KR20220149093A - Apparatus for predicting intradialytic hypotension using AI and method thereof - Google Patents

Abstract

The present invention relates to a device for predicting hypotension during hemodialysis using artificial intelligence and a method thereof. According to the present invention, the device for predicting hypotension during hemodialysis is provided which comprises: a data collection unit collecting each of fixed variable data including body information and comorbidity information related to each of a plurality of patients and variable data including blood pressure and dialysis conditions observed during hemodialysis by time; a model learning unit using the fixed variable data and the variable data by time and using systolic blood pressure (SBP), systolic blood pressure variance per unit time (ㅿSBP), and mean arterial blood pressure variance (ㅿMAP) of the patient after a set time elapses to train a prediction model; a data acquisition unit acquiring the fixed variable data and the variable data observed by time during hemodialysis in real time for a patient undergoing hemodialysis; and a prediction unit predicting SBP, ㅿSBP, and ㅿMAP values after the set time elapses from the present for each hour during hemodialysis by applying the fixed variable data and the variable data observed by time in real time to the prediction model. According to the present invention, based on artificial intelligence, possibility of hypotension in the patient which may occur during hemodialysis can be predicted and informed in advance. Accordingly, hypotension during hemodialysis can be prevented and prognosis of hemodialysis patients can be improved.

Description

Apparatus for predicting intradialytic hypotension using AI and method thereof

The present invention relates to an apparatus for predicting hypotension during hemodialysis using artificial intelligence and a method therefor, and more particularly, to an apparatus for predicting hypotension during hemodialysis capable of predicting in advance the possibility of occurrence of hypotension in a patient during hemodialysis using artificial intelligence, and the same it's about how

The mortality rate of hemodialysis patients is very high, approaching 40% in 5 years, and half of the causes of death are related to cardiovascular disease. Hypotension should not occur during hemodialysis in order to lower the mortality rate from cardiovascular disease, but there is no model for predicting it or a technique for preventing it yet.

Because there is no model, software, or device that predicts hypotension during hemodialysis, medical staff and nursing staff directly monitor the patient's blood pressure during hemodialysis, measuring the patient's blood pressure at intervals of approximately 30 minutes to 1 hour.

Since the number of patients is rapidly increasing, their workload is increasing, and when unpredictable hypotension suddenly occurs during hemodialysis, it is difficult to cope with it, and the patient's prognosis is deteriorated.

In particular, hypotension during hemodialysis could be dealt with only after hypotension had already occurred. As a coping method, a method of changing the patient's dialysis conditions and patient's position was selected. However, since it is to be dealt with after the blood pressure has already dropped, there is a problem that side effects related to hypotension appear, and the risk of cardiovascular disease and death of the patient is increased accordingly.

To solve this problem, clinical application of predictive models is essential, but the reason that predictive models have not yet been developed is that it is difficult to predict hypotension during dialysis with general mathematical or statistical models.

Therefore, there is a need for a new model that can effectively predict a patient's hypotension during hemodialysis.

The technology that is the background of the present invention is disclosed in Korean Patent Application Laid-Open No. 1995-0702440 (published on July 29, 1995).

An object of the present invention is to provide an apparatus and method for predicting hypotension during hemodialysis using artificial intelligence capable of predicting in advance the possibility of hypotension in a patient during hemodialysis based on artificial intelligence.

The present invention provides a data collection unit that collects fixed variable data including body information and accompanying disease information related to the patient for each patient, and variable variable data including blood pressure and dialysis conditions observed by time during hemodialysis, respectively; Using variable data and time-dependent variable data as input data, the patient's systolic blood pressure (SBP), systolic blood pressure change per unit time (SBP), and mean arterial blood pressure change (ㅿMAP) a model learning unit for learning a predictive model using the output data; a data acquisition unit for acquiring the fixed variable data and the variable variable data observed by time during current hemodialysis for a patient on hemodialysis in real time; and the fixed variable data and Hypotension prediction device during hemodialysis including a prediction unit that applies the variable variable data observed for each hour to the prediction model in real time, and predicts the SBP, SBP, and MAP values after a set time has elapsed from the present for every hour during hemodialysis provides

In addition, the apparatus for predicting hypotension during hemodialysis may further include a determination unit that compares SBP, ㅿSBP, and MAP values predicted for each hour with each preset threshold value to determine in real time whether hypotension is possible during hemodialysis. can

In addition, if the predicted SBP value is less than 90, the SBP value is 20 or more, or the MAP value is 10 or more, it may be determined that there is a possibility of hypotension of the patient after a set time has elapsed from the present.

In addition, the predictive model may be learned through a recurrent neural network (RNN) algorithm.

In addition, the set time may be less than half of the total dialysis time required for the hemodialysis operation.

In addition, the fixed variable data may further include at least one of taking drug information, blood test information, and a patient's hypotension risk.

In addition, the dialysis condition among the variable variable data may include at least one of blood flow, ultrafiltration rate, and dialysis fluid electrolyte concentration, and the variable variable data may further include at least one of a pulse rate and body temperature observed over time during hemodialysis.

In addition, the present invention provides a method for predicting hypotension using a hypotension prediction device during hemodialysis, fixed variable data including body information and accompanying disease information related to the patient for each of a plurality of patients, blood pressure observed by time during hemodialysis and dialysis conditions Collecting variable variable data including: the fixed variable data and the variable variable data for each time as input data, the patient's systolic blood pressure (SBP) after a set time has elapsed at the corresponding time for each time, the amount of change in the systolic blood pressure per unit time Learning a predictive model using (ㅿSBP) and mean arterial blood pressure change (ㅿMAP) as output data; obtaining, and applying the fixed variable data and the variable variable data observed by time to the predictive model in real time to predict the SBP, SBP, and MAP values after a set time has elapsed from the present for every hour during hemodialysis It provides a method for predicting hypotension during hemodialysis, comprising the steps of:

In addition, the method for predicting hypotension during hemodialysis may further include determining in real time whether hypotension is possible during hemodialysis by comparing the SBP, SBP, and MAP values predicted for each hour with each preset threshold value. have.

In addition, in the determining step, if the predicted SBP value is less than 90, the SBP value is 20 or more, or the MAP value is 10 or more, it can be determined that there is a possibility of hypotension of the patient after a set time has elapsed from the present.

According to the present invention, it is possible to predict in advance the possibility of hypotension in a patient, which may occur during hemodialysis, based on artificial intelligence. Through this, it is possible to treat and evaluate a patient before hypotension occurs, thereby preventing hypotension during hemodialysis and improving the prognosis of a hemodialysis patient.

In particular, in the case of the present invention, real-time analysis of the patient's biosignals and dialysis conditions, which vary in real time during hemodialysis, through an artificial intelligence model, predicts the possibility of hypotension in real time from the start of hemodialysis to the end of the patient's hemodialysis within 1 hour. can provide

1 is a diagram showing the configuration of an apparatus for predicting hypotension during hemodialysis according to an embodiment of the present invention.
FIG. 2 is a diagram briefly explaining the function and data flow of each part of FIG. 1 .
3 is a diagram illustrating a recurrent neural network used in an embodiment of the present invention.
4 is a diagram illustrating blood pressure parameters observed during hemodialysis in an embodiment of the present invention.
FIG. 5 is a view for explaining a method for predicting hypotension during hemodialysis using the apparatus of FIG. 1 .
6 is a diagram comparing performance results for each algorithm used.

Then, with reference to the accompanying drawings, embodiments of the present invention will be described in detail so that those of ordinary skill in the art to which the present invention pertains can easily implement them. However, the present invention may be embodied in various different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part is "connected" with another part, this includes not only the case of being "directly connected" but also the case of being "electrically connected" with another element interposed therebetween. . Also, when a part "includes" a certain component, it means that other components may be further included, rather than excluding other components, unless otherwise stated.

The present invention is a technique for predicting hypotension during hemodialysis using artificial intelligence. By analyzing biosignals and dialysis conditions that change in real time during hemodialysis through artificial intelligence, the possibility of hypotension occurring within 1 hour during hemodialysis can be predicted and provided in real time. and proposes a technique for preventing hypotension during hemodialysis and improving the patient's prognosis by enabling treatment in advance before the occurrence of hypotension.

1 is a diagram showing the configuration of an apparatus for predicting hypotension during hemodialysis according to an embodiment of the present invention, and FIG. 2 is a diagram briefly explaining the function and data flow of each part of FIG. 1 .

1 and 2 , an apparatus 100 for predicting hypotension during dialysis (hereinafter, referred to as a prediction apparatus) according to an embodiment of the present invention includes a data collection unit 110 , a model learning unit 120 , and a data acquisition unit 130 . ), a prediction unit 140 , a determination unit 150 , and an output unit 160 .

Here, the operation of each unit 110 to 160 and the data flow between each unit may be controlled by a controller (not shown).

The data collection unit 110 collects fixed variable data and variable variable data for each of a plurality of patients (hemodialysis patients) who have undergone hemodialysis in the past. Here, the fixed variable data may include patient-related body information (age, gender) and accompanying disease information, and may include information on medications taken, blood test information, and a patient's hypotension risk.

Variable variable data is a variable that varies with time, and includes the patient's blood pressure and the patient's dialysis conditions (blood flow rate, ultrafiltration rate, and dialysis fluid electrolyte concentration) periodically measured for each time during hemodialysis. It may further include at least one of a pulse and body temperature observed during dialysis.

The data collection unit 110 may collect basic patient data from an EMR or hospital server that manages and stores the patient's medical information.

The EMR, which is an electronic medical record system, may be included in a hospital server or may be operated in conjunction with the hospital server. The EMR or hospital server can manage the patient's medical information (medical information DB) history by time, including the patient's personal information, body information, medical records for the past several months or years, medical history, various examination information, hemodialysis measurement records, etc. . As such, basic data for each patient can be collected through EMR, hospital servers, etc., and big data necessary for predictive model learning can be built from this.

Here, for model learning, the data collection unit 110 may additionally collect SBP, SBP, and MAP values, which are blood pressure parameters for each time, obtained from blood pressure data records measured during hemodialysis of a patient in the past.

In this case, SBP is the systolic blood pressure, MBP is the mean arterial pressure, ㅿSBP is the change in systolic blood pressure per unit time, and MAP is the mean arterial pressure per unit time.

Therefore, among the blood pressure parameters over time, SBP is the patient's systolic blood pressure, sBP is the amount of change in systolic blood pressure per unit time, and MAP is the mean arterial blood pressure per unit time.

Here, ⅿSBP may be obtained in real time from the SBP value according to time, and may be obtained in real time from the MBP according to the time ⅿMAP. For example, ㅿSBP may correspond to a value obtained by subtracting the current SBP value (SBP) at time t for each hour from the SBP value (SBP') one hour before (ㅿSBP=SBP'-SBP). The value of ㅿ is derived as a positive number when the current value is lower than the value before the unit time.

In general, since hemodialysis takes a total of 4 hours, blood pressure parameter values may be periodically obtained for every hour (eg, several seconds, minutes) during the four hours of hemodialysis. The data collection unit 110 may additionally collect SBP, SBP, and MAP, which are blood pressure parameter values for each time measured in real time during hemodialysis, and provide it to the model learning unit 120 .

Past patient-specific data collected by the data collection unit 110 is used as learning data for building a model for predicting intradialytic hypotension during hemodialysis of a patient.

Among the collected data, the patient's fixed variable data (patient's age, sex, medications, comorbidities, etc.) and the variable variable data observed over time during dialysis (blood pressure, pulse, body temperature, blood flow, ultrafiltration rate, dialysis fluid electrolyte concentration, etc.) is used as input data when learning the predictive model, and blood pressure parameters (SBP, sBP, MAP) after a set time for each time (eg, 1 hour) are used as output data. The predictive model learns the relationship between these input data and output data based on deep learning.

According to this, by deep learning analysis of the fixed variable data corresponding to the patient during hemodialysis and the variable variable data measured by time, it is possible to derive the SBP, SBP, and MAP values expected one hour after the present in real time. Of course, it is also possible to derive the expected blood pressure parameter after 30 minutes instead of 1 hour, depending on which point in time the output data is applied during model training.

Specifically, the model learning unit 120 uses the patient's fixed variable data (time-fixed variable) and time-dependent variable data as input data, and the patient's systolic pressure (SBP) after a set time has elapsed at the corresponding time for each time. ), SBP change amount per unit time (ㅿSBP), and MAP change amount (ㅿMAP) are used as output data to train a predictive model.

The model learning unit 120 may generate a prediction model for predicting the possibility of occurrence of hypotension within 1 hour of a patient undergoing hemodialysis through artificial intelligence-based analysis.

Various techniques may be used to generate the predictive model, and a recurrent neural network (RNN), a multi-layer perceptron (MLP), LightGBM, a logistic regression (LR) algorithm, etc. may be used.

An embodiment of the present invention uses a recurrent neural network (RNN) as a representative example. The input data of the predictive model includes fixed variable data collected for patients who have undergone hemodialysis in the past and variable variable data observed during hemodialysis, and the output data is after a set time (eg, 1 hour) from each time. Observed patient blood pressure parameters, ie, SBP, SBP, and MAP. The set time is less than the total dialysis time required for hemodialysis, and through this, the possibility of hypotension within several hours or minutes is predicted even during hemodialysis.

The predictive model can learn the relationship between the input data and the output data through the RNN deep learning algorithm. That is, the predictive model is trained to derive the SBP, SBP, and MAP parameters after a set time for each expected hourly from the input of the patient's fixed variable data and the variable variable data that varies with time during dialysis, respectively.

The model learning unit 120 may modify and optimize the weights in the neural network by continuously analyzing the relationship between the input data and the output data by machine learning based on supervised learning. Of course, in the process, learning can be performed until the error between the measured value and the predicted value falls below the threshold.

When the predictive model is completed through learning, the possibility of hypotension during dialysis of the patient can be predicted simply by applying the fixed variable data for the currently dialysis patient and the variable variable data that varies according to time during dialysis to the model.

To this end, the data acquisition unit 130 acquires, in real time, fixed variable data and time-dependent variable variable data observed during hemodialysis for a hemodialysis patient.

The data acquisition unit 130 may acquire fixed variable data and variable variable data related to a hemodialysis patient by interworking with a hospital server, dialysis equipment, a medical information server, a computer, and the like. Here, in the case of fixed data, related data may be input from a manager or the like through an input interface, a wired/wireless interface, or the like. In addition, variable variable data observed in real time during hemodialysis may be acquired in real time from a user terminal such as a dialysis equipment or a computer connected thereto.

In addition, the prediction device 100 of the present invention may be implemented as an on/offline platform such as a web server or an app server that predicts a patient's hypotension during dialysis, or may be implemented in the form of an application, an application, etc. in a user terminal. have.

In the present invention, pre-obtained fixed variable data (age, sex, comorbidities, previous hypotension risk, blood tests, medications used, etc.) for patients undergoing hemodialysis and variable variable data (blood pressure, blood pressure, Using the pulse rate, body temperature, blood flow, ultrafiltration rate, and electrolyte concentration of the dialysate, etc.), the blood pressure parameter prediction results expected one hour after the present are continuously provided during hemodialysis, and the possibility of hypotension can be checked in real time.

The prediction unit 140 applies the fixed variable data and the variable variable data observed by time to the prediction model in real time, and predicts the SBP, SBP, and MAP values after a set time has elapsed from the present for every hour during hemodialysis. .

Here, as described above, the set time corresponds to less than half of the total dialysis time (eg, 4 hours) required for hemodialysis, and through this, the possibility of hypotension during dialysis can be predicted in advance. For example, the set time may be various, such as 10 minutes, 30 minutes, 1 hour, and the embodiment of the present invention exemplifies that it is 1 hour.

The prediction unit 140 includes fixed variable factors (age, sex, comorbidities, prior hypotension risk, blood tests, medications used, etc.) of the hemodialysis patient and variable variable factors (blood pressure, pulse, body temperature, blood flow, etc.) observed in real time during current dialysis. , ultrafiltration rate, dialysis fluid electrolyte concentration, etc.) are applied as input parameters to the RNN-based prediction model, and the blood parameter values (SBP, ㅿSBP, ㅿMAP) expected when 1 hour has elapsed from the present every hour are analyzed by artificial intelligence. It can be derived and provided in real time through

3 is a diagram illustrating a recurrent neural network used in an embodiment of the present invention. As shown in FIG. 3 , in the case of the present invention, output data is derived in real time in response to an input vector for each time input to the recurrent neural network.

At this time, the time-invariant feature and time-varying feature are normalized and transformed into a form that can be learned by the artificial intelligence model, and then passed through the RNN layer, and then the hidden layer Finally, hypotension during dialysis can be predicted. The predicted value may be repeatedly learned until it is possible to predict whether hypotension actually occurs.

Here, the model learning unit 120 and the prediction unit 140 may be included or embedded in a processor as shown in FIG. 2 . Such a processor may be equipped with a prediction model using artificial intelligence and provide related prediction results, and may be connected to external platforms and related servers to update and update AI learning data and prediction models.

In addition, the processor has built-in artificial intelligence algorithms and can perform operations based on machine learning. For example, a processor may use a machine learning algorithm based on a highly multi-layered artificial neural network in the pre-training of the algorithm. The processor may predict an expected result value in response to the patient's input data currently being analyzed based on the weights learned in advance through the artificial intelligence algorithm.

The determination unit 150 compares the SBP, SBP, and MAP values predicted for each hour with each preset threshold value to determine in real time whether hypotension is possible during hemodialysis.

Specifically, the determination unit 150, the currently predicted SBP value is less than 90 (SBP < 90), the ㅿSBP value is 20 or more (ㅿSBP ≥ 20), or the ㅿMAP value is 10 or more (ㅿMBP ≥ 10) ), it is determined that there is a possibility of hypotension in the patient during dialysis after a set time (eg: 1 hour) has elapsed from the present. Of course, such a prediction result can be considered as a situation where there is a possibility of hypotension within 1 hour from the present time.

According to this, hypotension during hemodialysis can be prevented in advance, one of the side effects of dialysis, by predicting the occurrence of hypotension one hour before the occurrence of hypotension and enabling pre-measures against it, enabling rapid response before emergency situations occur. can do.

4 is a diagram illustrating time-series blood pressure parameters observed during hemodialysis in an embodiment of the present invention.

The left figure of FIG. 4 shows a case in which hypotension did not occur during hemodialysis for 4 hours, and the right figure shows a case in which hypotension occurred at 3 hours. In the case of the right figure, the observed SBP value at 2 hours is 102 mmHg, and 1 hour after that, the observed SBP value at 3 hours is 81 mmHg, which is a case in which the SBP value exceeds 20.

According to an embodiment of the present invention, by applying an artificial intelligence-based prediction algorithm, it is possible to predict the possibility of hypotension one hour earlier (second hour) than when the actual blood pressure drops (the third hour in the right figure). . These algorithms provide clinicians with the opportunity to respond more quickly to hypotension and evaluate the patient's problems.

The output unit 160 may output and provide the real-time SBP, SBP, and MAP values derived through the predictive model and the resulting hypotension possibility classification result through a display or a network-connected user terminal. Here, the user terminal may include a desktop, a tablet PC, a smart phone, a smart pad, a notebook computer, and the like.

As described above, the prediction device 100 executes a process related to prediction of hypotension during hemodialysis and provides the corresponding result, thereby enabling efficient prognosis management as well as rapid treatment of hypotension that may occur during dialysis in a hemodialysis patient. .

3 is a view for explaining data related to occurrence of hypotension during hemodialysis in an embodiment of the present invention.

First, for each patient with a past hemodialysis history, the data collection unit 110 collects patient-related fixed variable data and time-observed variable variable data during hemodialysis as learning data, respectively ( S510 ).

In addition, the data collection unit 110 additionally collects SBP, SBP, and MAP, which are blood pressure parameters for each time, obtained from blood pressure data records measured during hemodialysis of a patient in the past. The data collection unit 110 builds big data for model learning based on this.

Next, the model learning unit 120 uses the collected fixed variable data and time-dependent variable variable data as input data, and the SBP, ㅿSBP, and MAP of the patient after the set time has elapsed for each time as output data to learn the predictive model. make it (S520).

In this case, the model learning unit 120 may train the model using a recurrent neural network (RNN), and may increase the reliability of the model through big data learning for related patients.

After the learning is completed, it is possible to predict and provide the possibility of hypotension within 1 hour from the current flow during hemodialysis in response to current factors for an actual hemodialysis patient.

Specifically, the data acquisition unit 130 acquires fixed variable data and variable variable data for a patient undergoing hemodialysis in real time ( S530 ), and transmits them to the prediction unit 140 .

In addition, the prediction unit 140 applies the fixed variable data received corresponding to the patient on hemodialysis and the variable variable data according to time to the prediction model in real time, and the SBP after a set time has elapsed from the present for every hour during hemodialysis, Predicting and deriving the ㅿSBP and ㅿMAP values (S540).

Thereafter, the determination unit 150 compares the predicted values with the respective threshold values to determine whether hypotension is possible during dialysis in real time ( S550 ). In addition, the output unit 160 may output the blood pressure parameter prediction result and the result of determining the possibility of hypotension during dialysis based thereon through a display or the like (S560).

The following describes the results of comparing the prediction performance according to the types of machine learning algorithms used to generate the prediction model.

6 is a diagram comparing performance results for each algorithm used. 6, IDH-1, IDH-2, and IDH-3 are performance results for the SBP, ㅿSBP, and ㅿMBP values, respectively, and in all cases, the case where the prediction model is constructed using the RNN algorithm shows the best performance. It was.

According to the present invention as described above, it is possible to predict and notify the possibility of hypotension in a patient that may occur during hemodialysis based on artificial intelligence. Through this, it is possible to treat and evaluate a patient before hypotension occurs, thereby preventing hypotension during hemodialysis and improving the prognosis of a hemodialysis patient.

In particular, in the case of the present invention, real-time analysis of the patient's biosignals and dialysis conditions, which vary in real time during hemodialysis, through the RNN artificial intelligence model, predicts the possibility of hypotension in real time from the start of hemodialysis of the patient to the end of hemodialysis within 1 hour This allows for rapid response to emergencies that may occur during hemodialysis.

Although the present invention has been described with reference to the embodiment shown in the drawings, which is only exemplary, those of ordinary skill in the art will understand that various modifications and equivalent other embodiments are possible therefrom. Accordingly, the true technical protection scope of the present invention should be defined by the technical spirit of the appended claims.

100: device for predicting hypotension during dialysis 110: data collection unit
120: model learning unit 130: data acquisition unit
140: prediction unit 150: judgment unit
160: output unit

Claims (12)

a data collection unit configured to collect, for each patient, fixed variable data including body information and comorbid disease information related to the patient, and variable variable data including blood pressure and dialysis conditions observed for each time during hemodialysis;
Using the fixed variable data and time-dependent variable data as input data, the patient's systolic pressure (SBP) after the set time has elapsed at the corresponding time for each time, the amount of SBP change per unit time (SBP), and the amount of change in MAP (ㅿMAP) are output a model learning unit for learning a predictive model using data;
a data acquisition unit for acquiring the fixed variable data and the variable variable data observed for each hour during current hemodialysis in real time from a patient on hemodialysis; and
Blood including a prediction unit that applies the fixed variable data and the variable variable data observed by time to the predictive model in real time to predict the SBP, SBP, and MAP values after a set time has elapsed from the present for each hour during hemodialysis Hypotension prediction device during dialysis. The method according to claim 1,
The apparatus for predicting hypotension during hemodialysis further comprising a determination unit configured to determine in real time whether hypotension is possible during hemodialysis by comparing the SBP, SBP, and MAP values predicted for each hour with respective preset threshold values. 3. The method according to claim 2,
The judging unit,
If the predicted SBP value is less than 90, the SBP value is 20 or more, or the MAP value is 10 or more, the hypotension prediction device during hemodialysis determines that the patient has hypotension after a set time has elapsed from the present. The method according to claim 1,
The predictive model is
A device for predicting hypotension during hemodialysis learned through a recurrent neural network (RNN) algorithm. The method according to claim 1,
The set time is
A device for predicting hypotension during hemodialysis, which is less than half of the total dialysis time required for hemodialysis operations. The method according to claim 1,
The fixed variable data may further include at least one of information on taking medications, blood test information, and a patient's risk of hypotension. The method according to claim 1,
Among the variable variable data, the dialysis condition includes at least one of blood flow, ultrafiltration rate, and dialysis fluid electrolyte concentration,
The variable variable data may further include at least one of a pulse and body temperature observed for each hour during hemodialysis. A method for predicting hypotension using an apparatus for predicting hypotension during hemodialysis, the method comprising:
Collecting, for each patient, fixed variable data including body information and accompanying disease information related to the corresponding patient, and variable variable data including blood pressure and dialysis conditions observed for each time during hemodialysis;
Using the fixed variable data and time-dependent variable variable data as input data, the patient's systolic blood pressure (SBP), systolic blood pressure change per unit time (SBP), and mean arterial blood pressure change (ㅿMAP) ) as output data to train a predictive model;
acquiring the fixed variable data and the variable variable data observed for each hour during current hemodialysis in real time for a patient undergoing hemodialysis; and
The step of applying the fixed variable data and the variable variable data observed by time to the predictive model in real time to predict the SBP, SBP, and MAP values after a set time has elapsed from the present for each hour during hemodialysis. Methods for predicting hypotension during dialysis. 9. The method of claim 8,
The method of predicting hypotension during hemodialysis further comprising the step of determining in real time whether hypotension is possible during hemodialysis by comparing the SBP, SBP, and MAP values predicted for each hour with each preset threshold value. 10. The method of claim 9,
The determining step is
When the predicted SBP value is less than 90, the SBP value is 20 or more, or the MAP value is 10 or more, the hypotension prediction method during hemodialysis is to determine that the patient is likely to have hypotension after a set time has elapsed. 9. The method of claim 8,
The predictive model is
A method for predicting hypotension during hemodialysis learned through a recurrent neural network (RNN) algorithm. 9. The method of claim 8,
The fixed variable data further includes at least one of information about taking medications, blood test information, and a patient's risk of hypotension,
Among the variable variable data, the dialysis condition includes at least one of blood flow, ultrafiltration rate, and dialysis fluid electrolyte concentration,
The method for predicting hypotension during hemodialysis, wherein the variable variable data further includes at least one of a pulse and body temperature observed for each hour during hemodialysis.

Images