Method and system for predicting acute renal disease

Abstract

To provide a method and a system for predicting acute renal diseases that accurately predict whether an acute renal disease is in progress.SOLUTION: An acute renal disease prediction system includes an input device, a storage device, a processor, and an output device. A prediction method of the acute renal disease prediction system includes: a step of inputting multiple pieces of immune cell group data, a blood serum creatinine value and a urea nitrogen value, and storing them in the storage device; a step of accessing the storage device, and constructing an acute renal disease prediction model via a decision tree algorithm using the multiple pieces of immune cell group data, the blood serum creatinine value and the urea nitrogen value as parameters; a step of acquiring the immune cell group data, the blood serum creatinine value and the urea nitrogen value that are evaluation objects, and acquiring a determination result procedure of acute renal diseases by executing a determination procedure; and a step of outputting a determination result of acute renal diseases.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method and system for predicting acute kidney disease (AKD), and more particularly to a method and system that can accurately predict whether a patient will progress to acute kidney disease by analyzing immune cell population data obtained by a flow cytometer together with serum creatinine (cre) and blood urea nitrogen (BUN) using a decision tree (DT) algorithm. [Background technology]

[0002] Acute kidney disease (AKD) is a common and serious condition in intensive care units (ICUs), but it is difficult to diagnose in the early stages of the ICU. This is because clinical judgment requires that patients' renal function remains abnormal even 7 days after the onset of acute kidney injury. Studies have shown that the prevalence of acute kidney disease in intensive care units accounts for approximately half of the incidence of acute kidney injury. Patients with acute kidney disease require intensive care because they may develop chronic kidney disease, which may require further medical interventions such as hemodialysis or kidney transplantation.

[0003] Furthermore, many patients develop sepsis during treatment due to bacterial infections through various routes, such as wounds, mucous membranes, and the respiratory tract, and there are countless cases of acute kidney injury. Furthermore, because the causes of acute kidney disease are complex and change rapidly, determining the timing of treatment is a major challenge for medical professionals. In particular, determining whether complications of acute kidney disease will occur and providing appropriate treatment in advance are major challenges.

[0004] On the other hand, serum creatinine and urea nitrogen are clinical indicators for determining whether acute kidney injury will occur, and in patients with symptoms of acute kidney disease, both levels are abnormally elevated, which can be used to determine whether the patient will develop acute kidney disease. However, due to the physical conditions of different patients, the accuracy of using only serum creatinine and urea nitrogen as clinical indicators is relatively low.

[0005] With the rapid development of technology, artificial intelligence is gradually being applied to various medical diagnostic techniques, and the use of computational methods is a powerful tool to support the research of related problems in the fields of medicine and biology.

[0006] As described above, in clinical practice of assessing the presence or absence of acute kidney disease complications, a preventive, effective, and accurate assessment method is urgently needed so that medical professionals can take appropriate medical measures in advance to reduce the risk of acute kidney disease complications and improve patient survival rates. Summary of the Invention [Problem to be solved by the invention]

[0007] In view of the above-mentioned conventional problems in assessing the presence or absence of acute kidney disease complications, an object of the present invention is to provide a method and system for predicting acute kidney disease that alleviates the problems of erroneous judgments due to artificial evaluations and the difficulty of making prompt medical decisions. [Means for solving the problem]

[0008] According to one object of the present invention, there is provided a method for predicting acute kidney disease, which includes the steps of inputting a plurality of immune cell population data, serum creatinine levels, and urea nitrogen levels via an input device and storing them in a storage device, accessing the storage device via a processor and constructing an acute kidney disease prediction model using a decision tree algorithm with the plurality of immune cell population data, serum creatinine levels, and urea nitrogen levels as parameters, acquiring the plurality of immune cell population data, serum creatinine levels, and urea nitrogen levels to be evaluated via the input device and executing a determination procedure in the processor to obtain a diagnosis of acute kidney disease, and accessing the storage device via an output device and outputting the diagnosis of acute kidney disease, all of which are performed by computer software.

[0009] The immune cell population data, the serum creatinine level, and the urea nitrogen level data are all obtained by analyzing a peripheral blood sample collected from a patient who may have complications of acute kidney disease. The immune cell population data are obtained by analyzing the peripheral blood sample with a flow cytometer.

[0010] The multiple immune cell group data are specifically 55 immune cell group data, so there are 55 feature points in the machine learning, and the types of the 55 immune cell group data are shown in Table 1 below.

[0011] Table 1 JPEG2025141825000002.jpg249170JPEG2025141825000003.jpg212170

[0012] Then, the multiple immune cell population data, serum creatinine levels, and urea nitrogen levels are used as parameters to train a decision tree algorithm. Through supervised learning, these parameters are divided into a training group, a validation group, and a test group, and the model is continuously trained in a loop approach. When the test group reaches the optimal accuracy rate, the key feature of the optimal decision tree, i.e., the key immune cell population data to be combined with serum creatinine levels and urea nitrogen levels, is identified to obtain an acute kidney disease prediction model.

[0013] Finally, the immune cell population data, serum creatinine level, and urea nitrogen level to be evaluated are predicted using the acute kidney disease prediction model to obtain a diagnosis of acute kidney disease. The diagnosis of acute kidney disease allows medical professionals to know whether the patient to be evaluated has acute kidney disease, and to take appropriate medical measures in advance.

[0014] According to another object of the present invention, there is provided an acute kidney disease prediction system including an input device, a storage device, a processor, and an output device. The input device inputs a plurality of immune cell group data, serum creatinine values, and urea nitrogen values, as well as a plurality of immune cell group data, serum creatinine values, and urea nitrogen values ​​to be evaluated. The storage device is connected to the input device and stores the plurality of immune cell group data, serum creatinine values, and urea nitrogen values, as well as a plurality of immune cell group data, serum creatinine values, and urea nitrogen values. The output device is connected to the storage device and is used for outputting an acute kidney disease assessment result. The processor is connected to the storage device and executes a plurality of instructions to construct an acute kidney disease prediction model using a decision tree algorithm with the plurality of immune cell group data, serum creatinine values, and urea nitrogen values ​​as parameters; executes a assessment procedure for the plurality of immune cell group data, serum creatinine values, and urea nitrogen values ​​to be evaluated based on the acute kidney disease prediction model, thereby obtaining an acute kidney disease assessment result; and accesses the storage device via the output device and outputs the acute kidney disease assessment result.

[0015] As described above, by using the acute kidney disease prediction method and prediction system of the present invention, it is possible to quickly and accurately determine whether a patient has concurrent acute kidney disease, thereby reducing the burden on nurses and doctors and alleviating problems caused by differences in judgment criteria between nurses and doctors, allowing nurses and doctors to more accurately assess whether a patient has concurrent acute kidney disease and to more quickly formulate appropriate medical measures. [Brief explanation of the drawings]

[0016] In order to more clearly show the technical features, contents, advantages and achievable effects of the present invention, the present invention will be described in detail below in the form of embodiments with reference to the accompanying drawings.

[0017] [Figure 1] FIG. 1 is a flowchart illustrating a method for predicting acute kidney disease according to one embodiment of the present invention. [Figure 2A] FIG. 1 is a schematic diagram of a decision tree structure obtained using serum creatinine levels and multiple immune cell population data as parameters in one embodiment of the present invention. [Figure 2B] FIG. 2B is an analysis diagram of the prediction accuracy of the acute kidney disease prediction model constructed using the decision tree structure of FIG. 2A. [Figure 3A] FIG. 1 is a schematic diagram of a decision tree structure obtained using urea nitrogen levels and multiple immune cell population data as parameters in one embodiment of the present invention. [Figure 3B] FIG. 3B is an analysis diagram of the prediction accuracy of the acute kidney disease prediction model constructed using the decision tree structure of FIG. 3A. [Figure 4A] FIG. 1 is a schematic diagram of a decision tree structure obtained using serum creatinine and urea nitrogen values ​​as parameters in one embodiment of the present invention. [Figure 4B] FIG. 4B is an analysis diagram of the prediction accuracy of the acute kidney disease prediction model constructed using the decision tree structure of FIG. 4A. [Figure 5A]FIG. 1 is a schematic diagram of a decision tree structure obtained using serum creatinine and urea nitrogen levels, as well as both Naive Treg and NK CD56d immune cells as parameters in one embodiment of the present invention. [Figure 5B] FIG. 5B is an analysis diagram of the prediction accuracy of the acute kidney disease prediction model constructed using the decision tree structure of FIG. 5A. [Figure 6] FIG. 1 is a schematic diagram of an acute kidney disease prediction system according to one embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0018] In order to allow the examiner to easily understand the technical features, contents, advantages, and achievable effects of the present invention, the present invention will be described in detail in the form of an embodiment with reference to the accompanying drawings. The drawings used herein are for illustration and to assist the description, and do not necessarily reflect the actual proportions or precise configurations of the present invention after implementation. Therefore, the scope of the present invention should not be interpreted based on the proportions or configurations of the accompanying drawings in actual implementation.

[0019] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by a person skilled in the art to which this invention belongs. Furthermore, terms as defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the relevant art and this invention, and should not be interpreted as having an idealized or overly formal meaning unless expressly defined as such in this specification.

[0020] 1, which is a flowchart of a method for predicting acute kidney disease according to one embodiment of the present invention, as shown in FIG. 1, the method for predicting acute kidney disease includes the following steps (S1 to S4).

[0021] Step S1: A plurality of immune cell population data, serum creatinine values, and urea nitrogen values ​​are input through an input device and stored in a storage device.

[0022] The collected immune cell population data, serum creatinine values, and urea nitrogen values ​​are input into the system's storage device via an input device. Here, the input device is not limited to a flow cytometer for acquiring immune cell population data or other analytical devices capable of acquiring serum creatinine values ​​and urea nitrogen values. The immune cell population data, serum creatinine values, and urea nitrogen values ​​stored in the medical institution's database can be transmitted via a physical line or a file in a storage device. Alternatively, the immune cell population data, serum creatinine values, and urea nitrogen values ​​can be input into the system's database via wired or wireless network transmission as training data for model construction.

[0023] Among them, the immune cell population data is obtained by collecting peripheral blood samples from patients with possible acute kidney disease complications and then analyzing the peripheral blood samples using a flow cytometer. Therefore, the peripheral blood samples from each patient with possible acute kidney disease complications all contain multiple immune cell population data, specifically 55 immune cell population data, as listed in Table 1 above.

[0024] Step S2: The storage device is accessed via the processor, and an acute kidney disease prediction model is constructed using a decision tree algorithm with multiple immune cell group data, serum creatinine levels, and urea nitrogen levels as parameters.

[0025] The processor reads the immune cell population data, serum creatinine levels, and urea nitrogen levels stored in the storage device, and uses the immune cell population data (the 55 immune cell population data) and values ​​such as serum creatinine and urea nitrogen levels as training parameters to train the decision tree algorithm, and continuously trains the model using a loop approach. When the test group achieves optimal accuracy, the key feature points (key immune cell population data) of the optimal decision tree are obtained to obtain an acute kidney disease prediction model.

[0026] Therefore, the above parameters obtained from patients with possible acute kidney disease complications are divided into a training group and a test group based on the total number of patients to build an acute kidney disease prediction model.The ratio of the training group to the test group is 8:2, that is, 80% training group and 20% test group.Since the actual number of patients is 72, the training group will have 57 patients and the test group will have 15 patients.

[0027] The acute kidney disease prediction model obtained through training was then retrained using another 106 patients who may have acute kidney disease as a training group, and the predictive accuracy of the retrained acute kidney disease prediction model was confirmed using another 32 patients who may have acute kidney disease as a validation group. The personal medical information of the 106 patients in the training group and the 32 patients in the validation group is as shown in Table 2 below. Of these, the 106 patients in the training group were recruited between 2020 and 2021, and the 32 patients in the validation group were recruited in 2022. Increasing the number of patients in the training group (number of samples) can further improve the predictive accuracy of the trained acute kidney disease prediction model.

[0028] Table 2 JPEG2025141825000004.jpg255170JPEG2025141825000005.jpg25170

[0029] Referring to Figures 2A and 3A, these are schematic diagrams of a decision tree structure obtained using serum creatinine values ​​and multiple immune cell group data as parameters, and a schematic diagram of a decision tree structure obtained using urea nitrogen values ​​and multiple immune cell group data as parameters, respectively, in one embodiment of the present invention.

[0030] In the schematic diagram of the decision tree structure, Cr represents serum creatinine, BUN represents urea nitrogen, and gini represents the Gini coefficient, which indicates the mean level of the classified groups (generally, a gini coefficient of 0 indicates perfect equality, less than 0.2 indicates high equality, 0.2–0.3 indicates equality, 0.3–0.4 indicates an acceptable range, 0.4–0.6 indicates large disparities, 0.6 or greater indicates high inequality, and 1.0 indicates complete inequality). samlpes is the total number of people, and value is the number of people classified by machine learning as likely to develop acute kidney disease. class is the classification group; in the case of sepsis-associated acute kidney disease (SA-AKD), it is classified as acute kidney disease accompanied by sepsis; if it is not SA-AKD, it is displayed as "Non."

[0031] For example, in the first layer of the decision tree structure in Figure 2A, serum creatinine value ≦ 1.185 is the classification threshold, the total number of people is 57, the value is [28, 29], the value 28 on the left represents non-SA-AKD, the value 29 on the right represents SA-AKD, and the Gini coefficient calculated by the ratio of the two is approximately 0.5. Because the number of people with SA-AKD is relatively high, this layer is classified as SA-AKD. Next, in the second layer, JPEG2025141825000006.jpg417Treg≦0.002 was the classification threshold, and 9 of the 57 patients JPEG2025141825000007.jpg417Treg>0.002, the value is [0, 9], and the Gini coefficient is 0.0, meaning that there is perfect equality and all nine of these individuals are SA-AKD patients, so classification is stopped. The remaining 48 individuals have a value of [28, 20], a Gini coefficient of 0.486, and because there are more non-SA-AKD individuals, they are classified as Non and continue to be classified. In this way, classification continues at different thresholds for immune cells until the Gini coefficient reaches 0.0, completing the decision tree structure.

[0032] Furthermore, the key features of the optimal decision tree were extracted using the classification nodes of the decision tree as weights. The higher the classification node, the greater the weight of the immune cells used, indicating a greater contribution and importance in classifying whether sepsis-related acute kidney disease will occur.

[0033] As can be seen from Figures 2A and 3A, the decision tree structure constructed by combining multiple immune cell population data, whether serum creatinine or urea nitrogen, has both in the second layer. JPEG2025141825000008.jpg417Treg, and in the third tier, both have NK CD56d, which means that these two immune cells account for a large weight. Therefore, these two immune cells are used as the main immune cell group data for the optimal decision tree, and serum creatinine and urea nitrogen levels are used as parameters to construct an accurate acute kidney disease prediction model.

[0034] Next, referring to Figures 2B and 3B, these are diagrams illustrating the analysis of the prediction accuracy of the acute kidney disease prediction model constructed using the decision tree structure of Figures 2B and 3B, respectively, in one embodiment of the present invention. As mentioned above, these two diagrams confirm the accuracy of the acute kidney disease prediction model trained on 106 additional patients as a training group and then retrained on 32 additional patients as a validation group. As shown in Figures 2B and 3B, the prediction accuracy of the training group (106 patients) was 89.62%, the sensitivity was 100% and 94.44%, respectively, and the specificity was 84.29% and 87.14%, respectively. However, the prediction accuracy of the validation group (32 patients) was 81.25%, the sensitivity was 100%, and the specificity was 77.78%. From the above, it can be seen that the prediction accuracy of the decision tree structures constructed by combining multiple immune cell group data was 80% or higher for both serum creatinine and urea nitrogen values.

[0035] On the other hand, to confirm that the predictive accuracy of an acute kidney disease prediction model constructed by combining serum creatinine or urea nitrogen levels with data on multiple immune cell groups is higher than when using only serum creatinine or urea nitrogen levels, which are known as clinical indicators of acute kidney disease, we similarly constructed an acute kidney disease prediction model using a decision tree algorithm with only serum creatinine and urea nitrogen levels as parameters and compared the predictive accuracy.

[0036] Referring to Figures 4A and 4B, in one embodiment of the present invention, a schematic diagram of a decision tree structure obtained using serum creatinine and urea nitrogen values ​​as parameters, and an analysis diagram of the predictive accuracy of an acute kidney disease prediction model constructed using the same are shown.

[0037] As can be seen from the decision tree structure in Figure 4A, using serum creatinine and urea nitrogen levels indeed allows for rapid classification of SA-AKD and non-SA-AKD patients. However, as can be seen from Figure 4B, the acute kidney disease prediction model obtained using only serum creatinine and urea nitrogen levels as training parameters had a prediction accuracy of 77.36%, a sensitivity of 66.67%, and a specificity of 82.86% in the training group (106 patients). The prediction accuracy in the validation group (32 patients) was 75%, a sensitivity of 100%, and a specificity of 70.37%. From these results, it can be seen that the prediction accuracy of the acute kidney disease prediction model constructed using only serum creatinine and urea nitrogen levels as training parameters was less than 80%, lower than the prediction accuracy when combined with multiple immune cell population data.

[0038] Furthermore, a decision tree algorithm was used to construct an acute kidney disease prediction model using multiple immune cell population data, serum creatinine levels, and urea nitrogen levels as parameters, and it was confirmed whether the measurement accuracy could be further improved by combining these three parameters.

[0039] Referring to Figures 5A and 5B, in one embodiment of the present invention, serum creatinine and urea nitrogen levels, as well as JPEG2025141825000009.jpg417This is a schematic diagram of the decision tree structure obtained using both Treg and NK CD56d immune cells as parameters, and an analysis of the predictive accuracy of the acute kidney disease prediction model constructed using this structure.

[0040] As can be seen from FIGS. 2A and 3A, JPEG2025141825000010.jpg417Since the weights of both Treg and NK CD56d immune cells are relatively large, these immune cells are combined with serum creatinine and urea nitrogen levels as training parameters to build an acute kidney disease prediction model. As can be seen from Figure 5B, serum creatinine and urea nitrogen levels, as well as JPEG2025141825000011.jpg417An acute kidney disease prediction model constructed using both Treg and NK CD56d immune cells as parameters had a prediction accuracy of 88.68%, a sensitivity of 94.44%, and a specificity of 85.71% in the training group (106 patients). The prediction accuracy of the validation group (32 patients) was 81.25%, a sensitivity of 100%, and a specificity of 77.78%. Based on the above, by using only two immune cell data with higher classification weights and combining them with serum creatinine and urea nitrogen levels, a prediction accuracy of over 80% can be achieved, eliminating the need to simultaneously acquire data on the 55 immune cell groups as parameters for constructing an acute kidney disease prediction model. Not only does this speed up the construction of prediction models, but it also further improves the analysis efficiency of peripheral blood samples. JPEG2025141825000012.jpg417All you need to do is obtain immune cell data for both Treg and NK CD56d.

[0041] Step S3: Data on multiple immune cell groups to be evaluated, serum creatinine levels, and urea nitrogen levels are obtained via the input device, and a determination procedure is performed in the processor to obtain a determination result of acute kidney disease.

[0042] The acute kidney disease prediction model constructed in step S2 is used to evaluate the immune cell population data, serum creatinine levels, and urea nitrogen levels of the patients to be evaluated, and an acute kidney disease assessment result is obtained. The acute kidney disease assessment result indicates whether the patients to be evaluated will develop acute kidney disease, allowing medical professionals to take appropriate medical measures in advance. The input device used here is the same as that described above, and will not be repeated here.

[0043] To confirm the accuracy of acute kidney disease prediction using different algorithms, peripheral blood samples from 106 patients were analyzed using multiple immune cell group data (i.e., the 55 immune cell group data), serum creatinine levels, and urea nitrogen levels as parameters. For example, acute kidney disease prediction models constructed using other machine learning algorithms, such as support vector machines (SVM) and k-nearest neighbor algorithms (KNN), were compared with acute kidney disease prediction models constructed using a decision tree (DT) algorithm. The prediction accuracies of the three models are shown in Table 3 below.

[0044] Table 3 JPEG2025141825000013.jpg20170

[0045] As can be seen from Table 3, the prediction accuracy of SVM is 70%, while that of KNN is only 50%, both of which are much lower than the 85% prediction accuracy of the decision tree algorithm. From the above, it can be seen that the acute kidney disease prediction model built using the decision tree algorithm is superior to other types of algorithms.

[0046] Step S4: The storage device is accessed via the output device to output the assessment result of acute kidney disease.

[0047] The diagnosis result of acute kidney disease obtained in step S3 can be output through an output device, which may include various display interfaces such as a computer screen, a display, or a display of a handheld device.

[0048] 6 is a schematic diagram of an acute kidney disease prediction system according to one embodiment of the present invention. As shown in FIG. 6, an acute kidney disease prediction system 20 may include an input device 21, a storage device 22, a processor 23, and an output device 24.

[0049] In this embodiment, the input device 21 may be configured with a flow cytometer and an analyzer for detecting serum creatinine and urea nitrogen. A peripheral blood sample collected from a patient with possible acute kidney disease is analyzed to obtain multiple immune cell population data, serum creatinine levels, and urea nitrogen levels. In other embodiments, the input device 21 is not limited to a flow cytometer and an analyzer for detecting serum creatinine and urea nitrogen. The input device 21 may include an input interface for an electronic device such as a personal computer, smartphone, or server, including a touchscreen, keyboard, mouse, etc., and transmits the multiple immune cell population data, serum creatinine levels, and urea nitrogen levels in the form of a file. Alternatively, the history data may be uploaded and stored in the memory of the storage device 22 via wireless network transmission, wireless communication transmission, or a standard wired internet connection. The memory may include a read-only memory, flash memory, disk, cloud database, etc.

[0050] Next, the acute kidney disease prediction system 20 accesses the storage device 22 through the processor 23, which may include a central processing unit, a graphics processor, a microprocessor, etc. in a computer or server, and may also include a multi-core processing device or a combination of multiple processing devices. The processor 23 executes instructions to access the multiple immune cell population data, serum creatinine values, and urea nitrogen values ​​stored in the storage device 22, access the multiple immune cell population data, serum creatinine values, and urea nitrogen values ​​to be evaluated, and perform a judgment procedure. Specifically, the training procedure uses the multiple immune cell population data, serum creatinine values, and urea nitrogen values ​​originally stored in the storage device 22 as parameters to perform calculations using a decision tree algorithm to construct a prediction model for acute kidney disease.

[0051] Then, the immune cell population data to be evaluated, serum creatinine level, and urea nitrogen level are calculated through the constructed acute kidney disease prediction model in a determination procedure to obtain a diagnosis of acute kidney disease. The output device 24 accesses the storage device 22 to output the diagnosis of acute kidney disease, and the output device 24 may include various display interfaces such as a computer screen, a display, or a handheld device display.

[0052] The method and system for predicting acute kidney disease of the present invention can significantly reduce the burden on medical professionals and doctors, reduce bias in the diagnosis of whether or not a patient has acute kidney disease due to artificial misjudgment, and can quickly and accurately determine whether or not a patient has acute kidney disease, thereby reducing the burden on nurses and doctors and alleviating the problem of artificially different judgment criteria between nurses and doctors.It also helps nurses and doctors to more accurately evaluate whether or not a patient has acute kidney disease and make appropriate medical decisions more quickly.

[0053] The above is for illustrative purposes only and is not intended to be limiting. Any equivalent modifications or variations to the present invention that do not depart from the spirit and scope of the present invention should be included in the appended claims. [Explanation of symbols]

[0054] 20: Acute kidney disease prediction system 21: Input device 22: Storage device 23: Processor 24: Output device S1~S4: Step

Claims

1. 1. A method for predicting acute kidney disease, comprising: Step S1: inputting a plurality of immune cell group data, serum creatinine values, and urea nitrogen values ​​via an input device and storing them in a storage device; a step S2 of accessing the storage device via a processor and constructing an acute kidney disease prediction model using a decision tree algorithm with the plurality of immune cell group data, the serum creatinine value, and the urea nitrogen value as parameters; a step S3 of acquiring a plurality of immune cell population data, serum creatinine value, and urea nitrogen value to be evaluated via the input device, and executing a determination procedure in the processor to obtain a determination result of acute kidney disease; a step S4 of accessing the storage device via an output device and outputting the determination result of acute kidney disease; Including, Steps S1 to S4 of this acute kidney disease prediction method are all performed by computer software.

2. The method for predicting acute kidney disease according to claim 1 , wherein the input device comprises a flow cytometer and an analyzer for detecting serum creatinine and urea nitrogen.

3. The immune cell group data includes data on B cells, T cells, helper T cells, activated helper T cells (HLADR+), naive helper T cells (CD62L+), naive helper T cells (CD45RA+), memory helper T cells (CD62L-HLADR+), memory helper T cells (CD45RO+), regulatory helper T cells, type 1 helper T cells, naive type 1 helper T cells, memory type 1 helper T cells, and type Type 2 helper T cells, naive type 2 helper T cells, memory type 2 helper T cells, regulatory helper T cells, naive regulatory helper T cells, memory regulatory helper T cells, type 17 helper T cells, naive type 17 helper T cells, memory type 17 helper T cells, type 22 helper T cells, naive type 22 helper T cells, memory type 22 helper T cells, follicular helper T cells, naive follicular helper T cells 2. The method for predicting acute kidney disease according to claim 1, wherein the immune cell group data is selected from 55 immune cell group data: immune T cells, memory follicular helper T cells, cytotoxic T cells, activated cytotoxic T cells, naive cytotoxic T cells, memory cytotoxic T cells, regulatory cytotoxic T cells, double positive T cells, double negative T cells, natural killer cells, natural killer cell CD56b, natural killer cell CD56d, CD56b cells, CD56d cells, natural killer T cells, natural killer T cells CD8, natural killer T cells CD4, double positive natural killer T cells, double negative natural killer T cells, CD56b natural killer T cells, CD56d natural killer T cells, CD56bCD16- cells, CD56dCD16- cells, dendritic cells, mature dendritic cells, immature dendritic cells, monocytes, classical monocytes, non-classical monocytes, and intermediate monocytes.

4. The acute kidney disease prediction method according to any one of claims 1 to 3, wherein the acute kidney disease prediction model includes an optimal decision tree, and the optimal decision tree has two major immune cell populations.

5. The two major immune cell populations are naive regulatory helper T cells ( The method for predicting acute kidney disease according to claim 4, wherein the target antigens are Treg (Treg) and natural killer cell CD56d (NK CD56d).

6. An acute kidney disease prediction system including an input device, a storage device, an output device, and a processor, the input device is used to input a plurality of immune cell group data, a serum creatinine value, and a urea nitrogen value, as well as a plurality of immune cell group data, a serum creatinine value, and a urea nitrogen value to be evaluated; the storage device is connected to the input device and is used to store the plurality of immune cell group data, serum creatinine value, and urea nitrogen value, as well as the plurality of immune cell group data, serum creatinine value, and urea nitrogen value to be evaluated; the output device is connected to the storage device and is used to output the assessment result of acute kidney disease; The processor is coupled to the storage device and executes a plurality of instructions to constructing an acute kidney disease prediction model using a decision tree algorithm with the immune cell group data, serum creatinine level, and urea nitrogen level as parameters; a step of performing a determination procedure on the plurality of immune cell group data, serum creatinine value, and urea nitrogen value to be evaluated using the acute kidney disease prediction model, and obtaining a determination result of acute kidney disease; and accessing the storage device via the output device and outputting the acute kidney disease assessment result.

7. The acute kidney disease prediction system according to claim 6, wherein the input device comprises a flow cytometer and an analyzer for detecting serum creatinine and urea nitrogen.

Images