TWI870263B - Method for assessing length of stay in patient with dengue fever and system thereof - Google Patents

Abstract

The present invention relates to a method for assessing length of stay in patient with dengue fever and system thereof. The method for assessing length of stay in patient with dengue fever comprises the following steps: inputting a plurality of cytokine data and a plurality of blood biomarker data obtained from dengue patients through an input device, and storing the plurality of cytokine data and the plurality of blood biomarker data in a storage device; performs biostatistical analysis on the plurality of cytokine data and the plurality of blood biomarker data, finds out cytokine data and blood biomarker data with significant differences, then applies the cytokine data and the blood biomarker data with significant differences to establish a dengue fever patient length of stay assessment model with principal component analysis(PCA) algorithm; obtain cytokine data and blood biomarker data to be discriminated through the input device, and use the processor to perform an interpretation program to obtain a interpretation result of length of stay in patient with dengue fever; the interpretation result of the length of stay in patient with dengue fever is output through a output device. The system includes an input device, a storage device, a processor, and an output device.

Description

Method and system for evaluating hospitalization time of dengue fever patients

The present invention relates to a method and system for evaluating the hospitalization time of dengue fever patients, and in particular to an evaluation method and system for accurately predicting the hospitalization time of dengue fever patients by analyzing cytokine data and blood marker data using biostatistical analysis combined with principal component analysis (PCA) algorithm.

Dengue fever is an acute viral infection caused by dengue virus (DENV). It is transmitted by the bites of mosquitoes carrying the dengue virus, mainly Aedes aegypti and Aedes albopictus. The disease is common in tropical and subtropical regions and is a major public health issue in these regions. Symptoms of dengue fever range from mild fever and joint pain to severe bleeding and shock, and severe cases may lead to death.

Symptoms of dengue fever usually appear within 4 to 10 days after infection, and include sudden high fever, headache, muscle and joint pain, rash and bleeding tendency. In some cases, dengue fever may progress to severe hemorrhagic tropical disease, in which case patients may experience life-threatening complications such as thrombocytopenia, organ failure and shock.

Dengue fever is graded according to severity, usually mild, moderate, and severe, based on the patient's symptoms and clinical manifestations. Mild cases may present with no symptoms or fever, headache, muscle and joint pain, which may be accompanied by a rash, but the patient usually remains awake and has no signs of organ damage; moderate to severe cases present with severe headache and fatigue, thrombocytopenia, and organ damage. If the disease is a little more severe, there may be bleeding tendency, low blood pressure, and organ failure, requiring hospitalization; and patients with severe cases may have shock and severe organ failure, requiring further treatment with blood transfusions or other supportive therapies while in hospital.

The current detection methods for grading the severity of dengue fever mainly include clinical symptoms and laboratory tests. The doctor will evaluate the patient's symptoms and test the patient's condition by testing prothrombin time (PT, in seconds), activated partial thromboplastin time (aPTT, in seconds), white blood cell count (WBC, in 10 ⁶ /L), hemoglobin concentration (Hb, in g/dL), hematocrit (Hct, in %), platelet count (platelet, in 10 ⁹ The severity of the disease is assessed by blood markers such as serum creatinine (creatinine, mg/dL), aspartate aminotransferase (AST, U/L), alanine aminotransferase (GPT, U/L), blood urea nitrogen (BUN, mg/dL), and

However, even if a patient is found to be infected with dengue fever, it is impossible to know whether the patient will further develop into moderate, severe or extremely severe illness. Therefore, when a patient is found to be infected with dengue fever, it is generally recommended that the patient be hospitalized for observation or further treatment, which in turn increases the burden on the medical system. If the patient does not develop into moderate, severe or extremely severe illness after hospitalization, or if the patient recovers well within a few days after subsequent hospitalization but continues to be hospitalized, such patients will not only consume medical manpower and resources, but will also prevent patients with moderate, severe or extremely severe illness from receiving effective care and treatment.

In summary, there is an urgent need for effective hospitalization time assessment methods in the clinical practice of dengue fever, providing doctors with additional patient cytokine-related data supplemented by blood marker data as a reference to accelerate the judgment of whether dengue patients need to be hospitalized for treatment, or whether they need short-term or long-term hospitalization, so that medical manpower and resources can be properly utilized to reduce the burden on the medical system.

In view of the above-mentioned problem of evaluating the hospitalization time of dengue fever patients, the purpose of the present invention is to provide a method and system for evaluating the hospitalization time of patients, so that medical staff can make quick medical decisions on whether dengue fever patients need to be hospitalized and the length of hospitalization time, improve the work efficiency of the medical system and reduce the burden of medical manpower, and enable patients to receive timely and appropriate treatment.

According to one purpose of the present invention, a method for evaluating the hospitalization time of dengue patients is proposed, which comprises the following steps: inputting a plurality of cytokine data and a plurality of blood marker data through an input device and storing them in a storage device; accessing the storage device through a processor, performing biostatistical analysis on the plurality of cytokine data and the plurality of blood marker data, and finding out the cytokine data and blood marker data with significant differences. Then, the cytokine data and blood marker data with significant differences are used to establish a dengue fever patient hospitalization time evaluation model using a principal component analysis (PCA) algorithm; the cytokine data and blood marker data to be evaluated are obtained through an input device, and a processor is used to perform a judgment program to obtain a dengue fever patient hospitalization time judgment result; the storage device is accessed through an output device to output the dengue fever patient hospitalization time judgment result.

Among them, the multiple cytokine data are the expression levels (pg/ml) of various cytokines, and the multiple blood marker data include prothrombin time, partial thromboplastin time, white blood cell count, hemoglobin concentration, hematocrit, platelet count, aspartate transaminase concentration, alanine transaminase concentration, blood urea nitrogen concentration and serum creatinine concentration; the multiple cytokine data and the multiple blood marker data are all obtained from the patient's blood sample.

There are 24 cytokines, and their types are shown in Table 1:

Table 1 Cytokines Chinese name IL-10 Interleukin-10 IFN-γ Interferon-γ IL-6 Interleukin-6 TNF-α Tissue necrosis factor alpha IL-8 Interleukin-8 IP-10 Interferon gamma-induced 10-kilodalton protein MCP-1 Monocyte tropism factor protein 1 IL-4 Interleukin-4 MIP-1β Macrophage inflammatory protein-1β IFN-α Interferon-α RANTES TNF-L5 GM-CSF Granular mononuclear cell colony stimulating growth factor IL-15 Interleukin-15 VEGF Vascular endothelial growth factor IL-13 Interleukin-13 IL-18 Interleukin-18 IL-1Rα IL-1Rα IL-12 Interleukin-12 MIP-1α Macrophage inflammatory protein-1α MIG Interferon gamma-induced mononuclear factor IL-17 Interleukin-17 IL-1β Interleukin-1β IL-7 Interleukin-7 IL-2 Interleukin-2

Since there are a large number of cytokines and blood markers, the efficiency and accuracy of directly training a large amount of cytokine data and blood marker data by machine learning are low. Therefore, a large amount of cytokine data and blood marker data are first subjected to biostatistical analysis to find out cytokines and blood markers with significant differences for machine learning, which can greatly improve the training efficiency and accuracy of machine learning.

The biostatistical analysis is classified according to whether hospitalization is required, hospitalization ≤ 7 days, and hospitalization > 7 days, and cytokine data and blood marker data with significant differences are found out from a large amount of cytokine data and blood marker data.

Then, the cytokine data and blood marker data with significant differences were trained using the principal component analysis (PCA) algorithm to obtain a dengue fever patient hospitalization time assessment model; the principal component analysis can find the maximum variance after weighted averaging of the values, and can transform many highly correlated variables into independent variables, and select fewer new variables that can explain most of the variance in the data, which are the so-called principal components, and then use the selected principal components to explain the comprehensive indicators of the original data.

Finally, the cytokine data and blood marker data to be evaluated are evaluated using the dengue fever patient hospitalization time evaluation model to obtain the dengue fever patient hospitalization time judgment result; from the dengue fever patient hospitalization time judgment result, it can be known whether the dengue fever patient to be evaluated needs hospitalization, or whether short-term or long-term hospitalization is required, thereby providing medical staff with treatment and handling methods for the dengue fever patients waiting for evaluation, so as to effectively reduce the burden on the medical system and further enable patients to receive timely and appropriate treatment.

According to another object of the present invention, a dengue fever patient hospitalization time evaluation system is provided, which includes an input device, a storage device, a processor and an output device. The input device is used to input a plurality of cytokine data and a plurality of blood marker data of dengue patients, as well as the cytokine data and blood marker data to be evaluated; the storage device is connected to the input device, and is used to store a plurality of cytokine data and a plurality of blood marker data, as well as the cytokine data and blood marker data to be evaluated; the output device is connected to the storage device, and is used to output the dengue fever patient hospitalization time evaluation result; the processor is connected to the storage device, and executes a plurality of instructions to perform the following steps: a plurality of cytokine data and a plurality of blood marker data are evaluated; The blood marker data is subjected to biostatistical analysis to find out the cytokine data and blood marker data with significant differences; the cytokine data and blood marker data with significant differences are calculated using the principal component analysis (PCA) algorithm to establish a dengue fever patient hospitalization time evaluation model; based on the dengue fever patient hospitalization time evaluation model, the cytokine data and blood marker data to be evaluated are subjected to a judgment procedure to obtain a dengue fever patient hospitalization time judgment result; the dengue fever patient hospitalization time judgment result is output by accessing the storage device through the output device.

As mentioned above, the dengue fever patient hospitalization time assessment method and system of the present invention can quickly and accurately determine whether the dengue fever patient needs to be hospitalized, or whether short-term or long-term hospitalization is required, by combining cytokine data with blood markers known to detect dengue fever, so that subsequent medical staff can quickly choose the corresponding medical decision for the patient. In addition to improving the use of medical resources, it can also effectively treat dengue fever patients and increase the survival rate of patients.

In order to help the review committee understand the technical features, contents and advantages of the present invention and the effects that can be achieved, the present invention is described in detail as follows with the accompanying drawings and in the form of embodiments. The drawings used therein are only for illustration and auxiliary description, and may not be the true proportions and precise configurations after the implementation of the present invention. Therefore, it should not be interpreted based on the proportions and configurations of the attached drawings to limit the scope of rights of the present invention in actual implementation.

Unless otherwise defined, all terms (including technical and scientific terms) used herein have the meanings commonly understood by those of ordinary skill in the art to which the present invention belongs. It will be further understood that those terms as defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meanings in the context of the relevant art and the present invention, and will not be interpreted as idealized or overly formal meanings unless expressly so defined herein.

Please refer to FIG. 1, which is a flow chart of the method for assessing the hospitalization time of a dengue fever patient according to an embodiment of the present invention. As shown in FIG. 1, the method for assessing the progress of dengue fever comprises the following steps (S1-S4):

Step S1: Input a plurality of cytokine data and a plurality of blood marker data of a dengue patient through an input device and store them in a storage device.

The cytokine data and multiple blood marker data of dengue patients that can be collected are input into the storage device of the system through the input device. The input device described here is not limited to a flow cytometer for obtaining cytokine data and a hematology analyzer for obtaining blood marker data. The cytokine data and blood marker data stored in the database of a medical institution can also be input into the system database through physical lines, files in a storage device, or files of cytokine data and blood marker data can be transmitted through a wired or wireless network to serve as training data for model construction.

The multiple cytokine data and the multiple blood marker data are all obtained from the patient's blood sample, and the data are all obtained from the blood sample obtained within 7 days when the patient has mild clinical manifestations (i.e., when the patient goes to the hospital for medical treatment); therefore, the blood sample of each dengue fever patient has multiple cytokine data and multiple blood marker data. The marker data, the multiple cytokine data are specifically 24 cytokine data (as shown in the aforementioned Table 1), and the multiple blood marker data are prothrombin time, partial thromboplastin time, white blood cell count, hemoglobin concentration, hematocrit, platelet count, aspartate aminotransferase concentration, alanine aminotransferase concentration, blood urea nitrogen concentration and serum creatinine concentration.

Step S2: The processor accesses the storage device to perform biostatistical analysis on a plurality of cytokine data and a plurality of blood marker data to find out the cytokine data and blood marker data with significant differences, and then uses the principal component analysis (PCA) algorithm to establish a dengue fever patient hospitalization time evaluation model.

Due to the large number of cytokines and blood markers, the efficiency and accuracy of directly training a large amount of cytokine data and blood marker data by machine learning are low; therefore, a large amount of cytokine data and blood marker data are first subjected to biostatistical analysis to find out cytokines and blood markers with significant differences for machine learning, which can greatly improve the training efficiency and accuracy of machine learning. Whether dengue fever patients need to be hospitalized, or whether they need short-term or long-term hospitalization

Please refer to Table 2, which shows the changes in the expression of 24 cytokines (concentration, unit is pg/ml) of dengue fever patients according to whether they need hospitalization, short-term hospitalization (≤7 days) or long-term hospitalization (＞7 days). The data are calculated using mean ± standard deviation, and the Spearman correlation method is used for data analysis, and the effective p value is ≤0.05; among them, the cytokines and their values in bold in Table 2 represent that there is a significant relationship between these cytokines and whether hospitalization is required and the length of hospitalization.

Table 2 Cytokine (mean ± SD) Hospitalization Length of hospital stay No hospitalization required (NO) ≦7 days >7 days P-value Correlation coefficient GM-CSF 11.58 ± 41.15 12.2 ± 31.59 154.72 ± 923.62 0.353 0.098 IFN- ⍺ 469.1 ± 1140.41 155.92 ± 204.81 201.36 ± 397.75 0.010 -0.266 IFN-γ 170.04 ± 195.82 104.04 ± 104.95 107.18 ± 201.01 0.002 -0.312 IL-1β 29.61 ± 93.43 5.5 ± 5.16 15.95 ± 75.09 0.052 -0.203 IL-1R⍺ 55.2 ± 93.93 229.25 ± 868.71 295.97 ± 873.66 0.390 0.091 IL-2 2.4 ± 6.9 0.5 ± 0.33 3.92 ± 18.78 0.149 -0.151 IL-4 0.95 ± 2.1 1 ± 1.45 1.15 ± 4.79 0.208 -0.132 IL-6 23.88 ± 33.97 67.69 ± 167 409.04 ± 1592.78 0.000 0.488 IL-7 67.15 ± 303.71 3.53 ± 5.22 4.64 ± 8.23 0.061 -0.196 IL-8 82.62 ± 148.62 114.69 ± 301.9 218.47 ± 424.41 0.011 0.265 IL-10 51.34 ± 100.03 223.47 ± 616.01 150.53 ± 321.49 0.184 0.14 IL-12 14.03 ± 39.17 2.12 ± 2.03 8.16 ± 43.22 0.025 -0.234 IL-13 20.74 ± 54.94 18.73 ± 19.77 53.57 ± 234.37 0.209 -0.132 IL-15 34.15 ± 19.6 32.58 ± 26.08 54.16 ± 59.68 0.017 0.248 IL-17 8.28 ± 27.03 2.39 ± 2.36 12.26 ± 68.18 0.015 -0.252 IL-18 53.86 ± 65.56 207.59 ± 215.09 154.8 ± 163.13 0.004 0.298 IP-10 182895 ± 478065 52472 ± 58699 51760 ± 55731 0.521 -0.068 MCP-1 6019 ± 15290 2194 ± 3010 2749 ± 2561 0.929 -0.009 MIG 8236 ± 5231 8642 ± 6122 19068 ± 21733 0.000 0.389 MIP-1⍺ 27.2 ± 19.93 39.98 ± 59.48 87.83 ± 283.17 0.742 0.035 MIP-1β 47.67 ± 33.31 69.35 ± 97.17 90.7 ± 117.49 0.183 0.14 RANTES 12405 ± 7554 8007 ± 6458 6687 ± 6234 0.001 -0.35 TNF- ⍺ 28.23 ± 18.93 45.6 ± 62.54 52.71 ± 48.6 0.032 0.223 VEGF 21.69 ± 20.36 46.96 ± 58.9 65.9 ± 122.94 0.310 0.107

As can be seen from Table 2, the expression levels of different cytokines vary depending on whether hospitalization is required and the length of hospitalization. In addition, depending on whether hospitalization is required and the length of hospitalization, the expression levels of some cytokines are significantly upregulated or downregulated.

Please refer to Table 3, which shows the changes in blood markers of dengue patients according to whether they need hospitalization, short-term hospitalization (≤7 days) or long-term hospitalization (>7 days). The data are calculated using mean ± standard deviation, and the Spearman correlation method is used for data analysis, and the effective p value is ≤0.05; among them, the blood markers and their values in bold in Table 3 represent that there is a significant relationship between these blood markers and whether hospitalization is required and the length of hospitalization.

Table 3 Blood markers (mean ± SD) Normal reference value Hospitalization Length of hospital stay No hospitalization required (NO) ≦7 days >7 days P-value Correlation coefficient Coagulation test Prothrombin time (seconds) 11.5~14.5 13 ± 1.6 12.1 ± 1.4 63.1 ± 196.8 0.002 0.477 Partial thromboplastin time (seconds) 28.6~38.2 41.6 ± 2.1 48.1 ± 8.6 47.7 ± 11.4 0.166 0.226 Complete blood count White blood cell count (10 ⁶ /L) 3800~10400 4700 ± 1818 5661 ± 4548 6798 ± 3986 0.005 0.291 Hemoglobin concentration (g/dL) Male: 13.6~16.9 Female: 11.9~14.8 13.7 ± 1.6 13.6 ± 1.6 12.8 ± 3.0 0.169 -0.149 Hematocrit (%) Male: 40~50 Female: 35~43 41.4 ± 5.9 41.1 ± 4.9 42.6 ± 24.9 0.345 -0.118 Platelet count (10 ⁹ /L) Male: 152~324 Female: 153~361 139 ± 49 91 ± 70 83 ± 98 0.001 -0.342 Liver and kidney function Aspartate aminotransferase concentration (U/L) 10~40 45 ± 17 155 ± 157 841 ± 2190 0.000 0.581 Alanine transaminase concentration (U/L) 10~40 25 ± 24 79 ± 106 224 ± 563 0.000 0.454 Blood urea nitrogen concentration (mg/dL) 8~20 21 ± 10 43 ± 44 29 ± 21 0.528 0.116 Serum creatinine concentration (mg/dL) Male: 0.5~1.1 Female: 0.7~1.3 1.0 ± 0.4 1.6 ± 2.4 1.5 ± 1.2 0.102 0.192

Next, refer to Figure 2A and Figure 2B, which are heat maps of the expression of different cytokines and different blood markers in accordance with the embodiments of the present invention, in terms of whether hospitalization is required, short-term hospitalization (≤7 days), or long-term hospitalization (>7 days). As can be seen from the figure, the expression of different cytokines and different blood markers are different depending on whether hospitalization is required, short-term hospitalization, or long-term hospitalization; for example, the expression of IP-10 in patients who do not need hospitalization tends to be upregulated compared to healthy subjects who are not infected with dengue fever (red part), while the expression of IP-10 in patients who need hospitalization tends to be downregulated compared to healthy subjects who are not infected with dengue fever (blue part).

From the above changes in the expression of cytokines and blood markers, it can be seen that the changes of different cytokines and different blood markers are different when hospitalization is required, short-term hospitalization, and long-term hospitalization. Then, the cytokines and blood markers were further analyzed by ROC analysis, as shown in Table 4 and Table 5. In Table 4 and Table 5, the values in bold indicate that after ROC analysis, the area under the curve (AUC) obtained has clinical significance (p≦0.05), that is, the expression has a significant change, and the value with an upward arrow indicates that the higher the value, the higher the positive rate, and conversely, the value with a downward arrow indicates that the lower the value, the higher the positive rate. In addition, in the table, the comparison between no hospitalization and hospitalization required includes short-term hospitalization and long-term hospitalization; and in the comparison between short-term hospitalization and long-term hospitalization, short-term hospitalization includes no hospitalization and hospitalization ≤ 7 days.

Table 4 Cytokines No hospitalization required vs. hospitalization required Short-term hospitalization (≤7 days) vs long-term hospitalization (＞7 days) GM-CSF 0.562 0.484 IFN-⍺ 0.689 ↓ 0.543 IFN-γ 0.656 ↓ 0.589 IL-1β 0.441 0.623 IL-1R⍺ 0.564 0.474 IL-2 0.432 0.522 IL-4 0.524 0.688 ↑ IL-6 0.703 ↑ 0.746 ↓ IL-7 0.653 ↓ 0.488 IL-8 0.608 0.398 IL-10 0.656 ↑ 0.553 IL-12 0.414 0.594 IL-13 0.496 0.606 IL-15 0.559 0.663 ↓ IL-17 0.621 ↓ 0.589 IL-18 0.793 ↑ 0.586 IP-10 0.48 0.518 MCP-1 0.427 0.39 MIG 0.66 ↑ 0.716 ↓ MIP-1⍺ 0.524 0.477 MIP-1β 0.563 0.43 RANTES 0.731 ↓ 0.581 TNF-⍺ 0.638 ↑ 0.404 VEGF 0.564 0.506

Table 5 Blood markers No hospitalization required vs. hospitalization required Short-term hospitalization (≤7 days) vs long-term hospitalization (＞7 days) White blood cell count (WBC) 0.59 0.606 Hemoglobin concentration (Hb) 0.384↓ 0.423 Hematocrit (Hct) 0.562 0.481 Platelet count 0.767 ↓ 0.426 Aspartate transaminase concentration (AST) 0.59 0.565 Alanine transaminase concentration (GPT) 0.709 ↑ 0.658 ↓ Blood urea nitrogen concentration (BUN) 0.642 ↑ 0.579 Serum creatinine concentration (Cr) 0.623 0.645 ↓ Prothrombin time (PT) 0.628 ↑ 0.546 Partial thromboplastin time (aPTT) 0.586 0.508

As can be seen from Tables 4 and 5, after ROC analysis, among the cytokines, the expression levels of IL-6 and MIG were significantly different under the conditions of hospitalization, short-term hospitalization, and long-term hospitalization; and among the blood markers, in addition to the significant differences in the expression levels of alanine aminotransferase and serum creatinine under the conditions of hospitalization, short-term hospitalization, and long-term hospitalization, their area under the curve (AUC) values were also higher than those of other blood markers; therefore, the data of the two cytokines and two blood markers with significant differences were used as parameters of the subsequent principal component analysis (PCA) algorithm to establish a dengue fever patient hospitalization time evaluation model.

Next, from the data of the two cytokines (IL-6 and MIG) and two blood markers (alanine transaminase and serum creatinine) with significant differences, the two cytokines or the two blood markers were used as parameters of the principal component analysis algorithm, or the two cytokines combined with the two blood markers were used as parameters of the principal component analysis algorithm to establish dengue fever patient hospitalization time evaluation models respectively; the parameter combinations used in the dengue fever patient hospitalization time evaluation models are shown in Table 6.

Table 6 PCA algorithm Parameter Type 2 parameters IL-6, MIG 2 parameters creatinine, GPT 4 parameters creatinine, GPT, IL-6, MIG

Step S3: Obtain the cytokine data and blood marker data to be evaluated through the input device, and use the processor to perform an interpretation process to obtain the dengue fever patient's hospitalization time interpretation result.

Through the above step S2, according to different cytokines or blood markers with significant differential expression as parameters, or the cytokines and blood markers with significant differential expression as parameters (see Table 6), the dengue fever patient hospitalization time evaluation model established by the principal component analysis algorithm is used to evaluate the cytokine data and blood marker data to be evaluated using the model to obtain the dengue fever progression interpretation results, and the results are shown in Table 7; in addition, in the table, the part of no need for hospitalization versus need for hospitalization, the need for hospitalization includes short-term hospitalization and long-term hospitalization; and in the part of short-term hospitalization versus long-term hospitalization, the short-term hospitalization includes no need for hospitalization and hospitalization ≤ 7 days.

Table 7 Prediction Accuracy Evaluation Model No hospitalization required vs. hospitalization required Short-term hospitalization (≤7 days) vs long-term hospitalization (＞7 days) 2 parameters (IL-6, MIG) 71% 71% 2 parameters (creatinine, GPT) 66% 59% 4 parameters (creatinine, GPT, IL-6, MIG) 96% 94%

As can be seen from Table 7, the dengue fever patient hospitalization time assessment model established using IL-6 and MIG as parameters has a prediction accuracy of 71% for whether hospitalization is required, short-term hospitalization, and long-term hospitalization; and the dengue fever patient hospitalization time assessment model established using alanine aminotransferase (GPT) and serum creatinine (creatinine) as parameters has a prediction accuracy of 66% and 59% for whether hospitalization is required, short-term hospitalization, and long-term hospitalization, respectively; and when IL-6, MIG, alanine aminotransferase, and serum creatinine are used as parameters to establish the dengue fever patient hospitalization time assessment model, the prediction accuracy for whether hospitalization is required, short-term hospitalization, and long-term hospitalization is 94% and 94%, respectively, which are significantly improved compared to the prediction accuracy of using only two cytokines or only two blood markers as parameters.

Therefore, when IL-6, MIG, alanine transaminase and serum creatinine are used as parameters to establish a dengue fever patient hospitalization time assessment model, it has an extremely high prediction accuracy of more than 90%.

Step S4: The output device accesses the storage device to output the dengue fever patient's hospitalization time interpretation result.

The dengue fever patient hospitalization time interpretation result obtained in the above step S3 can be further outputted through an output device. The output device disclosed in this embodiment can include various display interfaces, such as a computer screen, a display, or a handheld device display.

Finally, please refer to FIG. 3, which is a schematic diagram of a dengue fever patient hospitalization time evaluation system according to an embodiment of the present invention. As shown in FIG. 3, the dengue fever patient hospitalization time evaluation system 20 may include an input device 21, a storage device 22, a processor 23, and an output device 24.

In this embodiment, the input device 21 can be a flow cytometer for obtaining cytokine data and a hematology analyzer for obtaining blood marker data, both of which are obtained from a patient's blood sample. In another embodiment, the input device 21 is not limited to flow cytometers and hematology analyzers. The input device 21 may include input interfaces of electronic devices such as personal computers, smart phones, servers, etc., including touch screens, keyboards, mice, etc., to transmit cytokine data and blood marker data in file form; or upload historical data to the memory storage in the storage device 22 via wireless network transmission, wireless communication transmission or general wired Internet. The memory may include read-only memory, flash memory, disk or cloud database, etc.

Secondly, the dengue fever progress assessment system 20 accesses the storage device 22 through the processor 23. The processor 23 may include a central processing unit, an image processor, a microprocessor, etc. in a computer or server, and may include a multi-core processing unit or a combination of multiple processing units. The processor 23 executes instructions to access multiple cytokine data and multiple blood marker data of dengue patients in the storage device 22 for training procedures, and accesses the cytokine data and blood marker data to be evaluated for interpretation procedures. Specifically, the training procedure is to perform biostatistical analysis on a plurality of cytokine data and a plurality of blood marker data of dengue patients originally stored in the storage device 22, find out the cytokine data and blood marker data with significant differences, and calculate the cytokine data and blood marker data with significant differences using the principal component analysis (PCA) algorithm to establish a dengue patient hospitalization time assessment model.

Next, the cytokine data and blood marker data to be evaluated are calculated through the established dengue fever patient hospitalization time model through the interpretation program, and are classified according to whether hospitalization is required, short-term hospitalization (≤7 days) and long-term hospitalization (＞7), to obtain the dengue fever patient hospitalization time interpretation result; the output device 24 accesses the storage device 22 to output the dengue fever patient hospitalization time interpretation result, and the output device 24 can include various display interfaces, such as a computer screen, a display or a handheld device display.

Through the above-mentioned dengue fever patient hospitalization time assessment method and system, by combining cytokine data with blood markers known to detect dengue fever, it is possible to quickly and accurately determine whether a dengue fever patient needs hospitalization, or whether short-term or long-term hospitalization is required. The assessment accuracy rate can reach more than 90%, so that subsequent medical staff can choose the corresponding medical decision for the patient as soon as possible. In addition to improving the use of medical resources, it can also effectively provide corresponding treatment for dengue fever patients and increase the survival rate of patients.

The above description is for illustrative purposes only and is not intended to be limiting. Any equivalent modifications or changes made to the invention without departing from the spirit and scope of the invention shall be included in the scope of the attached patent application.

20: Dengue fever patient hospitalization time assessment system 21: Input device 22: Storage device 23: Processor 24: Output device S1~S4: Steps

In order to make the technical features, contents and advantages of the present invention and the effects that can be achieved more obvious, the present invention is described in detail as follows with reference to the accompanying drawings and in the form of embodiments:

Figure 1 is a flow chart of the dengue fever patient hospitalization time assessment method of the present invention; Figure 2A and Figure 2B are heat maps of different cytokine expression levels and different blood marker expression levels of the present invention, respectively, for whether hospitalization is required or short-term hospitalization (≦7 days) and long-term hospitalization (＞7 days); Figure 3 is a schematic diagram of the dengue fever patient hospitalization time assessment system of the present invention.

S1~S4: Steps

Claims (7)

A method for evaluating the hospitalization time of a dengue patient comprises the following steps: Step S1: inputting a plurality of cytokine data and a plurality of blood marker data of a dengue patient through an input device and storing them in a storage device; Step S2: accessing the storage device through a processor, performing a biostatistical analysis on the plurality of cytokine data and the plurality of blood marker data, finding out the cytokine data and the blood marker data with significant differences, and then storing the cytokine data and the blood marker data with significant differences in the storage device; A dengue fever patient hospitalization time evaluation model is established using a principal component analysis (PCA) algorithm for the cytokine data and blood marker data with significant differences; step S3: obtaining the cytokine data and blood marker data to be evaluated through the input device, and performing a judgment process with the processor to obtain a dengue fever patient hospitalization time judgment result; and step S4: accessing the storage device through an output device, and outputting the dengue fever patient hospitalization time judgment result, wherein The plurality of cytokine data are selected from IL-10, IFN-γ, IL-6, TNF-α, IL-8, IP-10, MCP-1, IL-4, MIP-1β, IFN-α, RANTES, GM-CSF, IL-15, VEGF, IL-13, IL-18, IL-1Rα, IL-12, MIP-1α, MIG, IL-17, IL-1β, IL-7 and IL-2; The plurality of blood marker data are selected from prothrombin time, partial thromboplastin time, white blood cell count, hemoglobin concentration, hematocrit, platelet count, aspartate aminotransferase concentration, alanine aminotransferase concentration, blood urea nitrogen concentration and serum creatinine concentration; the hospitalization time of the dengue fever patient is classified according to whether hospitalization is required, short-term hospitalization (≤7 days) and long-term hospitalization (>7); and steps S1 to S4 are all completed by computer software. A method for assessing hospitalization time of dengue fever patients as described in claim 1, wherein the input device comprises a flow cytometer and a hematology analyzer. The method for evaluating the hospitalization time of dengue fever patients as described in claim 1, wherein the biostatistical analysis is performed using the Spearman correlation method. The method for evaluating the hospitalization time of dengue fever patients as described in claim 1, wherein the cytokine data with significant differences are selected from IL-6 and MIG; and the blood marker data with significant differences are selected from alanine aminotransferase and serum creatinine. A dengue fever patient hospitalization time evaluation system comprises: an input device for inputting a plurality of cytokine data and a plurality of blood marker data of a dengue patient, as well as the cytokine data and blood marker data to be evaluated; a storage device connected to the input device for storing the plurality of cytokine data and the plurality of blood marker data, as well as the cytokine data and blood marker data to be evaluated; an output device connected to the storage device for outputting a dengue fever patient hospitalization time evaluation result; and a processor connected to the storage device, executing a plurality of instructions to perform the following steps: performing a biostatistical analysis on a plurality of cytokine data and a plurality of blood marker data of dengue patients to find out cytokine data and blood marker data with significant differences; performing a principal component analysis (PCA) algorithm on the cytokine data and blood marker data with significant differences to establish a dengue patient hospitalization time evaluation model; based on the dengue patient hospitalization time evaluation model, the cytokine data to be evaluated The invention relates to a method for performing an interpretation process on the dengue fever patient's hospitalization time interpretation result by using the blood marker data and the blood marker data; and accessing the storage device through the output device to output the dengue fever patient's hospitalization time interpretation result, wherein the plurality of cytokine data are selected from IL-10, IFN-γ, IL-6, TNF-α, IL-8, IP-10, MCP-1, IL-4, MIP-1β, IFN-α, RANTES, GM-CSF, IL-15, VEGF, IL-13, IL-18, IL -1Rα, IL-12, MIP-1α, MIG, IL-17, IL-1β, IL-7 and IL-2; the plurality of blood marker data are selected from prothrombin time, partial thromboplastin time, white blood cell count, hemoglobin concentration, hematocrit, platelet count, aspartate aminotransferase concentration, alanine aminotransferase concentration, blood urea nitrogen concentration and serum creatinine concentration; and the dengue fever patient's hospitalization time is classified according to whether hospitalization is required, short-term hospitalization (≤7 days) and long-term hospitalization (>7). A dengue fever patient hospitalization time assessment system as described in claim 5, wherein the input device includes a flow cytometer and a hematology analyzer. A dengue fever patient hospitalization time assessment system as described in claim 5, wherein the output device comprises a computer screen, a display or a handheld device display.