CN116682566B - Hemodialysis data processing method and system - Google Patents

Hemodialysis data processing method and system Download PDF

Info

Publication number
CN116682566B
CN116682566B CN202310966440.1A CN202310966440A CN116682566B CN 116682566 B CN116682566 B CN 116682566B CN 202310966440 A CN202310966440 A CN 202310966440A CN 116682566 B CN116682566 B CN 116682566B
Authority
CN
China
Prior art keywords
index
data
fluctuation
value
hemodialysis
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202310966440.1A
Other languages
Chinese (zh)
Other versions
CN116682566A (en
Inventor
李淑秀
王彩虹
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Qingdao Hospital of Traditional Chinese Medicine
Original Assignee
Qingdao Hospital of Traditional Chinese Medicine
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Qingdao Hospital of Traditional Chinese Medicine filed Critical Qingdao Hospital of Traditional Chinese Medicine
Priority to CN202310966440.1A priority Critical patent/CN116682566B/en
Publication of CN116682566A publication Critical patent/CN116682566A/en
Application granted granted Critical
Publication of CN116682566B publication Critical patent/CN116682566B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Medical Informatics (AREA)
  • Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Computational Mathematics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Evolutionary Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Primary Health Care (AREA)
  • General Health & Medical Sciences (AREA)
  • Operations Research (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Epidemiology (AREA)
  • Algebra (AREA)
  • Pathology (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • External Artificial Organs (AREA)

Abstract

The application relates to the field of medical health care informatics, in particular to a data processing method and system for hemodialysis, which are used for acquiring relevant data of hemodialysis of a patient in a previous period and preprocessing the data to obtain a data sequence corresponding to each index to be detected; acquiring the overall association degree between every two indexes according to the data sequence corresponding to each index; acquiring an upper limit and a lower limit of a predicted value corresponding to each index at a subsequent moment according to the overall association degree among indexes and the real-time data of hemodialysis; and determining data prediction of hemodialysis of the patient at the future moment according to the upper limit and the lower limit of the predicted value. The application realizes the data processing in the hemodialysis process and improves the accuracy of the future hemodialysis data prediction.

Description

Hemodialysis data processing method and system
Technical Field
The application relates to the field of electric digital data processing, in particular to a data processing method and system for hemodialysis.
Background
In the hemodialysis process of a patient, strict monitoring of the patient is an important guarantee for smooth progress of the dialysis process, and the dialysis effect and the safety and comfort of the dialysis process can be improved. The monitoring content in the hemodialysis process mainly comprises four aspects of patient condition monitoring, vascular access monitoring, extracorporeal circulation monitoring and dialysate path monitoring, generally, an alarm threshold is set for each detection index, and when a certain index exceeds the value of the alarm threshold, an alarm is sent out to stop the hemodialysis process until the medical staff checks the working state of the continuous machine.
However, the warning threshold is set according to experience values and the conditions of most patients, so the threshold is generally a warning value of dangerous conditions, and when the patients only deviate from normal conditions, the warning threshold cannot be identified, so that monitoring the dialysis condition of the patients by only the threshold set in advance is not enough, and a method for automatically processing data according to the monitored data is needed, so that the condition of the patients in the hemodialysis process is detected and predicted in real time. The existing prediction algorithm can predict the data at the future moment according to the past period data, however, in the process of hemodialysis on a patient, complex relations often exist among different indexes, when one index fluctuates, the other indexes lag in the index value acquired later and fluctuate accordingly. Therefore, the current prediction algorithm cannot accurately predict the index value to be monitored in the hemodialysis process, and a data processing method capable of considering direct influence and indirect influence among different indexes is needed, so that the accurate prediction of the index value to be monitored in the hemodialysis process is realized.
Disclosure of Invention
In order to solve the technical problems, the application provides a data processing method and system for hemodialysis, which are used for solving the existing problems.
The application relates to a data processing method and a system for hemodialysis, which adopt the following technical scheme:
one embodiment of the present application provides a data processing method for hemodialysis, the method comprising the steps of:
acquiring relevant data of hemodialysis on a patient in a previous period, and preprocessing the data to obtain a data sequence corresponding to each index to be detected;
acquiring the overall association degree between every two indexes according to the data sequence corresponding to each index;
acquiring an upper limit and a lower limit of a predicted value corresponding to each index at a subsequent moment according to the overall association degree among indexes and the real-time data of hemodialysis;
and determining data prediction of hemodialysis of the patient at the future moment according to the upper limit and the lower limit of the predicted value.
Further, the step of obtaining relevant data of hemodialysis on a patient in a previous period and preprocessing the data to obtain a data sequence corresponding to each index to be detected comprises the following steps:
the relevant data of hemodialysis on the patient in the past include, but are not limited to, monitoring vital signs and emergency complications of the patient and extracorporeal circulation monitoring, and each data acquired at the same interval corresponds to a plurality of indexes;
the data values corresponding to the indexes are formed into a group of data sequences according to the acquired time, and a group of data sequences corresponding to each index can be acquired after hemodialysis is carried out on a patient every time.
Further, the overall association degree between every two indexes is obtained according to the data sequence corresponding to each index, specifically:
the total association degree between every two indexes is recorded as the adjustment association degree between the two indexes, the adjustment association degree expression is specifically:
，
in the method, in the process of the application,is index->With index->The degree of association is adjusted between the two; />Is index->With index->A degree of synchronization association between the two; />Is index->With index->Degree of hysteresis correlation between。
Further, the method for acquiring the synchronous association degree comprises the following steps:
，
in the method, in the process of the application,is index->With index->A degree of synchronization association between the two; />According to the future period->Index of group data acquisition->With index->Pearson correlation coefficient therebetween; />The number of groups for the acquired past date data; />Is index->With index->Index to be monitored->And the two indexes are not the same index.
Further, the method for obtaining the hysteresis association degree comprises the following steps:
the hysteresis association degree is obtained by a fluctuation data sequence corresponding to each index and a fluctuation significance degree corresponding to each fluctuation data sequence, and the expression is specifically as follows:
，
in the method, in the process of the application,is index->With index->Hysteresis association between; />Is index->Corresponding fluctuation data sequence->Corresponding fluctuation saliency; />Is index->Corresponding fluctuation data sequence->Corresponding fluctuation saliency;for fluctuating data sequences->And fluctuation data sequence->A corresponding dtw distance; />Is index->With index->The number of corresponding fluctuation data sequences to be compared acts as the average value of the molecules of the division formula; />As a normalization function, it acts as a normalization value in brackets.
Further, the method for obtaining the fluctuation saliency comprises the following steps:
fitting each numerical value contained in the data sequence corresponding to each index by using a least square method to obtain a fitting curve corresponding to each data sequence, and further obtaining a fitting value corresponding to each numerical value in each data sequence;
taking the sum of the fitting value corresponding to each numerical value and the absolute value of the difference value of the numerical values as the fluctuation difference value of the numerical values;
acquiring the fluctuation degree corresponding to each numerical value in each data sequence according to the ratio of the fluctuation difference value corresponding to each numerical value in the data sequence to the average value of the fluctuation difference value of each numerical value in the data sequence;
dividing the fluctuation degree corresponding to each numerical value in each data sequence by using a maximum inter-class variance method to obtain an adaptive division threshold, selecting the fluctuation degree larger than the adaptive division threshold, and if the continuous fluctuation degree is larger than a preset number, marking the sequence consisting of the numerical values corresponding to the continuous fluctuation degree as a fluctuation data sequence;
the expression of the fluctuation significance corresponding to the fluctuation data sequence is specifically as follows:
，
in the method, in the process of the application,for fluctuating data sequences->Corresponding fluctuation saliency; />For the fluctuation data sequence->Summing the fluctuation degrees corresponding to all the numerical values in the range; />For fluctuating data sequences->Standard deviation of the waviness corresponding to all values in the range.
Further, the upper limit and the lower limit of the predicted value corresponding to each index at the subsequent moment are obtained according to the overall association degree between indexes and the real-time data of hemodialysis, specifically:
obtaining the same index data of a patient needing to monitor the hemodialysis state during dialysis according to the same data obtaining method from the beginning of dialysis;
when the acquired data is greater than or equal to the preset quantity, a differential autoregressive moving average model ARIMA is used, and a predicted value of the subsequent time is acquired according to a value with a relatively previous acquisition time;
and obtaining an estimated value corresponding to each index according to the adjustment association degree and the predicted value among the indexes, wherein the estimated value comprises an estimated value upper limit and an estimated value lower limit.
Further, the upper and lower limits of the predicted value are expressed as follows:
，
，
in the middle ofFor the next->Time index->A corresponding upper limit of the estimated value; />For the next->Time index->A corresponding lower limit of the estimated value; />Is index->The mean value of fluctuation difference values corresponding to the currently acquired data values; />For the next->Time index->A corresponding predicted value; />The number of groups for the acquired past date data;is index->With index->And (5) adjusting the association degree between the two.
In a second aspect, the present application provides a hemodialysis data processing system comprising:
the data sequence acquisition module is used for acquiring relevant data of hemodialysis on a patient in a future period and preprocessing the data to obtain a data sequence corresponding to each index to be detected;
the overall association degree acquisition module is used for acquiring the overall association degree between every two indexes according to the data sequence corresponding to each index;
the data processing module is used for acquiring the upper limit and the lower limit of the predicted value corresponding to each index at the subsequent moment according to the total association degree among the indexes and the real-time data of hemodialysis;
and the data application module is used for determining data prediction of hemodialysis of the patient at the future moment according to the upper limit and the lower limit of the predicted value.
The application has at least the following beneficial effects:
according to the application, through acquiring the data of the patient in the past period for hemodialysis, the direct correlation of each index is acquired, and the more accurate correlation degree between different indexes is acquired by combining the hysteresis correlation of the change of the other indexes when each index changes, so that the problem that the complex correlation exists between different index values and the correlation degree between different index values is not easy to evaluate is solved; and then, acquiring the upper limit and the lower limit of a predicted value corresponding to each index at the subsequent moment according to the correlation degree between the hemodialysis real-time data of the patient to be monitored and different indexes, taking direct influence and indirect influence among different indexes into consideration, and comparing the predicted value with an alarm threshold value, so that the accurate prediction of the index value required to be monitored in the hemodialysis process is realized, and the accuracy of the hemodialysis data prediction at the future moment is improved.
Drawings
In order to more clearly illustrate the embodiments of the application or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the application, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a data processing method for hemodialysis provided by the application;
fig. 2 is a flowchart of a hemodialysis data processing system provided by the present application.
Detailed Description
In order to further describe the technical means and effects adopted by the application to achieve the preset aim, the following detailed description refers to the specific implementation, structure, characteristics and effects of a data processing method and system for hemodialysis according to the application with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
The following specifically describes a specific scheme of a data processing method and system for hemodialysis provided by the present application with reference to the accompanying drawings.
The application provides a hemodialysis data processing method and system.
Specifically, the data processing method and system for hemodialysis of the present embodiment provide a data processing method for hemodialysis, referring to fig. 1, the method includes the following steps:
and S001, acquiring relevant data of hemodialysis on a patient in the previous period, and preprocessing the data to obtain a data sequence corresponding to each index to be detected.
First, the present embodiment will extract data for monitoring during hemodialysis of a patient during the lead time, including, but not limited to, monitoring of patient vital signs and emergency complications, and extracorporeal circulation monitoring. The monitoring of vital signs and emergency complications of patients comprises indexes such as pulse, blood pressure and respiration, and the extracorporeal circulation monitoring comprises monitoring of a blood pump, monitoring of arterial and venous pressure, air detection, monitoring of a heparin pump and the like.
Hemodialysis is typically carried out for a period of 3-5 hours, and for ease of analysis, data is collected from monitoring the patient during the course of the hospital's forward phase of hemodialysis, at intervals from the beginning of the dialysisAcquiring the numerical value of each index to be monitored once in minutes until hemodialysis is completed, forming a group of data sequences by the data values corresponding to each index according to the acquired time, and adding the index +.>The corresponding data sequence is recorded as data sequence +.>. Wherein (1)>Namely, each index to be monitored is the index; />The empirical value was 1 for a constant. After hemodialysis is carried out on a patient in the future, a group of data sequences corresponding to each index can be obtained.
When a small amount of data in the acquired data sequence is missing due to various reasons, the acquired data sequence is preprocessed by adopting a mean filling method. The mean filling method is a well-known technique, and is not described in detail in this embodiment.
So far, the data sequence corresponding to each index to be detected can be obtained and used for processing the data of each index in the subsequent hemodialysis process.
Step S002, the overall association degree between every two indexes is obtained according to the data sequence corresponding to each index.
In the hemodialysis process, a plurality of indexes need to be monitored, and complex correlations exist between different indexes, and the correlations are changed synchronously and have hysteresis feedback, so that the prediction of the index value at the future moment is not accurate only according to the index value acquired by a certain index at a relatively earlier time. Therefore, the embodiment analyzes the synchronous association degree and the hysteresis association degree between different index values respectively, and according to the correlation relations, more accurate prediction of each index value is realized.
According to the data sequence corresponding to each index obtained by the extracted hemodialysis forward data, the pearson correlation coefficient between the data sequences corresponding to every two indexes of the same group is calculated, and every two indexes in each group of data can obtain a corresponding pearson correlation coefficient. When the synchronization of the two indexes changes is more closely related, the pearson correlation coefficient obtained according to each group of data is larger.
According to the pearson correlation coefficient corresponding to each two indexes, the synchronous correlation degree between each two indexes is obtained, the tightness degree of the more accurate synchronous correlation degree between each two indexes is obtained, and the synchronous correlation degree expression is specifically as follows:
，
in the method, in the process of the application,is index->With index->A degree of synchronization association between the two; />Is based onThe first time->Index of group data acquisition->With index->Pearson correlation coefficient therebetween; />The number of groups for the acquired past date data;is index->With index->Index to be monitored->And the two indexes are not the same index.
The synchronous association degree between the two indexes is the average value of the pearson correlation coefficients of the two indexes in each group of sequences, and when the synchronous change of the two indexes is more closely related, the correlation coefficient obtained according to each group of data is larger, namely the synchronous association degree between the two indexes is larger.
Therefore, the synchronous association degree between every two indexes can be obtained according to the multiple groups of data corresponding to different indexes.
Further, a degree of hysteresis correlation between each two index values is obtained.
And analyzing each data sequence corresponding to each index. Fitting each data sequence by using a least square method and a model which is most consistent with the law according to the law of the change of the numerical value in the data sequence corresponding to each index in the hemodialysis process, obtaining an expression of a fitting curve, and obtaining a fitting value of each numerical value according to the expression. And taking the sum of absolute values of differences between the fitting value and the actual value of each numerical value in the data sequence as a fluctuation difference value of the numerical value. And obtaining the average value of the fluctuation difference value of each numerical value in the same data sequence according to the fluctuation difference value.
Obtaining the fluctuation degree corresponding to each numerical value according to the fluctuation difference value of each numerical value in the data sequence and the average value of the fluctuation difference value of each numerical value in the same data sequence, wherein the fluctuation degree expression is specifically as follows:
，
in the method, in the process of the application,values in the data sequence +.>Corresponding waviness; />Values in the data sequence +.>Corresponding fluctuation difference values; />Is a numerical value +.>The mean value of the fluctuation difference value of each numerical value in the data sequence.
When the difference between the value in the data sequence and the fitting value thereof is larger, the fluctuation degree corresponding to the value is larger as the fluctuation difference relative to other values in the data sequence is larger, namely the fluctuation degree of the value in the data sequence is larger.
Each value in the data sequence has a corresponding degree of fluctuation. Obtaining a divided threshold value for the fluctuation degree corresponding to each numerical value in the data sequence by using a maximum inter-class variance method, dividing the numerical value corresponding to each fluctuation degree larger than the threshold value, and recording the divided numerical valuesA value of a relatively large degree of fluctuation in the data sequence. When there is a succession of greater than or equal toWhen the values are selected, the successive selected values are used as a fluctuation data sequence. When no fluctuation data sequence exists in all data sequences corresponding to a certain index, calculating each +_ in each data sequence>The sum of fluctuation degrees corresponding to successive values is taken as the sum of all the values of +.>And taking a continuous numerical value as a fluctuation data sequence corresponding to the index. Wherein (1)>The empirical value of (c) is set to 10 in this embodiment, which can be set by the practitioner. In this embodiment, the maximum inter-class variance method is used to obtain the threshold value of the fluctuation degree division corresponding to each numerical value in the data sequence, and the implementer may select other methods for obtaining the adaptive threshold value of the division according to the specific implementation scenario.
Acquiring the fluctuation significance of each fluctuation data sequence according to the selected fluctuation data sequence, wherein the fluctuation significance expression is specifically as follows:
，
in the method, in the process of the application,for fluctuating data sequences->Corresponding fluctuation saliency; />For the fluctuation data sequence->Summing the fluctuation degrees corresponding to all the numerical values in the range; />For fluctuating data sequences->Standard deviation of the waviness corresponding to all values in the range.
When the fluctuation degree of each numerical value in the fluctuation data sequence is larger and the fluctuation degree difference is smaller, the corresponding fluctuation significance degree is larger, namely a part with larger fluctuation degree in the fluctuation data sequence appears in the data sequence.
The screened fluctuation data sequence is the position of the corresponding index, at which the data value fluctuates in the hemodialysis process. The association degree of hysteresis influence among the indexes can be obtained according to the screened fluctuation data sequences corresponding to different indexes.
So far, the fluctuation data sequences can be adaptively divided according to the fluctuation degree of the numerical value contained in each data sequence, and meanwhile, the fluctuation significance corresponding to each fluctuation data sequence is obtained.
According to the fluctuation data sequence corresponding to each index, obtaining the hysteresis association degree between the two indexes, wherein the hysteresis association degree expression is specifically as follows:
，
in the method, in the process of the application,is index->With index->Hysteresis association between; />Is index->Corresponding fluctuation data sequence->Corresponding fluctuation saliency; />Is index->Corresponding fluctuation data sequence->Corresponding fluctuation saliency;for fluctuating data sequences->And fluctuation data sequence->A corresponding dtw distance; />Is index->With index->The number of corresponding fluctuation data sequences to be compared acts as the average value of the molecules of the division formula;as a normalization function, it acts as a normalization value in brackets.
In this embodiment, normalization processing is performed by using a normalization function, and an implementer may select other normalization processing methods according to a specific implementation scenario.
When one index fluctuates, after the time of the fluctuation, if the other index fluctuates, and the fluctuation is more similar, the fluctuation significance between the two indexes is larger.
Thus, the hysteresis association degree between every two indexes can be obtained according to a plurality of fluctuation data sequences corresponding to every two indexes.
According to the synchronous association degree and the hysteresis association degree between every two indexes, the adjustment association degree between the two indexes is obtained, and the adjustment association degree expression is specifically:
，
in the method, in the process of the application,is index->With index->And (5) adjusting the association degree between the two.
When the synchronization association degree and the hysteresis association degree corresponding to the two indexes are larger, the adjustment association degree between the two indexes is larger, namely the association degree between the two indexes is larger.
So far, the adjustment association degree between the two indexes can be obtained according to the synchronization association degree and the hysteresis association degree between every two indexes.
Step S003, obtaining the upper limit and the lower limit of the predicted value corresponding to each index at the subsequent moment according to the total association degree between indexes and the real-time data of hemodialysis.
From the beginning of dialysis, the same respective index data of the patient for which the hemodialysis condition needs to be monitored at the time of dialysis are acquired according to the same data acquisition method. When the acquired data is greater than or equal toWhen it is, the differential autoregressive moving average model ARIMA root is usedAccording to the front->Acquisition of the number value is followed by->Predicted values for each time. Wherein (1)>The experimental values of (a) are respectively set to 15 and 30, and the specific values can be set by an practitioner according to the illness state and tolerance of the patient in the hemodialysis process.
And obtaining an estimated value corresponding to each index according to the adjustment association degree and the predicted value among the indexes.
，
，
In the middle ofFor the next->Time index->A corresponding upper limit of the estimated value; />For the next->Time index->A corresponding lower limit of the estimated value; />Is index->The mean value of fluctuation difference values corresponding to the currently acquired data values; />For the next->Time index->Corresponding predicted values.
When the fluctuation difference value of the numerical value at each moment corresponding to each index value obtained at present is larger, the difference between the upper limit and the lower limit of the estimated value corresponding to each index and the predicted value is larger.
So far, each index can be obtained in the following step according to the adjustment association degree among different indexesTime index->A corresponding upper and lower estimated value limit.
Step S004, data prediction of hemodialysis of the patient at the future moment is determined according to the upper limit and the lower limit of the predicted value.
Acquiring each index value at the subsequent stageAnd respectively comparing the predicted values with an upper limit warning threshold value and a lower limit warning threshold value corresponding to the index value according to the predicted values of the time, and considering that the condition of influencing the patient dialysis can occur in the subsequent hemodialysis process when the upper limit and the lower limit of the estimated value exceed the range of the threshold value, so that the set parameters in the patient dialysis process need to be adjusted in time, and the safety and the comfort degree of the patient in the dialysis process are ensured.
Further, the present embodiment also proposes a hemodialysis data processing system, which includes the following contents:
the data sequence acquisition module is used for acquiring relevant data of hemodialysis on a patient in a future period and preprocessing the data to obtain a data sequence corresponding to each index to be detected;
the overall association degree acquisition module is used for acquiring the overall association degree between every two indexes according to the data sequence corresponding to each index;
the data processing module is used for acquiring the upper limit and the lower limit of the predicted value corresponding to each index at the subsequent moment according to the total association degree among the indexes and the real-time data of hemodialysis;
and the data application module is used for determining data prediction of hemodialysis of the patient at the future moment according to the upper limit and the lower limit of the predicted value. The implementation of the above modules has been described in an embodiment of a data processing method for hemodialysis, and a detailed description of a data processing system for hemodialysis is omitted here.
In summary, according to the embodiment of the application, the obtained data of the patient in the future for hemodialysis is used for obtaining the direct correlation of each index, and the more accurate correlation degree between different indexes is obtained by combining the hysteresis correlation of the change of the other indexes when each index changes, so that the problem that the complex correlation exists between different index values and the correlation degree between different index values is not easy to evaluate is solved; then, according to the embodiment of the application, the upper limit and the lower limit of the predicted value corresponding to each index at the subsequent moment are obtained according to the real-time hemodialysis data of the patient to be monitored and the association degree between different indexes, and then the predicted value is compared with the warning threshold value, so that the condition of hemodialysis of the patient at the future moment is determined, the safety and the comfort of the patient in the dialysis process are ensured, and the accuracy of predicting the hemodialysis data at the future moment is improved.
It should be noted that: the sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and the same or similar parts of each embodiment are referred to each other, and each embodiment mainly describes differences from other embodiments.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; the technical solutions described in the foregoing embodiments are modified or some of the technical features are replaced equivalently, so that the essence of the corresponding technical solutions does not deviate from the scope of the technical solutions of the embodiments of the present application, and all the technical solutions are included in the protection scope of the present application.

Claims (3)

1. A method of data processing for hemodialysis, the method comprising the steps of:
acquiring relevant data of hemodialysis on a patient in a previous period, and preprocessing the data to obtain a data sequence corresponding to each index to be detected;
acquiring the overall association degree between every two indexes according to the data sequence corresponding to each index;
acquiring an upper limit and a lower limit of a predicted value corresponding to each index at a subsequent moment according to the overall association degree among indexes and the real-time data of hemodialysis;
determining data prediction of hemodialysis of the patient at a future moment according to the upper limit and the lower limit of the predicted value;
the overall association degree between every two indexes is obtained according to the data sequence corresponding to each index, specifically:
the total association degree between every two indexes is recorded as the adjustment association degree between the two indexes, the adjustment association degree expression is specifically:
，
in the method, in the process of the application,is index->With index->The degree of association is adjusted between the two; />Is index->With index->A degree of synchronization association between the two;is index->With index->Hysteresis association between;
the method for acquiring the synchronous association degree comprises the following steps:
，
in the method, in the process of the application,is index->With index->A degree of synchronization association between the two; />According to the future period->Index of group data acquisition->With index->Pearson correlation coefficient therebetween; />The number of groups for the acquired past date data; />Is index->With index->Index to be monitored->The two indexes are not the same index;
the method for acquiring the hysteresis association degree comprises the following steps:
the hysteresis association degree is obtained by a fluctuation data sequence corresponding to each index and a fluctuation significance degree corresponding to each fluctuation data sequence, and the expression is specifically as follows:
，
in the method, in the process of the application,is index->With index->Hysteresis association between; />Is index->Corresponding fluctuation data sequence->Corresponding fluctuation saliency; />Is index->Corresponding fluctuation data sequence->Corresponding fluctuation saliency; />For fluctuating data sequences->And fluctuation data sequence->A corresponding dtw distance; />Is index->With index->The number of corresponding fluctuation data sequences to be compared acts as the average value of the molecules of the division formula; />Acting as normalization value in brackets as normalization function;
the method for acquiring the fluctuation significance comprises the following steps:
fitting each numerical value contained in the data sequence corresponding to each index by using a least square method to obtain a fitting curve corresponding to each data sequence, and further obtaining a fitting value corresponding to each numerical value in each data sequence;
taking the sum of the fitting value corresponding to each numerical value and the absolute value of the difference value of the numerical values as the fluctuation difference value of the numerical values;
acquiring the fluctuation degree corresponding to each numerical value in each data sequence according to the ratio of the fluctuation difference value corresponding to each numerical value in the data sequence to the average value of the fluctuation difference value of each numerical value in the data sequence;
dividing the fluctuation degree corresponding to each numerical value in each data sequence by using a maximum inter-class variance method to obtain an adaptive division threshold, selecting the fluctuation degree larger than the adaptive division threshold, and if the continuous fluctuation degree is larger than a preset number, marking the sequence consisting of the numerical values corresponding to the continuous fluctuation degree as a fluctuation data sequence;
the expression of the fluctuation significance corresponding to the fluctuation data sequence is specifically as follows:
，
in the method, in the process of the application,for fluctuating data sequences->Corresponding fluctuation saliency; />For the fluctuation data sequence->Summing the fluctuation degrees corresponding to all the numerical values in the range; />For fluctuating data sequences->Standard deviation of fluctuation degrees corresponding to all values in the range;
the upper limit and the lower limit of the predicted value corresponding to each index at the subsequent moment are obtained according to the total association degree among indexes and the real-time data of hemodialysis, and specifically are as follows:
obtaining the same index data of a patient needing to monitor the hemodialysis state during dialysis according to the same data obtaining method from the beginning of dialysis;
when the acquired data is greater than or equal to the preset quantity, a differential autoregressive moving average model ARIMA is used, and a predicted value of the subsequent time is acquired according to a value with a relatively previous acquisition time;
acquiring an estimated value corresponding to each index according to the adjustment association degree and the predicted value among the indexes, wherein the estimated value comprises an estimated value upper limit and an estimated value lower limit;
the upper limit and the lower limit of the predicted value are expressed as follows:
，
，
in the middle ofFor the next->Time index->A corresponding upper limit of the estimated value; />For the next->Time index->A corresponding lower limit of the estimated value; />Is index->The mean value of fluctuation difference values corresponding to the currently acquired data values; />For the next->Time index->A corresponding predicted value; />The number of groups for the acquired past date data; />Is index->With index->And (5) adjusting the association degree between the two.
2. The method for processing hemodialysis data according to claim 1, wherein the steps of obtaining relevant data of hemodialysis of a patient in a future period and preprocessing the data to obtain a data sequence corresponding to each index to be detected are as follows:
the relevant data of hemodialysis on the patient in the past include, but are not limited to, monitoring vital signs and emergency complications of the patient and extracorporeal circulation monitoring, and each data acquired at the same interval corresponds to a plurality of indexes;
the data values corresponding to the indexes are formed into a group of data sequences according to the acquired time, and a group of data sequences corresponding to each index can be acquired after hemodialysis is carried out on a patient every time.
3. A hemodialysis data processing system comprising a processor and a memory, said processor for processing instructions stored in said memory to implement a hemodialysis data processing method of any one of claims 1-2.
CN202310966440.1A 2023-08-03 2023-08-03 Hemodialysis data processing method and system Active CN116682566B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310966440.1A CN116682566B (en) 2023-08-03 2023-08-03 Hemodialysis data processing method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310966440.1A CN116682566B (en) 2023-08-03 2023-08-03 Hemodialysis data processing method and system

Publications (2)

Publication Number Publication Date
CN116682566A CN116682566A (en) 2023-09-01
CN116682566B true CN116682566B (en) 2023-10-31

Family

ID=87782244

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310966440.1A Active CN116682566B (en) 2023-08-03 2023-08-03 Hemodialysis data processing method and system

Country Status (1)

Country Link
CN (1) CN116682566B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116886106B (en) * 2023-09-07 2024-01-23 青岛市中医医院(青岛市海慈医院、青岛市康复医学研究所) Hemodialysis data intelligent coding management system
CN117912711B (en) * 2024-03-19 2024-05-24 吉林大学 Hemodialysis data acquisition and analysis system and method
CN118430732B (en) * 2024-07-02 2024-09-10 深圳爱递医药科技有限公司 Hemodialysis data intelligent processing system based on smart phone application program

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101304773A (en) * 2005-11-11 2008-11-12 日机装株式会社 Hemodialysis apparatus and method for hemodialysis
CN108803528A (en) * 2018-07-13 2018-11-13 杭州电子科技大学 Process industry system prediction model based on multivariate correlation and time lag
CN108986419A (en) * 2018-10-17 2018-12-11 暨南大学 A kind of data alarm method for haemodialysis
CN110075378A (en) * 2019-05-08 2019-08-02 黄莉娟 A kind of haemodialysis data information monitoring system
CN110152087A (en) * 2019-04-19 2019-08-23 暨南大学 A kind of monitoring method of blood dialysis
CN112107752A (en) * 2019-06-20 2020-12-22 纬创资通股份有限公司 Blood pressure prediction method and electronic device using same
WO2021052156A1 (en) * 2019-09-18 2021-03-25 平安科技（深圳）有限公司 Data analysis method, apparatus and device, and computer readable storage medium
CN114550913A (en) * 2022-02-22 2022-05-27 深圳市裕辰医疗科技有限公司 Auxiliary diagnosis method for hemodialysis hypotension complication
CN116036399A (en) * 2021-10-28 2023-05-02 纬创资通股份有限公司 Analysis method in dialysis and analysis device for dialysis

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12040092B2 (en) * 2019-12-20 2024-07-16 Fresenius Medical Care Holdings, Inc. Real-time intradialytic hypotension prediction
US20230082362A1 (en) * 2021-09-09 2023-03-16 Nanowear Inc. Processes and methods to predict blood pressure

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101304773A (en) * 2005-11-11 2008-11-12 日机装株式会社 Hemodialysis apparatus and method for hemodialysis
CN108803528A (en) * 2018-07-13 2018-11-13 杭州电子科技大学 Process industry system prediction model based on multivariate correlation and time lag
CN108986419A (en) * 2018-10-17 2018-12-11 暨南大学 A kind of data alarm method for haemodialysis
CN110152087A (en) * 2019-04-19 2019-08-23 暨南大学 A kind of monitoring method of blood dialysis
CN110075378A (en) * 2019-05-08 2019-08-02 黄莉娟 A kind of haemodialysis data information monitoring system
CN112107752A (en) * 2019-06-20 2020-12-22 纬创资通股份有限公司 Blood pressure prediction method and electronic device using same
WO2021052156A1 (en) * 2019-09-18 2021-03-25 平安科技（深圳）有限公司 Data analysis method, apparatus and device, and computer readable storage medium
CN116036399A (en) * 2021-10-28 2023-05-02 纬创资通股份有限公司 Analysis method in dialysis and analysis device for dialysis
CN114550913A (en) * 2022-02-22 2022-05-27 深圳市裕辰医疗科技有限公司 Auxiliary diagnosis method for hemodialysis hypotension complication

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于甲亢临床指标的多维时间序列关联度分析;王致强 等;计算机与现代化(第04期);第12-15页 *

Also Published As

Publication number Publication date
CN116682566A (en) 2023-09-01

Similar Documents

Publication Publication Date Title
CN116682566B (en) Hemodialysis data processing method and system
CN107436993B (en) Method and server for establishing ICU patient condition evaluation model
US10342910B2 (en) Device and method for predicting intradialytic parameters
JP5728767B2 (en) Monitoring system, method and computer program for management of insulin delivery
US9345827B2 (en) Obtaining control settings for a dialysis machine
Papini et al. Photoplethysmography beat detection and pulse morphology quality assessment for signal reliability estimation
CN104689402A (en) Dialysis machine and device and method for signal processing
WO2009075796A1 (en) System and method for predicting septic shock
CN111297329B (en) Method and system for predicting dynamic onset risk of cardiovascular complications of diabetics
US20240082472A1 (en) Parameter Monitoring in Medical Treatment Systems
CN117542536B (en) Intelligent nursing method and system based on physical examination data
US20220047789A1 (en) Adequacy assessment method and system
CN110752002A (en) Medicine dosage prediction device
CN115862793A (en) Apparatus and method for cardiovascular medical data analysis, computer storage medium
WO2023226369A1 (en) Method, apparatus and device for predicting weaning from mechanical ventilation, and storage medium
CN117045889A (en) End stage renal disease hemodialysis blood pressure mode identification and intervention system and related device
CN118571513B (en) Method and system for monitoring patient sign in thyroid surgery process
WO2020253510A1 (en) Big data blood vessel image identification-based disease prediction system and method
CN117809833B (en) Remote intelligent peritoneal dialysis system
CN118571513A (en) Method and system for monitoring patient sign in thyroid surgery process
CN116439698B (en) Blood glucose monitoring and early warning method, system and equipment for intensive care unit
CN118197522B (en) Real-time vital sign monitoring and early warning method and system for emergency critical patients
WO2023017793A1 (en) Method and apparatus for adaptively detecting an acute hypotension episode event of a patient
CN118866365A (en) Remote diabetes monitoring management method and system based on artificial intelligence judgment
Zou et al. Intelligent analysis system in time series of smart health home on-line monitoring data

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant