KR102487864B1 - Method and system for predicting prognosis in patients Community-Acquired Pneumonia(CAP) - Google Patents

Abstract

A method for quickly and accurately predicting the prognosis of the patient's pneumonia risk and mortality from the blood test results of a patient with community-acquired pneumonia is provided. The method for predicting the prognosis of a pneumonia patient is to predict that the patient is at high risk of pneumonia when the average platelet volume of the patient is relatively large compared to other patients or the platelet count is relatively small compared to other patients according to the blood test results. can

Description

Method and system for predicting prognosis in patients Community-Acquired Pneumonia (CAP)}

The present invention relates to a method for predicting the mortality rate of a patient with community-acquired pneumonia (CAP), and in particular, a method for predicting the prognosis for pneumonia of a patient from platelet information extracted from a patient with community-acquired pneumonia, and the same It is about a predictive system for

Pneumonia is defined as infection of the lung parenchyma. Such pneumonia is largely classified into community-acquired pneumonia (CAP) and hospital-acquired pneumonia (HAP).

Community-acquired pneumonia is mainly caused by Streptococcus pneumoniae, Mycoplasma pneumoniae, and Hemophilus influenzae. In addition, symptoms of the disease include respiratory symptoms such as cough, phlegm, and difficulty breathing, and systemic symptoms such as headache, fatigue, fever, and chills.

Community-acquired pneumonia is the most common cause of death among infectious diseases. For example, in the case of Korea, the death rate from pneumonia was 9.4 per 100,000 population as of 2006, which is the highest among deaths caused by other infectious diseases.

In particular, the mortality rate due to pneumonia is high in the elderly, whose immunity is lower than that of young people, and it is expected that the mortality rate due to pneumonia in the elderly will increase in Korea, which has recently entered a rapidly aging society.

Accordingly, various methods for predicting the prognosis of pneumonia have been proposed in order to reduce mortality and medical costs due to community-acquired pneumonia, and among them, the Pneumonia Severity Index (PSI) is widely used.

The pneumonia severity index calculates scores based on gender, age, comorbidities, medical examination and physical examination results, etc. for patients with community-acquired pneumonia, and classifies them into 5 stages, and for each stage, the prognosis of pneumonia, such as mortality How to predict etc.

However, the conventional method for predicting the prognosis of community-acquired pneumonia requires a large number of predictors and takes a lot of time due to the increased complexity of prediction. In particular, in the case of the conventional pneumonia severity index, the relative There was a problem that the accuracy of prediction was lowered because a large score deviation occurred.

The present invention is to solve the above problems, and a prediction method capable of accurately predicting the prognosis of pneumonia in a short time with a simplified process by predicting the prognosis of the patient's pneumonia from platelet information extracted from patients with community-acquired pneumonia, and It is to provide a prediction system for this.

A method for predicting the prognosis of a patient with community-acquired pneumonia according to an embodiment of the present invention includes extracting platelet information from a blood test result of the patient; Calculating a platelet index from the platelet information; classifying the platelet index into one of a set first class to a fourth class; and predicting a prognosis for pneumonia of a patient corresponding to the platelet index of the class by analyzing each of the classified classes.

Here, the platelet information is characterized in that it includes each data of average platelet volume and platelet count.

A prediction system for predicting the prognosis of a patient with community-acquired pneumonia according to an embodiment of the present invention includes a test unit that performs a blood test on the patient; a calculation unit extracting platelet information of the patient from among blood test results obtained by the test unit and calculating a platelet index; a classification unit that classifies the platelet index into one of preset first to fourth classes; and a prediction unit that analyzes each class and predicts a prognosis of pneumonia for a patient of the corresponding class.

The present invention can predict the prognosis of pneumonia using minimal predictive variables, so it is possible to simply configure a prediction system compared to the prior art, and accordingly, it is possible to predict an accurate pneumonia prognosis for a patient in a short time.

In addition, the present invention can reduce the mortality rate of patients due to pneumonia by providing the medical staff with the result of predicting the prognosis of pneumonia of the patient, thereby enabling customized medical treatment and treatment of the patient by the medical staff, thereby reducing medical costs. .

1 is a diagram showing a prediction system for predicting the prognosis of patients with community-acquired pneumonia according to an embodiment of the present invention.
2 is a diagram showing a method for predicting the prognosis of patients with community-acquired pneumonia according to an embodiment of the present invention.
FIG. 3 is a diagram specifically illustrating the step of predicting the prognosis of FIG. 2 .
4 to 6 are embodiments according to the method for predicting the prognosis of patients with community-acquired pneumonia of the present invention.

Hereinafter, the configuration and operation of embodiments of the present invention will be described with reference to the accompanying drawings.

It should be noted that the same reference numerals and symbols refer to the same components in the drawings as much as possible, even if they are displayed on different drawings. In the following description of the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted. In addition, when a part "includes" a certain component, it means that it may further include other components without excluding other components unless otherwise stated.

In addition, the terms or words used in this specification and claims should not be interpreted in a conventional and dictionary sense, and the inventors can properly define the concept of the term in order to explain their invention in the best way. Based on the principle, it should be interpreted as meaning and concept consistent with the technical idea of the present invention. Therefore, the embodiments described in this specification and the configurations shown in the drawings are only preferred embodiments of the present invention, and do not represent all of the technical spirit of the present invention, so various equivalents that can replace them at the time of the present application. And there may be variations, and the scope of the present invention is not limited to the embodiments described below.

1 is a diagram showing a prediction system for predicting the prognosis of patients with community-acquired pneumonia according to an embodiment of the present invention.

Referring to FIG. 1, the prediction system 100 of this embodiment uses blood test information, for example, platelet information, for patients suspected of being infected with community-acquired pneumonia (CAP, hereinafter referred to as pneumonia) who have visited a hospital to detect pneumonia. It can be predicted by analyzing the prognosis for Here, the prediction of prognosis for pneumonia may include prediction of the risk of pneumonia for the patient, the possibility of death due to pneumonia, and the possibility of hospitalization in an intensive care unit due to pneumonia.

To this end, the prediction system 100 of this embodiment may include an inspection unit 110, a calculation unit 120, a classification unit 130, and a prediction unit 140. In addition, the prediction system 100 includes a first database 150 in which test results of the test unit 110 are stored and results of the calculation unit 120, the classification unit 130, and the prediction unit 140 are stored. 2 The database 160 may be further included.

The examination unit 110 may perform various examinations on patients and store the examination results in the first DB 150 for each patient.

The examination performed by the examination unit 110 may include a medical examination of the patient's age, gender, place of residence, etc., a physical examination such as measurement of the patient's physical index, a patient's blood examination, and a radiological examination such as X-ray.

The calculation unit 120 may extract the patient's platelet information from the blood test results of the patient's test results stored in the first DB 150 .

The platelet information may include data on mean platelet volume (MPV) and platelet count (PC) in blood. Here, the average platelet volume may be extracted in a unit of femto-liter (fL), and the number of platelets may be extracted in a unit of micro-liter (μL).

Also, the calculation unit 120 may calculate the patient's platelet index (MPR) from the extracted platelet information. The platelet index for each patient calculated by the calculation unit 120 may be stored in the second DB 160 .

The platelet index can be calculated as the ratio of mean platelet volume to platelet count. For example, the platelet index may be calculated by dividing the average platelet volume by the number of platelets. Therefore, the platelet index may be calculated as a value proportional to the average platelet volume and inversely proportional to the number of platelets.

The classification unit 130 may classify the patient's platelet index stored in the second DB 160 into a predetermined number of classes. The platelet index class classified by the classification unit 130 may be matched with the pre-stored platelet index for each patient and stored in the second DB 160 .

The classification unit 130 sets four classes, for example, the first class (class 1) to the fourth class (class 4), based on the size range of the total platelet index stored in the second DB 160, and each Patients can be classified into corresponding classes according to the size of their platelet index. The size range of the total platelet index can be determined by the platelet index having the minimum value and the platelet index having the maximum value.

Here, the classification unit 130 may set each class such that each of the first class (class1) to the fourth class (class4) has the same size range within the size range of the total platelet index.

In addition, the platelet index classified into each class by the classification unit 130 may increase as the class increases, that is, from the first class (class1) to the fourth class (class4).

The prediction unit 140 may analyze each patient's platelet index class stored in the second DB 160 and predict the patient's prognosis for pneumonia based on the analysis result. The prediction unit 140 may predict the prognosis of pneumonia for patients of each class through logistic regression analysis for all platelet index classes.

The pneumonia prognosis predicted by the prediction unit 140 may be matched and stored for each patient in at least one of the first DB 150 and the second DB 160 . In addition, the predicted pneumonia prognosis may be transmitted to an external device, for example, a terminal possessed by a medical staff such as a doctor, and may be referred to for patient treatment by them.

As such, the prediction system 100 of the present embodiment can predict prognosis such as risk, survival possibility, and possibility of admission to an intensive care unit based on blood test results, that is, patient's platelet information, among various test results of the patient.

Therefore, the prediction system 100 of the present invention can predict the prognosis of pneumonia using the minimum predictive variables, so that the prediction system can be simply configured compared to the prior art, and accordingly, accurate pneumonia prognosis for the patient in a short time can predict

In addition, the prediction system 100 of the present invention can reduce the mortality rate of patients due to pneumonia by providing the medical staff with the result of predicting the prognosis of pneumonia of the patient, enabling the medical staff to provide customized medical treatment and treatment of the patient. You can cut costs.

2 is a diagram showing a method for predicting the prognosis of patients with community-acquired pneumonia according to an embodiment of the present invention, and FIG. 3 is a diagram showing the step of predicting the prognosis in FIG. 2 in detail. 4 to 6 are examples of the method for predicting the prognosis of patients with community-acquired pneumonia of the present invention.

Hereinafter, for convenience of description, the prognosis prediction method of the present embodiment will be described in detail with reference to the previously described prediction system 100 of FIG. 1 .

Referring to FIG. 2 , first, the test unit 110 of the prediction system 100 may perform various tests on a patient suspected of being infected with pneumonia. Also, the calculation unit 120 may extract platelet information about the patient from a blood test among tests performed by the test unit 110 (S10).

The examination of the patient may include a medical examination of the patient's age, gender, place of residence, etc., a physical examination of the patient such as measuring a physical index of the patient, a blood test of the patient, and a radiological examination such as an X-ray. In addition, the platelet information may include data on the average platelet volume in blood and data on the number of platelets.

Next, the calculation unit 120 may calculate the platelet index from the extracted platelet information of the patient (S20). The calculated platelet index may be matched for each patient and stored.

The calculation unit 120 may calculate the platelet index according to the ratio of the average platelet volume and the number of platelets. For example, the platelet index may be calculated according to a value obtained by dividing the average platelet volume by the number of platelets. Accordingly, the platelet index may be calculated as a value proportional to the patient's average platelet volume and inversely proportional to the patient's platelet count.

Subsequently, the classification unit 130 may classify the calculated platelet index of the patient into one of a first class (class1) to a fourth class (class4) (S30).

The classification unit 130 divides the platelet index of all patients stored in the second DB 160 into four equal parts, such as the size range according to the minimum value and the maximum value, to obtain the first class (class1) to the fourth class (class4). ) can be set.

Also, the classification unit 130 may classify the platelet index into one of a first class (class1) to a fourth class (class4) based on the size of the platelet index of each patient. In this case, the size of the platelet index classified into the corresponding class may relatively increase as it goes from the first class (class1) to the fourth class (class4).

In addition, the platelet index class classified by the classification unit 130 may be matched to the platelet index and stored in the second DB 160 for each patient.

Next, the prediction unit 140 can predict the prognosis of pneumonia patients through analysis of the platelet index for each class (S40).

The prediction unit 140 may predict the prognosis of the pneumonia patient's risk, survival possibility, and intensive care unit admission possibility through class-specific analysis. In addition, the prediction unit 140 may store prediction results for each patient or transmit them to an external medical staff.

Referring to FIG. 3 , the prediction unit 140 may calculate an odds ratio for each platelet index class (S110).

The odds ratio may be expressed as a ratio value between all platelet index classes. In this embodiment, for convenience, the calculation of the ratio of the remaining classes based on the first class (class 1) will be described as an example.

Referring to FIG. 4 in more detail, first, the odds ratio for the first class (class 1) can be defined as 1. Then, the odds ratio for the second class (class 2) may be calculated by dividing the odds ratio of the first class (class 1) by the death rate (mortality rate/survival rate) value for patients in the second class (class 2). At this time, in an unadjusted state of the odds ratio, the odds ratio for the second class (class2) may be calculated as 1.16.

In addition, the odds ratio for the third class (class 3) can be calculated by dividing the odds ratio for the first class (class 1) by the death rate value for the third class (class 3) patients. At this time, in an unadjusted state of the odds ratio, the odds ratio for the third class (class3) may be calculated as 1.48.

In addition, the odds ratio for the fourth class (class 4) may be calculated by dividing the odds ratio of the first class (class 1) by the death ratio value for the fourth class (class 4) patients. At this time, in an unadjusted state of the odds ratio, the odds ratio for the fourth class (class4) may be calculated as 2.93.

Next, the prediction unit 140 may adjust the pre-calculated odds ratio for each class according to the first variable (S120). Here, the prediction unit 140 may adjust the odds ratio for each class one or more times according to the first variable.

The first variable may be a value corresponding to the patient's examination information stored in the first DB 150, for example, the patient's age or medical history. For example, the first variable may be at least one of values corresponding to the patient's age and previous medical history, such as high blood pressure. In addition, the first variable may be at least one of values corresponding to the patient's leukocyte index (WBC), delta neutrophil index (DNI), erythrocyte sedimentation index (ESR), and procalcitonin.

Therefore, as shown in FIG. 4, when the odds ratio is adjusted by the first variable, the odds ratio of the second class (class2) is adjusted to 1.20, and the odds ratio of the third class (class3) is calculated. It is calculated by adjusting to 1.49, and the odds ratio of the fourth class (class4) can be calculated by adjusting to 3.06.

At this time, the adjusted odds ratios of the second class (class2) and the third class (class3) have no significant difference compared to the unadjusted odds ratios, but the adjusted odds ratios of the fourth class (class4) are unadjusted odds ratios. Compared to the ratio, it can be calculated with a large increase.

Meanwhile, p shown in FIG. 4 means statistical significance, and is a value indicating statistical significance when p<0.05.

Continuing, the prediction unit 140 may predict the patient's pneumonia prognosis, that is, the prognosis according to the possibility of survival and mortality due to pneumonia, based on the pre-calculated odds ratio for each class (S130).

As described above, the odds ratio of each of the second to fourth classes may be a value indicating that the risk of death due to pneumonia is higher than that of the control class, that is, the first class. Further, the odds ratio of each class may be calculated as a relatively large value as the class increases, that is, as the size of the platelet index increases.

Thus, the prediction unit 140 may analyze the risk of death of patients due to pneumonia based on the odds ratio for each class, and predict the pneumonia prognosis for each patient accordingly.

More specifically, based on the unadjusted odds ratio shown in FIG. 4, the prediction unit 140 determines that the risk of death of patients classified as the second class (class2) is classified as the first class (class1). Compared to patients, it can be analyzed to be approximately 1.16 times higher. In addition, it can be analyzed that the risk of death of patients classified as the third class (class3) is approximately 1.48 times higher than that of patients classified as the first class (class1). In addition, it can be analyzed that the risk of death of patients classified as the fourth class (class4) is approximately 2.93 times higher than that of patients classified as the first class (class1).

In addition, the prediction unit 140 analyzes the risk of death of patients classified in the second class (class2) as approximately 1.20 times higher than that of the first class (class1) based on the adjusted odds ratio, and the third class (class3). The risk of death of patients classified as ) is analyzed to be approximately 1.49 times higher, and the risk of death of patients classified as class 4 is analyzed to be approximately 3.06 times higher.

Accordingly, the prediction unit 140 may predict the survival rate and death rate due to pneumonia of patients for each class based on the analyzed odds ratio for each class. In addition, the cumulative short-term survival rate for each class for all patients may be output according to the prediction result. Here, the short-term survival rate may mean the 60-day survival rate of pneumonia patients.

That is, as shown in FIG. 5, the prediction unit 140 has a survival rate of approximately 85.1% and a mortality rate of approximately 85.1% of patients classified into the first class (class1) to the third class (class3) among the total of 150 patients. It can be predicted at 14.9%. However, it can be predicted that the survival rate is approximately 62.5% and the mortality rate is approximately 37.5% for patients classified as class 4 (class 4). Mortality rates can be expected to be high.

In addition, as shown in FIG. 6, the prediction unit 140 outputs the cumulative short-term survival rate for patients of the first class (class1) as approximately 0.89, and the cumulative short-term survival rate for patients of the second class (class2) It is output as approximately 0.87, and the cumulative short-term survival rate for patients in the third class (class3) can be output as approximately 0.84. However, since the cumulative short-term survival rate for patients in the fourth class (class4) is output as approximately 0.76, it can be seen that the survival rate of patients classified as the fourth class (class4) is lower than that of patients in other classes.

As described above, the method for predicting the prognosis of pneumonia of the present invention classifies the patient according to the size of the platelet index and predicts the pneumonia prognosis of the patient based on the result of analyzing the risk of pneumonia for each class.

At this time, the risk of pneumonia in patients whose platelet index is calculated as a relatively large value, that is, patients classified as class 4, is analyzed higher than that of patients in other classes, and therefore, pneumonia in patients in class 4 (class 4) It can be predicted that the mortality rate due to this is greater than that of patients in other classes. As shown in FIG. 5, the prediction method of the present invention shows similar results to conventional pneumonia severity prediction (PSI).

That is, the method for predicting the prognosis of pneumonia of the present invention can predict the risk of pneumonia and the resulting mortality according to the size of the platelet index calculated according to the patient's blood test information.

In this case, the platelet index is calculated by the ratio of the patient's average platelet volume to the number of platelets, and it can be predicted that the higher the platelet index, the higher the risk of pneumonia. From this, the present invention is a patient with a high risk of pneumonia when the average platelet volume is relatively large compared to other patients or the platelet number is relatively small compared to other patients in a blood test for a patient suspected of having pneumonia infection. can be predicted as

Therefore, compared to the prior art, the method for predicting the prognosis of pneumonia of the present invention can show high predictive accuracy for the prognosis of pneumonia even when using the least predictive variables, that is, the patient's blood test results, compared to the prior art. With this configuration, it is possible to predict the patient's accurate pneumonia prognosis in a short time, and through this, optimal pneumonia treatment and treatment can be provided to the patient.

100: prediction system 110: inspection unit
120: calculation unit 130: sorting unit
140: prediction unit 150: first database
160: second database

Claims (12)

In the prognosis prediction method using a prediction system for predicting the prognosis of patients with community-acquired pneumonia,
extracting platelet information from the patient's blood test results;
Calculating a platelet index from the platelet information;
classifying the platelet index into one of a set first class to a fourth class; and
Analyzing each of the classified classes and predicting a prognosis for pneumonia of a patient corresponding to the platelet index of the corresponding class,
The platelet information includes data of average platelet volume and platelet count, respectively;
The platelet index is calculated as a value proportional to the average platelet volume and inversely proportional to the number of platelets;
The step of predicting the prognosis for pneumonia is a method for predicting the prognosis of patients with community-acquired pneumonia, characterized in that for predicting a 60-day survival rate.
delete delete According to claim 1,
The step of classifying the platelet index,
setting the platelet index size range of all patients to the first to fourth classes having the same range, respectively; and
Classifying into one of the first to fourth classes according to the size of the platelet index for each patient,
A method for predicting the prognosis of patients with community-acquired pneumonia, characterized in that the size of the platelet index classified from the first class to the fourth class is relatively increased. According to claim 1,
The step of predicting the prognosis for pneumonia of the patient,
Calculating an odds ratio of each of the first to fourth classes;
adjusting the odds ratio based on a first variable; and
Predicting the prognosis by analyzing the risk of death from pneumonia for each of the first to fourth classes of patients according to the adjusted odds ratio;
The prognosis is a method for predicting the prognosis of patients with community-acquired pneumonia, characterized in that it includes the survival rate and mortality of patients due to pneumonia. According to claim 5,
In the step of calculating the odds ratio,
defining an odds ratio of the first class as 1; and
Calculating the odds ratio of each of the second to fourth classes by dividing the odds ratio of the first class by the death ratio value for each patient of the second to fourth classes. A method for predicting the prognosis of patients with acquired pneumonia. According to claim 6,
The method for predicting the prognosis of patients with community-acquired pneumonia, characterized in that the odds ratio has a relatively large value from the first class to the fourth class. According to claim 5,
The first variable is
A community characterized by at least one of the patient's age, a value corresponding to a previous medical history, a value corresponding to a leukocyte index (WBC), a delta neutrophil index (DNI), an erythrocyte sedimentation index (ESR), and a value corresponding to procalcitonin A method for predicting the prognosis of patients with acquired pneumonia. A prediction system for predicting the prognosis of patients with community-acquired pneumonia according to any one of claims 1 and 4 to 8,
The prediction system,
a test unit that performs a blood test on the patient;
a calculation unit extracting platelet information of the patient from among blood test results obtained by the test unit and calculating a platelet index;
a classification unit that classifies the platelet index into one of preset first to fourth classes; and
Includes a prediction unit that analyzes each class and predicts the prognosis of pneumonia for patients in that class;
The platelet information includes data of average platelet volume and platelet count, respectively;
The platelet index is calculated as a value proportional to the average platelet volume and inversely proportional to the number of platelets;
The prediction system, characterized in that the step of predicting the prognosis for pneumonia predicts a 60-day survival rate.
delete According to claim 9,
Further comprising a database for storing the platelet index of all patients,
The prediction system, characterized in that the first to fourth classes are set to have the same range according to the size range of the platelet index of all the patients. According to claim 9,
The prediction unit,
The odds ratio of each of the first to fourth classes is calculated to predict the prognosis of pneumonia for patients in the corresponding class,
The prediction system, characterized in that the odds ratio is adjusted according to the first variable.

Images