KR20210055314A - Method and system for selecting new drug repositioning candidate - Google Patents

Abstract

According to an embodiment, a new drug re-creation candidate selecting method comprises the following steps of: obtaining test index values before and after exposure to a predetermined drug for each of patients; calculating a change amount of the test index values before and after exposure to the predetermined drug for each of obtained test indexes; calculating the total number of exposed drugs including the predetermined drug for each of the patients; correcting the calculated change amount for each of the patients using the calculated total number of drugs; and determining an effect other than a preset effect of the predetermined drug based on the corrected change amount. The embodiment selects a new drug re-creation candidate by using a large amount of information to increase selection reliability of the new drug re-creation candidate.

Description

Method and system for selecting candidates for new drug re-creation {METHOD AND SYSTEM FOR SELECTING NEW DRUG REPOSITIONING CANDIDATE}

The embodiment relates to a method and system for selecting a candidate for re-creation of a new drug, and more particularly, to a method and system for selecting a candidate for re-creation of a new drug for discovering the re-creation of a drug and evaluating its potential use.

Recently, multinational pharmaceutical companies are facing a big crisis due to deteriorating profitability due to expiration of patent term and increase in development cost of new drugs. Despite the accumulation of various new drug development technologies and large amounts of data in the field of biotechnology, the number of new drug licenses is declining.

This is a result of the increased cost required for discovery of new drug substances and the increased safety regulations. In order to overcome this crisis, a plan for developing new drugs with low cost and high efficiency is required. It is attracting attention as a method.

Drug repositioning is a method of finding new drug efficacy, that is, a new medical use (indication), by re-evaluating drugs already used for a specific purpose by re-evaluating drugs already used for a specific purpose by acknowledging their efficacy and safety for a specific disease through clinical trials. It can be said that it is a means of creating a new drug with a high probability of success as it can be expected to reduce costs and shorten the development period by using drugs that have been proven to be safe to some extent.

The demand for new drugs is increasing due to the aging population and chronic and intractable diseases. For example, cardiovascular disease accounts for 30% of deaths worldwide and is well known as a risk factor for high blood pressure, diabetes, and dyslipidemia. Many drugs have been developed and used clinically to properly manage major chronic diseases such as. In order to sufficiently overcome these chronic diseases or numerous intractable diseases, it is urgent to discover new drugs that can continuously replace or supplement existing drugs, and in order to meet these demands, a means to efficiently create new drugs within a short period of time. This is the time to develop.

However, in order to re-create new drugs through reevaluation of drugs already in use, a lot of information on drugs is required, and for this purpose, development of a method to obtain a large amount of useful information by effectively analyzing large-scale medical big data. This is a necessary situation.

In order to solve the above-described problem, the embodiment relates to a method and system for selecting a new drug re-creation candidate for selecting a new drug re-creation candidate with high reliability.

In addition, the embodiment relates to a method and system for selecting a candidate for re-creation of a new drug for providing an algorithm for analyzing a large amount of information.

The method of selecting a candidate for re-creation of a new drug according to an embodiment includes the steps of acquiring test index values before and after exposure to a predetermined drug for each of a plurality of patients; Calculating, calculating the total number of exposed drugs including the predetermined drug for each of the plurality of patients, and correcting the calculated amount of change for each of the plurality of patients using the calculated total number of drugs. And determining an effect other than a predetermined effect of the predetermined drug based on the corrected amount of change.

The test index values before and after the exposure may be obtained using blood data obtained before administration of the predetermined drug for each of the plurality of patients and blood data obtained after administration of the predetermined drug.

The test index may include low-density lipoprotein cholesterol levels, high-density lipoprotein cholesterol levels, triglyceride levels, and hemoglobin levels.

Correcting the amount of change may include dividing the calculated amount of change for each of the plurality of patients by the calculated total number of drugs.

Determining an effect other than the predetermined effect includes calculating an average of the corrected change amount to calculate an average amount of change for each test index, and the difference between a predetermined ideal change amount for each test index and the average amount of change. Based on the test index, it may include determining the effect of the predetermined drug.

In determining the effect of the predetermined drug, a distance may be calculated based on a difference between the ideal change amount and the average change amount, and may be determined in consideration of the direction of the calculated distance.

If the direction of the calculated distance is the same as the ideal change amount of the drug, a first weight is given, and if the direction of the calculated distance is different from the ideal change amount of the drug, a second weight greater than the first weight is applied. When a weight is assigned and the direction of the calculated distance is not known, a third weight between the first weight and the second weight may be assigned.

As for the effect of the predetermined drug, a value having the lowest value to which the first to third weights are applied may be ranked as the highest priority.

In addition, the new drug re-creation candidate selection system of the embodiment includes an acquisition unit for obtaining test index values before and after exposure to a predetermined drug for each patient, and the test index for each of the extracted test indexes before and after exposure to the predetermined drug. The plurality of patients using a change amount calculation unit that calculates the amount of change in value, a drug selection unit that calculates the total number of exposed drugs including the predetermined drug for each of the plurality of patients, and the total number of the selected drugs It may include a correction unit for correcting the calculated amount of change for each, and an effect determination unit for determining an effect other than a predetermined effect of the predetermined drug based on the corrected amount of change.

The test index values before and after the exposure may be obtained using blood data obtained before administration of the predetermined drug for each of the plurality of patients and blood data obtained after administration of the predetermined drug.

The test index may include low-density lipoprotein cholesterol levels, high-density lipoprotein cholesterol levels, triglyceride levels, and hemoglobin levels.

The correction unit may correct the calculated amount of change for each of the plurality of patients by dividing the calculated amount of change for each of the plurality of patients by the calculated total number of drugs.

The effect determination unit calculates an average of the corrected change amount for each of the plurality of patients, calculates an average change amount for each test index, and based on a difference between a predetermined ideal change amount for each test index and the average change amount. , It is possible to determine the effect of the predetermined drug for each test index.

The effect of the predetermined drug may be determined by calculating a distance based on a difference between the ideal change amount and the average change amount, and taking into account the direction of the calculated distance.

If the direction of the calculated distance is the same as the ideal change amount of the drug, a first weight is given, and if the direction of the calculated distance is different from the ideal change amount of the drug, a second weight greater than the first weight is applied. When a weight is assigned and the direction of the calculated distance is not known, a third weight between the first weight and the second weight may be assigned.

As for the effect of the predetermined drug, a value having the lowest value to which the first to third weights are applied may be ranked as the highest priority.

The embodiment selects candidates for re-creation of new drugs using a large amount of information, thereby improving the reliability of selection of candidates for re-creation of Shinyang.

In addition, the embodiment provides an analysis algorithm capable of using a large amount of information, thereby effectively selecting candidates for re-creation of new materials.

1 is a flowchart illustrating a method of selecting a candidate for re-creation of a new drug according to an embodiment.
2 to 6 are diagrams for explaining a method of selecting a candidate for re-creation of a new drug according to an embodiment.
7 is a flowchart illustrating a system for selecting candidates for re-creation of new drugs according to an embodiment.
8 is a diagram showing a sequence of drugs selected by a new drug re-creation candidate selection system according to an embodiment.

Hereinafter, embodiments will be described in detail with reference to the drawings.

1 is a flowchart illustrating a method of selecting a candidate for re-creation of a new drug according to an exemplary embodiment, and FIGS. 2 to 6 are diagrams for explaining a method of selecting a candidate for re-creation of a new drug according to an exemplary embodiment.

Referring to FIG. 1, in the method for selecting a candidate for re-creation of a new drug according to an embodiment, the step of obtaining a clinical index value before and after exposure to a predetermined drug for each patient within a predetermined period (S110), and the obtained clinical index Calculating the amount of change in the clinical index value before and after exposure to the predetermined drug for each (S120), and calculating the total number of exposed drugs including the predetermined drug for each of the plurality of patients (S130), Compensating the calculated amount of change for each of the plurality of patients using the calculated total number of drugs (S140), and effects other than the clinical effect and the predetermined effect of the predetermined drug based on the corrected amount of change. It may include a determining step S150 and a step S160 of combining various clinical effects of each drug to obtain a ranking with high clinical usefulness and the possibility of re-creation of a new drug. Here, the method of selecting a candidate for re-creation of a new drug may be performed in a system for selecting a candidate for re-creation of a new drug.

First, a step (S110) of acquiring clinical index values before and after exposure to a predetermined drug for each patient may be performed.

Clinical index values before and after exposure to a predetermined drug for each patient may be collected from the server. The server is a separate database that generally contains electronic medical record (EMR) data, and may include, for example, a clinical data warehouse (CDW) or common data mode (CDM) server. I can.

As an example, as shown in FIG. 2, the server 200 includes serial number (delimiter) for each patient, prescription information, test information (cholesterol level, glycated hemoglobin level, lung capacity, etc.), diagnostic information (diagnosis name, procedure, prescription). And the like may be included. The server 200 stores drug prescription history for each patient, laboratory tests, and execution dates and results for each test by combining them in the server 200. In addition to the server 200, a patient-specific visit (outpatient, hospitalization, emergency room) history and diagnosis, surgery/procedure history, and the like may be further included.

In the server 200, since the prescription history and laboratory test results for each patient with a prescription history are recorded, it is possible to collect data with high reliability.

Here, the clinical index value may include test results such as cholesterol level, glycated hemoglobin (HbAlc) level, lung capacity, etc., various clinical data items such as diagnosis name, procedure, prescription, or operational definition through a combination thereof.

The scope and criteria of the period of not being exposed to the drug and the period of exposure (the period affected by the drug) are assigned differently depending on the drug or clinical indicator, and the process of acquiring clinical data before and after exposure according to this criterion is included.

When data acquisition is complete, a step (S120) of calculating a change amount of the clinical index value before and after exposure to the predetermined drug for each acquired clinical index may be performed.

In the step S120 of estimating the amount of change, the amount of change in the value of the clinical index before and after exposure to a predetermined drug for each extracted clinical index may be calculated.

In order to calculate the amount of change in multiple clinical data in the pre-exposure section or post-exposure section, a representative value (eg, mean, median, frequency, logical value [presence/absence, etc.]) is obtained for each section and the difference (eg, change amount). , Ratio, logical value, etc.).

In the examples, after dosing in the same patient, it is to identify candidate drugs that have a significant impact on selected laboratory outcomes.

Therefore, the data obtained from the server 200 were exposed to a specific drug for each patient at least once, and the data of the patient who received the laboratory test result before and after the administration was used for research evaluation. As the data (exposed group), the test results performed during the use period of the drug were used, and the data at the time point when not exposed to the drug (non-exposed group) was used for comparative evaluation. Here, the threshold value was set as the threshold value for a period within one week from the start and end of drug use. The test results within 1 week after the start of drug use or 1 week after the drug withdrawal (stop) were set as the period of no exposure. Of course, the threshold value is not limited thereto, and may be changed to various reference values to reflect the characteristics of the target drug or the characteristics of clinical indicators. The maximum interval used to compare the unexposed time point and the exposed time point for an individual patient is determined within a separate threshold range, where the threshold value is set to a period within one year. This threshold value is not limited thereto, and may be changed according to various criteria.

As shown in FIG. 3, the exposed group (G1) used for the study evaluation and the non-exposed group (G2) used for the comparative evaluation may be paired. In FIG. 3, as an example, two pairing groups are shown. The left group may be referred to as a first pairing group P1, and the right group may be referred to as a second pairing group P2. The first pairing group P1 shows a state in which the exposed group G1 and the non-exposed group G2 are paired.

Returning to FIG. 1, a step (S120) of calculating the total number of exposed drugs including the predetermined drug for each patient may be performed.

Since a drug prescription for a patient includes several drugs at each time point, the cause of the drug's effect on the test results at each test time point may be ambiguous. In other words, a patient exposed to a specific drug may be simultaneously exposed to various drugs of different types for each time point. This can obscure the cause of the effect on the drug.

In order to adjust this ambiguity, a step (S130) of calculating the total number of exposed drugs may be performed. This can calculate the total number of drugs exposed for each time point by using data obtained from the server 200.

Subsequently, a step S140 of correcting the calculated amount of change for each of the plurality of patients by dividing by using the calculated total number of drugs or reflecting individual characteristic information (weight) of the drugs exposed at the same time may be performed.

As shown in FIG. 3, when looking at the second pairing group P2, assuming that the calculated number of drugs is two, the calculated amount of change may be corrected by grouping them and dividing them by the number of two drugs. Of course, when the number of prescribed drugs is n, it can be adjusted to reflect ambiguity by dividing the amount of change of the drug by n.

Subsequently, the step (S150) of estimating the effect of the drug on the paired group may be performed.

As shown in FIG. 4, the effect of the drug can be estimated through statistical analysis. For example, the effect of drugs can be estimated using the paired-sample T-test and Wilcoxon's signed rank test. The statistical analysis method is not limited thereto, and as an example, other statistical analysis methods may be used when the measured value is not a number. The paired-sample T-test, Wilcoxon's code ranking test, is one of the widely used statistical analysis methods to evaluate the significance of treatment effects by comparing the size of measurements before and after treatment.

It was set to allow an error of 5% significance level (P<0.05) by the matched sample T-test and Wilcoxon's coded ranking test. This is a standard that is commonly applied in statistical tests, and is a rather strict criterion to limit the likelihood of accidentally being assumed to be false positives, but if the test is repeated several times at the same time (multiple comparison, multiple tests), that is, 10,000 times. When the test was conducted, 500 of these means that it can be false positive even though it has no meaning.

Likewise, since there are multiple drugs in the analysis, multiple comparison problems can arise for selecting a meaningful drug among multiple candidates.

To reduce this false-positive error, the error was adjusted by applying Bonferroni and FDR (False Discovery Rate) calculation methods.

As shown in FIG. 5, the number of important drugs for glycated hemoglobin (HbA1c) was 265, and after adjustment, it could be reduced to 59 and 163 by Bonferroni and FDR calculation methods.

Through this, it is possible to determine whether a certain drug has a significant effect.

Subsequently, the step (S150) of selecting an effect other than a predetermined effect of a predetermined drug, that is, a candidate group for re-creation of a new drug, may be performed based on the amount of change corrected for drugs having a significant effect on a specific clinical index.

In the step of determining an effect other than the predetermined effect (S150), a drug having an effect of improving the clinical index is selected from among drugs having a significant effect on a specific clinical index, and then the corresponding clinical index is improved. It can be carried out in a way that excludes the prescribed drugs.

Determining the effect of improving a specific clinical indicator can be determined by comparing the corrected change amount for each drug, that is, the estimated effect with the ideal change amount, and whether the direction of the effect coincides.

Drugs with a known effect (i.e. use) for a specific clinical indicator can be identified through drug classification systems (e.g., Anatomical Therapeutic Chemical Classification) or country-specific drug approval information (e.g., U.S. Food and Drug Administration).

Subsequently, a step (S160) of prioritizing important drugs among the candidate groups for new drug re-creation selected as having significant effects other than the predetermined effects may be performed.

The step of prioritizing important drugs (S160) can be performed in two ways: a method for ranking individual clinical indicators and a complex advantageous ranking for a plurality of clinical indicators.

In order to prioritize drugs for individual clinical indicators, the ideal effect ranking of each drug for a specific clinical indicator can be determined based on the magnitude and direction of the effect of each drug's corrected variance (positive or negative).

For example, low-density lipoprotein (LDL) cholesterol must have a negative direction (<0) to be effective, and high-density lipoprotein (HDL) cholesterol must have a positive direction (>0) to be effective. Fat (TG) must have a negative direction <0) to be effective, and glycated hemoglobin (HbA1c) must have a negative direction (<0) to be effective. Based on these criteria, prioritization is given as the size of the effect increases along with the direction of the ideal effect (positive or negative), so that the drug with a higher possibility of re-creation is given priority to a specific clinical indicator among multiple candidate groups for new drug re-creation. It helps to make a choice.

In order to rank advantageously for multiple clinical indicators, it is possible to calculate the distance between the effect of the ideal drug and the effect of individual drugs for each clinical indicator.

As an example, each clinical indicator can be weighted. For example, if the calculated distance direction is the same as the abnormal change amount, a first weight is given, and if the calculated distance direction is different from the abnormal change amount, a second weight greater than the first weight is given, and the calculated distance direction is known. If not, a third weight between the first weight and the second weight may be assigned.

As shown in FIG. 6, drug A has a negative effect direction for low-density lipoprotein (LDL) cholesterol, a positive direction for high-density lipoprotein (HDL) cholesterol, and a negative direction for triglyceride (TG). And has a negative direction for glycated hemoglobin (HbA1c).

If calculated accordingly, drug A can have a total value of 0 as (0,0,0,0). Drug B may have a total value of 2 as (0,1,0,1). Drug C may have a total value of 2 as (0,0,0,2). From this, as for the effect of the drug, a value having the lowest value to which the first to third weights are applied can be ranked as a higher priority.

That is, drug A may be measured as a first priority, and drug B and drug C have the same value, and thus may be measured as a second priority. This means that Drug A has more ideal effects on multiple clinical indicators at the same time compared to Drug B or C, so it may be more advantageous for actual clinical use, and thus, the possibility of re-creation of new drugs is more likely. It can be evaluated as high.

7 is a block diagram showing a system for selecting a candidate for re-creation of a new drug according to an embodiment.

Referring to FIG. 7, the system 100 for selecting a candidate for re-creation of a new drug according to an embodiment includes an acquisition unit 110 for acquiring values of clinical indicators before and after exposure to a predetermined drug for each patient, and for each of the extracted clinical indicators. A change amount calculation unit 120 that calculates a change amount of the value of the clinical index before and after exposure to the prescribed drug, and a drug calculation unit that calculates the total number of exposed drugs including the prescribed drug for each of the plurality of patients ( 130), and a correction unit 140 for correcting the calculated amount of change for each of the plurality of patients using the total number of the selected drugs, and effects other than the predetermined effect of the predetermined drug based on the corrected amount of change. The effect determination unit 150 may include a ranking determination unit 160 that combines various clinical effects of each drug and obtains a ranking with high clinical usefulness and the possibility of re-creation of a new drug.

The acquisition unit 110 may collect patient, disease, and laboratory test result information using data existing in the server 200. Here, the server 200 is generally a separate database containing electronic medical record (EMR) data, for example, a clinical data warehouse (CDW) or a common data model (common data mode, CDM). ) May include a server. The clinical index value may include test results such as cholesterol level, HbAlc level, lung capacity, etc., as well as various clinical data items such as diagnosis name, procedure, prescription, or operational definition through a combination thereof.

The change amount calculation unit 120 may calculate a change amount of the value of the clinical index before and after exposure to a predetermined drug for each extracted clinical index.

In the examples, after dosing in the same patient, it is to identify candidate drugs that have a significant impact on selected laboratory outcomes. To this end, the exposed group and the non-exposed group may be paired, and grouping may be performed for a plurality of drugs among the exposed groups.

The drug calculation unit 130 may calculate the total number of a plurality of drugs exposed for each patient. The drug calculation unit 130 may calculate the total number of drugs using data provided from the server 200.

The correction unit 140 may divide by using the total number of drugs (ie, give the same weight) or correct the amount of change calculated for each of the plurality of patients by reflecting individual characteristic information (weight) of the drugs exposed at the same time.

The effect determination unit 150 may estimate the effect and significance of the drug through a statistical analysis method. Determining the effect of improving a specific clinical indicator can be determined by comparing the corrected change amount for each drug, that is, the estimated effect with the ideal change amount, and whether the direction of the effect coincides.

If the calculated distance direction is the same as the abnormal change amount, a first weight is given, and if the calculated distance direction is different from the abnormal change amount, a second weight greater than the first weight is given, and the calculated distance direction is unknown. In this case, a third weight between the first weight and the second weight is assigned, and a value having the lowest value to which the first to third weights are applied is assigned a higher priority.

The priority determination unit 160 may perform a step of prioritizing important drugs among the candidate group for new drug re-creation by combining various clinical effects of each drug. As a method of prioritizing important drugs, there are two methods: ranking individual clinical indicators and ranking multiple clinical indicators in combination.

In order to prioritize drugs for individual clinical indicators, the ideal effect ranking of each drug for a specific clinical indicator can be determined based on the magnitude and direction of the effect of each drug's corrected variance (positive or negative). In order to rank advantageously for multiple clinical indicators, it is possible to calculate the distance between the effect of the ideal drug and the effect of individual drugs for each clinical indicator.

8 is a diagram showing a sequence of drugs selected by a new drug re-creation candidate selection system according to an embodiment.

In the example, a candidate drug search algorithm was proposed based on new clinical big data in order to identify candidate drugs that have a significant influence on the laboratory results selected after administration in the same patient.

Through this research methodology, it is possible to establish rapid hypotheses and test initial concepts in the pre-experimental stage using a rapid and efficient approach through various clinical big data on various diseases. Accordingly, it has become possible to search for new candidate drug targets for various chronic diseases such as endocrine metabolic diseases and cardiovascular diseases.

Patients with dyslipidemia are judged by the test results of total cholesterol, low density lipoprotein (LDL) cholesterol, high density lipoprotein (HDL) cholesterol and triglyceride (TG), and diabetes is judged by the test results of glycated hemoglobin (HbA1c).

In Examples, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, triglyceride (TG), and glycated hemoglobin (HbA1c) were selected as clinical index values. We identified candidates for the creation of new drugs that have a significant effect on the test results of these clinical indicators.

The method and system for selecting a candidate for re-creation of a new drug according to an embodiment has an effect of achieving re-creation of a new drug according to the selection of various clinical indicators.

Although described above with reference to the drawings and embodiments, those skilled in the art will understand that the embodiments can be variously modified and changed without departing from the technical spirit of the embodiments described in the following claims. I will be able to.

100: New drug re-creation candidate selection system
110: acquisition unit
120: Change amount calculation government
130: drug calculation
140: correction unit
150: effect determination unit
160: rank determination unit

Claims (16)

Acquiring clinical index values before and after exposure to a predetermined drug for each patient within a predetermined period;
Calculating a change amount of the clinical index value before and after exposure to the predetermined drug for each of the acquired clinical indexes;
Calculating the total number of exposed drugs including the predetermined drug for each of the plurality of patients;
Correcting the calculated amount of change for each of the plurality of patients using the calculated total number of drugs or weights for each drug;
Determining an effect other than a predetermined effect of the predetermined drug based on the corrected amount of change; And
A method for selecting a candidate for re-creation of a new drug comprising the step of obtaining a ranking with a high probability of creating a new drug by combining the effects of each of the drugs. The method of claim 1,
The clinical index values before and after the exposure are,
A method of selecting a candidate for re-creation of a new drug obtained by using clinical data obtained before administration of the predetermined drug for each of the plurality of patients and clinical data obtained after administration of the predetermined drug. The method of claim 2,
The clinical indicators are cholesterol levels, glycated hemoglobin levels, test information including lung capacity or diagnostic information including diagnosis name, procedure, prescription, or a new drug re-creation candidate selection method including a combination thereof. The method of claim 1,
The step of correcting the amount of change comprises dividing the calculated amount of change for each of the plurality of patients by the calculated total number of drugs or correcting by assigning a weight for each drug. The method of claim 1,
The determining of effects other than the predetermined effect may include calculating an average of the corrected amount of change and calculating an average amount of change for each drug; And
A method for selecting a candidate for re-creation of a new drug, comprising determining an effect of the predetermined drug for each clinical index based on a difference between the predetermined ideal change amount for each clinical index and the average change amount. The method of claim 5,
The determining of the effect of the predetermined drug is a method for selecting a candidate for re-creation of a new drug by calculating a distance based on the difference between the ideal change amount and the average change amount, and determining the calculated distance direction The method of claim 6,
If the direction of the calculated distance is the same as the ideal change amount of the drug, a first weight is given,
If the direction of the calculated distance is different from the ideal change amount of the drug, a second weight greater than the first weight is given,
When the direction of the calculated distance is unknown, a method for selecting a candidate for re-creation of a new drug is assigned a third weight between the first weight and the second weight. The method of claim 7,
A method for selecting a candidate for re-creation of a new drug in which a value having the lowest effect of the predetermined drug applied to the first weight to the third weight is set as the highest priority. An acquisition unit for acquiring clinical index values before and after exposure to a predetermined drug for each patient;
A change amount calculation unit for calculating a change amount of the value of the clinical index before and after exposure to the predetermined drug for each of the extracted clinical indexes;
A drug calculation unit for calculating a total number of exposed drugs including the predetermined drugs for each of the plurality of patients;
A correction unit for correcting the calculated amount of change for each of the plurality of patients by using the total number of selected drugs or weights for each drug;
An effect determination unit determining an effect other than a predetermined effect of the predetermined drug based on the corrected amount of change; And
A ranking determination unit that combines the effects of each of the above drugs to obtain a ranking with a high possibility of creating a new drug.
New drug re-creation candidate selection system including. The method of claim 9,
The clinical indicator values before and after the above exposure are
A system for selecting candidates for re-creation of new drugs obtained by using clinical data obtained before administration of the predetermined drug for each of the plurality of patients and clinical data obtained after administration of the predetermined drug. The method of claim 10,
The clinical indicators are cholesterol levels, glycated hemoglobin levels, test information including lung capacity or diagnostic information including diagnosis name, procedure, prescription, or a new drug re-creation candidate selection method including a combination thereof. The method of claim 9,
The correction unit,
A new drug re-creation candidate selection system for correcting the calculated amount of change for each of the plurality of patients by dividing the calculated amount of change for each of the plurality of patients by the calculated total number of drugs. The method of claim 9,
The effect determination unit,
By calculating the average of the corrected amount of change for each of the plurality of patients, calculating the average amount of change for each drug for the clinical index,
A new drug re-creation candidate selection system for determining the effect of the predetermined drug for each clinical index based on a difference between a predetermined ideal change amount for each clinical index and the average change amount. The method of claim 13,
A new drug re-creation candidate selection system for determining the effect of the predetermined drug by calculating a distance based on a difference between the ideal change amount and the average change amount, and considering the direction of the calculated distance. The method of claim 14,
If the direction of the calculated distance is the same as the ideal change amount of the drug, a first weight is given,
If the direction of the calculated distance is different from the ideal change amount of the drug, a second weight greater than the first weight is given,
When the direction of the calculated distance is unknown, a new drug re-creation candidate selection system is assigned a third weight between the first weight and the second weight. The method of claim 15,
The effect of the predetermined drug is a new drug re-creation candidate selection system in which the lowest value to which the first weight to the third weight is applied is set as the highest priority.

Images