Hemodialysis data processing method and system

Abstract

The invention discloses a hemodialysis data processing method and system, and relates to the technical field of hemodialysis, and the method comprises the following steps: 1, collecting instrument core parameters in real time through a sensor arranged in a dialysis instrument, and collecting patient physiological data in real time through a circulation pipeline sensor arranged outside a patient body, environment interference data are collected in real time through a sensor arranged in the dialysis chamber and subjected to standardization processing; determining an error source according to a self-correlation analysis result, a cross-correlation analysis result and a longitudinal trend analysis result, carrying out layered correction on instrument core parameters, patient physiological data and environmental interference data, triggering a dialysis instrument self-inspection program for internal errors, and carrying out adaptive filtering on external errors to obtain a dialysis instrument self-inspection result; according to the method, the patient error model is established for individual specific interference, and the calibrated data is output, so that the error can be processed in a targeted manner, the accuracy of hemodialysis data processing is improved, and the treatment effect of hemodialysis is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of hemodialysis, and in particular to a hemodialysis data processing method and system. Background Art

[0002] Hemodialysis is a form of renal replacement therapy for patients with acute or chronic renal failure. It involves draining blood from the body and passing it through a dialyzer composed of numerous hollow fibers. Blood and a solution containing electrolytes similar to those in the body (dialysate) undergo a substance exchange process inside and outside the fibers through diffusion, ultrafiltration, adsorption, and convection. This process removes metabolic waste, maintains electrolyte and acid-base balance, and removes excess water from the body. The purified blood is then returned to the body for transfusion. In the traditional hemodialysis data processing process, due to the long-term operation of the dialysis instrument, its hardware often malfunctions, resulting in inaccurate collected instrument data. Different patients also have different physiological characteristics, and the environmental factors of the dialysis room will also affect the collected data. In the existing hemodialysis data processing process, it is not convenient to identify the source of errors in a timely and accurate manner. Due to the lack of effective error identification, traditional methods often treat all errors as the same type and are unable to take targeted corrective measures according to different error sources. As a result, when faced with complex and changeable error situations, the data processing results are prone to deviations, affecting the accurate assessment of dialysis adequacy.

[0003] In view of this, the present invention proposes a hemodialysis data processing method and system to remedy and improve the deficiencies of the prior art. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention provides a hemodialysis data processing method and system to solve the corresponding technical problems raised in the above background technology.

[0005] To achieve the above objectives, the present invention adopts a technical solution: a hemodialysis data processing method comprising the following steps: Step 1: The core parameters of the dialysis instrument are collected in real time by sensors installed inside the dialysis instrument, the patient's physiological data are collected in real time by sensors installed in the circulation pipeline outside the patient's body, and the environmental interference data are collected in real time by sensors installed in the dialysis room and standardized; Step 2: Obtain the acquisition frequency of the instrument core parameters, the acquisition frequency of the patient physiological data, and the acquisition frequency of the environmental interference data, and perform cubic spline interpolation on the low acquisition frequency parameters based on the highest acquisition frequency parameter as the reference time axis, generate virtual sampling values ​​at the time point of the highest acquisition frequency parameter, and associate the virtual sampling values ​​of the instrument core parameters, the virtual sampling values ​​of the patient physiological data, and the virtual sampling values ​​of the environmental interference data at the same time point to form a time series matrix; Step 3: Define the error types as internal errors caused by dialysis instrument hardware failures and external errors caused by environmental interference or individual patient differences. Based on the time series matrix, perform autocorrelation analysis on the virtual sampling values ​​of the instrument's core parameters. Combined with the historical maintenance records of the dialysis instrument, mark high-frequency fault points. Perform cross-correlation analysis on the virtual sampling values ​​of the environmental interference data and the virtual sampling values ​​of the patient's physiological data, calculate the interference coupling coefficient, perform longitudinal trend analysis on the data of multiple treatments for the same patient, identify individual-specific interference characteristics, and determine the error source based on the results of the autocorrelation analysis, cross-correlation analysis, and longitudinal trend analysis. Step 4: Based on the source of error, perform hierarchical corrections on the instrument core parameters, patient physiological data, and environmental interference data. Trigger the dialysis instrument self-test program for internal errors, use adaptive filtering for external errors, establish a patient error model for individual-specific interference, and output the calibrated data.

[0006] Preferably, the core parameters of the instrument include blood pump speed, transmembrane pressure and ultrafiltration rate, the patient physiological data include systolic blood pressure, venous pressure and activated clotting time, and the environmental interference data include room temperature, power frequency noise intensity and light intensity.

[0007] As a preferred method, the specific process of marking high-frequency fault points is as follows: S101, obtain the acquisition frequency of the instrument core parameters, the acquisition frequency of the patient's physiological data, and the acquisition frequency of the environmental interference data, and use the time axis corresponding to the highest acquisition frequency parameter as the reference time axis. The time points are ; The actual acquisition time points and corresponding values ​​of the low acquisition frequency parameters are used as known data points, and a cubic spline function is constructed between every two adjacent data points. Based on the constructed cubic spline function, the corresponding function values ​​are calculated at each time point on the reference time axis to obtain the virtual sampling values ​​of the low acquisition frequency parameters on the reference time axis. The virtual sampling values ​​of the instrument core parameters, the virtual sampling values ​​of the patient's physiological data, and the virtual sampling values ​​of the environmental interference data at the same time point are combined to form a data row. The data rows corresponding to all time points are arranged in sequence according to the time sequence of the reference time axis to form a time series matrix. S102. Calculate the autocorrelation coefficients at different lag orders k (k = 0, 1, 2, ...) based on the virtual sampling values ​​of the instrument core parameters. , and its calculation formula is: ; in, Represents the virtual sampling value of the instrument core parameter at the i-th time point in the time series matrix; It represents the average value of the virtual sampling values ​​of the instrument core parameters at all time points; Calculate the autocorrelation coefficients of blood pump speed a, transmembrane pressure b and ultrafiltration rate c at different lag orders respectively, and obtain a set of autocorrelation coefficient sequences for each 、 as well as ; S103, according to the preset autocorrelation coefficient threshold , for each instrument core parameter autocorrelation coefficient sequence, when the autocorrelation coefficient The absolute value exceeds When the lag order k is reached, it indicates that the value of the instrument core parameter is abnormally correlated with its own value at other time points. Record the corresponding lag order k and time point. ,in, is the starting position of the time point involved in the current calculation of the autocorrelation coefficient, Relative to the starting time point The time point after the lag of k time units; will exceed The time points are initially marked as potential failure points to form a preliminary failure point set ; S104, obtain historical maintenance records from the maintenance database of the dialysis instrument, and Every time point in , if in If there is a maintenance record related to the dialysis instrument hardware failure before and after, then the time point From the initial fault point collection Remove, and after screening, get the filtered fault point set ; For the filtered fault point set Every time point in , mark them in the time series matrix, and output the time series matrix with marked high-frequency fault points.

[0008] As an example, the specific process of calculating the interference coupling coefficient is as follows: S201. Obtain virtual sampling values ​​of environmental interference data and virtual sampling values ​​of patient physiological data at the same time point from the time series matrix and combine them. For each combination of virtual sampling values ​​of environmental interference data and virtual sampling values ​​of patient physiological data, calculate their cross-correlation coefficients at different lag orders m (m=0, 1, 2, ...). , and its calculation formula is: ; in, Represents the virtual sampling value of the environmental interference data at the qth time point in the time series matrix; Represents the average value of the virtual sampling values ​​of the environmental interference data at all n time points; Represents the virtual sampling value of the patient's physiological data at the q+mth time point in the time series matrix; Represents the virtual sampling value of the patient's physiological data at the qth time point in the time series matrix; Represents the average value of the virtual sampling values ​​of the patient's physiological data at all n time points; The cross-correlation coefficients between room temperature and systolic blood pressure, room temperature and venous pressure, room temperature and activated clotting time, power frequency noise intensity and systolic blood pressure, power frequency noise intensity and venous pressure, power frequency noise intensity and activated clotting time, light intensity and systolic blood pressure, light intensity and venous pressure, and light intensity and activated clotting time at different lag orders were calculated respectively, and multiple sets of cross-correlation coefficient sequences were obtained; S202: According to the preset mutual correlation coefficient threshold For each set of calculated cross-correlation coefficient sequences, find the value with the largest absolute value of the cross-correlation coefficient and its corresponding lag order. If the maximum absolute value exceeds the cross-correlation coefficient threshold , indicating that there is a high correlation between the environmental interference data and the patient's physiological data, the maximum absolute value is used as the interference coupling coefficient of the combination of the corresponding environmental interference data and the patient's physiological data.

[0009] Preferably, the specific process of identifying individual-specific interference features is as follows: S301. Obtain virtual sampling values ​​of instrument core parameters, patient physiological data, and environmental interference data at each time point for the same patient through a time series matrix, and organize the data of each treatment into a matrix format according to the chronological order of the treatments, with each row representing one treatment and each column corresponding to the values ​​of the aforementioned data at different time points. S302: Calculate the mean of each patient's physiological data for each treatment. , and its calculation formula is: ; Where u is the number of virtual sampling points of the patient's physiological data during the treatment process; is the value of the pth virtual sampling point; The above calculations were performed on the patient's systolic blood pressure, venous pressure, and activated clotting time for each treatment, and the mean values ​​of the patient's physiological data for each treatment were obtained; S303: Calculate the variance of each patient's physiological data for each treatment. , and its calculation formula is: ; The above calculations were performed for the systolic blood pressure, venous pressure, and activated clotting time for each treatment, and the variance of each patient's physiological data in each treatment was obtained; S304. Compare the means and variances of the same patient's physiological data between different treatment times. If a combination of the means and variances of the patient's physiological data appears repeatedly in multiple treatments, preliminarily determine that the combination is an individual-specific interference feature. With the number of treatments as the horizontal axis and the mean of the patient's physiological data as the vertical axis, a trend graph of the mean changing with the number of treatments was drawn to observe the changing trend of the mean. The linear regression method was used to fit the curve of the mean changing with the number of treatments and the regression coefficient was calculated. If the regression coefficient was not zero, it indicated that the mean had a linear changing trend. With the number of treatments as the horizontal axis and the variance of the patient's physiological data as the vertical axis, a trend graph of the variance changing with the number of treatments was drawn to observe the changing trend of the variance. The moving average method was used to observe the cyclical changes in the variance with the increase in the number of treatments. If the patient's physiological data gradually increased with the number of treatments and its variance also showed an increasing trend, and this change pattern persisted in multiple treatments, then the combination was further determined to be an individual-specific interference feature.

[0010] As an example, the specific process of determining the error source is as follows: S401. Obtain high-frequency fault points marked by autocorrelation analysis and combine them with the historical maintenance records of the dialysis instrument. If there is an abnormal change in the instrument core parameters at a certain time point, and there is a related maintenance record near this time point in the historical maintenance records, it is determined that the error is caused by an internal dialysis instrument hardware failure. S402. Obtaining the interference coupling coefficient calculated by cross-correlation analysis, and selecting the median as a preset standard threshold based on the distribution range of the interference coupling coefficient between the environmental interference data and the patient's physiological data under normal conditions. If the interference coupling coefficient between the environmental interference data and the patient's physiological data is greater than the preset standard threshold, it is determined to be an external error caused by environmental interference; S403. Obtain individual-specific interference characteristics identified by longitudinal trend analysis. If there are no abnormal changes in the instrument core parameters and there are no relevant maintenance records at that time point, and the interference coupling coefficient between the environmental interference data and the patient's physiological data does not exceed the preset standard threshold, and the patient's physiological data has individual-specific interference characteristics, then it is determined to be an external error caused by individual differences among patients.

[0011] As an example, the specific process of outputting the calibrated data is as follows: S501. When it is determined that the error source is an internal error caused by a dialysis instrument hardware failure, the dialysis instrument self-test program is triggered to repair the dialysis instrument hardware and replace the faulty parts. When it is determined that the error source is an external error caused by environmental interference, the environmental interference data is processed using an adaptive filtering method. S502. When the error source is determined to be external error caused by individual patient differences, a patient error model is established for individual-specific interference using the patient's age, weight, environmental interference data, and instrument core parameters as independent variables and the patient's physiological data as dependent variables through regression analysis. The formula is: ; Among them, Z is the patient's physiological data; is the patient's age; is the patient's weight; Y is the environmental interference data; X is the instrument core parameter; All are regression coefficients; is the error term; The patient error model is trained and optimized through historical patient physiological data, historical environmental interference data, and historical instrument core parameters, and the value of the regression coefficient is determined to obtain an updated patient error model; S503. According to the processing results of different error types, the instrument core parameters, environmental interference data and patient physiological data are corrected, and the corrected data are recorded in the data recording file. At the same time, the corresponding positions in the time series matrix are marked, and the calibrated data are output in the specified format.

[0012] A hemodialysis data processing system includes a data acquisition unit, a data processing unit, an error analysis unit, and an error calibration unit; The data acquisition unit is used to collect the core parameters of the dialysis instrument in real time through sensors installed inside the dialysis instrument, collect the patient's physiological data in real time through sensors installed in the circulation pipeline outside the patient's body, and collect environmental interference data in real time through sensors installed in the dialysis room and perform standardized processing; The data processing unit is used to obtain the acquisition frequency of the instrument core parameters, the patient physiological data, and the environmental interference data, and perform cubic spline interpolation on the low acquisition frequency parameters based on the highest acquisition frequency parameter as the reference time axis, generate virtual sampling values ​​at the time point of the highest acquisition frequency parameter, and associate the virtual sampling values ​​of the instrument core parameters, the virtual sampling values ​​of the patient physiological data, and the virtual sampling values ​​of the environmental interference data at the same time point to form a time series matrix; The error analysis unit is used to define the error type as internal error caused by dialysis instrument hardware failure and external error caused by environmental interference or individual patient differences. Based on the time series matrix, autocorrelation analysis is performed on the virtual sampling values ​​of the instrument core parameters to detect periodic deviations. In combination with the historical maintenance records of the dialysis instrument, high-frequency fault points are marked. Cross-correlation analysis is performed on the virtual sampling values ​​of the environmental interference data and the patient's physiological data to calculate the interference coupling coefficient. The data of multiple treatments for the same patient are longitudinally compared to identify individual-specific interference. The error source is determined based on the results of the autocorrelation analysis, cross-correlation analysis, and longitudinal comparison. The error calibration unit is used to perform hierarchical corrections on instrument core parameters, patient physiological data, and environmental interference data based on the source of the error, trigger the dialysis instrument self-test program for internal errors, use adaptive filtering for external errors, establish a patient-specific error model for individual-specific interference, and output calibrated data.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: by collecting instrument core parameters, patient physiological data and environmental interference data, and obtaining the acquisition frequency of the instrument core parameters, the acquisition frequency of the patient physiological data and the acquisition frequency of the environmental interference data, the highest acquisition frequency parameter is used as the reference time axis, cubic spline interpolation is performed on the low acquisition frequency parameters, virtual sampling values ​​are generated at the time point of the highest acquisition frequency parameter, the virtual sampling values ​​of the instrument core parameters, the virtual sampling values ​​of the patient physiological data and the virtual sampling values ​​of the environmental interference data at the same time point are associated to form a time series matrix, the error type is defined as the internal error caused by the dialysis instrument hardware failure and the external error caused by the environmental interference or the individual difference of the patient, the autocorrelation analysis is performed on the virtual sampling values ​​of the instrument core parameters, and the dialysis instrument history is combined with the dialysis instrument history. Maintain records, mark high-frequency fault points, perform cross-correlation analysis on virtual sampling values ​​of environmental interference data and virtual sampling values ​​of patient physiological data, calculate interference coupling coefficients, perform longitudinal trend analysis on data of multiple treatments of the same patient, identify individual-specific interference characteristics, determine the source of errors based on the results of autocorrelation analysis, cross-correlation analysis and longitudinal trend analysis, perform hierarchical corrections on instrument core parameters, patient physiological data and environmental interference data, trigger the dialysis instrument self-test program for internal errors, use adaptive filtering for external errors, establish a patient error model for individual-specific interference, and output calibrated data. This can perform targeted processing on errors, improve the accuracy of hemodialysis data processing, help improve the therapeutic effect of hemodialysis, increase the safety of hemodialysis treatment, and reduce the risk of complications. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a schematic diagram of the overall structure of a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0015] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0016] Embodiments of the present invention: Please refer to Figure 1 As shown, a hemodialysis data processing method includes the following steps: Step 1: The core parameters of the dialysis instrument are collected in real time by sensors installed inside the dialysis instrument, the patient's physiological data are collected in real time by sensors installed in the circulation pipeline outside the patient's body, and the environmental interference data are collected in real time by sensors installed in the dialysis room and standardized; Step 2: Obtain the acquisition frequency of the instrument core parameters, the acquisition frequency of the patient physiological data, and the acquisition frequency of the environmental interference data, and perform cubic spline interpolation on the low acquisition frequency parameters based on the highest acquisition frequency parameter as the reference time axis, generate virtual sampling values ​​at the time point of the highest acquisition frequency parameter, and associate the virtual sampling values ​​of the instrument core parameters, the virtual sampling values ​​of the patient physiological data, and the virtual sampling values ​​of the environmental interference data at the same time point to form a time series matrix; Step 3: Define the error types as internal errors caused by dialysis instrument hardware failures and external errors caused by environmental interference or individual patient differences. Based on the time series matrix, perform autocorrelation analysis on the virtual sampling values ​​of the instrument's core parameters. Combined with the historical maintenance records of the dialysis instrument, mark high-frequency fault points. Perform cross-correlation analysis on the virtual sampling values ​​of the environmental interference data and the virtual sampling values ​​of the patient's physiological data, calculate the interference coupling coefficient, perform longitudinal trend analysis on the data of multiple treatments for the same patient, identify individual-specific interference characteristics, and determine the error source based on the results of the autocorrelation analysis, cross-correlation analysis, and longitudinal trend analysis. Step 4: Based on the source of error, perform hierarchical corrections on the instrument core parameters, patient physiological data, and environmental interference data. Trigger the dialysis instrument self-test program for internal errors, use adaptive filtering for external errors, establish a patient error model for individual-specific interference, and output the calibrated data.

[0017] The core parameters of the instrument include blood pump speed, transmembrane pressure and ultrafiltration rate; the patient's physiological data include systolic blood pressure, venous pressure and activated clotting time; and the environmental interference data include room temperature, power frequency noise intensity and light intensity.

[0018] The specific process of marking high-frequency fault points is as follows: S101, obtain the acquisition frequency of the instrument core parameters, the acquisition frequency of the patient's physiological data, and the acquisition frequency of the environmental interference data, and use the time axis corresponding to the highest acquisition frequency parameter as the reference time axis. The time points are ; The actual acquisition time points and corresponding values ​​of the low acquisition frequency parameters are used as known data points, and a cubic spline function is constructed between every two adjacent data points. Based on the constructed cubic spline function, the corresponding function values ​​are calculated at each time point on the reference time axis to obtain the virtual sampling values ​​of the low acquisition frequency parameters on the reference time axis. The virtual sampling values ​​of the instrument core parameters, the virtual sampling values ​​of the patient's physiological data, and the virtual sampling values ​​of the environmental interference data at the same time point are combined to form a data row. The data rows corresponding to all time points are arranged in sequence according to the time sequence of the reference time axis to form a time series matrix. S102. Calculate the autocorrelation coefficients at different lag orders k (k = 0, 1, 2, ...) based on the virtual sampling values ​​of the instrument core parameters. , and its calculation formula is: ; in, Represents the virtual sampling value of the instrument core parameter at the i-th time point in the time series matrix; It represents the average value of the virtual sampling values ​​of the instrument core parameters at all time points; Calculate the autocorrelation coefficients of blood pump speed a, transmembrane pressure b and ultrafiltration rate c at different lag orders respectively, and obtain a set of autocorrelation coefficient sequences for each 、 as well as ; S103, according to the preset autocorrelation coefficient threshold , for each instrument core parameter autocorrelation coefficient sequence, when the autocorrelation coefficient The absolute value exceeds When the lag order k is reached, it indicates that the value of the instrument core parameter is abnormally correlated with its own value at other time points. Record the corresponding lag order k and time point. ,in, is the starting position of the time point involved in the current calculation of the autocorrelation coefficient, Relative to the starting time point The time point after the lag of k time units; will exceed The time points are initially marked as potential failure points to form a preliminary failure point set ; S104, obtain historical maintenance records from the maintenance database of the dialysis instrument, and Every time point in , if in If there is a maintenance record related to the dialysis instrument hardware failure before and after, then the time point From the initial fault point collection Remove, and after screening, get the filtered fault point set ; For the filtered fault point set Every time point in , mark them in the time series matrix, and output the time series matrix with marked high-frequency fault points.

[0019] The specific process of calculating the interference coupling coefficient is as follows: S201. Obtain virtual sampling values ​​of environmental interference data and virtual sampling values ​​of patient physiological data at the same time point from the time series matrix and combine them. For each combination of virtual sampling values ​​of environmental interference data and virtual sampling values ​​of patient physiological data, calculate their cross-correlation coefficients at different lag orders m (m=0, 1, 2, ...). , and its calculation formula is: ; in, Represents the virtual sampling value of the environmental interference data at the qth time point in the time series matrix; Represents the average value of the virtual sampling values ​​of the environmental interference data at all n time points; Represents the virtual sampling value of the patient's physiological data at the q+mth time point in the time series matrix; Represents the virtual sampling value of the patient's physiological data at the qth time point in the time series matrix; Represents the average value of the virtual sampling values ​​of the patient's physiological data at all n time points; The cross-correlation coefficients between room temperature and systolic blood pressure, room temperature and venous pressure, room temperature and activated clotting time, power frequency noise intensity and systolic blood pressure, power frequency noise intensity and venous pressure, power frequency noise intensity and activated clotting time, light intensity and systolic blood pressure, light intensity and venous pressure, and light intensity and activated clotting time at different lag orders were calculated respectively, and multiple sets of cross-correlation coefficient sequences were obtained; S202: According to the preset mutual correlation coefficient threshold For each set of calculated cross-correlation coefficient sequences, find the value with the largest absolute value of the cross-correlation coefficient and its corresponding lag order. If the maximum absolute value exceeds the cross-correlation coefficient threshold , indicating that there is a high correlation between the environmental interference data and the patient's physiological data, the maximum absolute value is used as the interference coupling coefficient of the combination of the corresponding environmental interference data and the patient's physiological data.

[0020] The specific process of identifying individual-specific interference features is as follows: S301. Obtain virtual sampling values ​​of instrument core parameters, patient physiological data, and environmental interference data at each time point for the same patient through a time series matrix, and organize the data of each treatment into a matrix format according to the chronological order of the treatments, with each row representing one treatment and each column corresponding to the values ​​of the aforementioned data at different time points. S302: Calculate the mean of each patient's physiological data for each treatment. , and its calculation formula is: ; Where u is the number of virtual sampling points of the patient's physiological data during the treatment process; is the value of the pth virtual sampling point; The above calculations were performed on the patient's systolic blood pressure, venous pressure, and activated clotting time for each treatment, and the mean values ​​of the patient's physiological data for each treatment were obtained; S303: Calculate the variance of each patient's physiological data for each treatment. , and its calculation formula is: ; The above calculations were performed for the systolic blood pressure, venous pressure, and activated clotting time for each treatment, and the variance of each patient's physiological data in each treatment was obtained; S304. Compare the means and variances of the same patient's physiological data between different treatment times. If a combination of the means and variances of the patient's physiological data appears repeatedly in multiple treatments, preliminarily determine that the combination is an individual-specific interference feature. With the number of treatments as the horizontal axis and the mean of the patient's physiological data as the vertical axis, a trend graph of the mean changing with the number of treatments was drawn to observe the changing trend of the mean. The linear regression method was used to fit the curve of the mean changing with the number of treatments and the regression coefficient was calculated. If the regression coefficient was not zero, it indicated that the mean had a linear changing trend. With the number of treatments as the horizontal axis and the variance of the patient's physiological data as the vertical axis, a trend graph of the variance changing with the number of treatments was drawn to observe the changing trend of the variance. The moving average method was used to observe the cyclical changes in the variance with the increase in the number of treatments. If the patient's physiological data gradually increased with the number of treatments and its variance also showed an increasing trend, and this change pattern persisted in multiple treatments, then the combination was further determined to be an individual-specific interference feature.

[0021] The specific process for determining the source of error is as follows: S401. Obtain high-frequency fault points marked by autocorrelation analysis and combine them with the historical maintenance records of the dialysis instrument. If there is an abnormal change in the instrument core parameters at a certain time point, and there is a related maintenance record near this time point in the historical maintenance records, it is determined that the error is caused by an internal dialysis instrument hardware failure. S402. Obtaining the interference coupling coefficient calculated by cross-correlation analysis, and selecting the median as a preset standard threshold based on the distribution range of the interference coupling coefficient between the environmental interference data and the patient's physiological data under normal conditions. If the interference coupling coefficient between the environmental interference data and the patient's physiological data is greater than the preset standard threshold, it is determined to be an external error caused by environmental interference; S403. Obtain individual-specific interference characteristics identified by longitudinal trend analysis. If there are no abnormal changes in the instrument core parameters and there are no relevant maintenance records at that time point, and the interference coupling coefficient between the environmental interference data and the patient's physiological data does not exceed the preset standard threshold, and the patient's physiological data has individual-specific interference characteristics, then it is determined to be an external error caused by individual differences among patients.

[0022] The specific process of outputting calibrated data is as follows: S501. When it is determined that the error source is an internal error caused by a dialysis instrument hardware failure, the dialysis instrument self-test program is triggered to repair the dialysis instrument hardware and replace the faulty parts. When it is determined that the error source is an external error caused by environmental interference, the environmental interference data is processed using an adaptive filtering method. S502. When the error source is determined to be external error caused by individual patient differences, a patient error model is established for individual-specific interference using the patient's age, weight, environmental interference data, and instrument core parameters as independent variables and the patient's physiological data as dependent variables through regression analysis. The formula is: ; Among them, Z is the patient's physiological data; is the patient's age; is the patient's weight; Y is the environmental interference data; X is the instrument core parameter; All are regression coefficients; is the error term; The patient error model is trained and optimized through historical patient physiological data, historical environmental interference data, and historical instrument core parameters, and the value of the regression coefficient is determined to obtain an updated patient error model; S503. According to the processing results of different error types, the instrument core parameters, environmental interference data and patient physiological data are corrected, and the corrected data are recorded in the data recording file. At the same time, the corresponding positions in the time series matrix are marked, and the calibrated data are output in the specified format.

[0023] A hemodialysis data processing system includes a data acquisition unit, a data processing unit, an error analysis unit, and an error calibration unit; The data acquisition unit is used to collect the core parameters of the dialysis instrument in real time through sensors installed inside the dialysis instrument, collect the patient's physiological data in real time through sensors installed in the circulation pipeline outside the patient's body, and collect environmental interference data in real time through sensors installed in the dialysis room and perform standardized processing; The data processing unit is used to obtain the acquisition frequency of the instrument core parameters, the patient's physiological data, and the environmental interference data, and perform cubic spline interpolation on the low acquisition frequency parameters based on the highest acquisition frequency parameter as the reference time axis, generate virtual sampling values ​​at the time point of the highest acquisition frequency parameter, and associate the virtual sampling values ​​of the instrument core parameters, the virtual sampling values ​​of the patient's physiological data, and the virtual sampling values ​​of the environmental interference data at the same time point to form a time series matrix; The error analysis unit is used to define error types as internal errors caused by dialysis instrument hardware failures and external errors caused by environmental interference or individual patient differences. Based on the time series matrix, it performs autocorrelation analysis on the virtual sampling values ​​of the instrument's core parameters to detect periodic deviations. Combined with the dialysis instrument's historical maintenance records, it marks high-frequency fault points. It also performs cross-correlation analysis on the virtual sampling values ​​of the environmental interference data and the patient's physiological data to calculate the interference coupling coefficient. It conducts longitudinal comparisons of data from multiple treatments on the same patient to identify individual-specific interference. The error source is determined based on the results of the autocorrelation analysis, cross-correlation analysis, and longitudinal comparisons. The error calibration unit is used to perform hierarchical corrections on the instrument's core parameters, patient physiological data, and environmental interference data based on the source of the error. It triggers the dialysis instrument's self-test program for internal errors, uses adaptive filtering for external errors, establishes a patient-specific error model for individual-specific interference, and outputs calibrated data.

[0024] By collecting instrument core parameters, patient physiological data and environmental interference data, and obtaining the acquisition frequency of instrument core parameters, patient physiological data and environmental interference data, cubic spline interpolation is performed on low acquisition frequency parameters based on the highest acquisition frequency parameter as the reference time axis, and virtual sampling values ​​are generated at the time point of the highest acquisition frequency parameter. The virtual sampling values ​​of instrument core parameters, patient physiological data and environmental interference data at the same time point are correlated to form a time series matrix. The error type is defined as internal error caused by dialysis instrument hardware failure and external error caused by environmental interference or individual differences among patients. Autocorrelation analysis is performed on the virtual sampling values ​​of instrument core parameters. Combined with the historical maintenance records of dialysis instruments, high-frequency errors are marked. Fault point, cross-correlation analysis is performed on the virtual sampling values ​​of environmental interference data and the virtual sampling values ​​of patient physiological data, the interference coupling coefficient is calculated, longitudinal trend analysis is performed on the data of multiple treatments of the same patient, individual-specific interference characteristics are identified, and the source of error is determined according to the results of autocorrelation analysis, cross-correlation analysis and longitudinal trend analysis. The core parameters of the instrument, patient physiological data and environmental interference data are layered and corrected. The dialysis instrument self-test program is triggered for internal errors, adaptive filtering is used for external errors, a patient error model is established for individual-specific interference, and the calibrated data is output. It can carry out targeted processing of errors, improve the accuracy of hemodialysis data processing, help improve the treatment effect of hemodialysis, increase the safety of hemodialysis treatment, and reduce the risk of complications.

[0025] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technical personnel in this field for each set of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0026] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by those skilled in the art according to actual conditions. In the two embodiments provided in this application, it should be understood that the disclosed devices and systems can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the module division is merely a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not implemented. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interface, and the indirect coupling or communication connection of devices or modules may be electrical, mechanical or other forms. The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A method for processing hemodialysis data, characterized in that: The following steps are involved: Step 1: The core parameters of the dialysis instrument are collected in real time by sensors installed inside the dialysis instrument, the patient's physiological data are collected in real time by sensors installed in the circulation pipeline outside the patient's body, and the environmental interference data are collected in real time by sensors installed in the dialysis room and standardized; Step 2: Obtain the acquisition frequency of the instrument core parameters, the acquisition frequency of the patient physiological data, and the acquisition frequency of the environmental interference data, and perform cubic spline interpolation on the low acquisition frequency parameters based on the highest acquisition frequency parameter as the reference time axis, generate virtual sampling values ​​at the time point of the highest acquisition frequency parameter, and associate the virtual sampling values ​​of the instrument core parameters, the virtual sampling values ​​of the patient physiological data, and the virtual sampling values ​​of the environmental interference data at the same time point to form a time series matrix; Step 3: Define the error types as internal errors caused by dialysis instrument hardware failures and external errors caused by environmental interference or individual patient differences. Based on the time series matrix, perform autocorrelation analysis on the virtual sampling values ​​of the instrument's core parameters. Combined with the historical maintenance records of the dialysis instrument, mark high-frequency fault points. Perform cross-correlation analysis on the virtual sampling values ​​of the environmental interference data and the virtual sampling values ​​of the patient's physiological data, calculate the interference coupling coefficient, perform longitudinal trend analysis on the data of multiple treatments for the same patient, identify individual-specific interference characteristics, and determine the error source based on the results of the autocorrelation analysis, cross-correlation analysis, and longitudinal trend analysis. Step 4: Based on the source of error, perform hierarchical corrections on the instrument core parameters, patient physiological data, and environmental interference data. Trigger the dialysis instrument self-test program for internal errors, use adaptive filtering for external errors, establish a patient error model for individual-specific interference, and output the calibrated data.

2. The hemodialysis data processing method according to claim 1, characterized in that: The core parameters of the instrument include blood pump speed, transmembrane pressure and ultrafiltration rate; the patient physiological data include systolic blood pressure, venous pressure and activated clotting time; and the environmental interference data include room temperature, power frequency noise intensity and light intensity.

3. The hemodialysis data processing method according to claim 2, characterized in that: The specific process of marking high-frequency fault points is as follows: S101, obtain the acquisition frequency of the instrument core parameters, the acquisition frequency of the patient's physiological data, and the acquisition frequency of the environmental interference data, and use the time axis corresponding to the highest acquisition frequency parameter as the reference time axis. The time points are ; The actual acquisition time points and corresponding values ​​of the low acquisition frequency parameters are used as known data points, and a cubic spline function is constructed between every two adjacent data points. Based on the constructed cubic spline function, the corresponding function values ​​are calculated at each time point on the reference time axis to obtain the virtual sampling values ​​of the low acquisition frequency parameters on the reference time axis. The virtual sampling values ​​of the instrument core parameters, the virtual sampling values ​​of the patient's physiological data, and the virtual sampling values ​​of the environmental interference data at the same time point are combined to form a data row. The data rows corresponding to all time points are arranged in sequence according to the time sequence of the reference time axis to form a time series matrix. S102. Calculate the autocorrelation coefficients at different lag orders k (k = 0, 1, 2, ...) based on the virtual sampling values ​​of the instrument core parameters. , and its calculation formula is: ； in, Represents the virtual sampling value of the instrument core parameter at the i-th time point in the time series matrix; It represents the average value of the virtual sampling values ​​of the instrument core parameters at all time points; Calculate the autocorrelation coefficients of blood pump speed a, transmembrane pressure b and ultrafiltration rate c at different lag orders respectively, and obtain a set of autocorrelation coefficient sequences for each 、 as well as ; S103, according to the preset autocorrelation coefficient threshold , for each instrument core parameter autocorrelation coefficient sequence, when the autocorrelation coefficient The absolute value exceeds When the lag order k is reached, it indicates that the value of the instrument core parameter is abnormally correlated with its own value at other time points. Record the corresponding lag order k and time point. ,in, is the starting position of the time point involved in the current calculation of the autocorrelation coefficient, Relative to the starting time point The time point after the lag of k time units; will exceed The time points are initially marked as potential failure points to form a preliminary failure point set ; S104, obtain historical maintenance records from the maintenance database of the dialysis instrument, and Every time point in , if in If there is a maintenance record related to the dialysis instrument hardware failure before and after, then the time point From the initial fault point collection Remove, and after screening, get the filtered fault point set ; For the filtered fault point set Every time point in , mark them in the time series matrix, and output the time series matrix with marked high-frequency fault points.

4. The hemodialysis data processing method according to claim 3, characterized in that: The specific process of calculating the interference coupling coefficient is as follows: S201. Obtain virtual sampling values ​​of environmental interference data and virtual sampling values ​​of patient physiological data at the same time point from the time series matrix and combine them. For each combination of virtual sampling values ​​of environmental interference data and virtual sampling values ​​of patient physiological data, calculate their cross-correlation coefficients at different lag orders m (m=0, 1, 2, ...). , and its calculation formula is: ； in, Represents the virtual sampling value of the environmental interference data at the qth time point in the time series matrix; Represents the average value of the virtual sampling values ​​of the environmental interference data at all n time points; Represents the virtual sampling value of the patient's physiological data at the q+mth time point in the time series matrix; Represents the virtual sampling value of the patient's physiological data at the qth time point in the time series matrix; Represents the average value of the virtual sampling values ​​of the patient's physiological data at all n time points; The cross-correlation coefficients between room temperature and systolic blood pressure, room temperature and venous pressure, room temperature and activated clotting time, power frequency noise intensity and systolic blood pressure, power frequency noise intensity and venous pressure, power frequency noise intensity and activated clotting time, light intensity and systolic blood pressure, light intensity and venous pressure, and light intensity and activated clotting time at different lag orders were calculated respectively, and multiple sets of cross-correlation coefficient sequences were obtained; S202: According to the preset mutual correlation coefficient threshold For each set of calculated cross-correlation coefficient sequences, find the value with the largest absolute value of the cross-correlation coefficient and its corresponding lag order. If the maximum absolute value exceeds the cross-correlation coefficient threshold , indicating that there is a high correlation between the environmental interference data and the patient's physiological data, the maximum absolute value is used as the interference coupling coefficient of the combination of the corresponding environmental interference data and the patient's physiological data.

5. The hemodialysis data processing method according to claim 4, characterized in that: The specific process of identifying individual-specific interference features is as follows: S301. Obtain virtual sampling values ​​of instrument core parameters, patient physiological data, and environmental interference data at each time point for the same patient through a time series matrix, and organize the data of each treatment into a matrix format according to the chronological order of the treatments, with each row representing one treatment and each column corresponding to the values ​​of the aforementioned data at different time points. S302: Calculate the mean of each patient's physiological data for each treatment. , and its calculation formula is: ； Where u is the number of virtual sampling points of the patient's physiological data during the treatment process; is the value of the pth virtual sampling point; The above calculations were performed on the patient's systolic blood pressure, venous pressure, and activated clotting time for each treatment, and the mean values ​​of the patient's physiological data for each treatment were obtained; S303: Calculate the variance of each patient's physiological data for each treatment. , and its calculation formula is: ； The above calculations were performed for the systolic blood pressure, venous pressure, and activated clotting time for each treatment, and the variance of each patient's physiological data in each treatment was obtained; S304. Compare the means and variances of the same patient's physiological data between different treatment times. If a combination of the means and variances of the patient's physiological data appears repeatedly in multiple treatments, preliminarily determine that the combination is an individual-specific interference feature. With the number of treatments as the horizontal axis and the mean of the patient's physiological data as the vertical axis, a trend graph of the mean changing with the number of treatments was drawn to observe the changing trend of the mean. The linear regression method was used to fit the curve of the mean changing with the number of treatments and the regression coefficient was calculated. If the regression coefficient was not zero, it indicated that the mean had a linear changing trend. With the number of treatments as the horizontal axis and the variance of the patient's physiological data as the vertical axis, a trend graph of the variance changing with the number of treatments was drawn to observe the changing trend of the variance. The moving average method was used to observe the cyclical changes in the variance with the increase in the number of treatments. If the patient's physiological data gradually increased with the number of treatments and its variance also showed an increasing trend, and this change pattern persisted in multiple treatments, then the combination was further determined to be an individual-specific interference feature.

6. The hemodialysis data processing method according to claim 5, characterized in that: The specific process for determining the source of error is as follows: S401. Obtain high-frequency fault points marked by autocorrelation analysis and combine them with the historical maintenance records of the dialysis instrument. If there is an abnormal change in the instrument core parameters at a certain time point, and there is a related maintenance record near this time point in the historical maintenance records, it is determined that the error is caused by an internal dialysis instrument hardware failure. S402. Obtaining the interference coupling coefficient calculated by cross-correlation analysis, and selecting the median as a preset standard threshold based on the distribution range of the interference coupling coefficient between the environmental interference data and the patient's physiological data under normal conditions. If the interference coupling coefficient between the environmental interference data and the patient's physiological data is greater than the preset standard threshold, it is determined to be an external error caused by environmental interference; S403. Obtain individual-specific interference characteristics identified by longitudinal trend analysis. If there are no abnormal changes in the instrument core parameters and there are no relevant maintenance records at that time point, and the interference coupling coefficient between the environmental interference data and the patient's physiological data does not exceed the preset standard threshold, and the patient's physiological data has individual-specific interference characteristics, then it is determined to be an external error caused by individual differences among patients.

7. The hemodialysis data processing method according to claim 6, characterized in that: The specific process of outputting calibrated data is as follows: S501. When it is determined that the error source is an internal error caused by a dialysis instrument hardware failure, the dialysis instrument self-test program is triggered to repair the dialysis instrument hardware and replace the faulty parts. When it is determined that the error source is an external error caused by environmental interference, the environmental interference data is processed using an adaptive filtering method. S502. When the error source is determined to be external error caused by individual patient differences, a patient error model is established for individual-specific interference using the patient's age, weight, environmental interference data, and instrument core parameters as independent variables and the patient's physiological data as dependent variables through regression analysis. The formula is: ； Among them, Z is the patient's physiological data; is the patient's age; is the patient's weight; Y is the environmental interference data; X is the instrument core parameter; All are regression coefficients; is the error term; The patient error model is trained and optimized through historical patient physiological data, historical environmental interference data, and historical instrument core parameters, and the value of the regression coefficient is determined to obtain an updated patient error model; S503. According to the processing results of different error types, the instrument core parameters, environmental interference data and patient physiological data are corrected, and the corrected data are recorded in the data recording file. At the same time, the corresponding positions in the time series matrix are marked, and the calibrated data are output in the specified format.

8. A hemodialysis data processing system, characterized in that: It includes a data acquisition unit, a data processing unit, an error analysis unit and an error calibration unit; The data acquisition unit is used to collect the core parameters of the dialysis instrument in real time through sensors installed inside the dialysis instrument, collect the patient's physiological data in real time through sensors installed in the circulation pipeline outside the patient's body, and collect environmental interference data in real time through sensors installed in the dialysis room and perform standardized processing; The data processing unit is used to obtain the acquisition frequency of the instrument core parameters, the patient physiological data, and the environmental interference data, and perform cubic spline interpolation on the low acquisition frequency parameters based on the highest acquisition frequency parameter as the reference time axis, generate virtual sampling values ​​at the time point of the highest acquisition frequency parameter, and associate the virtual sampling values ​​of the instrument core parameters, the virtual sampling values ​​of the patient physiological data, and the virtual sampling values ​​of the environmental interference data at the same time point to form a time series matrix; The error analysis unit is used to define the error type as internal error caused by dialysis instrument hardware failure and external error caused by environmental interference or individual patient differences. Based on the time series matrix, autocorrelation analysis is performed on the virtual sampling values ​​of the instrument core parameters to detect periodic deviations. In combination with the historical maintenance records of the dialysis instrument, high-frequency fault points are marked. Cross-correlation analysis is performed on the virtual sampling values ​​of the environmental interference data and the patient's physiological data to calculate the interference coupling coefficient. The data of multiple treatments for the same patient are longitudinally compared to identify individual-specific interference. The error source is determined based on the results of the autocorrelation analysis, cross-correlation analysis, and longitudinal comparison. The error calibration unit is used to perform hierarchical corrections on instrument core parameters, patient physiological data, and environmental interference data based on the source of the error, trigger the dialysis instrument self-test program for internal errors, use adaptive filtering for external errors, establish a patient-specific error model for individual-specific interference, and output calibrated data.

Images