JP2021174108A - Diagnosis support device - Google Patents

Abstract

To provide a diagnosis support device that can accurately determine the biological state of an analyte.SOLUTION: A diagnosis support device 1 has an acquisition function, a biological state determination function, and a normal range calculation function. The acquisition function is to acquire, from a storage circuit storing for every analyte an observed value of an analyte from a sensor and lifestyle/environment information including at least one of lifestyle information and environment information, an observed value and lifestyle/environment information in a predetermined analyte. The biological state determination function is to determine the biological state of the predetermined analyte for a disease based on the observed value and the lifestyle/environment information in the predetermined analyte. The normal range calculation function is to calculate a normal range of the observed value of the predetermined analyte in the lifestyle/environment information on the predetermined analyte, based on a change in the biological state of the predetermined analyte and a change in the lifestyle/environment information.SELECTED DRAWING: Figure 9

Description

The embodiments disclosed in this specification and drawings relate to a diagnostic support device.

Conventionally, gene diagnosis by next generation sequencer (NGS) or gene panel, IVD (In-Vitro Diagnostics) test, protein analysis, antibody test, histopathological diagnosis, radiographic image diagnosis, image diagnosis other than radiographic image diagnosis , And there are many diagnostic and testing techniques such as non-imaging testing.

In recent years, in the field of cancer, companion diagnostics, etc. for appropriately selecting therapeutic agents such as new anticancer agents, molecular-targeted therapeutic agents, immune checkpoint inhibitors, etc. by tests using NGS or gene panels, etc. Is coming to be done. In the oncogene panel test, the creation of a mechanism for profiling and other efforts using AI (Cognitive Computing System) such as IBM Watson based on tens of millions of papers is being promoted in Japan. In addition, Comprehensive Genomic Profiling efforts including TMB (tumor mutational burden) and MSI (microsatellite instability) by Roche-Foundation Medicine are in progress.

In the conventional diagnosis and test, for example, a diagnosis based on a gene profile, a diagnosis based on an individual diagnosis, a diagnosis based on an individual test result, or a clinical judgment is performed. Alternatively, the judgment is made in a complex manner based on the results of a plurality of diagnoses and based on the guidelines and the experience of the doctor.

However, it is difficult for a doctor to accurately determine the state (biological state) of a subject with respect to the living body only by these limited diagnoses and tests.

Japanese Unexamined Patent Publication No. 2011-152194 Japanese Unexamined Patent Publication No. 2009-181564 Japanese Unexamined Patent Publication No. 2000-67139 Japanese Unexamined Patent Publication No. 2016-154042 Japanese Unexamined Patent Publication No. 2005-192954 Japanese Unexamined Patent Publication No. 2005-122231 Japanese Unexamined Patent Publication No. 2013-12025 International Publication No. 2015/050174 Japanese Unexamined Patent Publication No. 2019-8812

"Development of a method to estimate a complex metabolic reaction network only from actual measurement data-It is possible to theoretically search for unknown metabolic pathways-", [online], January 11, 2013, RIKEN RIKEN, [Searched April 1, 2019], Internet <URL: http://www.riken.jp/pr/press/2013/20130111_1/>

One of the problems to be solved by the embodiments disclosed in the present specification and the drawings is to provide a diagnostic support device capable of accurately determining the biological state of a subject. However, the problems to be solved by the embodiments disclosed in the present specification and the drawings are not limited to the above problems. It is also possible to position the problem corresponding to each effect of each configuration shown in the embodiment described later as another problem.

The diagnosis support device according to the embodiment includes an acquisition unit, a determination unit, and a calculation unit. The acquisition unit obtains the observed value of the subject from the storage unit that stores the observed value of the subject by the sensor and the lifestyle / environmental information including at least one of the lifestyle information and the environmental information for each subject. Acquire lifestyle / environmental information. The determination unit determines the biological state of the predetermined subject for the disease based on the observed value in the predetermined subject and lifestyle / environmental information. The calculation unit determines the normal range of the observed value of the predetermined subject in the lifestyle / environmental information of the predetermined subject based on the change in the biological state of the predetermined subject and the change in the lifestyle / environmental information. calculate.

FIG. 1 is a schematic view showing a configuration example of a diagnostic support device according to the first embodiment. FIG. 2 is a block diagram showing the functions of the diagnostic support device according to the first embodiment. FIG. 3 is a diagram showing an example of a data structure of a database for determining a biological state in the diagnostic support device according to the first embodiment. FIG. 4 is a diagram for explaining an example of a method of calculating the degree of relationship from a quantitative score in the diagnostic support device according to the first embodiment. FIG. 5 is a diagram showing an example of displaying the degree of relationship and the total reliability in the diagnostic support device according to the first embodiment. FIG. 6 is a diagram showing the operation of the diagnosis support device according to the first embodiment as a flowchart. FIG. 7 is a diagram showing an example of warning information in the third modification of the diagnostic support device according to the first embodiment. FIG. 8 is a diagram for explaining an example of a method for estimating a biological state in the diagnostic support device according to the second embodiment. FIG. 9 is a block diagram showing the functions of the diagnostic support device according to the fourth embodiment. FIG. 10 is a diagram showing an example of the biological state of a predetermined subject for a disease in the diagnostic support device according to the fourth embodiment. FIG. 11 is a diagram showing a display example of a normal range in the diagnostic support device according to the fourth embodiment.

Hereinafter, embodiments of the diagnostic support device will be described in detail with reference to the drawings. The diagnostic support device according to the present application is not limited to the embodiments shown below.

(First Embodiment)
FIG. 1 is a schematic view showing a configuration example of a diagnostic support device according to the first embodiment.

FIG. 1 shows a diagnostic support device 1 according to the first embodiment. The diagnosis support device 1 is a device that supports the diagnosis of a disease (disease) of a subject by an operator (for example, a doctor). The diagnosis support device 1 is a medical image management device (image server), a workstation, an image interpretation terminal, or the like, and is provided on a medical image system connected via a network. The diagnosis support device 1 may be an offline device.

The diagnosis support device 1 includes a processing circuit 11, a storage circuit 12, an input interface 13, a display 14, and a network interface 15.

The processing circuit 11 controls the operation of the diagnostic support device 1 in response to an input operation received from the operator via the input interface 13. For example, the processing circuit 11 is realized by a processor. The function of the processing circuit 11 will be described later with reference to FIG.

The storage circuit 12 is composed of semiconductor memory elements such as a RAM (Random Access Memory) and a flash memory (Flash Memory), a hard disk, an optical disk, and the like. The storage circuit 12 may be composed of a portable medium such as a USB (Universal Serial Bus) memory and a DVD (Digital Video Disk). The storage circuit 12 stores various processing programs (including an OS (Operating System) and the like in addition to the application program) used in the processing circuit 11 and data necessary for executing the program. In addition, the OS may include a GUI (Graphical User Interface) that makes extensive use of graphics for displaying information on the display 14 to the operator and allows basic operations to be performed by the input interface 13. The storage circuit 12 is an example of a storage unit.

Further, as shown in FIG. 2, the memory circuit 12 includes gene expression / mutation information U1, epigenetic environmental effect information U2, protein expression information U3, signal transduction information U4, immune function information U5, and endocrine function. Information U6, pathological information U7, diagnostic imaging information U8, physiological information U9, physical findings / medical condition information U10, and a biological condition determination database (Data Base) U11 are stored. The information U1 to U10 may be acquired from a subject to be diagnosed by various sensors, may be input by an operator via an input interface 13, or may be input by an operator, or may include a network interface 15. It may be obtained from an external device via the device.

The gene expression / mutation information U1 is, for example, biological information indicating the expression level and the mutation amount of a specific gene of a subject to be diagnosed. For example, the gene expression / mutation information U1 is acquired by a genetic test or the like using a sequencer, a gene panel, or the like. Then, the acquired gene expression / mutation information U1 is stored in the storage circuit 12.

Epigenetic environmental impact information U2 is biological information indicating the regulation of gene expression by acquired environmental factors that affect the epigenome of a subject such as histone modification and methylation. For example, the operator asks the subject about the number of cigarettes the subject smokes per day, the amount of ultraviolet rays exposed per day, and the like. Alternatively, phenomena such as histone modification and methylation as a result are detected, and the obtained environmental factor information and information indicating the test result are stored in the storage circuit 12 as epigenetic environmental impact information U2.

Protein expression information U3 is biological information regarding protein expression of a subject. For example, the protein expression information U3 is information indicating the ratio of the amount of a specific protein such as in the blood of a subject to a reference amount of the specific protein. The amount of a particular protein in the subject is obtained by in vitro or spectroscopic analysis of the biomarker and a mass spectrometer (eg, mass spectrum scoop, etc.). In addition, the reference amount of a specific protein is, for example, the amount of a specific protein when the subject is in a healthy state. Then, the information indicating the ratio of the obtained specific protein amount to the reference amount of the specific protein is stored in the storage circuit 12 as the protein expression information U3.

Signal transduction information U4 is biological information regarding intercellular and intracellular signal transduction of a subject. The signal transduction information U4 is acquired by an antibody array or the like. Then, the acquired signal transmission information U4 is stored in the storage circuit 12.

The immune function information U5 is biological information regarding the immune function of the subject. For example, the immune function information U5 is information indicating the number of white blood cells per unit amount of blood of the subject. The number of such white blood cells is obtained by a blood test or the like. The immune function information U5 may be information indicating the expression level and mutation amount of a specific gene related to immunity, or the expressed antibody level itself. Such information is acquired by a sample test device, an antibody test, a gene chip, or the like. Then, the acquired immune function information U5 is stored in the storage circuit 12.

The endocrine function information U6 is biological information regarding the endocrine function of the subject. For example, the endocrine function information U6 is information indicating the amount of a specific hormone per unit amount of blood of a subject. Information indicating the amount of such a specific hormone is obtained by a blood test or the like.

Pathological information U7 is various biological information of a subject obtained by a pathological examination or a cytopathological examination. For example, in a pathological examination, cells of a tumor are collected, and it is determined whether or not the collected cells are malignant. Information indicating the result of such determination is stored in the storage circuit 12 as pathological information U7.

The diagnostic imaging information U8 is biological information obtained by image analysis of a radiological diagnostic image in which an organ of a subject is depicted and a radiological diagnostic image. For example, in an examination, a doctor makes an image diagnosis of a CT (Computed Tomography) image in which the heart of a subject is depicted. In diagnostic imaging, the doctor diagnoses whether there is an abnormality in the heart. For example, a doctor pays attention to the aortic valve and makes a diagnosis of whether or not it is valvular heart disease and whether or not it is aortic valve stenosis. Then, the information indicating the diagnosis result is stored in the storage circuit 12 as the image diagnosis information U8. The image diagnosis may include an image diagnosis of an image other than a CT image. For example, for diagnostic imaging, MR images obtained by an MRI (Magnetic Resonance Imaging) device, X-ray diagnostic images obtained by an X-ray diagnostic device, and ultrasonic diagnostic images obtained by an ultrasonic diagnostic device are used. The diagnostic imaging of at least one image out of the nuclear medical diagnostic images obtained by the nuclear medical diagnostic apparatus may be included.

Physiological information U9 refers to biological information about the heart based on the electrocardiographic waveform of the subject obtained by, for example, an electrocardiogram. In the electrocardiogram examination, the electrocardiographic waveform of the subject is obtained by an electrocardiograph. In addition, information about the heart such as the RR interval can be obtained from the electrocardiographic waveform. Then, such information about the heart is stored in the storage circuit 12 as physiological information U9. In addition to this, the physiological information U9 includes brain wave information, respiratory monitor information, body temperature, blood pressure, and other information indicating various physiological phenomena.

Physical findings / medical condition information U10 is biological information indicating physical findings and medical conditions obtained by interviewing a subject. For example, a doctor obtains physical findings and medical conditions by asking a subject. For example, physical findings include information at the human body level, such as the height and weight of the subject. The information indicating the physical findings and the medical condition thus obtained is stored in the storage circuit 12 as the physical findings / medical condition information U10.

As described above, the storage circuit 12 stores the information U1 to U10. In this way, the storage circuit 12 stores biological information at the gene level of the subject, biological information at the molecular / cell level, biological information at the organ level, and biological information at the human body level. That is, the storage circuit 12 stores a plurality of types of comprehensive biological information from the gene level of the subject to the human body level. To give a specific example, the memory circuit 12 includes biological information related to the gene of the subject, biological information related to protein, biological information related to signal transduction, biological information related to endocrine function, biological information related to immune function, and biological information related to environmental effects. In addition, a plurality of types of biological information including biological information regarding the human body are stored.

The biological state determination database U11 is a database used when determining the biological state (status) for a disease. The biological state means the degree of relationship of the subject to a certain disease, and is an index showing the degree to which the subject is related to a certain disease. More specifically, this degree of relationship is an index indicating the degree (possibility) of the subject suffering from a certain disease, that is, an index indicating the severity of the symptom of the disease A. For example, as shown in FIG. 5, the biological state includes "health state (no illness)", "intermediate state", and "disease state" according to the severity of the symptom of "disease A" of the subject. It is classified into four categories, "serious / dead". The details will be described later.

FIG. 3 is a diagram showing an example of the data structure of the biological state determination database U11.

In FIG. 3, in the item of "gene expression / mutation", a quantitative score to be described later calculated from the gene expression / mutation information U1 is registered. In the item of "epigenetic environmental impact", a quantitative score to be described later calculated from the epigenetic environmental impact information U2 is registered. In the item of "protein expression (biomarker)", a quantitative score to be described later calculated from the protein expression information U3 is registered. In the item of "signal transduction", a quantitative score to be described later calculated from the signal transduction information U4 is registered.

Further, in the item of "immune function", a quantitative score to be described later calculated from the immune function information U5 is registered. In the item of "endocrine function", a quantitative score to be described later calculated from the endocrine function information U6 is registered. In the item of "pathological change", a quantitative score to be described later calculated from the pathological information U7 is registered. In the item of "imaging diagnosis", a quantitative score to be described later calculated from the diagnostic imaging information U8 is registered. In the item of "electrocardiogram", a quantitative score to be described later calculated from the physiological information U9 is registered. In the item of "physical findings / medical conditions", a quantitative score, which will be described later, calculated from the physical findings / medical condition information U10 is registered.

In the biological state determination database U11, a correlation coefficient showing a correlation (correlation) between each of the plurality of diseases A to N and each of the plurality of types of biological information is registered. Further, in the biological state determination database U11, the reliability indicating the certainty of the correlation coefficient is registered. For example, in the example of FIG. 3, assuming that the correlation coefficient is "R" and the reliability is "S", the phase is expressed as "R / S" for each combination of disease and biological information. The number of relationships and the reliability are registered in the biological condition determination database U11.

For example, the biological state determination database U11 has gene expression / mutation information U1 in a combination of gene expression / mutation information U1 and "disease A", which are the sources of quantitative scores registered in the item of "gene expression / mutation". It is shown that the correlation coefficient showing the correlation between "disease A" and "disease A" is "0.8". The same is true for other combinations of disease and biometric information.

In the present embodiment, for example, the range of the correlation coefficient is a range of "-1" or more and "1" or less. For example, a case where the correlation coefficient indicating the correlation between a certain disease and a certain kind of biological information is a positive value will be described. In this case, it is considered that the closer the correlation coefficient is to "1", the higher the possibility of suffering from the disease due to the biological information of the type, and the more severe the symptoms of the disease.

In addition, a case where the correlation coefficient indicating the correlation between a certain disease and a certain type of biological information is a negative value will be described. In this case, it is considered that the closer the correlation coefficient is to "-1", the higher the possibility of becoming less susceptible to the disease due to the type of biological information.

Further, when the correlation coefficient indicating the correlation between a certain disease and a certain type of biometric information is "0" or a numerical value close to "0", this correlation coefficient is the correlation coefficient between the type of biometric information and the disease. Indicates that there is no correlation.

Further, for example, in the biological state determination database U11, the reliability indicating the certainty of the correlation coefficient “0.8” indicating the correlation between the gene expression / mutation information U1 and “disease A” is “A”. Is shown. The same applies to other reliability. Here, in the present embodiment, the reliability is indicated by any of "A", "B", and "C". For example, the reliability "A" is higher than the reliability "B", and the reliability "B" is higher than the reliability "C". The reliability "A", "B", and "C" indicate numerical values. For example, the reliability "A" is "1.0", the reliability "B" is "0.7", and the reliability "C" is "0.3". For example, confidence is also determined by the scientific basis for the correlation coefficient. In addition, the reliability may be determined from, for example, guidelines, the contents of clinical research registered in a database, the contents of a dissertation, and the like.

Returning to the description of FIG. 1, the input interface 13 includes an input device that can be operated by an operator and an input circuit that inputs a signal from the input device. Input devices include trackballs, switches, mice, keyboards, touchpads that perform input operations by touching the operation surface, touchscreens that integrate the display screen and touchpad, and non-contact input devices that use optical sensors. And it is realized by a voice input device and the like. When the input device is operated by the operator, the input circuit generates a signal corresponding to the operation and outputs the signal to the processing circuit 11. The diagnostic support device 1 may include a touch panel in which the input device is integrally configured with the display 14. Further, the input device is not limited to the one provided with physical operating parts such as a mouse and a keyboard. For example, an example of the input interface 13 is a configuration in which the input circuit receives an electric signal corresponding to the input operation from an external input device provided separately from the diagnostic support device 1 and outputs the electric signal to the processing circuit 11. include. The input interface 13 is an example of an input unit.

The display 14 is a display device such as a liquid crystal display panel, a plasma display panel, and an organic EL (Electro Luminescence) panel. The display 14 is connected to the processing circuit 11 and displays various information and images generated under the control of the processing circuit 11. The display 14 is an example of a display unit.

The network interface 15 is composed of connectors that meet the parallel connection specifications and the serial connection specifications. The network interface 15 has a function of performing communication control according to each standard and being able to connect to the network through a telephone line, whereby the diagnostic support device 1 can be connected to the network. The network interface 15 is an example of a communication unit.

Subsequently, the function of the diagnostic support device 1 will be described with reference to FIG.

FIG. 2 is a block diagram showing the functions of the diagnosis support device 1.

When the processing circuit 11 executes the program, the diagnosis support device 1 realizes the score calculation function 111, the acquisition function 112, the biological state determination function 113, and the display control function 114. All or part of the functions 111 to 114 may be realized in the diagnostic support device 1 as a function of a circuit such as an ASIC. The score calculation function 111 is an example of an analysis unit. The acquisition function 112 is an example of an acquisition unit. The biological state determination function 113 is an example of a determination unit. The display control function 114 is an example of a display control unit. The score calculation function 111, the acquisition function 112, the biological state determination function 113, and the display control function 114 will be described later.

Here, for example, the functions 111 to 114, which are the components of the processing circuit 11, are stored in the storage circuit 12 in the form of a program that can be executed by a computer. The processing circuit 11 reads each program from the storage circuit 12 and executes each read program to realize a function corresponding to each program. In other words, the processing circuit 11 in the state where each program is read has the functions 111 to 114 shown in the processing circuit 11.

It should be noted that all the processing functions of the functions 111 to 114 may be stored in the storage circuit 12 in the form of one program that can be executed by a computer. For example, such a program is also called a diagnostic support program. In this case, the processing circuit 11 reads the diagnosis support program from the storage circuit 12 and executes the read diagnosis support program to realize the functions 111 to 114 corresponding to the diagnosis support program.

The example of the configuration of the diagnosis support device 1 according to the first embodiment has been described above. As a result of diligent studies, the inventors of the present application have found that all diseases involve complex factors including hereditary factors and non-hereditary factors. Then, the inventors of the present application determine the biological state of the subject by using a plurality of types of comprehensive biological information from the gene level of the subject to the human body level, and obtain a determination result with good accuracy. I found that it was possible. Therefore, the diagnostic support device 1 according to the first embodiment will be described below using a plurality of types of comprehensive biological information from the gene level of the subject to the human body level in order to accurately determine the biological state. Perform various processes.

The score calculation function 111 performs an analysis for determining the degree of abnormality of the biometric information for each of the plurality of types of biometric information stored in the storage circuit 12. Explaining with a specific example, the score calculation function 111 performs an analysis for each of a plurality of types of biological information to obtain a score (quantitative score) indicating the degree of abnormality of the biological information, and quantitatively determines the analysis result. Get a score. For example, the score calculation function 111 calculates a quantitative score within the range of "0" or more and "1" or less. For example, the quantitative score indicates that the degree of the abnormal state indicated by the biological information increases as it approaches "1". The score calculation function 111 calculates such a quantitative score for each of the information U1 to U10. The quantitative score is used when the biological condition is determined by the biological condition determination function 113 described later.

To give an example, the score calculation function 111 has a larger value as the expression level and the mutation amount of a specific gene indicated by the gene expression / mutation information U1 increase as an analysis for the gene expression / mutation information U1. The quantitative score is calculated as the analysis result. Then, the score calculation function 111 registers the calculated quantitative score in the item of "gene expression / mutation" of the biological state determination database U11.

Further, as an analysis of the epigenetic environmental impact information U2, the score calculation function 111 has a value as the number of cigarettes or the amount of ultraviolet rays indicated by the epigenetic environmental impact information U2 increases, or as the methylation score or the like increases. A quantitative score that increases is calculated as the analysis result. Then, the score calculation function 111 registers the calculated quantitative score in the item of "epigenetic environmental impact" of the biological state determination database U11.

Further, as an analysis for the protein expression information U3, the score calculation function 111 calculates as an analysis result a quantitative score such that the value increases as the ratio indicated by the protein expression information U3 increases from "1". Then, the score calculation function 111 registers the calculated quantitative score in the item of "protein expression (biomarker)" in the biological state determination database U11.

Further, as an analysis for the signal transduction information U4, the score calculation function 111 calculates as an analysis result a quantitative score such that the value increases as the degree of abnormality indicated by the information related to the signal transduction indicated by the signal transduction information U4 increases. Then, the score calculation function 111 registers the calculated quantitative score in the item of "signal transduction" of the biological state determination database U11.

Further, as an analysis for the immune function information U5, the score calculation function 111 calculates as an analysis result a quantitative score such that the value increases as the degree of abnormality indicated by the number of white blood cells indicated by the immune function information U5 increases. When the immune function information U5 indicates the expression level and the mutation amount of a specific gene related to immunity, the score calculation function 111 increases the expression level and the mutation amount of the specific gene related to immunity. A quantitative score with a large value is calculated as the analysis result. Then, the score calculation function 111 registers the calculated quantitative score in the item of "immune function" of the biological state determination database U11.

Further, as an analysis for the endocrine function information U6, the score calculation function 111 calculates as an analysis result a quantitative score such that the value increases as the degree of abnormality indicated by the amount of the specific hormone indicated by the endocrine function information U6 increases. .. Then, the score calculation function 111 registers the calculated quantitative score in the item of "endocrine function" of the biological state determination database U11.

Further, as an analysis for the pathological information U7, the score calculation function 111 calculates as an analysis result a quantitative score such that the value increases as the degree of abnormality indicated by the determination result indicated by the pathological information U7 increases. Then, the score calculation function 111 registers the calculated quantitative score in the item of "pathological change" in the biological state determination database U11.

Further, as an analysis for the diagnostic imaging information U8, the score calculation function 111 calculates as an analysis result a quantitative score such that the value increases as the degree of abnormality indicated by the diagnostic result indicated by the diagnostic imaging information U8 increases. Then, the score calculation function 111 registers the calculated quantitative score in the item of "imaging diagnosis" of the biological state determination database U11.

Further, as an analysis for the physiological information U9, the score calculation function 111 calculates as an analysis result a quantitative score such that the value increases as the degree of abnormality indicated by the information about the heart indicated by the physiological information U9 increases. Then, the score calculation function 111 registers the calculated quantitative score in the item of "electrocardiogram" of the biological state determination database U11.

When the physiological information U9 indicates the electroencephalogram information, the score calculation function 111 calculates as an analysis result a quantitative score such that the value increases as the degree of abnormality indicated by the electroencephalogram information increases. Then, the score calculation function 111 registers the calculated quantitative score in the item (not shown) of the "electroencephalograph" provided in the same manner as the item of the "electrocardiogram" of the biological state determination database U11.

Further, as an analysis of the physical findings / medical condition information U10, the score calculation function 111 analyzes a quantitative score such that the value increases as the degree of the physical findings and the abnormalities indicated by the physical condition information U10 increases. Calculate as. Then, the score calculation function 111 registers the calculated quantitative score in the item of "physical findings / medical conditions" of the biological state determination database U11.

By the method as described above, the score calculation function 111 calculates a plurality of quantitative scores corresponding to a plurality of types of biological information, and registers the plurality of quantitative scores in the biological state determination database U11.

The score calculation function 111 does not have to automatically calculate the quantitative score as described above. For example, the score calculation function 111 may receive a quantitative score corresponding to biometric information from the operator via the input interface 13. Then, the score calculation function 111 may register the received quantitative score in the biological state determination database U11.

The acquisition function 112 acquires a correlation coefficient showing a correlation between each information of the plurality of information U1 to U10 and a plurality of diseases (disease A to N) from the biological state determination database U11 shown in FIG. In addition, the acquisition function 112 acquires a plurality of quantitative scores corresponding to a plurality of types of biometric information calculated by the score calculation function 111.

The biological state determination function 113 determines the biological state of the subject for each of the plurality of diseases (disease A to disease N) by using the biological state determination database U11.

Explaining with a specific example, the acquisition function 112 acquires a correlation coefficient “0.8” indicating the correlation between the gene expression / mutation information U1 and the disease A from the biological state determination database U11 shown in FIG. .. Then, the acquisition function 112 acquires a quantitative score calculated from the gene expression / mutation information U1 from the item of “gene expression / mutation”. Then, the biological state determination function 113 calculates the multiplication value r1 of the acquired quantitative score and the acquired correlation coefficient “0.8”.

The acquisition function 112 was calculated from the correlation coefficient showing the correlation between the other biological information and the disease A and the other biological information in the same manner for the biological information other than the gene expression / mutation information U1. Get a quantitative score and. Then, the biological state determination function 113 similarly calculates the multiplication value of the acquired correlation coefficient and the acquired quantitative score.

Specifically, the biological state determination function 113 calculates a multiplication value r2 of the quantitative score calculated from the epigenetic environmental impact information U2 and the correlation coefficient “0.5”. Further, the biological state determination function 113 calculates a multiplication value r3 of the quantitative score calculated from the protein expression information U3 and the correlation coefficient “0.65”. Further, the biological state determination function 113 calculates a multiplication value r4 of the quantitative score calculated from the signal transmission information U4 and the correlation coefficient “0.2”.

Further, the biological state determination function 113 calculates a multiplication value r5 of the quantitative score calculated from the immune function information U5 and the correlation coefficient “0.4”. Further, the biological state determination function 113 calculates a multiplication value r6 of the quantitative score calculated from the endocrine function information U6 and the correlation coefficient “0.01”. Further, the biological state determination function 113 calculates a multiplication value r7 of the quantitative score calculated from the pathological information U7 and the correlation coefficient “0.9”.

Further, the biological state determination function 113 calculates a multiplication value r8 of the quantitative score calculated from the diagnostic imaging information U8 and the correlation coefficient “0.3”. Further, the biological state determination function 113 calculates a multiplication value r9 of the quantitative score calculated from the physiological information U9 and the correlation coefficient “0.0”. Further, the biological state determination function 113 calculates a multiplication value r10 of the quantitative score calculated from the physical findings / medical condition information U10 and the correlation coefficient “0.1”.

Then, the biological state determination function 113 is a 10-dimensional coordinate space when points r (r1, r2, r3, r4, r5, r6, r7, r8, r9, r10) are arranged in the 10-dimensional coordinate space. Calculate the distance between the origin O and the point r. The calculated distance is also the magnitude of the vector having (r1, r2, r3, r4, r5, r6, r7, r8, r9, r10) as a component.

Then, the biological state determination function 113 normalizes the calculated distance so that the calculated distance is within the range of “0” or more and “1”, and calculates the normalized distance as the degree of relationship. This degree of relationship is an index showing the degree to which the subject is related to the disease A, and is an index showing the biological state of the subject with respect to the disease A. More specifically, this degree of relationship is an index showing the degree (possibility) of the subject suffering from the disease A and an index showing the severity of the symptoms of the disease A. The closer the relationship is to "1", the higher the degree of the subject suffering from the disease A, and the more severe the symptoms of the disease A. On the other hand, the closer the relationship is to "0", the lower the degree of the subject suffering from the disease A and the less severe the symptoms of the disease A.

Here, an example of a method of calculating the degree of relationship from the quantitative score will be described with reference to FIG. Using FIG. 3 above, 10 quantitative scores generated from 10 types of biological information are registered in the biological condition determination database U11, and the biological condition determination function 113 uses the 10 quantitative scores. The case of calculating the degree of relationship has been described. On the other hand, in the explanation using FIG. 4, for convenience of explanation, it is assumed that three quantitative scores SC1_1, SC2_1, SC3_1 generated from three types of biological information are registered in the biological condition determination database U11. .. Then, in the explanation using FIG. 4, the biological state determination function 113 will be described as calculating the degree of relationship for a single disease by using the three quantitative scores SC1-1, SC2_1, SC3_1.

FIG. 4 is a diagram for explaining an example of a method of calculating the degree of relationship from the quantitative score.

FIG. 4 shows a three-dimensional space formed by the axes SC1, the axis SC2, and the axis SC3. The axis SC1 is an axis indicating the magnitude of the quantitative score SC1_1. The axis SC2 is an axis indicating the magnitude of the quantitative score SC2_1. The axis SC3 is an axis indicating the magnitude of the quantitative score SC3_1. The three axes SC1, axis SC2, and axis SC3 are orthogonal to each other.

In the example of FIG. 4, the biological state determination function 113 calculates a multiplication value R1'(not shown) of the quantitative score SC1_1 and the correlation coefficient corresponding to the quantitative score SC1_1. Further, the biological state determination function 113 calculates a multiplication value R2'(not shown) of the quantitative score SC2_1 and the correlation coefficient corresponding to the quantitative score SC2_1. Further, the biological state determination function 113 calculates a multiplication value R3'(not shown) of the quantitative score SC3_1 and the correlation coefficient corresponding to the quantitative score SC3_1.

Then, the biological state determination function 113 arranges the origin O and the point R of the three-dimensional coordinate space when the points R'(R1', R2', R3') (not shown) are arranged in the three-dimensional coordinate space. Calculate the distance to ´. The calculated distance is also the magnitude of the vector having (R1', R2', R3') as a component.

Then, the biological state determination function 113 normalizes the calculated distance so that the calculated distance is within the range of “0” or more and “1”, and calculates the normalized distance D1 as the degree of relationship. Note that FIG. 4 shows points R (R1, R2, R3) corresponding to the normalized distance D1. The point R is the position where the position of the point R'is changed according to the normalization. That is, the point R is a point corresponding to the point R'.

Here, in FIG. 4, the sphere SP is a sphere having a radius of 1. In this embodiment, the point R is located inside and on the surface of the sphere SP.

Then, as shown in FIG. 4, the intersection of the line segment connecting the origin O and the point R with the surface of the sphere SP and the line segment further extended from the point R side is defined as the point P. At this time, the shorter the distance D2 between the point R and the point P, the higher the degree of the subject suffering from the disease, and the more severe the symptoms of the disease.

With reference to FIG. 4, the biological condition determination function 113 determines the position of the point R (R1, R2, R3) indicating the current biological condition of the subject in the sphere SP representing a single disease. By doing so, the case of determining the biological condition for a single disease has been described. Then, the biological state determination function 113 also determines the biological state for a plurality of other diseases in the same manner. That is, the biological state determination function 113 determines at which position the point indicating the current biological state of the subject is located in each of the plurality of spheres SP corresponding to each of the diseases A to N. , The biological state for each of the plurality of diseases arranged on the surface of the sphere SP is determined.

In addition, one sphere SP may represent all diseases A to N. For example, the biological state determination function 113 determines at which position the point corresponding to the degree of relationship for each of all diseases A to N is located in one sphere SP, thereby causing all diseases A to N. The biological condition for each of the diseases N may be determined.

Returning to the description of FIG. 2, the biological state determination function 113 determines the biological state of the subject with respect to the disease A based on the calculated degree of relationship. For example, the biological state determination function 113 determines the biological state by comparing a plurality of threshold values with the degree of relationship.

For example, when the degree of relationship is within the range of the threshold value “0” or more and less than the threshold value “0.4”, it is considered that the subject does not suffer from the disease A. Therefore, in the biological state determination function 113, when the degree of relationship is within the range of the threshold value “0” or more and less than the threshold value “0.4”, the biological state of the subject is not affected by the disease A ( It is determined that the patient is in good health. The health condition is an example of the first condition.

Further, for example, when the degree of relationship is within the range of the threshold value “0.6” or more and less than the threshold value “0.9”, the subject suffers from the disease A, and the symptom of the subject disease A. Is considered to be mild. Therefore, in the biological state determination function 113, when the degree of relationship is within the range of the threshold value “0.6” or more and less than the threshold value “0.9”, the biological state of the subject is affected by the disease A. , It is determined that the symptom of the disease A is a mild state (disease state). The disease state is an example of a second state.

Further, for example, when the degree of relationship is within the range of the threshold value "0.4" or more and less than the threshold value "0.6", the biological state of the subject is a state (intermediate state) between the health state and the disease state. ). The intermediate state refers to, for example, a state in which the subject is not affected by the disease A, but is more likely to shift to the disease state than the above-mentioned "health state". Therefore, the biological state determination function 113 determines that the biological state of the subject is an intermediate state when the degree of relationship is within the range of the threshold value “0.4” or more and less than the threshold value “0.6”. .. The intermediate state is an example of the third state.

Further, for example, when the degree of relationship is within the range of the threshold value “0.9” or more and the threshold value “1.0” or less, the subject suffers from the disease A, and the symptom of the subject disease A. Is considered to be severe or the subject is dead. Therefore, in the biological state determination function 113, when the degree of relationship is within the range of the threshold value “0.9” or more and the threshold value “1.0” or less, the biological state of the subject is affected by the disease A. In addition, it is determined that the symptom of the disease A is severe or dead (severe / dead). Severe / death is an example of the fourth condition.

By the method described above, the biological condition determination function 113 determines the biological condition of the subject with respect to the disease A based on the plurality of quantitative scores and the plurality of correlation coefficients. More specifically, the biological condition determination function 113 calculates the degree of relationship between the subject and the disease A based on a plurality of quantitative scores and a plurality of correlation coefficients, and based on the calculated degree of relationship, the disease. The biological condition of the subject with respect to A is determined. Then, the biological state determination function 113 calculates the degree of relationship between the subject and each of the diseases B to N by the same method, and based on the calculated degree of relationship, the subject's for each of the diseases B to N Determine the biological condition. As described above, the biological condition determination function 113 is applied to each of the diseases A to N based on a plurality of quantitative scores obtained by comprehensive analysis of a plurality of types of biological information from the gene level to the human body level of the subject. Determine the biological condition of the subject. Therefore, the diagnostic support device 1 according to the present embodiment can accurately determine the biological state of the subject.

In addition, the biological state determination function 113 uses the biological state determination database U11 to calculate the total reliability indicating the certainty of the biological state determined for each of the plurality of diseases (disease A to disease N).

First, an example of a method for calculating the total reliability indicating the certainty of the biological condition determined for the disease A will be described.

For example, the acquisition function 112 acquires the correlation coefficient “0.8” corresponding to the gene expression / mutation information U1 for the disease A from the biological state determination database U11. In addition, the acquisition function 112 acquires the reliability "A" indicating the certainty of the acquired correlation coefficient "0.8" from the biological state determination database U11. Then, the biological state determination function 113 calculates the multiplication value r11 of the acquired absolute value of the correlation coefficient “0.8” and the acquired reliability “A”.

The acquisition function 112 also uses the same method for other biological information other than the gene expression / mutation information U1 to obtain a correlation coefficient showing the correlation between the other biological information and the disease A and the certainty of the correlation coefficient. Get the confidence shown. Then, the biological state determination function 113 similarly calculates the multiplication value of the acquired absolute value of the correlation coefficient and the acquired reliability.

Specifically, the biological state determination function 113 calculates a multiplication value r12 of the absolute value of the correlation coefficient “0.5” corresponding to the epigenetic environmental impact information U2 and the reliability “B”. Further, the biological state determination function 113 calculates a multiplication value r13 of the absolute value of the correlation coefficient “0.65” corresponding to the protein expression information U3 and the reliability “A”. Further, the biological state determination function 113 calculates a multiplication value r14 of the absolute value of the correlation coefficient “0.2” corresponding to the signal transmission information U4 and the reliability “B”.

Further, the biological state determination function 113 calculates a multiplication value r15 of the absolute value of the correlation coefficient “0.4” corresponding to the immune function information U5 and the reliability “B”. Further, the biological state determination function 113 calculates a multiplication value r16 of the absolute value of the correlation coefficient “0.01” corresponding to the endocrine function information U6 and the reliability “C”. Further, the biological state determination function 113 calculates a multiplication value r17 of the absolute value of the correlation coefficient “0.9” corresponding to the pathological information U7 and the reliability “A”.

Further, the multiplication value r18 of the absolute value of the correlation coefficient “0.3” corresponding to the biological state determination function 113 and the diagnostic imaging information U8 and the reliability “B” is calculated. Further, the biological state determination function 113 calculates a multiplication value r19 of the absolute value of the correlation coefficient “0.0” corresponding to the physiological information U9 and the reliability “C”. Further, the biological state determination function 113 calculates a multiplication value r20 of the absolute value of the correlation coefficient “0.1” corresponding to the physical findings / medical condition information U10 and the reliability “B”.

Then, the biological state determination function 113 calculates the sum Q of 10 multiplication values r11 to r20 as the total reliability according to the following equation (1).
Q = r11 + r12 + r13 + r14 + r15 + r16 + r17 + r18 + r19 + r20 ... (1)

The biological state determination function 113 may use the sum Q as it is as the total reliability, or the value obtained by normalizing the sum Q to any of a plurality of stages A to E as the total reliability. You may use it. For example, the total reliability "A" has a higher reliability than the total reliability "B", and the total reliability "B" has a higher reliability than the total reliability "C". Further, the total reliability "C" has a higher reliability than the total reliability "D", and the total reliability "D" has a higher reliability than the total reliability "E". The overall reliability "A", "B", "C", "D", and "E" indicate numerical values. For example, the total reliability "A" is "1.0", the total reliability "B" is "0.8", and the total reliability "C" is "0.6". Further, for example, the total reliability "D" is "0.4", and the total reliability "E" is "0.2".

By the method described above, the biological condition determination function 113 calculates the total reliability indicating the certainty of the biological condition determined for the disease A based on the plurality of reliabilitys and the plurality of correlation coefficients. Then, the biological state determination function 113 calculates the total reliability indicating the certainty of the biological state determined for each of the diseases B to N by the same method.

The display control function 114 causes the display 14 to display the relationship degree and the total reliability calculated for each of the plurality of diseases A to N, and the biological state determined for each of the plurality of diseases A to N.

FIG. 5 is a diagram showing an example of display of the degree of relationship and the total reliability according to the first embodiment.

For example, as shown in FIG. 5, the display control function 114 associates the relationship degree “0.75” with the disease state for the disease A and displays the disease A in the display area 14a of the display 14. Thereby, the operator can easily determine that the biological state of the subject for the disease A is the disease state.

Further, the display control function 114 associates the degree of relationship "0.86" with the disease state for the disease N and displays the disease N in the display area 14a of the display 14. Thereby, the operator can easily determine that the biological state of the subject for the disease N is the disease state.

In the example of FIG. 5, although the relationship and the biological state of the diseases B to M are not shown, the relationship between the determined biological state and the calculated biological state is the same as that of the disease A and the disease N. Are associated with each other and displayed in the display area 14a.

In this way, it is possible to visualize how close the current biological state of the subject is to each of the diseases A to N. This allows the operator to provide health advice and support to the subject so as to reduce the degree of relevance.

In the present embodiment, as described above, the diagnostic support device 1 can accurately determine the biological state of the subject. Then, the diagnosis support device 1 displays the biological state determined with high accuracy on the display 14. Therefore, the diagnosis support device 1 can make the operator accurately diagnose the disease.

Further, for example, as shown in FIG. 5, the display control function 114 causes the total reliability “C” for the disease A to be displayed in the display area 14a of the display 14. As a result, the operator can grasp that the certainty of the "disease state" determined for the disease A is "C".

In addition, the display control function 114 causes the overall reliability "E" for the disease N to be displayed in the display area 14a of the display 14. As a result, the operator can grasp that the certainty of the "disease state" determined for the disease N is "E".

In the example of FIG. 5, although the illustration of the total reliability is omitted for the diseases B to M, the total reliability is displayed in the display area 14a as in the case of the disease A and the disease N.

As described above, the diagnostic support device 1 according to the present embodiment displays the total reliability indicating the certainty of the determined biological state for each disease. Therefore, according to the diagnostic support device 1 according to the present embodiment, the operator can grasp the certainty of the determined biological state. As a result, for example, the operator can determine whether or not additional inspections need to be performed, depending on the overall reliability.

FIG. 6 is a diagram showing the operation of the diagnosis support device as a flowchart. In FIG. 6, the reference numeral “ST” with a number indicates each step of the flowchart. For example, the process shown in FIG. 6 is executed when the operator inputs an instruction for determining a biological state to the process circuit 11 via the input interface 13.

As shown in FIG. 6, the score calculation function 111 calculates a quantitative score for each of a plurality of types of biological information stored in the storage circuit 12 (step ST1). Then, the biological state determination function 113 calculates the degree of relationship using a plurality of quantitative scores and a plurality of correlation coefficients for each of the diseases A to N, and determines the biological state based on the degree of relationship (step). ST2).

Then, the biological state determination function 113 calculates the total reliability for each of the diseases A to N by using the plurality of correlation coefficients and the plurality of reliabilitys (step ST3). Then, the display control function 114 displays the biological state, the degree of relationship, and the total reliability on the display 14 for each of the diseases A to N (step ST4), and ends the process. For example, in step ST4, as described above, the display control function 114 causes the display 14 to display the determined biological state in association with the degree of relationship.

The diagnostic support device 1 according to the first embodiment has been described above. According to the diagnostic support device 1 according to the first embodiment, as described above, the biological state of the subject can be accurately determined.

(Modification 1 of the first embodiment)
In the above-described embodiment, the biological state determination function 113 determines any of the four biological states of the health state, the intermediate state, the disease state, and the severe / death state for each of the diseases A to N. The case of determining whether or not the above is the case has been described. However, the method for determining the biological condition by the biological condition determining function 113 is not limited to this. Therefore, in the modified example 1, an example of another determination method of the biological state by the biological state determination function 113 will be described.

For example, the biological state determination function 113 according to the first modification may determine at least which of the healthy state, the intermediate state, and the disease state. That is, the biological condition determination function 113 may not determine the severity / death of a subject who is clearly not seriously ill.

According to the first modification, the load of the biological state determination process can be reduced because the severity / death is not determined. Further, according to the first modification, the same effect as that of the first embodiment can be obtained.

(Modification 2 of the first embodiment)
Further, in the first embodiment described above, the case where the biological state determination function 113 determines the biological state by using 10 correlation coefficients for each disease has been described. However, the biological state determination function 113 may determine the biological state using a correlation coefficient corresponding to a reliability equal to or higher than a specific reliability among 10 correlation coefficients for each disease. Therefore, such a modification will be described as a modification 2 of the first embodiment. In the description of the modified example 2, the points different from those of the first embodiment will be mainly described.

Hereinafter, the reliability "A" is "1.0", the reliability "B" is "0.7", the reliability "C" is "0.3", and the specific reliability is "0.3". The case of "0.8" will be described. In this case, among the plurality of reliability "A", "B", and "C", the reliability of the specific reliability "0.8" or higher is the reliability "A". In the second modification, the biological state determination function 113 determines the biological state for each of the diseases A to N by using the correlation coefficient corresponding to the reliability “A”.

For example, when determining the biological condition for the disease A, the biological condition determination function 113 uses a correlation coefficient corresponding to the reliability “A” from the biological condition determination database U11 shown in FIG. For example, the biological state determination function 113, in order from the top, has a correlation coefficient “0.8” corresponding to the reliability “A”, a correlation coefficient “0.65” corresponding to the reliability “A”, and a reliability “A”. The correlation coefficient "0.9" corresponding to "A" is used. That is, the biological state determination function 113 determines the distance between the origin O and the point r'in the three-dimensional coordinate space when the point r'(r1, r3, r7) is arranged in the three-dimensional coordinate space. calculate. The calculated distance is also the magnitude of the vector having (r1, r3, r7) as a component.

Then, the biological state determination function 113 normalizes the calculated distance so that the calculated distance becomes a value in the range of “0” or more and “1” or less, as in the first embodiment. Calculate the distance as the degree of relationship. Then, the biological state determination function 113 determines the biological state for the disease A based on the calculated degree of relationship, as in the first embodiment. Then, the biological state determination function 113 determines the biological state for each of the diseases B to N in the same manner.

According to the second modification, the diagnostic support device 1 determines the biological state by using a correlation coefficient corresponding to a reliability equal to or higher than a specific reliability. Therefore, according to the diagnostic support device 1 according to the modified example 2, the biological state of the subject can be determined more accurately.

(Modification 3 of the first embodiment)
In the first embodiment described above, the biological state determination function 113 compares the plurality of threshold values “0.4”, “0.6” and “0.9” shown in FIG. 5 with the degree of relationship. , The case of determining the biological condition for the disease A has been described. Here, for example, in FIG. 5, when the degree of relationship with respect to disease A is "0.39", it is determined that the biological state of the subject is a healthy state, but it is also a state in which it is easy to shift to an intermediate state. .. The same applies to other biological conditions. The same applies to other diseases. That is, when the absolute value of the difference between the threshold value and the degree of relationship is equal to or less than a specific value (for example, 0.2), the subject is in a biological state different from the biological state determined by the biological state determination function 113. It is in a state where it is easy to shift to.

Therefore, in the modified example 3, the biological state determination function 113 performs the same processing as in the first embodiment, and at the same time, the operator grasps that the biological state is likely to shift to a biological state different from the determined biological state. In order to make it happen, the process described below is executed. In the description of the modified example 3, the points different from those of the first embodiment will be mainly described.

For example, the biological state determination function 113 calculates the absolute value of the difference between each of the plurality of threshold values and the degree of relationship. As a result, the absolute value of a plurality of differences is calculated. Then, the biological state determination function 113 determines whether or not each of the absolute values of the plurality of differences is equal to or less than a specific value. Then, when it is determined that at least one absolute value among the absolute values of the plurality of differences is equal to or less than a specific value, the biological state determination function 113 shifts to a biological state different from the determined biological state. Information indicating that the state is easy to operate is displayed on the display 14 as warning information. That is, the biological state determination function 113 causes the display 14 to display warning information when the absolute value of the difference between at least one of the plurality of threshold values and the degree of relationship is equal to or less than a specific value. As a result, the operator can grasp that the state is likely to shift to a biological state different from the determined biological state.

FIG. 7 is a diagram showing an example of warning information according to the third modification of the first embodiment.

As illustrated in FIG. 7, when the absolute value of the difference between the threshold value “0.6” and the relationship degree “0.58” is equal to or less than a specific value, the biological state determination function 113 provides warning information. The text data "It is an intermediate state close to the disease state" is generated, and the generated text data is displayed in the display area 14a of the display 14. As a result, the operator can grasp that the state is likely to shift to a disease state different from the determined intermediate state.

Here, in particular, in chronic diseases and cancer, it is important to maintain homeostasis in an attempt to maintain a normal state. In general, biological systems have a redundant and stable maintenance mechanism for maintaining vital activity. For example, even if the doctor can judge the subject to be in the same state at first glance, in reality, it is maintained in a truly healthy state without any problems, and the maintenance mechanism works to the maximum and somehow functions. There is a state of maintaining.

A complex system (redundant mechanism) having a plurality of redundant paths is used to realize the state and function of a living body. While this redundant mechanism contributes to the maintenance of homeostasis, it often tries to maintain the state and function until it is about to collapse, and works in the direction of exacerbating the disease. Therefore, it is important to accurately determine the detailed state.

In the third modification, for example, even if the biological state of the subject is determined to be a disease state, the absolute value of the difference between the threshold value indicating the boundary between the disease state and severe / death and the degree of relationship is equal to or less than a specific value. In some cases, the operator can grasp that the condition is likely to lead to serious illness or death. That is, the operator can grasp the information that the maintenance mechanism works to the maximum and somehow maintains the function.

(Second Embodiment)
In the second embodiment, the biological state is determined without using the quantitative score. Therefore, in the second embodiment, the processing circuit 11 does not include the score calculation function 111 shown in FIG. 2, but includes an acquisition function 112, a biological state determination function 113, and a display control function 114. And the acquisition function 112 does not need to acquire the quantitative score.

In the second embodiment, the biological state determination function 113 estimates the biological state by using the equation of state and the observation equation.

In the second embodiment, for example, in estimating the current biological state of a subject, a Kalman filter suitable for estimating an amount that changes from moment to moment from observations with discrete errors is called a non-linear biological body. An example of applying a state space model applied to a dynamic system will be described. For example, in a time series analysis that predicts the future from past data, if the past data does not exist in the time series or there are missing values, the prediction cannot be made or the prediction accuracy is greatly deteriorated. However, there are cases where only discrete data exists, and there are cases where the observed results do not match the actual state. In the second embodiment, a flexible model is constructed in which the equation of state and the observation equation are modeled separately, the observed value is estimated, and the part where the measured value is present is reflected in the prediction level. For the equation of state, Markov property is assumed.

FIG. 8 is a diagram for explaining an example of a method for estimating a biological state in the second embodiment.

FIG. 8 conceptually shows the state space model. In the second embodiment, the biological state determination function 113 uses the following state equation (2) and observation equation (3) to show the biological state S of the subject at the current time t, as shown in FIG. Estimate _t and the observed value D _{t.
S t = G (S t-} 1, a t, u t) ... (2)
D _t = F (D _t-1 , _bt , w _t ) ... (3)

In the equation of state (2) described above, " _St-1 " is a variable indicating the biological state of the subject at the time t-1 one step before the current time t. "A _t" is an explanatory variable of the current time t. "U _t" is the system noise and biological fluctuation of the current time t (hereinafter, simply referred to as "system noise") is a variable indicating a.

For example, in the present embodiment, each of the four biological states (health state, intermediate state, disease state, severe / death) is quantified and used. Therefore, " _St-1 " and " _St " are treated as numerical values. "G", and "S _t-1", and "a _t", with the "u _t" is a function that outputs the biological condition S _t of the subject of the current time. Further, the initial value S ₀ is input to the processing circuit 11 by the operator via the input interface 13, and is stored in the storage circuit 12 by the biological state determination function 113.

In the above-mentioned observation equation (3), the observed value D _t is a variable indicating the biological information of the subject at the current time t. For example, the observed value D _t is biological information indicating "HbA1c" obtained by a blood test. "D _t-1 " is " _St ". “ _Bt ” is an explanatory variable at the current time t. "W _t" is a variable that indicates the observation noise at the current time t. "F" is a function that outputs the observed value D _{t by using "D t-1} ", " _bt ", and "w _t".

As shown in FIG. 8, by incorporating an unobservable state into the model, a more complicated time-series model of the biological state can be constructed. In addition, the biological state determination function 113 can correct the current biological state based on the past biological state based on the current observed value, and can determine the current biological state with higher accuracy.

For example, the biological condition determination function 113, using the following equation (4), to estimate more accurate, corrected biological condition S _A of the current time.
_{S A = S R + K (} D R -D P) ... (4)

In the above formula (4), "S _A" is a variable indicating the biological condition of the subject in the corrected current time. "S _B" is a variable indicating the biological condition of the subject at the current time before the correction. "K" is Kalman gain. "D _R" is the actual observed values. That is, "D _R" is the biological information indicating "HbA1c" of the current time obtained by the actual blood tests. "D _P" is a predicted observations. That is, "D _P" is biological information indicating "HbA1c" of the current time predicted by the above observation equation (3).

The biological state determination function 113 can determine the current biological state with higher accuracy by correcting the current biological state using the above formula (4).

Then, the biological state determination function 113 further corrects the current biological state so as to be consistent with a plurality of quantitative scores, and statistically estimates other unobserved missing values from that state. By doing so, the biological state at the current time is determined regardless of observation or non-observation.

If the biological condition at the current time is a disease state (disease state and severe / death), the operator makes a definitive diagnosis, and if the overall reliability is low, whether or not to perform an additional test, etc. Make a judgment.

The diagnostic support device 1 according to the second embodiment has been described above. The diagnostic support device 1 according to the second embodiment can accurately determine the biological state of the subject by using the equation of state and the observation method formula.

(Third Embodiment)
In the first embodiment, the case where the biological state determination function 113 calculates the degree of relationship using 10 quantitative scores and 10 correlation coefficients for each disease has been described. However, the biological state determination function 113 may derive the degree of relation from 10 quantitative scores by using a trained model instead of 10 correlation coefficients. Therefore, such an embodiment will be described as a third embodiment.

In the third embodiment, the memory circuit 12 stores the trained models corresponding to each of the diseases A to N. That is, 14 trained models are stored in the storage circuit 12. The trained model receives input of 10 quantitative scores and outputs the biological state.

For example, the learned model corresponding to the disease A may be generated by the diagnosis support device 1 or may be generated by a device other than the diagnosis support device 1. Hereinafter, a case where a device other than the diagnosis support device 1 generates a learned model corresponding to the disease A will be described. In the following description, such a device will be referred to as a trained model generator.

The trained model generator generates a trained model corresponding to the disease A by learning the relationship between the combination of a plurality of biological information of the subject and the biological state with respect to the disease A. Here, the plurality of biological information is at least two or more of the gene expression / mutation information U1 to U10.

For example, the trained model generator performs machine learning by inputting the combination of the above combination and the set of biological states into the machine learning engine as learning data (teacher data).

As a result of such machine learning, the trained model generator generates a trained model that outputs the biological state in response to the input of the above combination. In this way, the trained model generator generates a trained model using a plurality of types of comprehensive biological information from the gene level of the subject to the human body level. Therefore, the trained model generator can generate a trained model that outputs an accurate biological state. Then, the trained model generation device transmits the generated trained model to the diagnosis support device 1 via a network (not shown). The diagnosis support device 1 stores the received learned model in the storage circuit 12.

Then, the biological state determination function 113 of the diagnosis support device 1 derives biological information by inputting 10 quantitative scores into the learned model corresponding to the disease A. The biological state determination function 113 derives biological information for each of diseases B to N by inputting 10 quantitative scores into the trained model corresponding to each of diseases B to N in the same manner. do.

The diagnostic support device 1 according to the third embodiment has been described above. The diagnostic support device 1 according to the third embodiment can accurately determine the biological state of the subject by using the trained model.

(Fourth Embodiment)
The fourth embodiment is positioned as an additional function of the second embodiment described above. In the fourth embodiment, the biological state determination function 113 of the processing circuit 11 calculates the change in the biological state of the subject with respect to the disease based on the information U1 to U10 of the subject, and also changes the biological state of the subject. The normal range of observation values specific to a predetermined subject is calculated based on changes in lifestyle / environmental information.

Since the schematic configuration of the diagnostic support device 1A according to the fourth embodiment is the same as that shown in FIG. 1, the description thereof will be omitted.

FIG. 9 is a block diagram showing the functions of the diagnostic support device 1A according to the fourth embodiment.

When the processing circuit 11 executes the program, the diagnosis support device 1 realizes the acquisition function 112, the biological state determination function 113, the display control function 114, and the normal range calculation function 115. All or part of the functions 112 to 115 may be realized in the diagnostic support device 1A as a function of a circuit such as an ASIC. The normal range calculation function 115 is an example of a calculation unit. The acquisition function 112, the biological state determination function 113, the display control function 114, and the normal range calculation function 115 will be described later.

Here, for example, the functions 112 to 115, which are the components of the processing circuit 11, are stored in the storage circuit 12 in the form of a program that can be executed by a computer. The processing circuit 11 reads each program from the storage circuit 12 and executes each read program to realize a function corresponding to each program. In other words, the processing circuit 11 in the state where each program is read has the functions 112 to 115 shown in the processing circuit 11.

It should be noted that all the processing functions of the functions 112 to 115 may be stored in the storage circuit 12 in the form of one program that can be executed by the computer. For example, such a program is also called a diagnostic support program. In this case, the processing circuit 11 reads the diagnosis support program from the storage circuit 12 and executes the read diagnosis support program to realize the functions 112 to 115 corresponding to the diagnosis support program.

The storage circuit 12 stores a plurality of types of biological information among the information U1 to U10 regarding a predetermined subject for each subject. In addition, the storage circuit 12 stores information (hereinafter, referred to as “lifestyle / environmental information”) U12 indicating at least one of lifestyle information and environmental information for each subject.

The information U1 to U10 and the biological state determination database U11 are as described in the first embodiment. The lifestyle / environmental information U12 is, for example, the amount of salt intake in the diet of the subject to be diagnosed, the medicinal effect in the medication, and the like. For example, the lifestyle / environment information U12 is stored in the storage circuit 12.

The acquisition function 112 acquires a plurality of types of biological information among the information U1 to U10 regarding a predetermined subject, and lifestyle / environmental information U12.

The biological state determination function 113 determines the biological state of a predetermined subject for a disease based on a plurality of analysis results obtained by analysis of a plurality of types of biological information in a predetermined subject acquired by the acquisition function 112. Also, the change in the biological state is judged. A change in biological state means a difference between a plurality of biological states determined at different times. For example, a change in biological state, the biological state S _t at time t, which is the difference between the previous time t-1 of the biological state S _t-1 "△ S _t". Biological condition S _t of a predetermined object at time t is calculated from the equation (2). In addition, the biological state determination function 113 estimates changes in the biological state of the predetermined subject in the future based on changes in the past lifestyle / environmental information of the predetermined subject by a state transition simulation related to the transition of the biological state. You can also do it. The biological state determination function 113 is an environment based on changes in past lifestyle / environmental information regarding a predetermined subject in a state transition simulation that reflects individualized parameters (for example, the above equations (2) and (3)). Estimate future changes in biological conditions based on the individual subject's robustness score for changes, modify future personalization parameters used for prediction, and make predictions by state transition simulation using the modified personalization parameters. It is also possible to do it. The individualized parameter means the above-mentioned explanatory variables, system noise, observation noise, and the like.

Or biological state S _t at time t, it is also possible to determine in another manner. For example, the acquisition function 112 receives a predetermined subject from the storage circuit 12 that stores the observed value of the subject by the sensor and the lifestyle / environmental information including at least one of the lifestyle information and the environmental information for each subject. Obtain the observed values and lifestyle / environmental information. Then, the biological state determination function 113 determines the biological state of the predetermined subject with respect to the disease based on the observed value in the predetermined subject and the lifestyle / environmental information acquired by the acquisition function 112. In that case, the biological state S _t of a predetermined object at time t, and a lifestyle and environmental information C _t at time t and the observed values D _t, is calculated from the following equation (5).
_St = f (C _t , D _t ) ... (5)

FIG. 10 is a diagram showing an example of the biological state of a predetermined subject for a disease.

Figure 10 is a plurality of diseases (e.g., diseases A through C, N) in which is disposed on the spherical surface, in a state arrangement diagram showing how to position where the predetermined subject biological state S _t at the time t be. Biological condition S _t of a given object is uniquely determined at a predetermined position inside of the sphere by the distance from all diseases. It should _{be noted that the center of the sphere in which the biological state St} is the position where the correlation with any disease is “0”. The distance from each disease indicates the correlation coefficient (-1 to 1), and if it is closer than the radius, it means a positive side, and if it is farther away, it means a negative correlation. For example, in the example shown in FIG. 10, the biological condition S _t of a given subject is close to the disease A, meaning it is far state from disease N.

Returning to the explanation of FIG. 9, the normal range calculation function 115 includes a change in the biological state of a predetermined subject evaluated by the biological state determination function 113, and a change in lifestyle / environmental information acquired by the acquisition function 112. Based on the above, the normal range of the observed value of the predetermined subject in the lifestyle / environmental information of the predetermined subject is calculated. Changes in lifestyle / environmental information mean differences between multiple lifestyle / environmental information acquired at different times. For example, a change of lifestyle and environmental information, lifestyle and environmental information C _t of the time t, which is the difference between the lifestyle and environmental information C _t-1 of the previous time t-1 "△ C _t".

For example, the normal range calculation function 115, based on the rate of change of the biological condition with respect to a change in lifestyle, environment information _{(△ S t / △ C t} ), in a given subject, to changes in lifestyle and environment information Score robustness to get a robustness score. The normal range calculation function 115, based on the robustness score, and calculates the normal range of observations D _t of a given subject. For example, since a plurality of robustness scores for a predetermined subject are calculated at a plurality of times, the normal range calculation function 115 normalizes the _{observed value D t from the variation (standard deviation or variance) of the robustness scores.} The range may be calculated.

The display control function 114 can display the normal range of the predetermined subject-specific observation value calculated by the normal range calculation function 115 on the display 14 together with the observation value of the predetermined subject. In addition, the display control function 114 can also display the normal range of the observed value of the predetermined subject and the observed value based on the future lifestyle / environmental information on the display 14.

FIG. 11 is a diagram showing a display example of a normal range.

11 shows temporal change of a given object observed value D _t, and a normal range. At time t, the normal range of the observed values of the subject is calculated. Then, at the next time, that is, when a certain amount of salt is to be ingested in a meal at a future time t + 1 (or immediately after a certain amount of salt is ingested in a meal at a future time t + 1), the salt intake is performed. It is possible to present to the subject _{how the blood pressure value whose amount is the observed value D t + 1} of the subject changes, and whether the blood pressure value D _{t + 1} is within the normal range for the subject.

In the case shown in FIG. 11, it is predicted that the salt intake in the meal at time t + 1 is large and the blood pressure value of the subject rises, but the predicted blood pressure value is within the normal range for the subject. show. With this indication, the subject can safely ingest salt in the meal at time t + 1.

Next, the significance of calculating the normal range of the observed values of a predetermined subject and the method thereof will be described.

Due to differences in subjects, that is, individual differences, the characteristics (tendencies) of constitution such as increase / decrease in blood glucose level and lipid may differ even if the same food is ingested, and even if the same drug is administered, the medicinal effect (For example, the efficacy of drug metabolism) may be different. In addition, there have been many discussions and studies on the characteristics of the constitution of the subject and the individual differences caused by it. According to one disease guideline, clinical judgment is made by summarizing the individual differences that vary, and statistically defining the range of characteristics that are distributed by many people as "normal", although there are exceptions. I came.

On the other hand, with the recent advances in genome analysis technology and protein / metabolome analysis technology, the differences in characteristics and the factors that cause the differences have been gradually elucidated. Along with this, the idea of personalized medicine and precision medicine, which select treatments and medications according to the characteristics of the subject, has advanced.

For example, companion diagnostics is one example. Companion diagnostics include glucose tolerance tests, cardiac stress tests that observe the function of the heart under exercise load, prediction of drug efficacy and side effects based on genetic predisposition, and appropriate medication for the subject. Is to treat.

In order to perform appropriate treatment according to the characteristics of the subject, what kind of biological characteristics (for example, information U1 to U10) the subject has, and the subject eats or takes medicine. It is necessary to correctly understand, predict, and formulate a treatment policy for what kind of reaction will occur when drinking.

On the other hand, in the living body, there is a "homeostasis maintenance mechanism" that tries to maintain the life activity so as not to shift to a pathological state. Due to the "homeostasis maintenance mechanism", the salt concentration in the body, various minerals, amino acids, other biomarkers, and various states such as body temperature, blood pressure, and heart rate are short-term due to daily activities such as diet and exercise. However, it is controlled to stay within a certain normal range in the long term.

In this way, in order to formulate an appropriate treatment policy, the characteristics of the individual subject are estimated by taking into account lifestyle / environmental information related to at least one of the lifestyle and environment related to the “homeostasis maintenance mechanism”. Is appropriate.

Here, assuming that the drug reactivity of each subject, which means the characteristic of the drug efficacy of the subject, is "R" (Response), the drug reactivity R is, as is well known, based on the coefficient f by the following equation (6). expressed. Here, "C" is the drug concentration (Concentration) of the action site of the subject, and can be defined as lifestyle / environmental information. Further, "E" is the tissue sensitivity "Sensitivity" of the site of action of the subject.
R = f (CE) ... (6)

The drug concentration C site of action of the subject in the above formula (6), can be defined as lifestyle and environmental information C _t of the subject at the current time t. Also, tissue sensitivity E of the site of action of the subject in the above formula (6) can be defined as the change in the biological condition of the subject "△ S _{t /} △ C _t" at the current time t. Further, the above equation drug response R of the subject in (6), can be defined as a biological state S _t of the subject at the current time t.

In the fourth embodiment, the normal range calculation function 115, the value of the change in the biological condition of the subject "△ S _{t /} △ C _t" drug response R of the subject, i.e., the relative change in the biological condition It evaluates the robustness (robustness) of a sample and calculates the normal range of individualized observations of the subject. Further, although the normal range of the individualized observation values of the subject with respect to the medication to the subject has been described using the above formula (6), the present invention is not limited to that case. The same applies to the normal range of individualized observations of the subject with respect to the salt intake of the subject.

For example, the racial specificity and genetic predisposition of salt sensitivity will be described. Normal range calculation function 115, based on the salt intake from the diet of the subject (which may be included snacking) and "C _t", and the blood pressure value of the subject by the IoT and wearable device "S _t" Te, calculates a change in the biological condition of the subject "△ S _{t /} △ C _t". As a result, the normal range calculation function 115 can routinely evaluate the robustness of the subject to changes in the biological state and present it to the subject.

Specifically, the normal range calculation function 115 calculates the amount of salt contained in one meal according to the recipe. For example, even when eating out, the normal range calculation function 115 can calculate the amount of salt contained in the ordered meal according to the recipe and present it to the subject who is the customer. Moreover, the normal range calculation function 115, based on the order history, the salt content in the automatic meal its options change △ C _t lifestyle given to the subject (or environment), ingested a meal Calculated as "lifestyle / environmental information" over time.

The same can be said for the case where the subject ingests a certain confectionery at a certain time. Information on "Nutrition" is given to the candy bag. Therefore, the normal range calculation function 115 acquires the information with an IC (Integrated Circuit) chip, an RFID (Radio Frequency Identification) reader, or the like, and recognizes that the subject has eaten it as necessary, and then " It calculates a change amount of salt contained in the nutrient "as △ C _t.

Furthermore, in the meal by the self-contained by the subject at home, the normal range calculation function 115 may calculate a change △ C _t of the amount of seasoning used. In addition, in meals for a plurality of people, the normal range calculation function 115 was cooked by personal input of meal intake, estimation of individual intake by image recognition, or estimation based on standard intake. from the ratio of the amount of salt contained in the entire meal, estimates the ingested amount of salt subject individual, it is also possible to calculate the variation △ C _t. As for the seasoning, it is possible to obtain a few grams per swing, or if it is packaged, the amount of salt contained in the seasoning can be obtained by reading it from an IC chip attached to the package.

On the other hand, the biological state determination function 113, by continuously measuring the biological condition S _t of a subject by a wearable device such as can be evaluate changes △ S _t of the biological condition of the subject (calculated) be. For example, change △ S _t of the biological condition of the subject can be considered a change in blood pressure, and by heart rate variation △ D _t like. Here, for convenience, change △ S _t of the biological condition of the subject is described as due to changes in blood pressure △ D _t. The time to measure blood pressure varies depending on, for example, whether you want to observe changes in the acute phase one hour after a meal, or changes in the chronic phase on the order of several months or years. Is similar in such a glucose tolerance test, the biological state determination function 113, a change △ S _t of the biological condition of the subject, and the blood pressure just before salt intake, influence from the difference △ D _t and the timing to be observed blood pressure It can be calculated from the above formula (6).

The blood pressure of the subject changes due to the influence of activities such as exercise of the subject, stress and other mental and physical activities. Therefore, the biological state determination function 113 can also be calculated as a multivariate effect with an event change that affects the state to be observed. Further, the biological condition determination function 113, the influence factors other than salt intake at time t assuming other system noise u _{t (the} formula (2)), observing changes factors occurring in the observation system such as a wearable device noise w _A state equation or an observation equation can be set as t (the above equation (3)). Also, a change △ C _t of salt intake, it is possible to observe the change as a long-term total intake in.

And this way the observed salt intake of change △ C _t, the blood pressure change △ S _t at the time span to be observed changes, changes in the biological state "△ S _{t /} △ C _t" is calculated. The smaller the change in the biological condition "△ S _{t /} △ C _t", it can be said that the change in observations in load due to salt intake enough to set small, it has a resistance of robustness and salt load.

The value of change "△ S _{t /} △ C _t" of thus obtained biological condition is a numerical value that reflects the resistance-fastness relative to the subject individual salt load, the accumulation of this data, the individualization of personal In addition, a load range that does not show an abnormal value can be defined as a normal range.

In the previous embodiments, the description has focused on changes in salt intake and changes in blood pressure, but the changes in state and tolerance / robustness to various substances, not limited to salt intake, are described, for example, at rest (wake up) of an individual. When the numerical value of time) is used as a reference, it is possible to score each subject as follows and define the normal range for each individualized change in lifestyle / environmental information as a robust range.
(1) Robust load area: Load area within the fluctuation range of 10% (2) Semi-robust load area: 30% fluctuation range (3) Fluctuation boundary area: 30 to 50%
(4) High load area: 50% or more

Also, the change in the biological condition "△ S _{t /} △ C _t" and, based on the scored subject individual robustness score by the foregoing description of the time t associated with lifestyle and environment information C variable a _t When the state equation set the system noise u _t at time t (the formula (2)) and the explanatory variable b _t at time t, the observation equation by the setting of the observation noise w _t at time t (the above formula (3 )) Makes it possible to estimate and predict the state change (effect) of the individual subject due to the state change.

In fact, the blood pressure fluctuations mentioned in the example include not only short-term and long-term changes due to salt intake, but also other minerals and drugs that affect the fluctuations, exercise load, changes in ambient temperature, stress, etc. such as changes due to the effect of the autonomic nervous system, and affects the multiparametric, by weighting the impact on their system variations, system noise u _t of the subject person has to be set in real time. That is, the system noise u _t at time t is a weighted numerical lifestyle and environmental information C _t for each factor influencing the output.

The diagnostic support device 1 according to the fourth embodiment has been described above. According to the diagnostic support device 1 according to the fourth embodiment, it is possible to predict an individualized biological state that reflects environmental changes such as meals and individual differences in drug responsiveness.

According to at least one embodiment and modification described above, the biological state of the subject can be accurately determined.

In the above description, an example in which the "processor" reads a program corresponding to each function from the storage circuit and executes the program has been described, but the embodiment is not limited to this. The word "processor" is used, for example, as a CPU (central processing unit), a GPU (Graphics Processing Unit), an application specific integrated circuit (ASIC), and a programmable logic device (for example, a simple programmable logic device). (Simple Programmable Logic Device: SPLD), Complex Programmable Logic Device (CPLD), Field Programmable Gate Array (FPGA), and other circuits. When the processor is, for example, a CPU, the processor realizes a function by reading and executing a program stored in a storage circuit. On the other hand, when the processor is, for example, an ASIC, the function is directly incorporated as a logic circuit in the circuit of the processor instead of storing the program in the storage circuit. It should be noted that each processor of the present embodiment is not limited to the case where each processor is configured as a single circuit, and a plurality of independent circuits are combined to be configured as one processor so that its function is realized. May be good. Further, the plurality of components in FIG. 2 may be integrated into one processor to realize the function.

Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other embodiments, and various omissions, replacements, changes, and combinations of embodiments can be made without departing from the gist of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, as well as in the scope of the invention described in the claims and the equivalent scope thereof.

1 Diagnosis support device 11 Processing circuit 111 Score calculation function 112 Acquisition function 113 Biological condition judgment function 114 Display control function 115 Normal range calculation function

Claims (4)

From the storage unit that stores the observed value of the subject by the sensor and the lifestyle / environmental information including at least one of the lifestyle information and the environmental information for each subject, the observed value and the life of the predetermined subject. The acquisition department that acquires customs / environmental information,
A determination unit that determines the biological state of the predetermined subject with respect to a disease based on the observed value in the predetermined subject and the lifestyle / environmental information.
Based on the change in the biological state of the predetermined subject and the change in the lifestyle / environmental information, the normal range of the observed value of the predetermined subject in the lifestyle / environmental information of the predetermined subject is set. Calculation unit to calculate and
Diagnostic support device equipped with. The calculation unit
The normal range of the observed value of the predetermined subject is calculated based on the ratio of the change in the biological state to the change in the lifestyle / environmental information.
The diagnostic support device according to claim 1. The determination unit estimates a change in the biological state of the predetermined subject in the future based on the past changes in the lifestyle / environmental information of the predetermined subject.
The diagnostic support device according to claim 1 or 2. A display control unit that displays the normal range of the observed values of the predetermined subject and the observed values based on future lifestyle / environmental information on the display unit.
The diagnostic support device according to any one of claims 1 to 3, further comprising.

Images