JP2006302222A - Cancer onset risk prediction system and method, and cancer derivative method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a system and a method allowing execution of prediction of possibility of future clinical cancer affection of a preclinical cancer state, or a loss amount such as a cost burden required for treatment in the case of preclinical cancer, to complement usefulness of high-level cancer examination to promote early treatment by executing prescribed derivative compensation to the preclinical cancer state, to design the derivative compensation with execution of optimum treatment to the preclinical cancer as a requirement, and to digitalize risk of the future clinical cancer affection in each kind of the treatment. <P>SOLUTION: This prediction system having a computer and a storage device is provided with a high-accuracy cancer medical examination clustering data storage part 201 and a high-accuracy cancer medical re-examination result data storage part 202, and they are connected to a computer. The computer is provided with a clinical cancer state distribution calculation means 203, a simulation means 204, a condition-based correlation analysis means 205, a premium rate generation means 212, and a compensation condition impartment means 214. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

The present invention relates to a system and method for predicting cancer development risk that changes based on the result of a specific cancer screening under a specific condition, and a trigger bond, and an optimal treatment is performed based on the result of the cancer screening. The present invention relates to a cancer derivative method in which the risk of more serious cancer is reduced by derivative compensation, and this effect is reflected in the compensation conditions.

In recent years, various trigger bonds have been devised (financial derivative products that automatically pay pre-contracted compensation money without triggering an accident assessment or proof of damage when a trigger event occurs that meets certain conditions). Have been commercialized. A typical example is the so-called weather derivative. The impact of the weather on the business is predicted from a certain simulation based on the correlation between the past weather and business results such as sales, and scenario creation and damage A system and method for generating insurance premiums, etc., and a method for assigning derivatives are put into practical use.
Trigger Bonds are observations to promote the sale of derivative products in areas where earthquake damage, such as earthquake derivatives, is difficult to obtain damage compensation for existing earthquakes, or in areas where there has been no earthquake in recent years and customers are less appreciable. There are methods such as analysis of results and risk prediction, notification of the results to customers, calculation of insurance premium rate, and provision of derivative compensation conditions.

Japanese Patent Laid-Open No. 2003-349591

On the other hand, cancer screening has also progressed, and high-accuracy cancer screening such as high-accuracy tumor marker screening and PET (positron emission tomography) has been carried out, and the number of cancer early detection cases has increased. Yes. The cancer test by PET is a test using the property of consuming nearly 20 times glucose when the number of cancer cells is several times higher than that of normal cells. After injecting a drug called FDG (fluorodeoxyglucose) with a structure very similar to glucose, it is diagnosed by imaging how it is gathered. In general, CT and MRI capture the “size and shape” of cancer, etc. On the other hand, the PET test is a test that captures an increased metabolic state of cancer cells and the like, and in some cases, a cancer lesion of several millimeters has been discovered.

In PET examinations, it is possible to examine a wide range of the body at once without aiming at specific parts or organs, including organs that were difficult to examine by conventional examinations, benign / malignant judgment, metastasis and recurrence It is also effective for grasping.
In addition, new methods and evaluation methods have been developed for high-accuracy tumor marker screening, and it has become possible to detect millicancers that have been difficult to detect. This high-accuracy tumor marker screening makes it possible to further subdivide the apparent healthy state and evaluate the possibility of cancer by stage.

Japanese Patent Laid-Open No. 2001-289861

Among methods for evaluating cancer using tumor markers, a method using ALP (alkaline phosphatase) isozyme has attracted attention. ALP is an enzyme that has the function of cutting out phosphate, and there are different molecular weights for each organ such as bone, liver, placenta, and small intestine. Further, when there are a plurality of enzymes in each organ, they are called isozymes. Cancer patients are known to produce isozymes that have different molecular weights from healthy individuals, and by examining and evaluating the pattern of this isozyme, it is possible to discover small cancers that have occurred in bones, liver, intestines, etc. It is.

In recent years, the accuracy of tumor marker screening using this ALP isozyme has been improved. For example, it is possible to detect a microcancer of 10 μg to 1 g from a change in a tumor marker value for a certain period of time and a pattern change of an isozyme. Technology to do has appeared.

JP 2002-181825 A

However, in order to receive these high-accuracy cancer screenings, an expensive consultation fee is required, and the results of the consultation also indicate that a patient who has been diagnosed as having a healthy condition so far may be a future cancer carrier. Therefore, a large economic expenditure is required for the burden of treatment costs. In addition, it is assumed that insurance for clinical cancer is not applied to preclinical cancer conditions (they are not clinical cancers but there are microscopic cancers that are not urgent or are likely to have cancer in the future). Since the economic burden on the elderly will be greater, it is also expected that the social significance brought about by the highly accurate screening will be impaired.

In addition, the risk rate related to the clinical cancer is not dependent on one factor such as age, but needs to be obtained as a sum of various risk factors. Of particular note is the future risk transition when a preclinical cancer condition is diagnosed. In the past, when trying to set up insurance against cancer, the insurance premium rate was often determined uniformly by age at the time of enrollment, but pre-clinical cancer status is suspected when regular screening is performed. The risk of developing cancer varies greatly depending on a number of factors, such as the status of its progression, whether it is treated or not, and the content of the treatment.

In conventional technology, especially in the era when there was no high-accuracy cancer screening, there is no method for assessing the risk of future cancer of such preclinical cancers, and future risk predictions for multiple possible future factors There was no way to stand up. In addition to the case where the derivative compensation conditions are terminated by a single trigger event, there are cases where multiple compensation conditions are required due to advancement of medical care. The system for this is undeveloped.

In addition, if such high-accuracy cancer screening is widely implemented, it can be assumed that the scope of insurance for preclinical cancer is expanded, but an increase in the insurance premium rate cannot be denied.
First of all, the first issue is to devise a system and method that can predict the amount of damage such as the possibility of preclinical cancer in the future and the cost of treatment for preclinical cancer. Become.

On the other hand, the second challenge is to correct the insurance premium rate of people with preclinical cancer conditions in conjunction with the results of high-accuracy cancer screening, and to suppress the increase in the conventional insurance premium rate due to advances in cancer screening technology. There is also a third challenge for preclinical cancer to provide basic data to compensate for preclinical cancer conditions and to provide early data for supplementing the usefulness of advanced cancer tests. The fourth challenge is to design a derivative compensation that requires optimal treatment to be performed and to provide a quantification of the risk of future clinical cancers for each type of treatment.

In addition, as a method for correcting insurance premiums for those who are in a healthy state, it is a fifth problem to realize various calculations for providing compensation for a checkup fee on the condition that it is classified as a healthy state.

In terms of the balance of insurance premiums, first, those who are classified as healthy will be compensated for screening fees, and those who are classified as preclinical cancer will be given specific treatments instead of an increase in insurance premiums. The purpose of the fifth problem is to enable a condition simulation of a derivative to reduce the risk of clinical cancer and to compensate for the treatment cost.

[Configuration 1] In order to solve the above problems, the present invention is a prediction system that includes at least computing means and storage means, and predicts the risk of developing clinical cancer under specific conditions. Is
Based on specific numerical values indicating the existence of cancer based on past screening results in specific high-accuracy cancer screening that can detect preclinical cancer, it is apparent that the existence of the cancer cannot be denied. A high-accuracy cancer screening clustering database, which is classified into healthy states and clinical cancer states, and records the data for each classification.
In the high-accuracy cancer screening clustering database, the reexamination result of the high-accuracy cancer screening received after a specific period for each of the classifications is recorded so that data can be referred to for each of the classifications. A results database,
Calculate the correlation between the high-accuracy cancer screening clustering data and the high-accuracy cancer screening re-examination result data, and calculate the clinical cancer state distribution using the result of the high-accuracy cancer screening or the specific period as a variable Distribution calculation means;
And a simulation means for generating a prediction of the possibility of suffering from clinical cancer after a specific period, which is calculated based on the result of the medical checkup, which is received by the high-accuracy cancer screening.

Here, clinical cancer refers to a cancer-affected state that requires medically urgent medical treatment, such as early stage cancer, advanced cancer, and end stage cancer. Although clinical cancer depends on the site, it is generally clarified by finding a cancer of 1 g or more by various medical examinations by medical institutions and diagnosing cancer. It is known that cancer cells are present in minute amounts in healthy humans, and it usually takes several to several decades before these minute cancer cells are diagnosed as clinical cancer. General cancer insurance compensates for this clinical cancer, and it is often ambiguous how far the target of compensation is for the discovery of minute cancer cells, and highly accurate testing techniques have been developed. While smaller cancers will need to be treated, the number of cases where compensation is limited increases.

Here, the prediction system includes one or a plurality of computer systems, a server client type computer system connected via a network, or the like, which is hardware such as a device and software such as a program, and is operated by a CPU or the like. It consists of a device and a program provided there. The storage means is configured by a program for performing various recordings in a storage unit such as a hard disk or a memory.
In addition to the prediction system, there is a means to generate multiple cancer occurrence scenarios from the results of simulation means to calculate the risk rate of clinical cancer occurrence and derivative compensation conditions (derivative target conditions, premium fees, etc.) , Means for extracting correlation data that matches the conditions of this cancer occurrence scenario, means for generating values that match the clinical cancer state distribution of the correlation data, calculating cancer risk rates, and generating derivative compensation conditions, etc. Can be provided.

Here, the high-accuracy cancer screening is required to be a screening capable of detecting a preclinical cancer state together with at least detection of a healthy state and a clinical cancer state. Preferably, preclinical cancer states can be further classified into multiple stages. Here, the apparent healthy state means a state in which this preclinical cancer is detected or suspected.

The high-accuracy cancer screening reexamination result database is data generated based on a result of receiving a plurality of at least the same high-accuracy cancer screening, and is used for statistical processing. For correlation analysis, a method such as data mining can be used. Various configurations are conceivable for the simulation means, for example, one that calculates cancer risk at a plurality of specific points in time, and one that visualizes cancer risk in a multidimensional manner based on cancer state distribution.

[Configuration 2] The prediction system according to Configuration 1, wherein means for storing a premium rate of derivative compensation with a clinical cancer incidence as a trigger event for a customer before receiving the high-accuracy cancer screening, and the simulation According to another aspect of the present invention, there is further provided a class-specific insurance rate correction unit that generates an appropriate insurance rate correction condition for each classification from the simulation result generated by the unit.

In this configuration, the trigger event is that the medical institution that conducts the high-accuracy cancer screening recognizes the fact that clinical cancer is afflicted based on a clear reference value. It is also possible to add additional conditions such as to do. In this configuration, before conducting a specific high-accuracy cancer screening, the conventional cancer insurance is set at a conventional rate, and correction conditions such as a rate are assigned to the initial setting based on the result of the specific high-accuracy cancer screening This makes it possible to reset the appropriate insurance premium rate.

[Configuration 3] The prediction system according to any one of Configurations 1 to 2, wherein in the plurality of high-accuracy cancer screenings, the high-accuracy cancer screening clustering database for each of the high-accuracy cancer screenings;
The high-accuracy cancer screening reexamination result database,
A clinical cancer state distribution calculating means for each screening for calculating the clinical cancer state distribution using the plurality of high-accuracy cancer screenings as variables is further provided.

In this configuration, it is possible to quantify the increase or decrease in the number of subjects to receive treatment depending on the accuracy of cancer screening, and calculate appropriate derivative compensation conditions for each screening received. If the preclinical cancer status can be detected in detail, derivative compensation for treatment at the time of preclinical cancer diagnosis will increase, but on the other hand, the number of people suffering from clinical cancer will decrease, and the derivative compensation will be reduced. It is possible to statistically calculate the difference in derivative compensation conditions depending on the calculation of conditions and the accuracy of cancer screening, and to generate an evaluation of the cancer screening.

[Configuration 4] The prediction system according to any one of Configurations 1 to 3, wherein the apparent healthy state further includes a plurality of stages according to an abnormal level of cancer by the high-accuracy cancer screening,
Furthermore, the clinical cancer state distribution is configured to be able to classify the results of the simulation means for each of the plurality of stages and generate a prediction of the likelihood of suffering clinical cancer for each classification. did.

In this configuration, just as clinical cancers are classified according to their abnormal levels, such as early stage cancer, advanced cancer, and end stage cancer, pre-clinical cancers are also classified according to their abnormal level, and the risk rate for each category. It is possible to provide a prediction system that can correct the error.

[Configuration 5] The prediction system according to any one of Configurations 1 to 4, wherein the plurality of stages has an ideal state in which a minute cancer cannot be detected as Stage I, a precancerous state at a microgram level as Stage II, A stage of tumor growth with a pre-cancerous state of milligram level as stage III, a preclinical cancer state of less than 1 g as stage IV, and a clinical cancer state where cancer of 1 g or more is estimated as stage V,
The simulation means generates a prediction of the possibility of suffering from a future clinical cancer for each tumor growth degree stage.

In this configuration, a system for generating ideal compensation conditions is provided based on the assumption that subdivision centering on preclinical cancer is necessary to reduce the risk of stage V, which is the final stage. The detection of cancer is not linked to the actual medical condition, and is based on optimal compensation and optimizes the overall compensation conditions of the test examinee. It is possible to manage.

[Configuration 6] The prediction system according to any one of Configurations 1, 2, 4, and 5, wherein the high-accuracy cancer screening is performed by electrophoresis, fractionating, separating proteins in serum, and ALP isozyme It is characterized by being a tumor marker screening including at least a method for evaluating the risk of cancer by detecting ALP isozyme by developing color with a staining solution.

It is preferable that the high-accuracy cancer screening can detect at least millicancer, preferably microcancer. A specific cancer evaluation method based on detection of ALP isozymes is known in Patent Document 3 described in the technical background of the present specification as a screening with high detectability of microcancer.

[Configuration 7] The prediction system according to any one of Configurations 1 to 6, wherein the specific period is the regular high period that is performed every predetermined period from the first high-accuracy cancer screening. It is an implementation interval between an accurate cancer screening and the next high-accuracy cancer screening.

It is possible to generate more detailed derivative compensation conditions by regularly conducting high-accuracy cancer screening, statisticalizing the history, and managing the risk rate. The specific period is preferably once a year in consideration of the growth of preclinical cancer.

[Configuration 8] The prediction system according to any one of Configurations 1 to 7, wherein the clinical cancer state distribution calculating unit approximates and calculates the distribution of the clinical cancer state by a normal distribution. .

When the correlation is defined as a normal distribution, it is possible to extract a numerical value that fits the normal distribution and apply it to the correlation to find the corresponding cancer risk, especially based on multiple conditions This is effective when analyzing

[Configuration 9]
A prediction method for predicting a risk of developing clinical cancer under specific conditions using a prediction system comprising at least a calculation means and a storage means,
Calculate the correlation between the high-accuracy cancer screening clustering data and the high-accuracy cancer screening re-examination result data stored in the storage means, and the clinical cancer state using the result of the high-accuracy cancer screening or the specific period as a variable Calculate the distribution,
A specific person who has received the high-accuracy cancer screening generates a prediction of the possibility of suffering from clinical cancer after a specific period, which is calculated based on the result of the medical examination.

[Configuration 10] The prediction method according to Configuration 9, the step of storing an insurance premium rate of derivative compensation with a clinical cancer occurrence as a trigger event for a customer before receiving the high-accuracy cancer screening, and the simulation And a step of generating an appropriate premium rate correction for each classification from the simulation result generated by the means.

[Configuration 11] The prediction method according to any one of Configuration 9 or Configuration 10, wherein a plurality of the high-precision cancer screenings are further used as variables from the plurality of high-precision cancer screening result databases. Is calculated.

[Configuration 12] The prediction method according to any one of Configuration 9 to Configuration 11,
The apparent healthy state further has a plurality of stages according to the abnormal level of cancer by the high-accuracy cancer screening,
Further, the clinical cancer state distribution is classified for each of the plurality of stages, and a prediction of the possibility of suffering from clinical cancer for each classification is generated.

[Configuration 13] The prediction method according to any one of Configurations 9 to 12,
The plurality of stages include an ideal state in which a minute cancer cannot be detected, stage I, a microgram level precancerous state, stage II, a milligram level precancerous state, stage III, a preclinical cancer state of less than 1 g, stage IV, And a tumor growth stage having a stage V as a clinical cancer state in which cancer of 1 g or more is estimated, and generating a prediction of the possibility of suffering from a future clinical cancer for each tumor growth stage .

[Configuration 14] The prediction method according to any one of Configurations 9, 10, 12, and 13,
The high-accuracy cancer screening comprises at least a method for fractionating by electrophoresis, separating proteins in serum, developing color with an ALP isozyme staining solution, detecting ALP isozyme, and evaluating the risk of the cancer. Including tumor marker screening.

[Configuration 15] The prediction method according to any one of Configurations 9 to 14,
The specific period is an interval between the regular high-accuracy cancer screening performed at regular intervals and the next high-accuracy cancer screening, starting from the first high-accuracy cancer screening. Features.

[Configuration 16] The prediction method according to any one of Configurations 9 to 15,
The method further comprises a treatment prognostic vortex risk simulation step for generating a prediction of the possibility of suffering from a future clinical cancer for each stage when specific treatment is performed for each of the plurality of stages. .

Here, the treatment prognosis risk simulation refers to a simulation of the risk of developing clinical cancer or the risk of recurrence of preclinical cancer that changes depending on the number of years since the treatment when a specific treatment is performed. In addition, it is possible to use the stage of the preclinical cancer state at the time of treatment as a variable, and to perform simulation for each content of treatment.

[Configuration 17] Configuration A derivative whose trigger event is that a patient who receives the high-accuracy cancer screening after a certain period of time is classified into the stages II to V in the plurality of stages and receives the specific treatment A method,
The derivative compensation condition is set or changed from the prediction generated in the treatment prognostic risk simulation step for the specific treatment and the treatment cost of the specific treatment.

In the present invention, the derivative method is a method for calculating a rate, setting conditions, and generating various simulations using an electronic computer, a calculation device, and the like. It has a relation.

The trigger event here is compensation for treatment costs, not compensation for pre-clinical cancer, and it is assumed that this treatment will affect the risk of future clinical cancer.

[Configuration 18] The derivative method according to Configuration 17, in which at least one stage selected from the plurality of stages in the history of the derivative-compensated person receiving the high-accuracy cancer screening has been classified, And analyzing the correlation between the classification and the trigger event occurrence rate based on the medical examination result history information including the above information, and calculating an appropriate insurance premium rate.

If the trigger event occurrence rate and compensation amount can be calculated, it is possible to calculate an appropriate insurance premium rate, and the trigger event occurrence rate can be simulated using any relevant variable according to past statistics. .

[Configuration 19] The prediction method according to Configuration 16, comprising the treatment prognosis progress risk simulation step based on a plurality of treatment details for each specific treatment, and a course based on the treatment details for each of the specific treatments It was characterized by obtaining a risk assessment.

Multiple treatment methods to be compensated in advance are identified, and if there are differences in the evaluation of the subsequent course and the possibility of clinical cancer, depending on the treatment method, a prognostic evaluation can be obtained numerically for the treatment content It becomes.

[Configuration 20] Configuration After a predetermined period of time, a trigger event is that a patient who receives the high-accuracy cancer screening is classified into the stages II to V in the plurality of stages and receives the specific treatment. A derivative method,
A cost-effectiveness is calculated from the progress risk evaluation for the treatment contents for each of the plurality of specific treatments and the treatment cost for each of the specific treatments,
Extracting the specific treatment from which the examinees classified in the stages II to IV selected from the plurality of stages can obtain the maximum cost-effectiveness;
The method further includes calculating a derivative compensation condition for the treatment.

In this configuration, the objective is to establish derivative compensation on the condition that the treatment content can be statistically reduced without establishing a medical evaluation, and from another viewpoint, It is also possible to statistically realize the evaluation of treatment content different from medical practice.

[Configuration 21] A derivative method configured to be able to give a compensation condition according to a result of the high-accuracy cancer screening that a medical examinee consults,
After a certain period of time, the examinee who receives the high-accuracy cancer screening is:
A step of giving a healthy classification derivative compensation condition for performing derivative compensation on a reasonable cost of the high-accuracy cancer screening as a trigger event that is classified into the stage I in the plurality of stages;
A preclinical cancer classification derivative that calculates the amount corresponding to the treatment cost required for the specific treatment and compensates for the derivative by using the specific event as the trigger event in the stages II to IV. A compensation condition assigning step;
A clinical cancer classification derivative compensation condition granting step for performing derivative compensation for a certain compensation amount among treatment costs, with triggering being classified into the stage V in the plurality of stages,
A step of applying a selective derivative compensation condition that is applied in accordance with the result of the high-accuracy cancer screening that is received by the examinee.

Derivative compensation conditions in this configuration can be realized through extraction of trigger events, generation of premium rates, etc. using the prediction system implemented in the present invention. It is possible to implement by calculating with and displaying on a display or the like.

[Configuration 22] A derivative method for granting derivative compensation conditions based on a certain condition, wherein a normal classification insurance for calculating a current insurance premium rate from the probability of applying the healthy classification derivative compensation condition granting step and the compensation amount Charge calculation step,
From the probability of applying the preclinical cancer classification derivative compensation condition grant step and its compensation amount, a preclinical cancer classification premium calculation step for calculating a current insurance premium rate,
A clinical cancer classification premium calculation step for calculating a current insurance premium rate from the probability of applying the clinical cancer classification derivative compensation condition granting step and its compensation amount,
The health insurance premium step, the preclinical cancer classification insurance premium calculation step, the clinical cancer classification insurance premium calculation step, and a step of determining a derivative compensation condition by summing current insurance premium rates calculated from It was characterized by that.

In this configuration as well, the definition of compensation for healthy classification is used as a high-accuracy cancer screening fee. In order to design derivative compensation with this configuration, it is required to set each derivative compensation condition for each classification, which is the result of high-accuracy cancer screening, and add up each insurance premium to calculate.

According to the present invention, in contrast to the monotonous definition of risk for those who have not previously suffered from clinical cancer, it is possible to realize subdivision by extremely rational risk factors.
In addition, treatment compensation for preclinical cancer that has not been certified as a clinical cancer that was previously excluded from derivative compensation, and optimization of derivative compensation conditions can be achieved.
In addition, the invention of advanced screening that increases the scope of derivative compensation reduces the uniform increase in the insurance premium rate, contributes to early treatment before the onset of cancer, reduces the number of people who die from cancer, and the resulting death Insurance premiums can be reduced.

Hereinafter, the present invention will be described in detail with reference to the drawings showing the best mode for carrying out the invention. The embodiment shown here is a non-limiting example, and in addition to the configuration shown here, the device type, contents, configuration, program type, configuration, system configuration, arrangement, etc. are arbitrarily combined within the scope of the present invention. It is possible and is not limited to the configuration.

FIG. 1 is a block diagram showing a configuration of a prediction system according to an embodiment of the present invention. A prediction system according to an embodiment of the present invention includes a computer 101 capable of correlation analysis and calculation of clinical cancer state distribution, an input device 103 for inputting an initial set value, a correlation index, and the like, and a display for displaying calculation results 102, the computer 101 is provided with a main control unit 111 for executing prediction calculations and various processes, and a high-accuracy cancer screening clustering data storage unit 121 for storing data used for the prediction calculations, A cancer screening reexamination result data storage unit 122, a correlation analysis result data storage unit 123 that stores the calculation results, and an analysis result data storage unit 124 are connected. Each of the data storage units 121 to 124 can be provided in a plurality of storage devices, and can also be provided separately in one storage device.

FIG. 2 is a block diagram showing in detail the configuration of the computer of this embodiment. The main control unit (FIG. 1: 101) is provided with clinical cancer state distribution calculating means 203 for executing correlation analysis, simulation means 204, and conditional correlation analysis means 205, and a high-accuracy cancer screening clustering data storage unit. The correlation analysis is executed with reference to 201 and the data recorded in the high-accuracy cancer screening reexamination result data storage unit 202, and is recorded in the correlation analysis result data storage unit 207 by the correlation analysis result storage unit 206.

In the correlation analysis result data storage unit 207, a reexamination correlation data storage unit 208 that records correlation data between the classification stage and cancer incidence using the elapsed years as a parameter, and a transition pattern of the classification stage is extracted from a plurality of examination results Then, the analysis data storage unit 209 for each classification that records the correlation analysis result regarding the cancer incidence for each pattern, and the analysis data for each therapy that records the correlation data of the cancer incidence using the classification stage, the treatment content classification, and the elapsed years as parameters. A storage unit 210 and an analysis data storage unit 211 for each screening that records correlation data calculated for each type of screening are provided.

The data in the correlation analysis data storage unit 207 is referred to the insurance rate generation unit 212, the classified insurance rate correction unit 213, and the compensation condition assigning unit 214. The insurance rate generation unit 212 calculates the risk rate and the compensation condition. Further, the classified insurance rate correction unit 213 executes insurance rate correction processing for each parameter of the correlation analysis data, and is recorded in the analysis data storage unit 217 by the analysis result storage unit 215. When the insurance premium rate is calculated for each defined condition, a derivative compensation condition is generated by the compensation condition adding unit 214 that generates a compensation condition by combining the insurance rate with the condition, and is analyzed by the compensation condition storage unit 216. It is recorded in the data storage unit 217. The analysis result data storage unit 217 further includes a clinical cancer state distribution data storage unit 218, a risk rate data storage unit 219, a compensation condition data storage unit 220, and an extraction result data storage unit 221.

Many structures of data mining systems including correlation analysis and clustering are open to the public, and the explanation is omitted here. The major feature is that statistical treatment can be extracted statistically, and further, correction of data by treatment that will occur in the future can be predicted in advance.

FIG. 3 is a flowchart showing processing for generating derivative compensation conditions according to the embodiment of the present invention. First, in the initial setting (step S11), setting conditions necessary for generating compensation conditions are input. The setting conditions include the setting of a required period, the type of examination, the basic classification of premium rates, the type of treatment, the interval between examinations, and the like. Here, it is preferable to select an optimal inspection extracted by the prediction system of the present invention as the inspection type. The input data can be stored on a computer memory (temporary storage device).

Next, past data is referred to (step S12) and temporarily stored together with the input data. Then, it is determined whether the comparison conditions for approximating the distribution average value, distribution standard deviation value, etc. of the array are the same (step S12). If not (step S12: No), the same condition If the data is the same (step S12: Yes), multidimensional coordinate values are generated (step S15) and stored in the memory as array data.

Next, an average value of the distribution is generated from the multidimensional coordinate values (step S16), a distribution standard deviation is generated (step S17), and the array is stored in the correlation analysis result data storage unit (step S18). Further, each process from S15 to S18 is repeated for each condition according to the input of the initial setting (step S11) (step S19).

Next, a simulation of the ratio of each stage classified by the test result and the treatment cost for each stage is generated (step S20). When there are a plurality of conditions input in the initial setting (step S11), a simulation for each condition is further generated (S21), and an optimal treatment method is extracted under the conditions (step S23). Derivative compensation conditions with triggering this extracted treatment as a trigger event are generated (step S24), and the result is stored in the storage unit or displayed on the display, and the process is terminated.

FIG. 4 is an example of the contents when the contents of the high-accuracy cancer screening clustering database are displayed on the display. Based on the examination results, the corresponding person corresponding to each stage synthesized and displayed the prediction of the possibility of moving to stage V under the condition setting of 7 years later.

FIG. 5 is a part of a diagram showing an example of the contents of a database storing derivative compensation conditions by treatment contents. Compensation condition data is stored for each treatment content classification and each reference stage.

FIG. 6 is an example of a probability distribution diagram in which the correlation between clinical cancer detection and the correlation analysis index is approximated by a normal distribution. In addition, when there are a plurality of data relating to the same cancer distribution, it is possible to plot a plurality of data in a multidimensional space and approximate a region where the data exists by a figure.

FIG. 7 is a diagram showing an example of a course distribution of future clinical cancer cases as a result of high-accuracy cancer screening. It has been shown that as the stage gets higher, the time to cancer development becomes shorter and the rate increases, and the risk rate becomes higher. This data is based on cancer screening using the tumor marker comprehensive analysis method from 1984 to 1986 for 2126 residents in 15 randomly selected regions in Japan.

FIG. 8 is a diagram showing cut-off values for determining tumor marker screening for detecting cancer by detecting ALP isozymes. The explanation of each tumor marker is described in detail below.
CEA (carcinoembryonic antigen) is obtained by measuring CEA by enzyme immunoassay.
HSAP (heat-resistant alkaline phosphatase) is obtained by heat treating 50 ml of serum at 65 ° C. for 7 minutes, then reacting with a fluorescent substrate at 37 ° C. for 20 minutes, and deproteinizing it with acetone to increase the tendency strength in serum. Measured with a fluorimeter.
Ferritin is measured by radioimmunoassay.
IAP is characterized by inhibiting both lymphocyte formation induced by phytohemoglutinin and lymphocyte mixing test in vitro, and 5 μl of each sample was added to an agar gel containing anti-IAP serum. The diameter of the sedimentation ring was measured after placing in a plate hole and incubating at 37 ° C. for 48 hours.
Ribonuclease is measured using polycytidylic acid as a substrate.
Sialic acid is measured by an automatic analyzer based on an enzyme measurement method.
Alkaline phosphatase isozyme uses three parameters (ALP1, ALP2 / 3, APA) for densitization of ALP isozymes using the fact that serum ALP isozymes show significant variation with pathology within the normal range of ALP activity. It is calculated from the pattern by the measurement.
CA19-9 (sugar chain antigen) was measured by radioimmunoassay.
TPA (tissue polypeptide antigen) is measured by radioimmunoassay.

This comprehensive analysis method of tumor markers makes it possible to detect preclinical cancers that have been impossible in the past. The results of cancer screening using highly accurate tumor markers are often variable data, and are particularly suitable for setting derivative compensation conditions by correlation analysis.

FIG. 9 is a slow chart showing a flow of derivative compensation condition generation according to the embodiment of the present invention. In this embodiment, the purpose is to generate data for correction of compensation conditions based on the result of the consultation in particular, and it is also possible to receive the application for the trigger bond by transmitting the data generated here to the customer's terminal. is there. In that case, the initial setting (step S81) is performed by the insurance company, the input of the consultation result (step S82) is performed by the medical institution that performed the screening, and each compensation condition generation (steps S85, S88, S91, S94, It is reasonable that the derivative compensation conditions generated by S96) are sent to the customer.

In the initial setting (step S81), setting conditions necessary for generating the compensation conditions are input, and further, a result of receiving a high-accuracy cancer screening is input (step S82). When the input result is stage I (step S83: Yes), the stage I classification possibility, the amount of damage at that time, the data such as the optimal treatment are referred to (step S84), and the compensation condition for stage I Is generated (step S85). Further, when the classification is not stage I (step S83: No), it is determined whether the classification is stage II (step S86). Similarly, whether the stage II data reference and compensation condition generation or the process is sent to stage III. (Steps S87 to S89) are executed. Thereafter, similarly, determination is made as stage III, stage IV, and stage V, and compensation conditions are generated (steps S89 to S96). Compensation conditions for each stage can be stored in the storage unit or displayed on the display.

Compensation conditions include setting of trigger event, prediction of compensation contents and damage amount, various compensation for treatment prognosis when compensation is treatment cost, compensation amount and insurance fee, change of compensation condition after trigger event occurrence, special contract etc. Generated based on a request for initial setting (step S81).

By using the present invention, it is possible to promote medical institutions to conduct advanced cancer screenings that are not covered by medical insurance. As a result, advanced medical technologies can be widely spread and cancer deaths can be reduced. it can.
On the other hand, it is possible to provide insurance companies with option design for cancer insurance and offer various values of risk-divided insurance, which can be used to develop new trigger bonds. In addition, the implementation of the present invention promotes treatment in preclinical cancer states, so that the mortality rate of cancer, which is the first mortality rate, can be reduced, so that the payment of death insurance payments is reduced, and expenditures by subsidizing medical expenses are increased. It is possible to simulate a decrease in death insurance payments and use it to revise insurance premiums.

It is a block diagram which shows the structure of the prediction system of embodiment of this invention. It is a block diagram which shows the structure of the computer of this embodiment in detail. It is a flowchart which shows the process of the derivative compensation condition production | generation which is embodiment of this invention. It is the example of a content at the time of displaying the content of the high precision cancer screening clustering database on a display. It is a part of figure which shows an example of the content of the database which memorize | stored the derivative compensation conditions according to treatment content. It is an example of the probability distribution diagram which approximated the correlation with clinical cancer detection and a correlation analysis parameter | index with normal distribution. It is a figure which shows an example of the course distribution of the future clinical cancer onset by the result of a high precision cancer screening. It is a figure which shows the cutoff value for the determination of the tumor marker screening which detects ALP isozyme and evaluates cancer. It is a slow chart which shows the flow of derivative compensation condition generation which is an embodiment of the present invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 Computer 102 Display 103 Input device 111 Main control part 121 High precision cancer screening clustering data storage part 122 High precision cancer screening reexamination result data storage part 123 Correlation analysis result data storage part 124 Analysis result data 203 Clinical cancer state distribution calculation means 204 Simulation means 205 Conditional correlation analysis means 206 Correlation analysis result storage means 212 Insurance rate generation means 213 Classification-specific insurance rate correction means 214 Compensation condition assignment means 215 Analysis result storage means 216 Compensation condition storage means S11 Initial setting step S12 Data reference step S13 Condition comparison step S14 Data creation step under the same condition S15 Multidimensional coordinate value generation step S16 Distribution average value generation step S17 Distribution standard deviation generation step S18 Array storage step S19 Conditional repetition step Generation step S23 optimal therapy extracting step S24 derivatives compensation condition generating step of-up S20 simulation step S21 Conditional Simulation step S22 the Conditional predicted value

Claims (22)

A prediction system comprising at least a calculation unit and a storage unit, and predicting a clinical cancer risk under a specific condition, wherein the storage unit includes:
Based on specific numerical values indicating the existence of cancer based on past screening results in specific high-accuracy cancer screening that can detect preclinical cancer, it is apparent that the existence of the cancer cannot be denied. A high-accuracy cancer screening clustering database, which is classified into healthy states and clinical cancer states, and records the data for each classification.
In the high-accuracy cancer screening clustering database, the reexamination result of the high-accuracy cancer screening received after a specific period for each of the classifications is recorded so that data can be referred to for each of the classifications. A results database,
The terminal cancer that calculates the correlation between the high-accuracy cancer screening clustering data and the high-accuracy cancer screening re-examination result data and calculates the clinical cancer state distribution using the result of the high-accuracy cancer screening or the specific period as a variable A state distribution calculating means;
A prediction system comprising: a simulation unit that generates a prediction of a possibility that a specific person who has undergone the high-accuracy cancer screening will suffer from clinical cancer after a specific period, which is calculated based on the result of the medical examination. From the simulation result generated by the simulation means generated from the simulation result generated by the simulation means generated by the means for storing the premium rate of the derivative compensation with the clinical cancer onset as the trigger event for the customer before receiving the high-accuracy cancer screening. The prediction system according to claim 1, further comprising classification-specific insurance rate correction means for generating a correct insurance rate correction condition. In the multiple high-precision cancer screening, the high-precision cancer screening clustering database for each of the high-precision cancer screening,
The high-accuracy cancer screening reexamination result database,
3. The clinical cancer state distribution calculating means for each screening for calculating the clinical cancer state distribution using a plurality of the high-accuracy cancer screenings as further variables, further comprising: Prediction system. The apparent healthy state further has a plurality of stages according to the abnormal level of cancer by the high-accuracy cancer screening,
The clinical cancer state distribution is further classified for each of the plurality of stages, and a prediction of the possibility of suffering from clinical cancer for each classification is generated. The prediction system described in. The plurality of stages include an ideal state in which a minute cancer cannot be detected, stage I, a microgram level precancerous state, stage II, a milligram level precancerous state, stage III, a preclinical cancer state of less than 1 g, stage IV, And a tumor growth stage in which a clinical cancer state in which cancer of 1 g or more is estimated is stage V,
The prediction system according to any one of claims 1 to 4, wherein the simulation unit generates a prediction of a possibility of suffering from a future clinical cancer for each tumor growth degree stage. The high-accuracy cancer screening comprises at least a method for fractionating by electrophoresis, separating proteins in serum, developing color with an ALP isozyme staining solution, detecting ALP isozyme, and evaluating the risk of the cancer. The prediction system according to claim 1, wherein the prediction system is a tumor marker screening. The specific period is an interval between the regular high-accuracy cancer screening performed at regular intervals and the next high-accuracy cancer screening, starting from the first high-accuracy cancer screening. The prediction system according to any one of claims 1 to 6, wherein the prediction system is characterized. The prediction system according to any one of claims 1 to 7, wherein the clinical cancer state distribution calculating means calculates the clinical cancer state distribution by approximating it with a normal distribution. A prediction method for predicting a risk of developing clinical cancer under specific conditions using a prediction system comprising at least a calculation means and a storage means,
Calculate the correlation between the high-accuracy cancer screening clustering data and the high-accuracy cancer screening re-examination result data stored in the storage means, and the clinical cancer state using the result of the high-accuracy cancer screening or the specific period as a variable Calculate the distribution,
A prediction method for generating a prediction of a possibility that a specific person who has received the high-accuracy cancer screening will suffer from clinical cancer after a specific period, which is calculated based on the result of the medical examination. The step of storing the premium rate of the derivative compensation with the clinical cancer onset as a trigger event for the customer before receiving the high-accuracy cancer screening, and the simulation result generated by the simulation means, And a step of generating a correct insurance rate correction condition. The prediction method according to any one of claims 9 and 10, wherein the clinical cancer state distribution is calculated from a plurality of high-accuracy cancer screening result databases using the plurality of high-accuracy cancer screenings as variables. The apparent healthy state further has a plurality of stages according to the abnormal level of cancer by the high-accuracy cancer screening,
The prediction method according to any one of claims 9 to 11, further comprising: classifying the clinical cancer state distribution for each of the plurality of stages and generating a prediction of a possibility of suffering from clinical cancer for each classification. The plurality of stages include an ideal state in which a minute cancer cannot be detected, stage I, a microgram level precancerous state, stage II, a milligram level precancerous state, stage III, a preclinical cancer state of less than 1 g, stage IV, And a tumor growth degree stage in which a clinical cancer state in which cancer of 1 g or more is estimated is stage V, and a prediction of a possibility of suffering a future clinical cancer is generated for each tumor growth stage. The prediction method according to any one of Items 12. The high-accuracy cancer screening comprises at least a method for fractionating by electrophoresis, separating proteins in serum, developing color with an ALP isozyme staining solution, detecting ALP isozyme, and evaluating the risk of the cancer. The prediction method according to claim 9, wherein the prediction method is a tumor marker screening. The specific period is an interval between the regular high-accuracy cancer screening performed at regular intervals and the next high-accuracy cancer screening, starting from the first high-accuracy cancer screening. 15. The prediction method according to claim 9, wherein the prediction method is characterized in that The treatment prognosis progress risk simulation step of generating a prediction of the possibility of suffering from a future clinical cancer for each stage when specific treatment is performed for each of the plurality of stages. The prediction method according to any one of 15 A derivative method that uses a specific event as a trigger event when a patient who receives the high-accuracy cancer screening after a certain period of time is classified into the stages II to V in the plurality of stages and receives the specific treatment,
A derivative method characterized in that a derivative compensation condition is set or changed from the prediction generated in the treatment prognostic risk simulation step for the specific treatment and the treatment cost of the specific treatment. The classification and the trigger are based on the diagnosis result history information including information on at least which stage selected from the plurality of stages in the history that the derivative recipient received the high-accuracy cancer screening in the past. The derivative method according to claim 17, further comprising a step of analyzing a correlation with an event occurrence rate and calculating an appropriate insurance premium rate. The prediction method according to claim 16, further comprising the treatment prognosis progress risk simulation step based on a plurality of treatment details for each specific treatment, and obtaining a progress risk evaluation based on the treatment details for each of the specific treatments. A derivative method in which a person who receives the high-accuracy cancer screening after a certain period of time is classified into the stages II to V in the plurality of stages and receives the specific treatment as a trigger event. ,
A cost-effectiveness is calculated from the progress risk evaluation for the treatment contents for each of the plurality of specific treatments and the treatment cost for each of the specific treatments,
Extracting the specific treatment from which the examinees classified in the stages II to IV selected from the plurality of stages can obtain the maximum cost-effectiveness;
A derivative method further comprising calculating a derivative compensation condition for the treatment. After a certain period of time, the examinee who receives the high-accuracy cancer screening is:
A step of giving a healthy classification derivative compensation condition for performing derivative compensation on a reasonable cost of the high-accuracy cancer screening as a trigger event that is classified into the stage I in the plurality of stages;
A preclinical cancer classification derivative that calculates the amount corresponding to the treatment cost required for the specific treatment and compensates for the derivative by using the specific event as the trigger event in the stages II to IV. A compensation condition assigning step;
A clinical cancer classification derivative compensation condition granting step for performing derivative compensation for a certain compensation amount among treatment costs, with triggering being classified into the stage V in the plurality of stages,
A derivative method comprising a selective derivative compensation condition providing step of applying one step selected from the above in accordance with the result of the high-accuracy cancer screening that the examinee receives. From the probability of applying the normal category derivative compensation condition granting step and its compensation amount, a normal category premium calculating step of calculating a current insurance premium rate,
From the probability of applying the preclinical cancer classification derivative compensation condition grant step and its compensation amount, a preclinical cancer classification premium calculation step for calculating a current insurance premium rate,
A clinical cancer classification premium calculation step for calculating a current insurance premium rate from the probability of applying the clinical cancer classification derivative compensation condition granting step and its compensation amount,
The health insurance premium step, the preclinical cancer classification insurance premium calculation step, the clinical cancer classification insurance premium calculation step, and a step of determining a derivative compensation condition by summing current insurance premium rates calculated from Derivative method.

Images