5,2,2.5,3 ]]In the followingIn the above three formulas, x' represents patient data into which a simulation error is introduced, x represents raw patient data (i.e., sample data),/i>Represents the average value of the dataset, σ represents the standard deviation of the dataset, and n is a factor used to control the magnitude of the introduced error. The above-mentioned TEa refers to a total allowable error in clinic, which is a sum of a random error (BE) and a Systematic Error (SE) for reflecting a difference between a measurement result and a true value, and the total error of the detection method selected must BE within a clinically acceptable level range.
Step 13: and (3) calculating the control limit based on the normal sample in the step (12), firstly obtaining the average value data of the normal sample, and then calculating the control limit according to different methods, wherein the calculation methods of the control limit are specifically classified into a symmetrical method, an integral dividing method and a daily dividing method. The symmetry method is to select the numerical value of the standard deviation width from left to right as the control limit by taking the mean value as the symmetry axis according to the distribution condition of the mean value sequence of the normal sample; the integral positioning method is as follows: taking the 0.5% quantile of the average value sequence data of all normal samples as a lower control limit and the 99.5% quantile as an upper control limit; daily positioning means: and (3) independently counting the minimum value and the maximum value of the average value sequence of the normal sample every day, wherein the minimum value set takes the 0.5% quantile as the lower control limit, and the maximum value set takes the 99.5% quantile as the upper control limit.
Step 14: for each control limit provided in step 13, 1150 samples per day are tested according to the control limit, the test mode is as shown in fig. 4, 1000 pieces of deviation data (i.e. the error data described above) are introduced after the maximum step size 150, when deviation data exist in the window is detected, the floating average value under each sliding average value algorithm is calculated according to step size sliding until the control limit is exceeded, an alarm is given, and the number of samples undergone by the window is recorded as Nped. If deviation data is not identified, then statistics are taken at 1100. Because of the large number of tests, and therefore, not listed one by one, the present application presents some of the test parameters as follows:
Table 1 test parameters
| Algorithm | Filtration mode | Filtering limit | Change to normal mode | Step size | Control limit calculation method |
| EWMA | 2 | 5 | 2 | 7 | 3 |
| MA | 2 | 5 | 2 | 7 | 3 |
| MM | 2 | 5 | - | 7 | 3 |
| Hd50 | - | - | 2 | 7 | 3 |
| MovSD | 2 | 5 | 2 | 7 | 3 |
| MovSO | - | - | - | 7 | 3 |
As can be seen from table 1, in the present embodiment, under the scale error scenario, the 4 floating average algorithms EWMA, MA, MM, movSD are considered as alternative floating average algorithms, the filtering mode represents that a removal mode is adopted, the conversion normal mode is Box-Cox transformation, the control limit calculation method is a daily quantile method, the window length is 50, and the preset evaluation index may include: false positive rate, true positive rate, MNped, ANped, and 95NPed. In this example, the embodiment of the present application performs a corresponding test on the four permutations and combinations of the above examples, to obtain a corresponding index analysis result. The MNped curves formed by the 5 detection items under the 4 permutation and combination are shown in fig. 5, the shapes of the curves are used for reflecting the identification capability of the adopted floating average algorithm under different types and different sizes of deviations, and the performance of each algorithm can be intuitively judged. The ideal shape is: the two sides are low, narrow and bilaterally symmetrical. It follows that the algorithms applicable to different detection items are different.
The priority order of the preset evaluation index may be: 1) Ordering the false positive columns under each permutation and combination, and reserving permutation and combination with false positive rate less than 5%; 2) Under the condition of 1), the permutation and combination with the true positive rate being more than or equal to 90% is reserved; 3) And sorting according to the accumulated MNPed, selecting an permutation and combination corresponding to the minimum value, and if the screening result is not unique, further comparing ANped with 95NPed, and selecting the permutation and combination corresponding to the minimum value. It should be noted that all 3 calculation methods of the calculation mode of the floating mean value, the setting of the window length and the control limit need to be enumerated in a certain range, and an optimal permutation and combination is selected through the optimization judgment rule (i.e. the preset evaluation index) under the final summary table. In the result optimizing process, the embodiment can sort the permutation and combination according to the priority order of the preset evaluation index, and take the optimal permutation and combination as the optimal solution of the grid search experiment. By controlling the quality of the various detection items in the above manner, an effect map of quality control under the various detection items in fig. 6 can be obtained.
Fig. 7 is a schematic structural diagram of a quality control device according to an embodiment of the present application. The apparatus of this embodiment may be in the form of software and/or hardware. As shown in fig. 7, the quality control apparatus provided in this embodiment includes: a data acquisition module 71, a data analysis module 72, and an anomaly identification module 73. Wherein, the liquid crystal display device comprises a liquid crystal display device,
a data acquisition module 71 for acquiring sample data of a detection item;
the data analysis module 72 is configured to determine a quality control model corresponding to the detection item according to a variation coefficient of the detection item and an error scene in which the detection item is located, where the quality control model includes a floating average algorithm and a control limit, the variation coefficient is a value formulated by the detection item according to biological variation, and is used to reflect a degree of dispersion of the detection item, and quality control models corresponding to the same detection item under different error scenes are different;
an anomaly identification module 73 for determining whether the specimen data is a monitoring result of uncontrolled data by the quality control model; the quality control model is used for calculating the specimen data by utilizing the floating average algorithm to obtain a floating average value, and determining whether the specimen data is a monitoring result of out-of-control data or not by judging whether the floating average value exceeds the control limit.
In a possible implementation manner, the quality control device further includes: the method comprises the steps of obtaining a determining module and a grid searching experiment module, wherein:
the acquisition determining module is used for acquiring the variation coefficient of the detection item and determining a plurality of alternative floating average algorithms according to the variation coefficient of the detection item;
and the grid search experiment module is used for carrying out grid search experiments based on the multiple alternative floating mean algorithms and error scenes where the detection items are located, and establishing a quality control model corresponding to the detection items based on experimental results.
In one possible implementation, the grid search experiment module includes: the device comprises a combination unit, a statistics unit and a selection unit, wherein:
the combination unit is used for providing a plurality of control limit calculation methods and a plurality of window lengths under the error scene of the detection item, and forming a plurality of permutation and combination based on the plurality of alternative floating average value algorithms, the plurality of control limit calculation methods and the plurality of window lengths;
the statistics unit is used for counting index analysis results of the permutation and combination under each preset evaluation index aiming at each permutation and combination;
And a selection unit, configured to select, based on the index analysis results of the permutation and combination under all the preset evaluation indexes, a target permutation and combination from the permutation and combination according to the priority order of all the preset evaluation indexes, as a quality control model corresponding to the detection item.
In a possible implementation manner, the quality control device further includes:
a reinforcement learning module 74 for updating a quality control model corresponding to the detection item when the specimen data is accumulated to a preset number; wherein the updated quality control model has a re-optimized control limit.
In a possible implementation, the data acquisition module 71 includes: a filtering unit, a transforming unit and a determining unit, wherein:
the filtering unit is used for acquiring original sample data, and filtering the original sample data to obtain filtered sample data;
the transformation unit is used for carrying out data transformation on the filtered sample data to obtain transformed sample data;
and a determining unit configured to determine the transformed sample data as sample data of the detection item.
In a possible implementation manner, the filtering unit includes a setting subunit and an intercepting subunit, where:
A setting subunit, configured to set an upper cutoff value and a lower cutoff value when the filtering process is an interception process;
and the interception subunit is used for intercepting the abnormal data in the original sample data according to the upper cutoff value and the lower cutoff value to obtain intercepted sample data.
In a possible implementation, the transformation unit is further configured to: and carrying out data transformation on the filtered sample data by using a Box-Cox transformation mode or a prototype transformation mode to obtain transformed sample data.
The quality control device provided in this embodiment may be used to execute the quality control method provided in any of the above method embodiments, and its implementation principle and technical effects are similar, and will not be described herein.
In the technical scheme of the application, the related processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the personal information of the user accord with the regulations of related laws and regulations, and the public order colloquial is not violated.
According to embodiments of the present application, an electronic device and a readable storage medium are also provided.
Fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present application. The electronic device includes a receiver 80, a transmitter 81, a processor 82 and a memory 83, and the electronic device formed by the above components may be used to implement the specific embodiments described in the present application, which are not described herein again.
The embodiments of the present application further provide a computer readable storage medium, where computer instructions are stored, and when the processor executes the computer instructions, each step in the method in the above embodiments is implemented.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the disclosure of the present application may be performed in parallel, sequentially, or in a different order, provided that the desired results of the technical solutions disclosed in the present application are achieved, and are not limited herein.
The above embodiments do not limit the scope of the application. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application are intended to be included within the scope of the present application.
Claims (8)
1. A quality control method, comprising:
acquiring specimen data of a detection item;
determining a quality control model corresponding to the detection item according to a variation coefficient of the detection item and an error scene where the detection item is located, wherein the quality control model comprises a floating average algorithm and a control limit, the variation coefficient is a numerical value formulated by the detection item according to biological variation and is used for reflecting the discrete degree of the detection item, and the quality control models corresponding to the same detection item under different error scenes are different;
Determining whether the specimen data is a monitoring result of uncontrolled data or not through the quality control model; the quality control model is used for calculating the specimen data by utilizing the floating average algorithm to obtain a floating average value, and determining whether the specimen data is a monitoring result of out-of-control data or not by judging whether the floating average value exceeds the control limit;
before the quality control model corresponding to the detection item is determined according to the variation coefficient of the detection item and the error scene in which the detection item is positioned, the method further comprises:
acquiring a variation coefficient of the detection item, and determining a plurality of alternative floating average algorithms according to the variation coefficient of the detection item;
performing a grid search experiment based on the multiple alternative floating mean algorithms and an error scene where the detection item is located, and establishing a quality control model corresponding to the detection item based on an experiment result;
the grid search experiment is performed based on the plurality of alternative floating mean algorithms and the error scene where the detection item is located, and a quality control model corresponding to the detection item is established based on an experiment result, including:
Providing a plurality of control limit calculation methods and a plurality of window lengths under an error scene where the detection item is located, and forming a plurality of permutation and combination based on the plurality of alternative floating average algorithms, the plurality of control limit calculation methods and the plurality of window lengths;
counting index analysis results of the permutation and combination under each preset evaluation index aiming at each permutation and combination;
and selecting one target permutation and combination from the permutation and combination according to the priority order of all the preset evaluation indexes based on index analysis results of the permutation and combination under all the preset evaluation indexes as a quality control model corresponding to the detection item.
2. The quality control method according to claim 1, further comprising, after the determination of whether the specimen data is a monitoring result of uncontrolled data by determining whether the floating average exceeds the control limit:
when the specimen data are accumulated to a preset quantity, updating a quality control model corresponding to the detection item; wherein the updated quality control model has a re-optimized control limit.
3. The method according to claim 1, wherein the acquiring sample data of the test item comprises:
Acquiring original sample data, and filtering the original sample data to obtain filtered sample data;
performing data transformation on the filtered sample data to obtain transformed sample data;
and determining the transformed sample data as sample data of the detection item.
4. A quality control method according to claim 3, wherein the filtering the raw sample data to obtain filtered sample data comprises:
when the filtering process is interception process, setting an upper cut-off value and a lower cut-off value;
and according to the upper cut-off value and the lower cut-off value, intercepting the abnormal data in the original sample data to obtain intercepted sample data.
5. A quality control method according to claim 3, wherein said data transforming said filtered specimen data to obtain transformed specimen data comprises:
and carrying out data transformation on the filtered sample data by using a Box-Cox transformation mode or a prototype transformation mode to obtain transformed sample data.
6. A quality control apparatus, comprising:
The data acquisition module is used for acquiring specimen data of the detection item;
the data analysis module is used for determining a quality control model corresponding to the detection item according to the variation coefficient of the detection item and an error scene where the detection item is located, the quality control model comprises a floating average algorithm and a control limit, the variation coefficient is a numerical value formulated by the detection item according to biological variation and is used for reflecting the discrete degree of the detection item, and the quality control models corresponding to the same detection item under different error scenes are different;
the abnormality identification module is used for determining whether the specimen data is a monitoring result of uncontrolled data or not through the quality control model; the quality control model is used for calculating the specimen data by utilizing the floating average algorithm to obtain a floating average value, and determining whether the specimen data is a monitoring result of out-of-control data or not by judging whether the floating average value exceeds the control limit;
the acquisition determining module is used for acquiring the variation coefficient of the detection item and determining a plurality of alternative floating average algorithms according to the variation coefficient of the detection item;
The grid search experiment module is used for carrying out a grid search experiment based on the multiple alternative floating mean algorithms and an error scene where the detection item is located, and establishing a quality control model corresponding to the detection item based on an experiment result;
the grid search experiment module comprises: the device comprises a combination unit, a statistics unit and a selection unit;
the combination unit is used for providing a plurality of control limit calculation methods and a plurality of window lengths under the error scene of the detection item, and forming a plurality of permutation and combination based on the plurality of alternative floating average value algorithms, the plurality of control limit calculation methods and the plurality of window lengths;
the statistics unit is used for counting index analysis results of the permutation and combination under each preset evaluation index aiming at each permutation and combination;
the selecting unit is configured to select, based on the index analysis results of the permutation and combination under all the preset evaluation indexes, one target permutation and combination from the permutation and combination according to the priority order of all the preset evaluation indexes, as a quality control model corresponding to the detection item.
7. An electronic device, comprising: at least one processor and memory;
The memory stores computer-executable instructions;
the at least one processor executing computer-executable instructions stored in the memory causes the at least one processor to perform the quality control method of any one of claims 1 to 5.
8. A computer readable storage medium having stored therein computer executable instructions which when executed by a processor are for implementing the quality control method of any of claims 1 to 5.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2021105840994 | 2021-05-27 | ||
| CN202110584099.4A CN113539397A (en) | 2021-05-27 | 2021-05-27 | Real-time quality control system only using patient sample |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114944208A CN114944208A (en) | 2022-08-26 |
| CN114944208B true CN114944208B (en) | 2023-07-28 |
Family
ID=78094799
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110584099.4A Pending CN113539397A (en) | 2021-05-27 | 2021-05-27 | Real-time quality control system only using patient sample |
| CN202210575556.8A Active CN114944208B (en) | 2021-05-27 | 2022-05-25 | Quality control method, quality control device, electronic equipment and storage medium |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110584099.4A Pending CN113539397A (en) | 2021-05-27 | 2021-05-27 | Real-time quality control system only using patient sample |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN113539397A (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115691722B (en) * | 2022-04-15 | 2023-09-29 | 首都医科大学附属北京朝阳医院 | Quality control method, device, equipment, medium and program product for medical data detection |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007164357A (en) * | 2005-12-12 | 2007-06-28 | Fujitsu Ltd | Quality control method, quality control program and quality control system |
| CN110827935A (en) * | 2019-10-21 | 2020-02-21 | 清华大学 | Data quality control method and system for Chinese and western medical big data |
| CN111696655A (en) * | 2020-06-12 | 2020-09-22 | 上海市血液中心 | Internet-based real-time sharing blood screening indoor quality control system and method |
| CN112102903A (en) * | 2020-08-25 | 2020-12-18 | 中国医学科学院北京协和医院 | Quality control system based on clinical laboratory testing result |
| CN113609216A (en) * | 2021-07-08 | 2021-11-05 | 中南民族大学 | Block chain-based product quality diagnosis method, device, equipment and storage medium |
| CN114330859A (en) * | 2021-12-23 | 2022-04-12 | 复旦大学附属中山医院 | Optimization method, system and equipment for real-time quality control |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2022026810A (en) * | 2020-07-31 | 2022-02-10 | シスメックス株式会社 | Accuracy management support method, accuracy management support system, accuracy management support device, and program |
-
2021
- 2021-05-27 CN CN202110584099.4A patent/CN113539397A/en active Pending
-
2022
- 2022-05-25 CN CN202210575556.8A patent/CN114944208B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007164357A (en) * | 2005-12-12 | 2007-06-28 | Fujitsu Ltd | Quality control method, quality control program and quality control system |
| CN110827935A (en) * | 2019-10-21 | 2020-02-21 | 清华大学 | Data quality control method and system for Chinese and western medical big data |
| CN111696655A (en) * | 2020-06-12 | 2020-09-22 | 上海市血液中心 | Internet-based real-time sharing blood screening indoor quality control system and method |
| CN112102903A (en) * | 2020-08-25 | 2020-12-18 | 中国医学科学院北京协和医院 | Quality control system based on clinical laboratory testing result |
| CN113609216A (en) * | 2021-07-08 | 2021-11-05 | 中南民族大学 | Block chain-based product quality diagnosis method, device, equipment and storage medium |
| CN114330859A (en) * | 2021-12-23 | 2022-04-12 | 复旦大学附属中山医院 | Optimization method, system and equipment for real-time quality control |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113539397A (en) | 2021-10-22 |
| CN114944208A (en) | 2022-08-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pum | A practical guide to validation and verification of analytical methods in the clinical laboratory | |
| US6556951B1 (en) | System and method for intelligent quality control of a process | |
| CA2650872C (en) | Methods and apparatus for identifying disease status using biomarkers | |
| Raghavan et al. | A review of cutoffs for nutritional biomarkers | |
| Badrick et al. | A primer on patient-based quality control techniques | |
| US7584162B2 (en) | Manufacture data analysis method and manufacture data analyzer apparatus | |
| JP2002022748A (en) | Clinical inspection system | |
| US20220373565A1 (en) | Aptt prolongation factor estimation system | |
| CN115691722B (en) | Quality control method, device, equipment, medium and program product for medical data detection | |
| CN114944208B (en) | Quality control method, quality control device, electronic equipment and storage medium | |
| CN112435756B (en) | Intestinal flora associated disease risk prediction system based on multi-dataset difference interaction | |
| KR102470763B1 (en) | Data outlier detection apparatus and method | |
| CN109030803A (en) | Biochemistry detection quality control method | |
| Santini et al. | Accuracy of diagnostic tests | |
| CN113140325A (en) | Method for establishing prediction model of incidence probability of Marneffei pannieri disease of HIV patient | |
| US6787361B1 (en) | Clinical assay calibration adjustment method | |
| Leen et al. | Statistical error detection for clinical laboratory tests | |
| CN112102903A (en) | Quality control system based on clinical laboratory testing result | |
| Baveja et al. | Statistical analysis of microbiological diagnostic tests | |
| Laszczyńska et al. | Estimation of missing baseline serum creatinine for acute kidney injury diagnosis in hospitalized patients | |
| Gruenberg et al. | Determining the utility of creatinine delta checks: a large retrospective analysis | |
| Sampson et al. | CUSUM-Logistic Regression analysis for the rapid detection of errors in clinical laboratory test results | |
| CN115620819A (en) | Biomarker reference interval calculation method and device | |
| Vineis et al. | Technical variability in laboratory data | |
| JP2011007686A (en) | Device and method for analyzing clinical inspection result |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |