Data comparison analysis method, device and equipment based on retest result

Abstract

The invention provides a data comparison analysis method, device and equipment based on a retest result. The method is applied to medical data processing. The method comprises the steps that a retest result of any first detection platform for a target item is acquired, the target item is an item with an abnormal initial test result, and the first detection platform is different from a second detection platform generating the abnormal initial test result; calculating a difference value based on the abnormal initial measurement result and the remeasurement result; calculating difference distribution data based on the plurality of accumulated difference values, wherein the difference distribution data comprises a difference distribution range and a first confidence interval; and analyzing the consistency of the retest result and the abnormal initial test result based on the difference distribution data. According to the invention, the consistency of the retest result and the abnormal initial test result is accurately and rapidly detected.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of medical data processing, in particular to a data comparison analysis method and device based on retest results. BACKGROUND

[0002] Clinical laboratories are an important part of the medical field, mainly responsible for detecting and analyzing various samples of patients, such as blood, urine, tissue, etc. With the continuous growth of medical needs, the amount of data faced by clinical laboratories is increasingly large, and the difficulty of detection is also increasing. In order to efficiently and timely complete the detection task of a large number of samples, laboratories usually use multiple same detection devices to detect the same detection project at the same time. The bias between different devices is an important factor affecting the detection quality. At present, clinical laboratories use periodic inter-instrument comparison to monitor whether there is a bias between different devices. That is, within a specified time interval, the same standard sample or reference material is used to detect multiple instruments in the laboratory, and the detection results are statistically analyzed and evaluated.

[0003] However, the process of periodic inter-instrument comparison of clinical laboratories is complex and cumbersome, not only a large number of standard samples or reference materials need to be prepared, but also multiple detections of the instruments are required, and the operation must strictly follow the established process and method, resulting in a large consumption of manpower, time and energy. In addition, due to the above cumbersome process, periodic inter-instrument comparison is difficult to carry out frequently, resulting in a low comparison frequency, which cannot timely detect the bias of the instrument during the interval between two comparisons, thereby affecting the accuracy and reliability of the detection results, and bringing potential risks to clinical diagnosis and treatment. SUMMARY

[0004] Therefore, the first aspect of the present application provides a data comparison analysis method based on medical retest results, comprising: obtaining retest results of any first detection platform for a target project, the target project being a project with abnormal initial test results, and the first detection platform being different from a second detection platform that produces abnormal initial test results; calculating a difference value based on the abnormal initial test results and the retest results; calculating difference distribution data based on a plurality of accumulated difference values, the difference distribution data including a difference distribution range and a first confidence interval; and analyzing the consistency of the retest results and the abnormal initial test results based on the difference distribution data.

[0005] Optionally, calculating the difference distribution data based on the plurality of accumulated difference values comprises: calculating the mean and standard deviation of the plurality of difference values; calculating the upper limit and lower limit of the 95% difference distribution range based on the mean and standard deviation, respectively, to obtain the difference distribution range; and calculating the upper limit and lower limit of the 95% confidence interval based on the mean, standard deviation and number of difference values, respectively, to obtain the first confidence interval.

[0006] Optionally, the consistency of the retest result and the abnormal preliminary test result is analyzed based on the difference distribution data, including: difference data of the retest result and the abnormal preliminary test result of each first detection platform is calculated respectively, the difference data includes a difference value or a relative difference value; whether each difference data belongs to a difference distribution range and a first confidence interval is judged; if the difference data belongs to the difference distribution range and the first confidence interval, it is determined that the retest result is consistent with the abnormal preliminary test result; if the difference data does not belong to the difference distribution range or the first confidence interval, it is determined that the retest result is inconsistent with the abnormal preliminary test result.

[0007] Optionally, the data comparison and analysis method based on the retest result provided by the application further comprises: in response to the number of difference distribution data being greater than or equal to a preset number, a linear regression model is constructed based on the retest result of each first detection platform and the abnormal preliminary test result of the second detection platform, the linear regression model includes a regression straight line and a second confidence interval of the regression straight line; whether the retest result falls within the second confidence interval of the regression straight line is judged; if the retest result does not fall within the confidence interval of the regression straight line, it is determined that the retest result is inconsistent with the abnormal preliminary test result; if the retest result falls within the confidence interval of the regression straight line, it is determined that the retest result is consistent with the abnormal preliminary test result.

[0008] Optionally, the linear regression model is constructed based on the retest result of each first detection platform and the abnormal preliminary test result of the second detection platform, including: the least square method is used to calculate the regression coefficient and the intercept based on the retest result of each first detection platform and the abnormal preliminary test result of the second detection platform; the regression straight line is generated based on the regression coefficient and the intercept; the residual standard deviation is calculated based on the retest result of each first detection platform and the abnormal preliminary test result of the second detection platform; the second confidence interval of the regression straight line is calculated based on the residual standard deviation.

[0009] Optionally, the data comparison and analysis method based on the retest result provided by the application further comprises: the difference distribution data and / or the linear regression model are visually displayed.

[0010] The second aspect of the application provides a data comparison and analysis device based on retest results, comprising: an acquisition module for acquiring retest results of any first detection platform for target projects, the target project being a project with an abnormal preliminary test result, and the first detection platform being different from the second detection platform that produces the abnormal preliminary test result; a first calculation module for calculating difference values based on accumulated abnormal preliminary test results and retest results; a second calculation module for calculating difference distribution data based on a plurality of accumulated difference values, the difference distribution data including a difference distribution range and a first confidence interval; and an analysis module for analyzing the consistency of the retest result and the abnormal preliminary test result based on the difference distribution data.

[0011] The application provides a data comparison and analysis device based on retest results, and further comprises a construction module configured to, in response to the number of difference distribution data being greater than or equal to a preset number, construct a linear regression model based on the retest results of each first detection platform and the abnormal preliminary test results of the second detection platform, wherein the linear regression model comprises a regression straight line and a second confidence interval of the regression straight line; a judgment module configured to judge whether the retest results fall within the second confidence interval of the regression straight line; if the retest results do not fall within the confidence interval of the regression straight line, it is determined that the retest results are inconsistent with the abnormal preliminary test results; and if the retest results fall within the confidence interval of the regression straight line, it is determined that the retest results are consistent with the abnormal preliminary test results.

[0012] The application provides a data comparison and analysis device based on retest results, and further comprises a display module configured to visually display the difference distribution data and / or the linear regression model.

[0013] The application provides a data comparison and analysis device based on retest results, and further comprises a display module configured to visually display the difference distribution data and / or the linear regression model.

[0014] The application only needs to replace the retest of the abnormal preliminary test results by the detection platform, and performs real-time analysis based on the retest results, so as to quickly and timely find out whether the detection results are consistent. Specifically, the retest results of the abnormal preliminary test results are retested by a first detection platform different from the first detection platform, and then the difference values are calculated based on the abnormal preliminary test results and the retest results, so that the difference between the test results of the two detection platforms is clearly and accurately obtained. Then, the difference distribution data is calculated based on the continuously accumulated difference values, and the data is continuously updated. Finally, the consistency of the retest results and the abnormal preliminary test results is analyzed based on the difference distribution data. This method fully excavates the data value of the retest sample in the abnormal result review, which is originally used for single detection result verification, and deeply analyzes and reuses the data, so that the data breaks through the limitation of single function. The process is simple, only the abnormal preliminary test results need to be retested, and the consistency of the results of the two devices needs to be compared, without manual comparison, so that the deviation of the detection platform can be quickly and timely found out, the accuracy of the detection results is improved, and the treatment risk in the subsequent detection can be avoided.

[0015] The present application can quickly compare the consistency between platforms by constructing a linear regression model containing a regression straight line and its second confidence interval based on the retest results of each first detection platform and the abnormal preliminary test results of the second detection platform when the number of difference distribution data is greater than or equal to a preset number, and updating the model in time with the increase of retest results. Then, it is judged whether the retest results fall within the second confidence interval of the regression straight line. If not, it is determined that the corresponding first detection platform and the second detection platform are inconsistent. If yes, it is determined that the detection is consistent. This provides a basis for potential bias correction, can timely find the consistency of test results of each first detection platform and the second detection platform, timely find abnormal data and detection platforms, and provides a reliable basis for ensuring detection quality, which helps to improve the reliability and credibility of overall detection. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the specific embodiments or prior art of the present application, the drawings needed in the specific embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0017] Figure 1 The flow chart of the difference analysis method based on medical detection results in the embodiment of the present application; Figure 2 The display diagram of difference distribution data in the embodiment of the present application; Figure 3 The display diagram of linear regression model in the embodiment of the present application; Figure 4 The structure diagram of a difference analysis device based on medical detection results in the embodiment of the present application; Figure 5 The structure diagram of another difference analysis device based on medical detection results in the embodiment of the present application. DETAILED DESCRIPTION

[0018] The technical solutions of the present application will be described clearly and completely in combination with the drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor belong to the protection scope of the present application.

[0019] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second", "third" are only for the purpose of description and cannot be understood as indicating or implying relative importance.

[0020] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connecting" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication between two elements inside, it can be wireless connection, or it can be wired connection. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0021] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as there is no conflict between them.

[0022] As Figure 1 shown, the embodiment of the present application provides a data comparison and analysis method based on retest results, which is executed by a computer or a server and the like electronic device, and specifically includes: S1, obtaining the retest result of any first detection platform for the target project, the target project being the project with abnormal initial test result, and the first detection platform being different from the second detection platform generating the abnormal initial test result.

[0023] In a medical laboratory, multiple detection platforms of the same model are equipped for detecting the same item, one of which is a standard detection platform with excellent performance and always in good maintenance state. These detection platforms are used for medical detection, i.e. obtaining initial test results, and then auditors will periodically review these initial test results, and if a certain initial test result does not conform to the patient's clinical symptoms or lacks correlation with other indicators, it will be marked as an abnormal initial test result. Once an abnormal initial test result occurs, a detection platform (first detection platform) different from the detection platform (second detection platform) that produced the abnormal initial test result will be selected to quickly retest the same item sample. The item involved here is commonly used in medical laboratories, such as blood sample testing. The above process occurs in real time, and as long as there is an abnormal initial test result, the device of the present application will immediately respond to obtain the retest result of another detection platform for the abnormal item sample in real time, ensuring the efficiency and timeliness of the detection process.

[0024] S2, calculating a difference value based on the abnormal initial test result and the retest result.

[0025] wherein, is the difference value of the i th sample, is the retest result of the i th sample, is the abnormal initial test result of the i th sample.

[0026] The calculation of the difference value is the difference calculation of the results of two different detection platforms, which can be a standard detection platform and a non-standard detection platform, or two non-standard detection platforms.

[0027] S3, calculating difference distribution data based on the accumulated multiple difference values, the difference distribution data including a difference distribution range and a first confidence interval.

[0028] The difference value of each item sample with an abnormality is continuously obtained, and when a certain number of difference values are collected, the difference distribution range and the first confidence interval are calculated based on these difference values, and the difference distribution range and the confidence interval are updated and calculated once for each retest result obtained subsequently.

[0029] S4, analyzing the consistency of the retest result and the abnormal initial test result based on the difference distribution data.

[0030] The retest result obtained each time is compared with the updated difference distribution data to determine the consistency of the retest result and the abnormal initial test result of the corresponding two detection platforms.

[0031] ​The embodiment only needs to replace the test platform to retest the items of the abnormal preliminary test results, and based on the real-time analysis of the retest results, whether the test results are consistent is quickly and timely found. The retest results of the items of the preliminary test results are retested by the first detection platform different from the first detection platform that produces the abnormal preliminary test results, and then the difference values are calculated based on the abnormal preliminary test results and the retest results, respectively, so that the difference between the test results of the two detection platforms is clearly and accurately obtained. Then, the difference distribution data is calculated based on the continuously accumulated difference values, and the data is continuously updated. Finally, the consistency of the retest results and the abnormal preliminary test results is analyzed based on the difference distribution data. This method fully excavates the data value of the retest sample in the abnormal result review, which is originally used, and deeply analyzes and reuses the data which is originally only used for single detection result verification, so that these data break through the limitation of single function. The process is simple, only needs to retest the abnormal preliminary test results, and compare the consistency of the results of the two devices, without manual comparison, so that the deviation of the detection platform can be quickly and timely found, the accuracy of the detection results is improved, and the treatment risk in the subsequent detection is avoided In one embodiment, the difference distribution data is calculated based on the accumulated plurality of difference values in step S3, comprising: S31, calculating the mean and standard deviation of the plurality of difference values.

[0032] Each time a difference value is updated, the mean and standard deviation of all difference values are calculated.

[0033] S32, calculating the upper limit and lower limit of the 95% difference distribution range based on the mean and standard deviation, respectively, to obtain the difference distribution range.

[0034] Exemplarily, the difference distribution range can be calculated by the following method: , , wherein, represents the lower limit of the difference distribution range, represents the upper limit of the difference distribution range, represents the mean of all difference values, represents the standard deviation of all difference values.

[0035] S33, calculating the upper limit and lower limit of the 95% confidence interval based on the mean, standard deviation and number of difference values, respectively, to obtain the first confidence interval.

[0036] Exemplarily, the first confidence interval can be calculated by the following method: , , wherein, represents the lower limit of the 95% confidence interval, represents the upper limit of the 95% confidence interval, represents the number of first retest results. This embodiment reflects the trend of the difference between the retest results and the abnormal initial test results of the two detection platforms in real time by dynamically calculating the mean and standard deviation of all difference values. Then, based on the mean and standard deviation, the upper and lower limits of the 95% difference distribution range are calculated to determine the approximate range of the difference. Finally, the upper and lower limits of the 95% confidence interval are calculated based on the mean and standard deviation. Compared with the traditional subjective judgment, this statistical evaluation method has stronger objectivity and reliability. It can effectively reduce human errors in management and provide a solid and scientific basis for subsequent analysis of the consistency of the retest results and the abnormal initial test results of the two detection platforms based on difference distribution data, making the entire difference analysis process more accurate.

[0037] In addition, the first confidence interval needs to be judged, specifically: whether the first confidence interval contains 0.

[0038] When the first confidence interval contains 0, that is, when ≤0≤ , it indicates that there is no significant difference between the retest results and the abnormal initial test results at the current confidence level, and then the step S4 can be executed to analyze the consistency of the retest results and the abnormal initial test results again.

[0039] When the first confidence interval does not contain 0, that is, when 0 < , it means that the confidence interval is completely on the right side of 0, indicating that the retest results are generally larger than the abnormal initial test results, that is, there is a "positive bias"; when < <0, it indicates that the confidence interval is completely on the left side of 0, indicating that the retest results are generally smaller than the abnormal initial test results, that is, there is a "negative bias".

[0040] When the first confidence interval does not contain 0, the above also needs to be compared with the preset value to determine whether the deviation of the retest results and the abnormal initial test results is within an acceptable range (such as ±5). If the first confidence interval completely exceeds this acceptable range (i.e., the lower limit is less than -5 and the upper limit is greater than 5), it indicates that the difference between the retest results and the abnormal initial test results is too large and exceeds the acceptable degree. In this case, it can be considered that the difference reflected by the first confidence interval is not reliable, and then the step S4 does not need to be executed.

[0041] This embodiment can timely identify the reliability of the first confidence interval by first judging the first confidence interval, avoid unnecessary subsequent analysis, improve efficiency, evaluate the consistency from multiple dimensions, reduce misjudgment and omission, and enhance the reliability and credibility of the analysis results.

[0042] In one embodiment, analyzing the consistency of the retest result and the abnormal initial test result based on the difference distribution data in step S4 includes: S41, respectively calculating the difference data between the retest results of each first detection platform and the abnormal initial test results, where the difference data includes a difference value or a relative difference value.

[0043] When comparing the retest results with the differential distribution data, the first retest data needs to be processed, that is, the retest results of the first detection platform are linked to the abnormal initial test results of the second detection platform. For example, the retest result is 4 and the abnormal initial test result is 5. Then the calculated difference is 5-4=1, which means that the retest result of a first detection platform is 1 lower than the abnormal initial test result of the second detection platform based on the abnormal initial test result of the second detection platform. The calculated relative difference is (5-4) / 5=0.2, which means that the retest result of a first detection platform is 0.2 lower than the abnormal initial test result of the second detection platform based on the abnormal initial test result of the second detection platform.

[0044] S42, determine whether each difference data belongs to the difference distribution range and the first confidence interval; if the difference data belongs to the difference distribution range and the first confidence interval, execute step S43; if the difference data does not belong to the difference distribution range or the first confidence interval, execute step S44.

[0045] S43, determining that the retest result is consistent with the abnormal initial test result; S44, determining that the retest result is inconsistent with the abnormal initial test result.

[0046] Compare the retest result of a first detection platform represented by the difference or relative difference with the difference distribution range and the first confidence interval. If the result is within both the difference distribution range and the first confidence interval, it means that the gap between the retest result of the corresponding first detection platform and the abnormal initial test result of the second detection platform is within a reasonable range, and it can be determined that the detections of the two detection platforms are consistent and normal. If it is not within the difference distribution range or the first confidence interval, it means that the gap between the retest result of the corresponding first detection platform and the abnormal initial test result of the second detection platform is beyond a reasonable range, and it can be determined that the detections of the two detection platforms are inconsistent and abnormal.

[0047] In addition, when it is determined that the two detection platforms are inconsistent, a warning can be issued to the corresponding two detection platforms, and relevant personnel can quickly carry out in-depth investigation on the measured platform according to the warning information, such as checking whether the equipment is normally running, whether there is a loophole in the detection process, whether the operator is operating normally, and the like. Through timely discovery and solution of problems, a series of erroneous decisions and adverse consequences that can be caused by inaccurate detection results can be avoided, and the detection performance of the measured platform can be further optimized, and the accuracy and stability of detection can be improved.

[0048] The embodiment associates the retest results of each detection platform with the abnormal preliminary measurement results, calculates the difference or relative difference as difference data, and then takes the abnormal preliminary measurement results as a standard to clearly show the specific gap between the first detection platform and the second detection platform. Then, the difference data is compared with the difference distribution range and the first confidence interval to determine whether it is within a reasonable range. Finally, whether the two detection platforms are consistent is determined according to the comparison result. This way, the data of the two detection platforms is analyzed for consistency, the data originally used only for single detection result verification is deeply analyzed and reused, the data breaks through the limitation of single function, the application efficiency of the data is maximized, the detection inaccurate situation is found in time, a reliable basis for ensuring detection quality is provided, and the reliability and credibility of the overall detection are improved.

[0049] As shown in FIG. 6, the difference distribution data calculated in step S3 can also be visually displayed. Figure 2

[0050] The horizontal coordinate is the average of the retest result of each first detection platform and the abnormal preliminary measurement result of the second detection platform, and the vertical coordinate is the difference or relative difference calculated in step S41. The yellow data points are the difference or relative difference between the retest result of each first detection platform and the abnormal preliminary measurement result of the second detection platform. The two yellow straight lines respectively represent the upper limit of the 95% difference distribution range and the lower limit of the 95% difference distribution range. The green dotted line represents the upper limit of the 95% confidence interval, the blue dotted line represents the lower limit of the 95% confidence interval, and the red dotted line represents the average of the difference value.

[0051] The embodiment can intuitively reflect the detection difference change trend of the same item between different platforms through real-time dynamic visual display, and provide real-time information support for laboratory managers.

[0052] As shown in FIG. 6, the difference distribution data calculated in step S3 can also be visually displayed. Figure 1 As shown in FIG. 6, the difference distribution data calculated in step S3 can also be visually displayed. ​S5, in response to the number of difference distribution data being greater than or equal to the preset number, constructing a linear regression model based on the retest result of each first detection platform and the abnormal preliminary test result of the second detection platform, the linear regression model including a regression straight line and a second confidence interval of the regression straight line.

[0053] When the difference distribution data obtained by analyzing the same target item exceeds a certain number (such as greater than or equal to 10), a linear regression model of each first detection platform and the second detection platform can be constructed, and the linear regression model can be updated each time a retest result is added, so that consistency comparison between platforms can be quickly performed, and a basis for potential system bias correction is provided.

[0054] S6, judging whether the retest result falls within the second confidence interval of the regression straight line; if the retest result does not fall within the confidence interval of the regression straight line, executing step S7, and if the retest result falls within the confidence interval of the regression straight line, executing step S8.

[0055] S7, determining that the retest result is inconsistent with the abnormal preliminary test result; S8, determining that the retest result is consistent with the abnormal preliminary test result.

[0056] The retest result obtained each time is compared with the second confidence interval of the corresponding regression straight line. If the retest result does not fall within the second confidence interval of the regression straight line, it indicates that the retest result of the first detection platform deviates from the reasonable range expected based on the regression model, and at this time, it is determined that the corresponding first detection platform and the second detection platform are inconsistent in detection, which helps to timely find the detection platform that may have detection bias. If the retest result falls within the second confidence interval of the regression straight line, it indicates that the detection result of the first detection platform is within a reasonable fluctuation range, and it is determined that the corresponding first detection platform and the second detection platform are consistent in detection.

[0057] The embodiment constructs a linear regression model including a regression straight line and its second confidence interval based on the retest result of each first detection platform and the abnormal preliminary test result of the second detection platform when the number of difference distribution data is greater than or equal to the preset number, and the model can be updated in time with the increase of the retest result, so that consistency comparison between platforms can be quickly performed. Then, it is judged whether the retest result falls within the second confidence interval of the regression straight line. If it does not fall within, it is determined that the corresponding first detection platform and the second detection platform are inconsistent in detection, and if it falls within, it is determined that the detection is consistent, which provides a basis for potential bias correction, can timely find the consistency of the test results of each first detection platform and the second detection platform, timely find abnormal data and detection platforms, and provides a reliable basis for ensuring detection quality, which helps to improve the reliability and credibility of the overall detection.

[0058] Further, when the first detection platform and the second detection platform are determined to be inconsistent, an alarm can be sent to the corresponding detection platform, and relevant personnel can quickly conduct in-depth investigation on the two detection platforms according to the alarm information, such as checking whether the equipment is normally running, whether there is a loophole in the detection process, whether the operator is operating normally, and the like. By discovering and solving the problems in time, a series of wrong decisions and adverse consequences caused by inaccurate detection results can be avoided, and the detection performance of the detection platform can be further optimized, and the accuracy and stability of the detection can be improved.

[0059] Further, the step S5 of constructing the linear regression model based on the retest results of each first detection platform and the abnormal preliminary test results of the second detection platform comprises: S51, calculating the regression coefficient and the intercept based on the retest results of each first detection platform and the abnormal preliminary test results of the second detection platform by using the least square method.

[0060] S52, generating a regression straight line based on the regression coefficient and the intercept.

[0061] Exemplarily, the regression straight line can be calculated in the following manner: , y represents the retest result of the first detection platform, x represents the abnormal preliminary test result of the second detection platform, represents the regression coefficient, represents the intercept.

[0062] S53, calculating the residual standard deviation based on the first retest result of each measured platform and the standard retest result of the standard platform.

[0063] S54, calculating the second confidence interval of the regression straight line based on the residual standard deviation.

[0064] According to the regression straight line, the retest result of the corresponding first detection platform can be predicted according to the abnormal preliminary test result of the second detection platform, but the predicted result may deviate from the actual first detection platform retest result, and because the predicted value of the second detection platform retest result based on the regression straight line may have a fluctuation interval, the residual standard deviation of the retest result and the abnormal preliminary test result is calculated, and then the second confidence interval of the regression straight line is calculated according to the residual standard deviation, so as to determine whether the actual retest result is within a reasonable fluctuation range. If the retest result falls within the second confidence interval, it means that the difference between the first retest result and the predicted result is within an acceptable range; if the retest result falls outside the second confidence interval, the difference between the retest result and the predicted result is too large. By calculating the second confidence interval, the detection deviation of the two detection platforms can be found in time, the accuracy and reliability of the detection result can be enhanced, and the stability and effectiveness of the entire detection system can be ensured.

[0065] For example, Figure 3As shown, the linear regression model constructed in step S5 can also be visually displayed.

[0066] The horizontal coordinate is the predicted retest result of the first detection platform, the vertical coordinate is the abnormal preliminary test result of the second detection platform, the yellow data points are the retest results of each first detection platform, the red straight line is the regression straight line, and the gray area is the second confidence interval of the regression straight line.

[0067] The embodiment displays the regression result in a graphical manner. By updating the model in real time, the manager can intuitively understand the consistency of the detection results between the first detection platform and the second detection platform, and provide a scientific basis for the quantitative analysis and correction of the system bias between the platforms.

[0068] As shown, Figure 4 The embodiment of the application also provides a data comparison and analysis device based on retest results, which specifically comprises: The acquisition module 101 is configured to acquire retest results of any first detection platform for target items, the target items being items with abnormal preliminary test results, and the first detection platform being different from the second detection platform that generates the abnormal preliminary test results; The first calculation module 102 is configured to calculate a difference value based on the abnormal preliminary test result and the retest result; The second calculation module 103 is configured to calculate difference distribution data based on the accumulated plurality of difference values, the difference distribution data including a difference distribution range and a first confidence interval; The analysis module 104 is configured to analyze the consistency of the retest result and the abnormal preliminary test result based on the difference distribution data.

[0069] In the embodiment, the specific processing of the acquisition module 101, the first calculation module 102, the second calculation module 103, and the analysis module 104 in the data comparison and analysis device based on retest results and the technical effects brought by the specific processing can be respectively referred to the related descriptions of steps 101-104 in the corresponding embodiments, which will not be repeated here. Figure 1 The corresponding embodiments in the corresponding embodiments, the specific processing of the acquisition module 101, the first calculation module 102, the second calculation module 103, and the analysis module 104 in the data comparison and analysis device based on retest results and the technical effects brought by the specific processing can be respectively referred to the related descriptions of steps 101-104 in the corresponding embodiments, which will not be repeated here.

[0070] The embodiment provided in the embodiment of the method exists as a device embodiment corresponding to the above method embodiment. The device for data comparison and analysis based on retest results provided in this embodiment only needs to replace the retest of the abnormal primary test result project, and performs real-time analysis based on the retest result to quickly and timely find out whether the test result is consistent. The retest result of the project of the primary test result anomaly is obtained through a first detection platform different from the first detection platform that produces the abnormal primary test result. Then, the difference value is calculated based on the abnormal primary test result and the retest result, respectively, and the difference between the test results of the two detection platforms is clearly and accurately obtained. Then, the difference distribution data is calculated based on the continuously accumulated difference value, and the data is continuously updated. Finally, the consistency of the retest result and the abnormal primary test result is analyzed based on the difference distribution data. This method fully excavates the data value of the retest sample in the abnormal result review, which is originally used for the purpose of the original use, and deeply analyzes and reuses the data that is originally only used for single detection result verification, so that these data break through the limitation of single function. The process is simple, only the retest of the abnormal primary test result is needed, and the consistency of the results of the two devices is compared, without manual comparison. The deviation of the detection platform can be quickly and timely found out, the accuracy of the detection result is improved, and the treatment risk that may occur in subsequent detection is avoided.

[0071] As shown in Figure 5 Further, the device for data comparison and analysis based on retest results provided in the embodiment of the application further comprises: The construction module 105 is configured to, in response to the number of difference distribution data being greater than or equal to a preset number, construct a linear regression model based on the retest result of each first detection platform and the abnormal primary test result of the second detection platform, wherein the linear regression model comprises a regression straight line and a second confidence interval of the regression straight line. The judgment module 106 is configured to judge whether the retest result falls within the second confidence interval of the regression straight line. If the retest result does not fall within the confidence interval of the regression straight line, it is determined that the retest result is inconsistent with the abnormal primary test result. If the retest result falls within the confidence interval of the regression straight line, it is determined that the retest result is consistent with the abnormal primary test result.

[0072] Further, the device for data comparison and analysis based on retest results provided in the embodiment of the application further comprises: The display module 107 is configured to visually display the difference distribution data and / or the linear regression model.

[0073] Those skilled in the art will appreciate that embodiments of the present application can be readily used as software, hardware, or a combination of software and hardware. In a software embodiment, the methods can be tangibly embodied by a computer-readable storage medium having stored

[0074] The present application is described in reference to the drawings using a flowchart and / or a block diagram of the method, apparatus (system) and computer program product according to embodiments of the application. It will be understood that each block of the flowchart and / or block diagram, and combinations of blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 Figure 1

[0075] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart and / or block diagram block or blocks. Figure 1 Figure 1

[0076] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 Figure 1

[0077] Obviously, the embodiments described above are only examples and are not intended to limit the present application. Other variations and modifications of the embodiments disclosed can be apparent to those of ordinary skill in the art, and this application is to be accorded the widest scope as can be reasonably encompassed by the language and concepts of the claims.​​​​​​

Claims

1. A data comparison and analysis method based on retest results, characterized in that: include: Obtaining a retest result of a target item on any first testing platform, the target item being an item with an abnormal initial test result, the first testing platform being different from a second testing platform that produced the abnormal initial test result; Calculating a difference value based on the abnormal initial test result and the retest result; Calculating difference distribution data based on the accumulated plurality of difference values, the difference distribution data including a difference distribution range and a first confidence interval; The consistency of the retest result and the abnormal initial test result is analyzed based on the difference distribution data.

2. The method according to claim 1, characterized in that The calculating of the difference distribution data based on the accumulated plurality of difference values ​​includes: Calculating the mean and standard deviation of the plurality of difference values; Based on the mean and standard deviation, the 95% difference distribution range upper limit and the 95% difference distribution range lower limit are calculated to obtain the difference distribution range; Based on the mean, standard deviation, and number of difference values, the 95% confidence interval upper limit and the 95% confidence interval lower limit are calculated respectively to obtain the first confidence interval.

3. The method according to claim 1, characterized in that The analyzing the consistency of the retest result and the abnormal initial test result based on the difference distribution data includes: Calculating the difference data between the retest results of each of the first detection platforms and the abnormal initial test results respectively, wherein the difference data includes a difference value or a relative difference; Determining whether each of the difference data falls within the difference distribution range and the first confidence interval; If the difference data falls within the difference distribution range and the first confidence interval, then the retest result is determined to be consistent with the abnormal initial test result; If the difference data does not fall within the difference distribution range or the first confidence interval, it is determined that the retest result is inconsistent with the abnormal initial test result.

4. The method according to claim 1, wherein Also includes: In response to the number of the differentially distributed data being greater than or equal to a preset number, constructing a linear regression model based on the retest results of each of the first detection platforms and the abnormal initial test results of the second detection platform, the linear regression model including a regression line and a second confidence interval of the regression line; Determining whether the retest result falls within a second confidence interval of the regression line; If the retest result does not fall within the confidence interval of the regression line, it is determined that the retest result is inconsistent with the abnormal initial test result; If the retest result falls within the confidence interval of the regression line, it is determined that the retest result is consistent with the abnormal initial test result.

5. The method according to claim 4, characterized in that The constructing of a linear regression model based on the retest results of each of the first detection platforms and the abnormal initial test results of the second detection platform includes: Calculating the regression coefficient and the intercept based on the retest results of each of the first detection platforms and the abnormal initial test results of the second detection platform using the least squares method; generating the regression line based on the regression coefficient and the intercept; Calculating the residual standard deviation based on the retest results of each of the first detection platforms and the abnormal initial test results of the second detection platform; A second confidence interval of the regression line is calculated based on the residual standard deviation.

6. The method according to claim 4, characterized in that Also includes: The differential distribution data and / or linear regression model are visualized.

7. A data comparison and analysis device based on retest results, characterized in that: include: An acquisition module, configured to acquire a retest result of a target item on any first detection platform, the target item being an item with an abnormal initial test result, the first detection platform being different from a second detection platform that produced the abnormal initial test result; A first calculation module is used to calculate the difference value based on the abnormal initial test result and the retest result respectively; a second calculation module, configured to calculate difference distribution data based on the accumulated difference values, wherein the difference distribution data includes a difference distribution range and a first confidence interval; An analysis module is used to analyze the consistency of the retest result and the abnormal initial test result based on the difference distribution data.

8. The device according to claim 7, characterized in that Also includes: a construction module, configured to construct, in response to the number of the differentially distributed data being greater than or equal to a preset number, a linear regression model based on the retest results of each of the first detection platforms and the abnormal initial test results of the second detection platform, the linear regression model comprising a regression line and a second confidence interval of the regression line; A judgment module, configured to judge whether the retest result falls within a second confidence interval of the regression line; If the retest result does not fall within the confidence interval of the regression line, it is determined that the retest result is inconsistent with the abnormal initial test result; If the retest result falls within the confidence interval of the regression line, it is determined that the retest result is consistent with the abnormal initial test result.

9. The device according to claim 8, characterized in that Also includes: A display module is used to visually display the difference distribution data and / or linear regression model.

10. A data comparison and analysis device based on retest results, characterized in that: include: A difference analysis device and a memory connected to the difference analysis device; wherein the memory stores instructions that can be executed by the difference analysis device, and the instructions are executed by the difference analysis device to enable the difference analysis device to perform the data comparison and analysis method based on retest results as described in any one of claims 1 to 6.

Images