Immune dynamic generation device and immune dynamic generation method

Abstract

The application provides an immune dynamic generation device which is helpful for health status management, early disease detection, early treatment and optimal treatment. The immune dynamic generation device comprises at least one immune dynamic processing unit, a memory unit for storing clinical data, and a display unit. The immune dynamic processing unit obtains clinical basic characteristics and immune subpopulation dynamic information based on the clinical data, forms a classification model through self-learning based on the immune subpopulation dynamic information and the classification model, and simultaneously performs monitoring; and based on the classification model, evaluates the immune dynamic information of a subject by comparing the overall relationship between the clinical basic characteristics and the immune subpopulation dynamic information, and simultaneously controls the display of the evaluation result of the immune dynamic information on the display unit.

Description

Technical Field

[0001] This invention relates to the field of dynamic immune generation technology, and in particular to an immune dynamic generation device and a method for immune dynamic generation. Background Technology

[0002] Following the 2018 Nobel Prize in Physiology or Medicine awarded to Tasuku Honjo, a specially appointed professor at Kyoto University, and the market launch of the immune checkpoint inhibitor Opdivo, fourth-generation cancer treatments, represented by immunotherapy, and various minimally invasive radiation-based cancer treatments—namely, minimally invasive surgery (MIS)—have gradually gained widespread attention, and related treatment methods are constantly evolving. A common characteristic of these treatment methods is that they all rely on the body's own immune function for treatment (hereinafter collectively referred to as immunotherapy).

[0003] In the process of immunotherapy, the analysis and evaluation of immune dynamics are indispensable. The analysis of immune dynamics not only helps in the early detection of diseases, but also serves as an important indicator for evaluating the efficacy and side effects of disease prevention and treatment. It is gradually becoming an important research area for early disease detection and treatment in modern clinical medicine.

[0004] For example, the evaluation of clinical cancer treatment efficacy typically involves assessing imaging findings (changes in cancer burden, i.e., cancer volume) at a specific time point (especially before and after a change in treatment regimen) while simultaneously analyzing immune dynamics at that time point. For immunotherapy methods, the simultaneous evaluation of imaging findings and immune dynamics is considered significant. The aforementioned imaging findings include, for example,... Figure 5 The image shows the results obtained by measuring cancer volume using positron emission tomography-computed tomography (PET-CT). The aforementioned immune dynamics are, for example, as shown below. Figure 6 The image shows the dynamic immune information obtained based on routine blood count and white blood cell differential count results from conventional blood component testing.

[0005] A wide variety of immune cells are involved in regulating blood immune function, primarily derived from lymphoid stem cells and myeloid stem cells. Furthermore, blood cell components such as erythrocytes and platelets also originate from the same type of hematopoietic stem cells (pluripotent hematopoietic stem cells), just like immune cells. Lymphoid stem cells, differentiated from hematopoietic stem cells, further differentiate into helper T cells, αβ killer T cells, γδ T cells, B cells, NKT cells, and NK cells. Myeloid stem cells, also differentiated from hematopoietic stem cells, further differentiate into macrophages, neutrophils (comprising over 90% of granulocytes), eosinophils, basophils, as well as erythrocytes and platelets.

[0006] In recent years, the inventors have proposed a dynamic information analysis technology for immune subpopulations in research on immune dynamics. This technology analyzes the number of B cells, NK cells, NKT cells, and the number and proportion of immune lymphocyte subpopulations (also known as subsets), including at least the number of helper T cells, cytotoxic T cells, NK cells, NKT cells, and B cells, to analyze changes in the immune status of the subject (see Patent Document 1).

[0007] Existing technical documents Patent documents Patent Document 1: Japanese Patent Application Publication No. 2021-189081 Summary of the Invention The problem that the invention aims to solve However, while methods exist for comparative analysis of imaging changes based on cancer (changes in cancer burden) or for generating only immune dynamics information, there is currently no objective judgment standard that can simultaneously consider both and guide clinical treatment. Furthermore, in general social life, there is a lack of an objective judgment standard that can comprehensively evaluate an individual's health status and immune dynamics information. Especially in modern society, with the frequent occurrence of infectious diseases such as the novel coronavirus, establishing a judgment standard that considers immune dynamics to objectively evaluate the effectiveness of clinical treatment has become an urgent need. In other words, there is an urgent need to establish a system that can objectively assess immune dynamics based on information from lymphocyte subsets, including helper T cells, αβ killer T cells, γδ T cells, B cells, NKT cells, and NK cells.

[0008] To overcome the problems existing in the prior art, the inventor has long been committed to studying the impact of immune dynamics on the treatment effects of diseases such as cancer. Through extensive and in-depth research, the inventor has successfully discovered a comprehensive judgment standard that can simultaneously combine clinical data and immune dynamic information to objectively evaluate the health status of examinees, and based on this, completed this invention. The relevant research results are now disclosed.

[0009] Solution for solving the problem To achieve the above objectives, the present invention provides the following technical solutions.

[0010] This invention provides an immune dynamics generation device, characterized in that it comprises: at least one immune dynamics processing unit; at least one memory unit for storing clinical data; and a display unit; wherein the immune dynamics processing unit: acquires clinical baseline characteristics and immune subpopulation dynamic information based on the clinical data, performs self-learning (self-training) based on the immune subpopulation dynamic information and the classification model to form a classification model and simultaneously monitors it; and based on the classification model, evaluates the immune dynamics information of the subject by comparing the overall relationship between the clinical baseline characteristics and the immune subpopulation dynamic information, and simultaneously controls the display of the evaluation results of the immune dynamics information on the display unit.

[0011] Preferably, the aforementioned clinical data includes at least health checkup data for normal individuals and health checkup data for individuals with abnormal health conditions.

[0012] Preferably, the aforementioned dynamic information on immune subpopulations includes: the number of B cells, the number of NK cells, the number of NKT cells, and the number and proportion of immune lymphocyte subpopulations, including at least the number of helper T cells, the number of killer T cells, the number of αβkiller-T cells, the number of γδT cells, the number of NK cells, the number of NKT cells, and the number of B cells.

[0013] Preferably, the immune dynamic processing unit: acquires the above-mentioned basic clinical characteristics, the above-mentioned dynamic information of immune subpopulations and the above-mentioned classification model based on the health examination data of the above-mentioned normal individuals, and obtains normal immune status information by comparing the overall relationship between the above-mentioned basic clinical characteristics and the above-mentioned dynamic information of immune subpopulations.

[0014] Preferably, the immune dynamic processing unit: acquires the above-mentioned clinical basic characteristics, the above-mentioned immune subpopulation dynamic information and the above-mentioned classification model based on the health examination data of the above-mentioned abnormal individuals, and obtains abnormal immune status information by comparing the overall relationship between the above-mentioned clinical basic characteristics and the above-mentioned immune subpopulation dynamic information.

[0015] Preferably, the immune dynamic processing unit: acquires the above-mentioned basic clinical characteristics, the above-mentioned dynamic information of immune subpopulations and the above-mentioned classification model based on the above-mentioned clinical data including cancer cell data, and obtains cancer incidence information by comparing the overall relationship between the above-mentioned basic clinical characteristics and the above-mentioned dynamic information of immune subpopulations based on the above-mentioned abnormal immune status information.

[0016] Preferably, the immune dynamic processing unit determines that cancer cells are in a deteriorating state when the proportion of the cancer incidence information is greater than 1.

[0017] Preferably, the immune dynamic processing unit determines that the cancer cells are in a prolongation state when the proportion of the cancer incidence information is equal to 1.

[0018] Preferably, the immune dynamic processing unit determines that the cancer cells are in an improved state when the proportion of the cancer incidence information is less than 1.

[0019] In addition, the present invention also provides a method for generating immune dynamics, characterized by comprising: a step of acquiring clinical data of a subject; a step of acquiring dynamic information of immune subpopulations of the subject; and a step of monitoring the average data of the normal majority and the abnormal data that have changed, calculated using the basic data and lymphocyte subpopulation data stored in the memory unit, and evaluating the immune dynamics of the subject by comparing the overall relationship between the clinical data and the dynamic information of immune subpopulations based on the average data and the abnormal data.

[0020] Invention Effects According to the present invention, an immune dynamics generation method and apparatus are provided, which can establish a comprehensive judgment standard for objectively evaluating immune dynamics based on information such as clinical data, and can timely and accurately grasp the body's immune status and the progression of diseases. In particular, by utilizing the immune dynamics generation method and apparatus of the present invention, based on the comprehensive judgment standard for objectively evaluating immune dynamics, health status management, early detection of diseases, early treatment, and optimized treatment can be achieved, and it can help improve and inhibit the development of serious diseases. Attached Figure Description

[0021] Figure 1 This is a conceptual schematic diagram of the immune dynamic generation device in a specific embodiment of the present invention.

[0022] Figure 2 This is a conceptual schematic diagram illustrating the structure of the memory section of the immune dynamic generation device.

[0023] Figure 3 A Gompertzian tumor growth curve representing the change in cancer cell proliferation over time.

[0024] Figure 4 This is a schematic diagram illustrating one embodiment of the classification extraction table in this implementation.

[0025] Figure 5 This is a schematic diagram illustrating the results of cancer volume measurement obtained via PET-CT.

[0026] Figure 6 This is a schematic diagram illustrating the results of routine blood component analysis and white blood cell differential tests.

[0027] Figure 7 This is a schematic diagram illustrating an embodiment of the learning mechanism functionality of the present invention.

[0028] Figure 8 This is a flowchart illustrating the learning process used to form a classification model.

[0029] Figure 9 This is a schematic diagram illustrating one embodiment of the learning data used in this implementation.

[0030] Figure 10 This is a flowchart illustrating the process of performing immunodynamic evaluation.

[0031] Figure 11 This is a schematic diagram illustrating one embodiment of the immune dynamic table in this implementation.

[0032] Figure 12 This is a schematic diagram illustrating the basic data of Example 1.

[0033] Figure 13 This is a schematic diagram illustrating the quantity of each component in Example 1 calculated based on Formula 1 and Formula 2 above.

[0034] Figure 14 This is a schematic diagram illustrating the number of granulocytes, single cells, lymphocytes, helper T cells, cytotoxic T cells, αβ cytotoxic T cells, γδ T cells, NK cells, NKT cells, and B cells in the classification data of Example 1.

[0035] Figure 15 This is a schematic diagram illustrating the proportions of granulocytes, unicellular cells, and lymphocytes in Example 1.

[0036] Figure 16 This is a schematic diagram illustrating the proportions of helper T cells, killer T cells, αβ killer T cells, γδ T cells, NK cells, NKT cells, and B cells in Example 1.

[0037] Figure 17 This is a schematic diagram illustrating one embodiment of the classification model in Example 1.

[0038] Figure 18 This is a schematic diagram illustrating an embodiment of comparing the evaluation model with the benchmark model in Example 1.

[0039] Figure 19 This is a schematic diagram illustrating the number of granulocytes, single cells, lymphocytes, helper T cells, cytotoxic T cells, αβ cytotoxic T cells, γδ T cells, NK cells, NKT cells, and B cells in the basic data and classification data of Example 2.

[0040] Figure 20 This is a schematic diagram illustrating the proportions of granulocytes, unicellular cells, and lymphocytes in Example 2.

[0041] Figure 21 This is a schematic diagram illustrating the proportions of helper T cells, killer T cells, αβ killer T cells, γδ T cells, NK cells, NKT cells, and B cells in Example 2.

[0042] Figure 22 This is a schematic diagram illustrating one embodiment of the classification model in Example 2.

[0043] Figure 23 This is a schematic diagram illustrating the number of granulocytes, single cells, lymphocytes, helper T cells, cytotoxic T cells, αβ cytotoxic T cells, γδ T cells, NK cells, NKT cells, and B cells in the basic data and classification data of Example 3.

[0044] Figure 24 This is a schematic diagram illustrating the proportions of granulocytes, unicellular cells, and lymphocytes in Example 3.

[0045] Figure 25 This is a schematic diagram illustrating the proportions of helper T cells, killer T cells, αβ killer T cells, γδ T cells, NK cells, NKT cells, and B cells in Example 3.

[0046] Figure 26 This is a schematic diagram illustrating an embodiment of comparing the evaluation model with the benchmark model in Example 3. Detailed Implementation

[0047] The embodiments of the immune dynamic generation device and immune dynamic generation method of the present invention will be described below with reference to the accompanying drawings.

[0048] (Overall composition) As an example, such as Figure 1As shown, the immune dynamics generation device 1 includes an immune dynamics processing unit 2, a memory unit 3, an information input unit 4, and a display unit 5. The memory unit 3 can be connected to an external information source 6 via a network 7 and can communicate with it. The immune dynamics processing unit 2 includes a learning mechanism 21 and an immune dynamics evaluation unit 22.

[0049] The immune dynamic processing unit 2 can be, for example, a workstation or a personal computer, or a server computer connected to it via a network. The immune dynamic processing unit 2 contains the immune dynamic processing program and learning program (training program) of the present invention. The immune dynamic processing program and learning program can be stored in a storage device or network storage device of the server computer connected to the network 7, and stored in a manner accessible externally; when needed, they can be downloaded to and installed on the immune dynamic processing unit 2 used by the doctor or patient. Alternatively, the aforementioned immune dynamic processing program and learning program can also be recorded on a recording medium such as a Digital Versatile Disc (DVD) or a Compact Disc Read Only Memory (CD-ROM) for distribution, and installed from that recording medium onto the immune dynamic processing unit 2.

[0050] The immune dynamics processing program of the present invention has at least the functions of forming a basic immune dynamics model, learning, and evaluating immune dynamics. For example, based on the above-mentioned immune dynamics processing program, the immune dynamics processing unit 2 uses relevant clinical data as the numerator and relevant immune subpopulation dynamic information as the denominator, and forms a basic immune dynamics model with a ratio of 1 based on the correlation between the two.

[0051] The memory unit 3 can be composed of hardware (circuit, dedicated logic, etc.), software (software running on a general-purpose computer system or dedicated device, etc.), or a combination thereof, and has the function of storing various information. For example, the memory unit 3 can be a hard disk drive (HDD), a solid-state drive (SSD), flash memory, or other storage devices. Alternatively, the memory unit 3 can also be a computer system for storing and managing various data, including a large-capacity external storage device and database management software. The memory unit 3 communicates with other devices via wired or wireless networks. For example, the memory unit 3 is connected to and able to communicate with an external information source 6 via network 7 to send and receive basic data related to immune dynamics, lymphocyte subpopulation data, and other information. In addition, the memory unit 3 also directly or via the network acquires various data generated by the immune dynamics processing unit 2, including immune dynamics evaluation results, and stores and manages them in recording media such as a large-capacity external storage device. The storage formats of various data and the communication between devices via the network are based on protocols such as Digital Imaging and Communications in Medicine (DICOM).

[0052] The information input unit 4 can be an input device such as a keyboard or mouse, or a device with information input function and connected to a network. The display unit 5 has the function of displaying analytical information such as immune dynamic evaluation results, and can be a common liquid crystal display such as a touch screen or display panel. Alternatively, a touch screen display that integrates the display unit 5 and the information input unit 4 can also be used. The network 7 is used for wired or wireless communication with an external information source 6.

[0053] External information source 6 is a terminal for enterprises, medical personnel, medical institutions, and ordinary users to send and receive basic data, lymphocyte subpopulation data, and other immune data to memory unit 3. It can be a mobile terminal such as a personal computer, laptop, smartphone, or tablet, or a wearable terminal such as a head-mounted display or smartwatch. There can be multiple enterprises, medical personnel, medical institutions, and ordinary users.

[0054] like Figure 2As shown, the memory unit 3 stores at least basic data 31, a classification extraction table 33, a classification model box 34, and an immune dynamic table 35. The basic data 31 includes at least normal health examination data and abnormal health examination data. Normal health examination data includes at least health examination data of healthy individuals. Abnormal health examination data includes examination data of individuals infected with infectious diseases such as influenza, cancer cell data of various cancer patients, and examination data of various other conditions. The information and data involved in this invention are in various forms, including but not limited to images, tables, numerical values, and text. Furthermore, the memory unit 3 also stores learning programs and immune dynamic processing programs. The basic data 31 and the classification extraction table 33 together constitute the clinical data of this invention. In addition, the memory unit 3 may also store immune subpopulation dynamic information (lymphocyte subpopulation data 32).

[0055] Health check-ups and the items examined include, for example, blood test results, urine test results, imaging results (X-ray, Magnetic Resonance Imaging (MRI), CT scan, etc.), and biopsy results. Blood test results include, for example, information obtained by analyzing blood components, such as white blood cell count, percentage of basophils, eosinophils, neutrophils, lymphocytes, monocytes, CD3 antigen, CD56 antigen (and / or CD16 antigen), CD4 antigen, and CD8 antigen.

[0056] Cancer cell data for various cancer patients includes information on the amount of cancer cells in the body and information on changes in the amount of cancer cells in the body. Information on the amount of cancer cells in the body and information on changes in the amount of cancer cells in the body can be obtained based on the results of biochemical markers, imaging technology, histological examination, liquid biopsy, molecular biology methods, etc., or it can be obtained based on Gompertzian tumor growth curves.

[0057] Biochemical markers are methods of indirectly estimating the amount of cancer cells by measuring the levels of specific biochemical markers released into the blood or other bodily fluids by certain types of cancer. For example, prostate-specific antigen (PSA) is considered a marker for prostate cancer.

[0058] Imaging technology refers to methods that use images obtained through imaging techniques such as CT scans, MRI, and PET scans to estimate the extent of cancer spread and the amount of cancer cells. This method helps to visually determine the location and size of cancerous tissue within the body.

[0059] Histological examination refers to the examination of tissue samples obtained through biopsy or surgery under a microscope to directly observe the presence and number of cancer cells. This method is indispensable for cancer detection and staging.

[0060] Liquid biopsy is a method called "liquid biopsy" that involves detecting cancer cells or fragments of their DNA from a blood sample. Liquid biopsies are less invasive and can help monitor the progression of cancer.

[0061] Molecular biology methods refer to methods that use polymerase chain reaction (PCR) or other molecular techniques to detect gene mutations or expression patterns specific to cancer cells.

[0062] The Gompertzian tumor growth curve refers to, for example, Figure 3 As shown, the Gompertzian tumor growth curve is obtained by plotting the proliferation of cancer cells over time. Typically, the Gompertzian curve uses the vertical axis to represent the amount of cancer cell proliferation and the horizontal axis to represent time, presenting an S-shaped curve.

[0063] When a cancer cell develops in the body at a certain point in time, it is not visible to the naked eye. Over time, the cancer cell gradually divides and multiplies, eventually forming a cluster of cancer cells. Cancer cells proliferate slowly when they first develop, but when the number of cancer cells in the body reaches a certain level, they proliferate rapidly. According to current reports, after cancer cells proliferate to more than about 100 (10^2) cells, their proliferation rate increases dramatically, rapidly forming a cluster of about 100 million (10^9) cancer cells, which then exhibits slow proliferation again. The Gompertzian tumor growth curve is a curve that demonstrates this proliferative characteristic of cancer cells.

[0064] With current medical technology, cancer cells are only detected as "early-stage cancer" when their cell count reaches approximately 100 million (10^9). At this stage, the tumor is about 1 cm in diameter and weighs about 1 gram. After reaching approximately 100 million cells, the rate of proliferation slows down. Cancer cells continue to proliferate from 100 million, reaching a lifespan limit when their cell count reaches 100 billion (10^12).

[0065] The lymphocyte subpopulation data 32 includes basic immune cell data, immune dynamic pattern data, and immune dynamic model data. The basic immune cell data includes at least the number of white blood cells, and the quantity and proportion of basophils, eosinophils, neutrophils, lymphocytes, monocytes, CD3 antigen, CD56 antigen (and / or CD16 antigen), CD4 antigen, and CD8 antigen among white blood cells. Lymphocytes include at least helper T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells. The basic immune cell data, immune dynamic pattern data, and immune dynamic model data included in the lymphocyte subpopulation data 32 can be obtained using data based on the technical solution disclosed in Japanese Patent No. 6796737, depending on the actual situation.

[0066] The classification model box 34 is a database used to store the classification model 600. The classification model 600, referring to the classification extraction table 33, extracts weighted and vectorized primality and type values ​​from the input data and inputs them. By repeatedly adjusting the type value and other function values ​​or immune dynamic features, the ratio of the type value and other function values ​​(in one example, Beck, the quantity of basic feature 341 in this embodiment) to the immune dynamic features (in one example, the quantity of immune dynamic feature 39 in this embodiment) is within the range of the comprehensive state index, thereby determining the final type value and the final primality, which are then output as the classification result. In this embodiment, the final type value refers to the total amount of basic feature 341 and basic feature weight 381 (described later); the final primality value refers to the total amount of immune dynamic feature 39 and immune dynamic weight 391. The input data in this embodiment is information composed of basic data 31 or basic data 31 and lymphocyte subpopulation data 32. Primality is a quantity representing the characteristics of a lymphocyte subpopulation, such as the number of cytotoxic T cells. The typological quantity is a measure that represents the categorical characteristics of the examinee; for example, for a cancer patient, it could be the number of cancer cells. The comprehensive state index is a numerical value or range characterized by the ratio between the primacy quantity and the typological quantity. For example, for an examinee, it could be a range with a primacy quantity to typological quantity ratio of 1 as a baseline.

[0067] The classification model learned by the learning institution 21 in this embodiment will be explained below. Figure 7 This illustrates an example of the functional composition of learning institution 21 during the learning phase. For example... Figure 7As shown, the learning unit 21 includes a data acquisition unit 210, a data classification unit 211, and a learning unit 212. The data acquisition unit 210 acquires the basic data 31 and lymphocyte subpopulation data 32 from the memory unit 3. The data classification unit 211 extracts corresponding types and primalities from the basic data 31 and lymphocyte subpopulation data 32 acquired by the data acquisition unit 210 according to the classification extraction table 33. The learning unit 212 performs machine learning based on the basic data 31 and lymphocyte subpopulation data 32 acquired by the data acquisition unit 210, and the types and primalities extracted by the data classification unit 211, and generates a classification model 600 and stores it in a classification model box 34.

[0068] Furthermore, in this embodiment, "learning" of the classification model refers to learning the classification model 600 using learning data composed of basic data 31 and lymphocyte subpopulation data 32. Additionally, in this embodiment, learning the classification model 600 refers to generating a classification model by training it using the learning data based on a machine learning model.

[0069] Figure 8 This illustrates one form of the machine learning model in this embodiment. The data acquisition unit 210 acquires basic data 31 or basic data 31 and lymphocyte subpopulation data 32 from the memory unit 3, and classifies and combines the basic data 31 or basic data 31 and lymphocyte subpopulation data 32 according to their characteristics and properties, thereby forming categorized data. The data classification unit 211, based on the acquired basic data 31 and lymphocyte subpopulation data 32, extracts corresponding elements from the classification extraction table 33 as basic features 341 (an example of clinical basic features). Figure 9 An example of learning data is shown. In this embodiment, for example, such as... Figure 9 As shown, the data classification unit 211 associates predetermined identity identification numbers (IDs) with the learning data. For example, when the ID corresponding to the learning data is "00001" to "09999", the data classification unit 211 combines the basic data 31 associated with IDs in the range of "00001" to "09999" with the lymphocyte subpopulation data 32 to form the learning data. Figure 9In the example shown, the combination of basic data 31 (ID "00001"), containing type and examination data, basic features 341, and lymphocyte subpopulation data 32, is used as learning data. The data classification unit 211 vectorizes the extracted basic features 341 using weights assigned to them. The data classification unit 211 outputs the vectorized basic features 341, along with the basic data 31 and lymphocyte subpopulation data 32 obtained from the memory unit 3, to the learning unit 212. Furthermore, the data classification unit 211 can also output the basic features 341 and the basic data 31 obtained from the memory unit 3 as basic elements 38 (an example of clinical basic features), and the lymphocyte subpopulation data 32 as immune dynamic features 39 (an example of immune subpopulation dynamic information) to the learning unit 212.

[0070] like Figure 9 As shown, for example, the combination of basic data 31, basic features 341, and lymphocyte subpopulation data 32 associated with healthy individuals is used to generate learning data with ID "00001". The combination of basic data 31, basic features 341, and lymphocyte subpopulation data 32 associated with influenza patients is used to generate learning data with ID "00002".

[0071] The learning unit 212 learns the classification model 600 based on the basic data 31, basic features 341, and lymphocyte subpopulation data 32, which are input as learning data from the data classification unit 211. In this embodiment, the learning unit 212 uses the basic data 31, basic features 341, and lymphocyte subpopulation data 32 as input, and aims to learn the classification model 600 with the proportion (or ratio) of the output results as the overarching state index. The machine learning method is not limited to a specific type; any method capable of classification is acceptable, such as the well-known Support Vector Machine (SVM) or Random Forest. The learning unit 212 stores the trained classification model 600 in the classification model box 34.

[0072] When the learning unit 21 learns from the classification model 600, the learning program is executed through the immune dynamic processing unit 2, thereby performing... Figure 8 The learning process is shown below. Figure 8 The learning process shown is the learning model of this embodiment. For example, the learning process can be executed at a preset learning time, or when the user issues a start execution command via the information input unit 4 or when other conditions are met.

[0073] exist Figure 8In step S1, the data acquisition unit 210, as described above, acquires basic data 31 from the memory unit 3, or acquires basic data 31 and lymphocyte subpopulation data 32, and forms classification data. An example of classification data formation is described below.

[0074] As an example, the basic data 31 includes at least the number and proportion of white blood cells, basophils, eosinophils, neutrophils, lymphocytes, monocytes, CD3 antigen, CD56 antigen (and / or CD16 antigen), CD4 antigen, and CD8 antigen.

[0075] Specifically, basic data 31 refers to, for example Figure 6 The table shown is a summary of the blood components of a typical examinee and the results of a routine blood test and white blood cell differential count. It includes data such as the number of white blood cells, the percentage of basophils (%), eosinophils (%), neutrophils (%), lymphocytes (%), and monocytes (%). Figure 6 In the text, "WBC" represents the number of white blood cells in 1 μl. In the white blood cell image (white blood cell differential count information) section, "Baso" represents the percentage of basophils (%), "Eosino" represents the percentage of eosinophils (%), "Neutro" represents the percentage of neutrophils (%), "Lympho" represents the percentage of lymphocytes (%), and "Mono" represents the percentage of single cells (%).

[0076] In addition, the basic data 31 also includes the CD numbering analysis results of leukocyte surface antigens such as CD3 antigen, CD56 antigen (and / or CD16 antigen), CD4 antigen and CD8 antigen (hereinafter sometimes referred to as CD classification analysis information).

[0077] The lymphocyte subpopulation data 32 includes: the number of B cells, the number of NK cells, the number of NKT cells, and the number and proportion of immune lymphocyte subpopulations, including at least the number of helper T cells, the number of killer T cells, the number of αβkiller-T cells, the number of γδT cells, the number of NK cells, the number of NKT cells, and the number of B cells.

[0078] The lymphocyte subpopulation data 32 is acquired through external acquisition or by calculation by the data acquisition unit 210. In the case of calculation by the data acquisition unit 210, the lymphocyte subpopulation data 32 is calculated based on the following steps.

[0079] In the first step, based on information related to the number of white blood cells, the percentage of basophils (%), eosinophils (%), neutrophils (%), lymphocytes (%), and monocytes (%), the total number of white blood cells (immune cells) in the subject and the total number of each of the white blood cell subgroups (immune cell subgroups) are calculated.

[0080] Specifically, the total number of white blood cells (immune cells) and the white blood cell subgroups (immune cell subgroups: granulocytes [eosinophils, basophils, neutrophils], monocytes (monocytes), lymphocytes [helper T cells, killer T cells, B cells, NK cells, NKT cells]) are calculated (detected) using a high-speed calculation program installed in the data acquisition unit 210. This high-speed calculation program includes at least the following calculation formulas (1) to (5). In addition, the total number of white blood cells is calculated using formula (1) when the subject is male, and calculated using formula (2) when the subject is female.

[0081] [Mathematical Expression 1] Total white blood cell count = White blood cell count ( / μl) × 5000 (blood volume) ml × 1000 (1000μl = 1ml) (1) Total white blood cell count = White blood cell count ( / μl) × 4500 (blood volume) ml × 1000 (1000μl = 1ml) (2) Total number of granulocytes = Total number of white blood cells × Granulocytes (basophils + eosinophils + neutrophils) (cells) (%) (3) Total number of single cells = Total number of white blood cells × Single cells (%) (4) Total number of lymphocytes = Total number of white blood cells × Lymphocytes (%) (5) Next, the data acquisition unit 210 detects the total number of T cells in the lymphocyte subpopulation based on the total number of lymphocytes and CD classification analysis information obtained from the above detection.

[0082] Specifically, the total number of T cells in the lymphocyte subpopulation is calculated using a high-speed calculation program installed in the data acquisition unit 210, which includes the following calculation formula (6).

[0083] Next, the data acquisition unit 210 calculates the total number of helper T cells and the total number of killer T cells based on the calculated total number of T cells and CD classification analysis information.

[0084] The total number of helper T cells and the total number of killer T cells are calculated by using a high-speed computing program installed in the data acquisition unit 210, which includes the following calculation formulas (7) and (8).

[0085] Next, the data acquisition unit 210 detects the total number of non-T cells other than T cells based on the calculated total number of T cells, total number of lymphocytes, and CD classification analysis information.

[0086] The total number of non-T cells is calculated using a high-speed calculation program installed in the data acquisition unit 210, which includes the following calculation formula (9).

[0087] Next, the data acquisition unit 210 detects the total number of B cells, the total number of NK cells, and the total number of NKT cells based on the calculated total number of non-T cells, the total number of T cells, and CD classification analysis information.

[0088] The total number of B cells, the total number of NK cells, and the total number of NKT cells are calculated using a high-speed calculation program installed in the data acquisition unit 210. The high-speed calculation program includes the following calculation formulas (10), (11), and (12).

[0089] [Mathematical Expression 2] Total number of T cells = Total number of lymphocytes × CD3(+) (%) (6) Total number of CD4(+) T cells = Total number of T cells × CD4(+) (%) (7) Total number of CD8(+) T cells = Total number of T cells × CD8(+) (%) (8) Total number of non-T cells = Total number of lymphocytes × (100) CD3(+) (%) (9) Total number of B cells = Total number of non-T cells × (CD16) ) / CD56( )) (%) (10) Total number of NK cells = Total number of non-T cells × ((CD16(+) / CD56(+)) + (CD16(+) / CD56(+)) )))(11) Total number of NKT cells = Total number of T cells × (CD16( ) / CD56(+)) (%) (12) Furthermore, the number of each component in 1 μl can be calculated using the above calculation formulas (1) to (12). The data acquisition unit 210 saves the calculated data to the memory unit 3.

[0090] In addition, the data acquisition unit 210 outputs lymphocyte subpopulation data 32, which includes information on the number and proportion of immune lymphocyte subpopulations, including the number of B cells, the number of NK cells, the number of NKT cells, and at least the number and proportion of helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells and B cells, as well as basic data 31, as classification data to the data classification unit 211.

[0091] In the next step S2, as described above, the data classification unit 211 extracts the corresponding learning type and basic feature 341 from the classification data acquired by the data acquisition unit 210, referring to the classification extraction table 33. The data classification unit 211 outputs the acquired classification data, extracted learning type, and basic feature 341 as learning data to the learning unit 212. That is, the data classification unit 211 uses the basic feature 341 and the basic data 31 acquired from the memory unit 3 as basic elements 38, and the lymphocyte subpopulation data 32 as immune dynamic features 39, and outputs them to the learning unit 212. The basic feature 341 is... Figure 4 The basic feature data is represented as "34 points...", but the minimum score for basic feature 341 is set to 34 points, and is accumulated based on the specific items in basic data 31 and lymphocyte subpopulation data 32. That is, due to the following... Figure 11 The sum of the highest scores for all main items in the shown immune dynamic table 35 is 34; therefore, it is assumed that the lowest score for basic feature 341 is 34. In this invention, the learning institution 21 can determine the lowest score for basic feature 341 based on the learning progress. The classification extraction table 33 in this embodiment is only one example of the present invention; the learning institution 21 can also learn based on the immune status to form the classification extraction table 33.

[0092] In the next step S3, the learning unit 212 learns the classification model 600 or the basic model 601 based on the above learning data, and in step S4, it makes a judgment on the trained classification model 600 or the basic model 601.

[0093] Here, the learning of classification model 600 refers to using basic feature 341 as the numerator and immune dynamic feature 39 as the denominator, making the ratio (or proportion) between the two approach (or equal to) 1, and adjusting and comparing relevant elements accordingly. The adjustment and comparison of the aforementioned relevant elements can also be achieved by adjusting the weights described later and performing correlation calculations.

[0094] In step S4, when the learning unit 212 determines that the training result meets the benchmark range, that is... Figure 8If the condition "yes" is met, the learning process is considered complete, and a classification model 600 is formed. In step S4, when the learning unit 212 determines that the training result does not meet the baseline range, i.e. Figure 8 If the result is "No", the process returns to step S1 and the learning process is repeated. As an example, in this embodiment, the learning unit 212 learns the classification model 600 using a predetermined amount of learning data, that is, the predetermined number of executions of steps S1 to S3 is used as the predetermined learning process termination condition. Before the predetermined baseline range is met, the judgment result of step S4 is negative, and the process returns to step S1, repeating steps S1 to S3. On the other hand, when the predetermined learning process termination condition is met, the judgment result of step S4 is positive. When the judgment result of step S4 is positive, Figure 8 The learning process shown has ended.

[0095] Therefore, during the learning phase, learning institution 21 implements... Figure 8 The learning process shown enables the trained classification model 600 to be stored in the classification model box 34.

[0096] The evaluation phase (or assessment phase) in this embodiment will be described below. Figure 10 An example of the functional configuration of the immunodynamic evaluation unit 22 during the evaluation phase is shown. When the input data is evaluated using the classification model 600, the CPU of the immunodynamic evaluation unit 22 functions as the data acquisition unit 210 and the data classification unit 211 by executing the evaluation program.

[0097] When the immunodynamic evaluation unit 22 performs the immunodynamic evaluation process (hereinafter sometimes referred to as "evaluation process"), the immunodynamic evaluation unit 22 executes the evaluation procedure. Figure 10 The immunodynamic evaluation treatment shown is as follows. Figure 10 The process shown is the immunodynamic evaluation process in this embodiment. For example, the immunodynamic evaluation process is executed at a pre-set evaluation time, when the user issues an execution start command through the information input unit 4, or when other predetermined conditions are met.

[0098] As one implementation method, when there is no data related to the immune dynamics of the same user in memory unit 3, Figure 10 In step S10, the data acquisition unit 210 can acquire the basic data 31 corresponding to the classification model 600 from the memory unit 3 and use it as the evaluation basic data 41. In this case, the data acquisition unit 210 can also output the basic data 31 corresponding to the classification model 600 as the evaluation basic data 41 to the data classification unit 211.

[0099] As another implementation method, in Figure 10In step S10, the data acquisition unit 210 acquires the evaluation basis data 41 of the user who actually undergoes immunodynamic evaluation from the memory unit 3 or the information input unit 4. In this embodiment, the evaluation basis data 41 may also be the basis data 31 obtained by the user corresponding to the above-described classification model 600 after the formation of the classification model 600, such as blood tests.

[0100] Next, the data acquisition unit 210 generates evaluation data based on the evaluation baseline data 41. The following is one embodiment of the evaluation data generation, employing the same method as described above. Figure 8 The evaluation data is generated using the same method as the classification data in step S1.

[0101] That is, the data acquisition unit 210 adopts the same method as described above. Figure 8 Using the same method as the classification data formation method in step S1, evaluation lymphocyte subpopulation data 42 is formed based on the evaluation baseline data 41. This evaluation data includes information such as the number of B cells, NK cells, NKT cells, and at least the number and proportion of immune lymphocyte subpopulations including helper T cells, killer T cells, αβ killer-T cells, γδ T cells, NK cells, NKT cells, and B cells. Then, the evaluation lymphocyte subpopulation data 42 and the evaluation baseline data 41 are output as evaluation data to the data classification unit 211. Alternatively, when the data acquisition unit 210 acquires the evaluation lymphocyte subpopulation data 42 simultaneously with the evaluation baseline data 41 during the initial data acquisition stage, the evaluation lymphocyte subpopulation data 42 and the evaluation baseline data 41 can also be directly output as evaluation data to the data classification unit 211.

[0102] In step S11, the data classification unit 211, based on the evaluation base data 41 and the evaluation lymphocyte subpopulation data 42, obtains the classification model 600 for the same user from the classification model box 34 and uses it as the base model 601. Next, the data classification unit 211 outputs the base model 601, the evaluation base data 41, and the evaluation lymphocyte subpopulation data 42 to the learning unit 212. When there is no classification model 600 for the same user, a benchmark model 602 is obtained as the base model 601. The benchmark model 602 is a classification model formed by the learning institution 21 based on the average data of a majority of normal healthy individuals. The learning institution 21 of the present invention can form a benchmark model 602 through the above-described learning stage. Furthermore, the learning institution 21 can also, based on the embodiments of the present invention, according to... Figure 4 The disease categories shown form the baseline model 602.

[0103] In the next step S12, the learning unit 212 compares the overall relationship between the evaluation baseline data 41 and the evaluation lymphocyte subpopulation data 42 related to immune dynamics based on the baseline model 601 or the benchmark model 602.

[0104] As an example, the overall relationship comparison of immune dynamics refers to comparing the basic data 31 and lymphocyte subpopulation data 32 corresponding to the basic model 601 with the evaluation basic data 41 and evaluation lymphocyte subpopulation data 42, and determining whether they meet the specified range.

[0105] That is, the learning unit 212 compares the basic data 31 and lymphocyte subpopulation data 32 (hereinafter sometimes referred to as "basic immune dynamic data") corresponding to the basic model 601 with the evaluation basic data 41 and evaluation lymphocyte subpopulation data 42 (hereinafter sometimes referred to as "evaluation immune dynamic data"), and determines whether they meet the specified range. When it is determined that the specified range is met, that is... Figure 10 If the condition is "yes", the comparison process ends. The learning unit 212 compares the basic data 31 and lymphocyte subpopulation data 32 related to the basic model 601 with the evaluation basic data 41 and evaluation lymphocyte subpopulation data 42. If the condition does not meet the specified range, i.e. Figure 10 If the condition is "No", the process returns to step S10 and performs the comparison process again. For example, in this embodiment, the learning unit 212 can use a predetermined number of evaluation data to perform the comparison process, i.e., execute steps S10 to S12 a predetermined number of times, as a predetermined end condition for the comparison process. Before the specified range is met, the judgment result of step S12 is a negative judgment, and the process returns to step S10, repeating the processing of steps S10 to S12. On the other hand, when the predetermined end condition for the comparison process is met, the learning unit 212 performs the calculations of the learning phase steps S1 to S4 described above, thereby calculating a classification model 600 corresponding to the evaluation base data 41 and the evaluation lymphocyte subpopulation data 42. Then, the learning unit 212 uses the calculated classification model 600 as the evaluation model 603 and transfers it to step S13. That is, when the judgment result of step S12 is a positive judgment, the evaluation model 603 (hereinafter sometimes referred to as "immunodynamic evaluation data") is formed based on the calculation results of the ratio of basic characteristics and immune dynamic characteristics obtained from the evaluation basic data 41 and the evaluation lymphocyte subpopulation data 42.

[0106] On the other hand, in step S12, if it is ultimately determined that the specified range is not met, the learning unit 212 performs the calculations of steps S1 to S4 in the above-mentioned learning phase to calculate a new classification model 600. In this case, the newly calculated classification model 600 is used as the evaluation model 603, and the process moves to step S13. Next, in step S13, an immunodynamic evaluation is performed using the baseline model 602 as the base model 601 and the newly calculated classification model 600 as the evaluation model 603.

[0107] In step S13, the learning unit 212 compares the evaluation model 603 with the base model 601. When the evaluation model 603 is less than the calculation result of the base model 601, the learning unit 212 evaluates that the immune dynamics have improved. When the evaluation model 603 is greater than the calculation result of the base model 601, the learning unit 212 evaluates that the immune dynamics have deteriorated. When the calculation results of the evaluation model 603 and the base model 601 are equal, the learning unit 212 evaluates that the immune dynamics have not changed. The learning unit 212 stores the comparison result in the classification model box 34, displays the comparison result on the display unit 5, and ends the immune dynamics evaluation process.

[0108] In addition, Figure 10 The illustrated immunodynamic evaluation describes an implementation method in which the evaluation lymphocyte subpopulation data 42 is calculated and sequentially acquired in step S10, but the present invention is not limited thereto. For example, in the data acquisition step S10, the evaluation lymphocyte subpopulation data 42 can also be acquired by the user directly inputting it.

[0109] Thus, during the evaluation phase, learning institution 21 implements... Figure 10 The immunodynamic evaluation process shown compares and evaluates the overall relationship between the evaluation baseline data 41 and the evaluation lymphocyte subpopulation data 42 and the baseline model 601. After the comparison results are associated with the baseline model 601, they are stored in the classification model box 34 and displayed on the display unit 5.

[0110] The immune dynamics generation device 1 of the present invention includes: at least one immune dynamics processing unit 2, at least one memory unit 3 storing clinical data, and a display unit 5. The immune dynamics processing unit 2 acquires basic clinical characteristics and immune subpopulation dynamic information based on clinical data, and forms a classification model through self-learning based on the immune subpopulation dynamic information and the classification model, while simultaneously monitoring it; furthermore, based on the classification model, it evaluates the subject's immune dynamics information by comparing the overall relationship between basic clinical characteristics and immune subpopulation dynamic information, and simultaneously controls the display unit 5 to display the evaluation results of the immune dynamics information.

[0111] Furthermore, the immune dynamics generation device 1 of the present invention, through the aforementioned functions, can at least obtain information on normal immune status, abnormal (or aberrant) immune status, and cancer incidence rate. Therefore, it is possible to establish a comprehensive judgment standard capable of objectively evaluating immune dynamics, and to provide an immune dynamics generation method and device that can timely and accurately grasp the immune status and the progression of diseases, etc.

[0112] [Example 1] Example 1 describes an instance of obtaining dynamic immune information related to normal immune status based on health examination data of normal individuals. Specifically, in Example 1, the function of the immune dynamics processing unit 2 will be explained based on the construction of a basic model for healthy individuals and the evaluation of immune dynamics.

[0113] In Example 1, firstly, as a process corresponding to step S1, the data acquisition unit 210 acquires the basic data 31 of the subject (user) stored in the memory unit 3. Next, based on the acquired basic data 31, the lymphocyte subpopulation data 32 corresponding to the basic data 31 is calculated according to the processing flow of step S1 described above. The lymphocyte subpopulation data 32 includes the number of B cells, NK cells, NKT cells, and information such as the number and proportion of immune lymphocyte subpopulations, including at least the number of helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells. Furthermore, if the lymphocyte subpopulation data 32 is already stored in the memory unit 3, or if the user inputs the lymphocyte subpopulation data 32 through the information input unit 4, the above calculation processing operation is not required. Next, the data acquisition unit 210 outputs the lymphocyte subpopulation data 32 and the basic data 31 as classification data to the data classification unit 211.

[0114] Figure 11 This is a schematic diagram illustrating an embodiment of the immune dynamics table 35. Figure 12 This is a schematic diagram illustrating the basic data 31 of Example 1 (as an example, a routine blood test form from a Japanese hospital). Figure 13 This is a schematic diagram illustrating the number of each component in Example 1 calculated based on Formula 1 and Formula 2 above. Figure 14 This is a schematic diagram illustrating the number of granulocytes, single cells, lymphocytes, helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells in the classification data used in Example 1. Figure 15 This is a schematic diagram illustrating the proportions of granulocytes, unicellular cells, and lymphocytes in Example 1. Figure 16This is a schematic diagram illustrating the proportions of helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells in Example 1.

[0115] Data acquisition unit 210 acquires data from memory unit 3. Figure 12 The basic data 31 shown. Furthermore, the subject (user) can also directly input the basic data 31 into the data acquisition unit 210 via the information input unit 4. Following the processing flow of step S1 described above, the data acquisition unit 210 calculates the number of each component in the blood using formulas 1 and 2 based on the acquired basic data 31, such as... Figure 13 As shown.

[0116] Next, the data acquisition unit 210 will conduct a process of... Figure 13 By readjusting the number of each component in the blood shown, the lymphocyte subpopulation data 32 of the present invention can be obtained. In Example 1, by... Figure 13 Each component corresponding to lymphocyte subpopulation data 32 in the diagram is multiplied by the subject's hemoglobin (Hb) value of 15.5 to calculate the results as shown below. Figure 14 The data shown are the number of granulocytes, unicellular cells, lymphocytes, helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells used for classification.

[0117] Then, based on Figure 14 The data used for classification, including the number of granulocytes, monocytes, lymphocytes, helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells, were calculated to obtain... Figure 15 The proportions of granulocytes, monocytes, and lymphocytes in Example 1 shown, and Figure 16 The proportions of helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells in Example 1 are shown.

[0118] The lymphocyte subpopulation data 32 in Example 1 consists of the following data: Figure 14 The data shown are the number of granulocytes, unicellular cells, lymphocytes, helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells used for classification. Figure 15 The proportions of granulocytes, monocytes, and lymphocytes in Example 1 shown; and Figure 16The proportions of helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells in Example 1 are shown.

[0119] The data acquisition unit 210 outputs the above-mentioned lymphocyte subpopulation data 32 and basic data 31 as classification data to the data classification unit 211.

[0120] Data classification department 211, following step S2 above, refers to Figure 4 The classification extraction table 33 shown extracts the corresponding learning type and basic feature 341 from the classification data acquired by the data acquisition unit 210. The data classification unit 211 outputs the acquired classification data, the extracted learning type, and the basic feature 341 as learning data to the learning unit 212.

[0121] Specifically, the data classification unit 211 determines the learning type and basic features 341 based on the examinee's age, gender, health status, type of examination data, and characteristic immune elements. In Example 1, the examinee was a healthy male in his 50s, and the examination data consisted of blood test data with a CD4 value greater than or equal to 400 (above 400), thus similar to the healthy individuals in the classification extraction table 33. Therefore, for the examinee in Example 1, the data classification unit 211 extracts "healthy individual" as the learning type and extracts 34 points as basic features 341, which are then used as part of the learning data.

[0122] The data classification unit 211 outputs the above-mentioned lymphocyte subgroup data 32, basic data 31, healthy individuals as learning types, and 34 points as basic characteristics 341 as learning data to the learning unit 212.

[0123] Learning unit 212, following steps S3 and S4 above, learns the classification model 600 based on the aforementioned learning data. Then, based on... Figure 10 The evaluation phase shown determines the classification model 600 after training.

[0124] The learning unit 212 first performs the learning of the classification model 600. In Example 1, the learning unit 212 uses the basic feature 341 as the numerator and the immune dynamic feature 39 as the denominator, and adjusts the weights (basic feature weight 381, immune dynamic weight 391) and compares them to make the ratio (proportion) of the two approach 1 (the ratio of the two is 1).

[0125] Specifically, the learning unit 212 can use "34" contained in the aforementioned learning data as basic feature 341. Based on the basic data 31 and lymphocyte subpopulation data 32 contained in the aforementioned learning data, the learning unit 212 refers to... Figure 11The immune dynamic table 35 shown undergoes self-learning to form basic feature weights 381, immune dynamic features 39, and immune dynamic weights 391.

[0126] Figure 11 An embodiment of the immune dynamics table 35 is shown. The immune dynamics table 35 of the present invention includes immune dynamics master items (master disclosure items) and immune dynamics sub-items (sub-disclosure items). Figure 11 The main item shown is one embodiment of the immune dynamics main item. Figure 11 The sub-item shown is one example of the immune dynamics sub-item.

[0127] The main immune dynamics parameters include the number of helper T cells, the ratio of granulocytes:monocytes:lymphocytes, the proportion of activated T cells (HLA-DR / CD38%), Th1 immune intensity, and other immune cell indicators.

[0128] In this embodiment, helper T cells are classified into six levels, from Level 1 to Level 6, based on their number, and a scoring method is used for classification. For example, when the number of helper T cells is greater than or equal to 500 / μl, it is classified as Level 6, with the highest score of 12 points; when the number of helper T cells is 400-500 / μl, it is classified as Level 5, with a score of 10 points; when the number of helper T cells is 300-400 / μl, it is classified as Level 4, with a score of 8 points; when the number of helper T cells is 200-300 / μl, it is classified as Level 3, with a score of 4 points; when the number of helper T cells is 100-200 / μl, it is classified as Level 2, with a score of 2 points; and when the number of helper T cells is less than 100 / μl, it is classified as Level 1, with a score of 0 points.

[0129] The ideal ratio of granulocytes:monocytes:lymphocytes is 60:5:35. Therefore, when the ratio is approximately 60–65:5–10:30 or higher, a maximum score of 8 is given. When the ratio is approximately 65–75:5–10:20–30, a score of 6 is given. When the ratio is approximately 65–85:5–10:10–20, a score of 4 is given. When the ratio is approximately 85 or higher:5–10:less than 10, a score of 0 is given.

[0130] The score is 6 points when the proportion of activated T cells (HLA-DR / CD38%) is greater than or equal to 15%; 4 points when it is between 10% and 15%; 2 points when it is between 5% and 10%; and 0 points when it is less than 5%.

[0131] Th1 immune strength is scored as follows: 4 points for a value greater than 1, 2 points for a value equal to 1, and 0 points for a value less than 1. Similarly, other immune cell indicators can be graded using a similar method and evaluated using a scoring system before being added to the immune dynamics table 35.

[0132] The immune dynamics sub-item includes indicators that affect immune dynamics, such as Alb value, ChE value, Hb value, Ccr value, hyaluronic acid value, presence of liver dysfunction, changes in tumor markers, and changes in HLA-DR relative to CD38.

[0133] An Alb value greater than 3.8 g / dl is rated as 2 points; an Alb value between 3.2 and 3.8 g / dl is rated as 1 point; and an Alb value less than 3.2 g / dl is rated as 0 points. A ChE value greater than 300 U / l is rated as 2 points; a ChE value between 250 and 300 U / l is rated as 1 point; and a ChE value less than 250 U / l is rated as 0 points. A Hb value greater than 10 g / dl is rated as 2 points; an Hb value between 8 and 10 g / dl is rated as 1 point; and an Hb value less than 8 g / dl is rated as 0 points. A Ccr value greater than 80 ml / min is rated as 2 points; a Ccr value between 50 and 80 ml / min is rated as 1 point; and a Ccr value less than 50 ml / min is rated as 0 points. In addition, indicators such as hyaluronic acid value may also be included. For example, a hyaluronic acid level less than 50 ng / ml can be scored as 2 points; a hyaluronic acid level greater than 50 ng / ml can be scored as 0 points. Liver function impairment can also be scored. For example, the presence of liver function impairment can be scored as 2 points; the absence of liver function impairment can be scored as 0 points.

[0134] Furthermore, based on the evaluation of immunodynamics, the main immunodynamic items can be used as mandatory items, while the sub-items can be used as optional items. In this embodiment, the learning unit 212 learns from the learning data, including the basic data 31 and lymphocyte subpopulation data 32 acquired by the data acquisition unit 210, and refers to the necessary items corresponding to the immunodynamic table 35, thereby forming the basic features 341 and the immunodynamic features 39.

[0135] For example, in Example 1, the number of helper T cells was 492 / μl, corresponding to Level 5 in Immunomodulatory Table 35, which can be evaluated as 10 points. Next, the learning unit 212 will... Figure 15 The shown ratio of granulocytes, monocytes, and lymphocytes is compared to the ideal value of 60:5:35%. Figure 15 As shown, in Example 1, the ratio of granulocytes, monocytes and lymphocytes was 56:8:37%, corresponding to Level 4 in Immunodynamic Table 35, which can be evaluated as 8 points.

[0136] In Example 1, since the number of helper T cells was 492 / μl, corresponding to Level 5 in Immunodynamic Table 35, and rated as 10 points, the Th1 immune strength could be presumed to be greater than 1. Therefore, the learning unit 212 determined that the Th1 immune strength was greater than 1, corresponding to Level 3 in Immunodynamic Table 35, and rated as 4 points.

[0137] In Example 1, since the number of helper T cells was 492 / μl, corresponding to Level 5 in Immunodynamic Table 35, and can be evaluated as 10 points, it can be inferred that the proportion of activated T cells is greater than 15%. Therefore, the learning unit 212 judged that the proportion of activated T cells was greater than 15%, corresponding to Level 4 in Immunodynamic Table 35, and can be evaluated as 6 points.

[0138] In Example 1, since the number of helper T cells was 492 / μl, corresponding to Level 5 in Immunomodulatory Table 35, and can be evaluated as 10 points, therefore, as Figure 16 As shown, the proportions of the core elements in lymphocyte subpopulation data 32 can be presumed to be within the normal range. Therefore, the proportions of helper T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells, which are the core elements of lymphocyte subpopulation data 32, can be determined to correspond to Level 3 of "Other Immune Cell Indicators" in Immunodynamic Table 35, and are rated as 4 points.

[0139] That is, when helper T cells reach Level 5 or above in the immune dynamics table 35, the evaluation of the proportion of activated T cells, Th1 immune strength, and other immune cell indicators can be considered as each indicator being at its highest level and used for learning calculations. In this case, sub-items can be disregarded.

[0140] Alternatively, when the levels of helper T cells, granulocyte / monocyte / lymphocyte ratio, and activated T cell ratio all reach Level 4 or higher, the evaluation of Th1 immune strength and other immune cell indicators can be considered as each indicator being at its highest level and used for learning calculations. In this case, sub-items can also be disregarded.

[0141] The learning unit 212 forms basic feature 341 by referring to basic element 38 and immune dynamic feature 39. The subject in Example 1 was a healthy male in his 50s whose examination data consisted of blood test data and a CD4 value greater than 400, thus similar to the healthy individuals in classification extraction table 33. Accordingly, the data classification unit 211, for the subject in Example 1, determined the learning type to be healthy and set the score of basic feature 341 to 34 points, extracting it as part of the learning data. That is, when the subject is healthy, the sub-items in immune dynamic table 35 can be disregarded, therefore 34 points are used as the minimum basic score (or minimum number of basic points).

[0142] In this embodiment, when the learning unit 212 learns using the basic feature 341 as the numerator and the immune dynamic feature 39 as the denominator, and makes the ratio between the two approach 1, it can introduce a basic feature weight 381 into the numerator and / or an immune dynamic weight 391 into the denominator. Furthermore, in a single learning calculation, only one of the basic feature weight 381 or the immune dynamic weight 391 can be used, or both can be used simultaneously.

[0143] In this embodiment, when the helper T cells reach Level 5 or higher in the immune dynamics table 35, the learning unit 212 can select the immune dynamics weight 391 from a range of 4 to 22 points for self-learning. That is, when the subject is healthy and the helper T cells reach Level 5 or higher in the immune dynamics table 35, since the evaluation of the proportions of granulocytes, monocytes, and lymphocytes, the proportion of activated T cells, the Th1 immune strength, and other immune cell indicators can all be considered as reaching the highest level for each indicator, the learning unit 212 can determine the immune dynamics weight 391 within the range of the sum of the scores (points) corresponding to the highest level. Of course, if there is actual test data of the subject regarding any of the proportions of granulocytes, monocytes, and lymphocytes, the proportion of activated T cells, the Th1 immune strength, and other immune cell indicators, then that actual test data can also be used.

[0144] That is, when helper T cells reach Level 5 or above in the immune dynamics table 35, but the main immune dynamics item is only 10 points, the learning section 212 can select 22 points as the immune dynamics weight 391.

[0145] When helper T cells reach Level 5 or higher in Immunodynamics Table 35, the proportions of granulocytes, monocytes, and lymphocytes correspond to Level 4 in Immunodynamics Table 35, and the total score of the main Immunodynamics item is 18 points, Learning Unit 212 can select 14 points as the Immunodynamics weight 391. Similarly, when helper T cells reach Level 5 or higher in Immunodynamics Table 35, the proportions of granulocytes, monocytes, and lymphocytes correspond to Level 3 in Immunodynamics Table 35, and the total score of the main Immunodynamics item is 16 points, Learning Unit 212 can also select 14 points as the Immunodynamics weight 391.

[0146] That is, in the immune dynamics table 35, when the score extracted by the learning section 212 as the main item of immune dynamics is greater than the score corresponding to the specified level of each item, the score corresponding to the highest level of the next item can be subtracted (or deducted) from the immune dynamics weight 391. The "specified level" referred to here, for helper T cells, refers to Level 5 in the immune dynamics table 35; for other items, it can be defined as the highest level of that item.

[0147] In this embodiment, when the helper T cells reach Level 5 or above in the immune dynamics table 35, the learning unit 212 can use only the immune dynamics weight 391 instead of the basic feature weight 381.

[0148] In this embodiment, when the levels of the three indicators—helper T cells, granulocyte / monocyte / lymphocyte ratio, and activated T cell ratio—all reach Level 4 or above, the learning unit 212 may use only the immune dynamic weight 391 instead of the basic feature weight 381.

[0149] In Example 1, since the immune dynamic weight 391 can be selected from 4 to 22 points, the learning unit 212 can use only the immune dynamic weight 391 without using the basic feature weight 381.

[0150] In Example 1, the learning unit 212 can select 34 points of the aforementioned basic feature 341 as the numerator and 18 points of the aforementioned immune dynamic main item as the denominator. To make the ratio (proportion) of the two approach 1, an appropriate immune dynamic weight 391 is selected from 4 to 22 points for self-learning. When the predetermined benchmark range in step S4 is met, a classification model 600 is formed. The predetermined benchmark range in Example 1 refers to the range used to determine whether the value of the immune dynamic weight 391 has reached its maximum value, taking into account factors such as actual detection data. In Example 1, assuming that there is no actual data on the proportion of activated T cells, Th1 immune intensity, and other immune cells, and since the helper T cells reach Level 5 or above in the immune dynamic table 35, the sum of the highest scores (points) corresponding to the proportion of activated T cells, Th1 immune intensity, and other immune cell indicators, which is 14 points, can be defined as the aforementioned predetermined range.

[0151] That is, in Example 1, it is assumed that the learning unit 212, through self-learning, makes the basic feature 341 34 points, and the sum (total) of the immune dynamic feature 39 and the immune dynamic weight 391 reaches 32 points. Specifically, since the helper T cells reach Level 5 or above in the immune dynamic table 35, the proportions of granulocytes, monocytes, and lymphocytes correspond to Level 4 in the immune dynamic table 35, and the total score of the main immune dynamic items is 18 points, the learning unit 212 will repeatedly execute steps S1 to S3 to optimize (converge) the immune dynamic weight 391 to 14 points. Finally, when the immune dynamic weight 391 reaches 14 points, a classification model 600 is formed and stored in the classification model box 34.

[0152] like Figure 17 As shown, in Example 1, the learning unit 212 calculates based on the above elements and obtains: the basic feature 341 has 34 points; the sum (total) of the immune dynamic feature 39 and the immune dynamic weight 391 is 32 points; the ratio of the two is 1.06. Next, the learning unit 212 will... Figure 17 The calculation results shown are stored as a classification model 600 (an embodiment of normal immune status information) in the classification model box 34. Furthermore, while storing the classification model 600 in the classification model box 34, the learning unit 212 can also display it on the display unit 5.

[0153] Next, the evaluation phase of Example 1 will be performed. In Example 1, when there is no data related to the immune dynamics of the same user in memory unit 3, Figure 10 In step S10, the data acquisition unit 210 acquires the basic data 31 corresponding to the classification model 600 from the memory unit 3 and uses it as the evaluation basic data 41. Furthermore, in Figure 10In step S11, since the memory unit 3 does not store a classification model for the same user, the baseline model 602 is obtained as the base model 601. In step S12, the basic immune dynamic data corresponding to the baseline model 602 is compared with the evaluation immune dynamic data corresponding to the classification model 600. When a positive judgment is obtained, the classification model 600 can be output as the evaluation model 603. Then, in step S13, the learning unit 212 compares the evaluation model 603 with the baseline model 602 and performs an immunodynamic evaluation. Figure 18 As shown, in Example 1, the learning unit 212 determines that the ratio of the evaluation model 603 is 1.06, which is a ratio greater than the ratio of the baseline model 602 "1". Since its ratio is greater than 1, it is determined that the immune status of the subject is lower than the normal immune status, and the result is displayed on the display unit 5.

[0154] [Example 2] In Example 2, the function of the immune dynamics processing unit 2 will be explained based on the construction of a basic model of colorectal cancer patients and the evaluation of immune dynamics. That is, in Example 2, an example of obtaining immune dynamics information related to the immune status information of abnormal populations (abnormal immune status information) will be described. The parts in Example 2 that are the same as in Example 1 will not be repeated here.

[0155] Figure 19 This is a schematic diagram illustrating the basic data 31 in Example 2 and the number of granulocytes, single cells, lymphocytes, helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells used for classification in Example 2. Figure 20 This is a schematic diagram illustrating the proportions of granulocytes, unicellular cells, and lymphocytes in Example 2. Figure 21 This is a schematic diagram illustrating the proportions of helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells in Example 2.

[0156] In Example 2, the same processes as in Example 1 are omitted here. Step S1 in Example 2 is performed using the same method as in Example 1.

[0157] The lymphocyte subpopulation data 32 in Example 2 consists of the following data: Figure 19 The data shown are the number of granulocytes, unicellular cells, lymphocytes, helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells used for classification. Figure 20The proportions of granulocytes, monocytes, and lymphocytes in Example 2 shown; and the proportions of helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells.

[0158] The data acquisition unit 210 outputs the above-mentioned lymphocyte subpopulation data 32 and the basic data 31 of Example 2 (including basic data containing cancer cell data) as classification data to the data classification unit 211.

[0159] Data classification department 211, following step S2 above, refers to Figure 4 The classification extraction table 33 shown extracts the corresponding learning type and basic feature 341 from the classification data acquired by the data acquisition unit 210. The data classification unit 211 outputs the acquired classification data, the extracted learning type, and the basic feature 341 as learning data to the learning unit 212.

[0160] Specifically, the data classification unit 211 determines the learning type and basic feature 341 based on the examinee's age, gender, health status, type of examination data, and characteristic immune elements. In Example 2, the examinee was a male in his 50s with colorectal cancer. His examination data consisted of blood test data, and his CD4 value was less than his CD8 value (CD4 < CD8), thus similar to the colorectal cancer (colonial cancer) case in the classification extraction table 33. Therefore, for the examinee in Example 2, the data classification unit 211 determined the learning type to be colorectal cancer and set the score of basic feature 341 to 34 points, extracting it as part of the learning data.

[0161] The data classification unit 211 outputs the above-mentioned lymphocyte subpopulation data 32, basic data 31, colorectal cancer patients as learning types, and 34 points as basic features 341 as learning data to the learning unit 212.

[0162] The learning unit 212 learns the classification model 600 based on the learning data according to the above steps S3 and S4, and makes a judgment on the trained classification model 600.

[0163] The learning unit 212 first performs the learning of the classification model 600. In Example 2, the learning unit 212 uses the basic feature 341 as the numerator and the immune dynamic feature 39 as the denominator, and performs adjustments and comparison calculations on the relevant elements so that the ratio (proportion) between the two is 1.

[0164] Specifically, the learning unit 212 uses the 34 points contained in the aforementioned learning data as basic features 341. Based on the basic data 31 and lymphocyte subpopulation data 32 contained in the aforementioned learning data, the learning unit 212 refers to... Figure 11The immune dynamic table 35 shown undergoes self-learning to form basic feature weights 381, immune dynamic features 39, and immune dynamic weights 391.

[0165] In Example 2, the number of helper T cells was 127 / μl, corresponding to Level 2 in Immunomodulatory Dynamics Table 35, which can be evaluated as 2 points. Next, the learning unit 212 will... Figure 20 The shown ratio of granulocytes, monocytes, and lymphocytes is compared to the ideal value of 60:5:35%. Figure 20 As shown, in Example 2, the ratio of granulocytes, monocytes and lymphocytes was 68:9:23%, corresponding to Level 3 in Immunodynamic Table 35, which can be evaluated as 6 points.

[0166] In Example 2, since the number of helper T cells was 127 / μl, corresponding to Level 2 in Immunodynamic Table 35, and can be evaluated as 2 points, it can be inferred that the Th1 immune strength is less than 1. Therefore, the learning unit 212 judges that the Th1 immune strength is less than 1, corresponding to Level 1 in Immunodynamic Table 35, and can be evaluated as 0 points.

[0167] In Example 2, since the number of helper T cells was 127 / μl, corresponding to Level 2 in Immunodynamic Table 35, and can be evaluated as 2 points, it can be inferred that the proportion of activated T cells is less than 5%. Therefore, the learning unit 212 judges that the proportion of activated T cells is less than 5%, corresponding to Level 1 in Immunodynamic Table 35, and can be evaluated as 0 points.

[0168] In Example 2, since the number of helper T cells was 127 / μl, corresponding to Level 2 in Immunomodulatory Table 35, and can be evaluated as 2 points, therefore, as Figure 20 As shown, it can be inferred that the proportions of the core elements in lymphocyte subpopulation data 32 are in an abnormal state. Therefore, the proportions of helper T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells, which are the core elements of lymphocyte subpopulation data 32, can be determined to correspond to Level 1 of "Other Immune Cell Indicators" in Immunodynamic Table 35, and are rated as 0 points.

[0169] That is, when helper T cells are at or below Level 2 in the Immunodynamics Table 35 (less than or equal to Level 2), the evaluation of the proportion of activated T cells, Th1 immune intensity, and other immune cell indicators can be considered as each indicator being at the lowest level and used for learning calculations. However, when actual values ​​(measured values) of the proportion of activated T cells, Th1 immune intensity, and other immune cell indicators exist, these actual values ​​can also be used.

[0170] Alternatively, when the levels of helper T cells, granulocyte / monocyte / lymphocyte ratio, and activated T cell ratio are all below Level 2 (less than or equal to Level 2), the evaluation of Th1 immune strength and other immune cell indicators can be considered as each indicator being at its lowest level and used for learning calculations. However, when actual values ​​for the activated T cell ratio, Th1 immune strength, and other immune cell indicators exist, these actual values ​​can also be used.

[0171] Furthermore, in blood tests for colorectal cancer, changes in tumor markers and HLA-DR relative to CD38 are essential indicators affecting immune dynamics. In this case, indicators affecting immune dynamics, such as changes in tumor markers and HLA-DR relative to CD38, can be used for self-learning. In this case, these indicators are considered when forming the baseline feature weight 381 and the immune dynamics weight 391. In Example 2, refer to... Figure 11 The scores corresponding to changes in tumor markers and changes in HLA-DR relative to CD38, i.e., 6 points, can be calculated (or summed) into a basic feature weight of 381.

[0172] In Example 2, the detection value of tumor marker CA72-4 (below 6.9 U / ml) was 14.9, and the detection value of anti-P53 antibody (below 1.3 U / ml) was 22.73. Both were greater than the standard values, therefore corresponding to Level 1 in Immunomodulatory Table 35, which can be evaluated as 0 points.

[0173] In this embodiment, when the helper T cell is below Level 2 in the immune dynamic table 35, the learning unit 212 can select basic feature weights 381 from the range of 0 to 22 points for self-learning and form basic feature weights 381.

[0174] Learning Unit 212 is based on the actual detection data in Example 2 and Figure 11 The immune dynamics table 35 shown undergoes self-learning to appropriately form the basic feature weights 381 and immune dynamics weights 391. In Example 2, due to... Figure 11 The sub-projects include detection data for three items: tumor markers, HLA-DR(+)CD38, and HLA-DR(-)CD38. Therefore, the data for these three items can be included in the immunodynamic evaluation. Thus, as the basic feature weight of 381, the sum of the highest scores of the three items—tumor markers, HLA-DR(+)CD38, and HLA-DR(-)CD38—can be selected, which can be considered as 6 points.

[0175] In Example 2, since helper T cells correspond to Level 2 or below (less than or equal to Level 2) in the immune dynamics table 35, the proportions of granulocytes, monocytes and lymphocytes correspond to Level 3 in the immune dynamics table 35, and the main immune dynamics item is 8 points, it can be considered that the learning unit 212 selects 0 points as the immune dynamics weight 391.

[0176] like Figure 22 As shown, in Example 2, the learning unit 212 calculates based on the above-mentioned elements and obtains: the total of basic feature 341 and basic feature weight 381 is 40 points; the total of immune dynamic feature 39 and immune dynamic weight 391 is 8 points; the ratio of the two is 5. Next, the learning unit 212 will... Figure 22 The calculation results shown are stored as classification model 600 in classification model box 34. Furthermore, the learning unit 212 can display the classification model 600 on the display unit 5 simultaneously with storing it in the classification model box 34. Additionally, the learning unit 212 can also process the data using the same method as the evaluation stage processing in Example 1. For example, the classification model 600 from Example 2 can be used as evaluation model 603 and compared with the baseline model 602 to obtain cancer incidence information. The comparison result between evaluation model 603 and the baseline model 602 shows that the ratio of evaluation model 603 is 5, which is greater than the ratio of the baseline model 602 (1). Since its ratio is greater than 1, it is determined that the subject's immune status is in a dangerous state, and this result is displayed on the display unit 5. Furthermore, the evaluation model 603 from Example 2 can be compared with the classification model 600 (normal immune status information) from Example 1 to obtain cancer incidence information. That is, the ratio result 5 of the evaluation model 603 in Example 2 can be compared with the ratio result 1.06 of the classification model 600 in Example 1, and the ratio between the two can be used as cancer incidence information.

[0177] [Example 3] Example 3 describes the function of the immune dynamics processing unit 2 based on the same basic model construction and immunodynamic evaluation of the second examination of a colorectal cancer patient as in Example 2 above. Specifically, Example 3 will describe an example of obtaining immune dynamics information related to cancer incidence information.

[0178] The following combination Figure 10 The evaluation phase shown will be explained for Example 3. In step S10, the data acquisition unit 210 acquires the evaluation basic data 41 corresponding to the same user as in Example 2 from the memory unit 3 or the information input unit 4.

[0179] Next, the data acquisition unit 210 generates evaluation data based on the evaluation baseline data 41. The following is an example of the generation of evaluation data, using... Figure 8 Evaluation data is generated using the same method as the classification data generation method in step S1. The evaluation lymphocyte subpopulation data 42 can also be acquired simultaneously from the memory unit 3 or the information input unit 4 along with the evaluation baseline data 41. Then, the data acquisition unit 210 outputs the evaluation lymphocyte subpopulation data 42 and the evaluation baseline data 41 as evaluation data to the data classification unit 211.

[0180] In step S11, the data classification unit 211 obtains the classification model 600 of the same user as the base model 601 from the classification model box 34 based on the evaluation base data 41 and the evaluation lymphocyte subpopulation data 42. Then, the data classification unit 211 outputs the base model 601, the evaluation base data 41, and the evaluation lymphocyte subpopulation data 42 to the learning unit 212.

[0181] In the next step S12, the learning unit 212 compares the overall relationship between the evaluation baseline data 41 and the evaluation lymphocyte subpopulation data 42 regarding immune dynamics based on the baseline model 601 or the benchmark model 602.

[0182] As an example, the overall relationship comparison regarding immune dynamics involves comparing the baseline data 31 and lymphocyte subpopulation data 32 corresponding to the baseline model 601 with the evaluation baseline data 41 and evaluation lymphocyte subpopulation data 42 to determine whether the specified range is met.

[0183] That is, when the learning unit 212 determines that the basic data 31 and lymphocyte subpopulation data 32 corresponding to the basic model 601 meet the specified range compared with the evaluation basic data 41 and evaluation lymphocyte subpopulation data 42, that is... Figure 10 If the condition is "yes", the comparison process ends. When the learning unit 212 determines that the basic data 31 and lymphocyte subpopulation data 32 corresponding to the basic model 601 do not meet the specified range compared to the evaluation basic data 41 and evaluation lymphocyte subpopulation data 42, i.e. Figure 10 If the condition is "No", the process returns to step S10 and performs the comparison process again. The predetermined conditions for ending the comparison process are, for example, situations where at least the following conditions are met: the learning type (evaluation type) of the same user is the same; the type of health check (examination) is the same; and the lymphocyte subpopulation data are within a similar range.

[0184] When the predetermined comparison processing termination condition is met, the judgment in step S12 is affirmative, and the calculations in steps S1 to S4 of the learning phase described above are executed to calculate the classification model 600 corresponding to the evaluation baseline data 41 and the evaluation lymphocyte subpopulation data 42. Then, the calculated classification model 600 is used as the evaluation model 603, and the process proceeds to step S13. That is, when the judgment in step S12 is affirmative, the evaluation model 603 (hereinafter also referred to as "immunodynamic evaluation data") is formed based on the ratio calculation result of the baseline feature 48 and the immune dynamic feature 49 obtained from the evaluation baseline data 41 and the evaluation lymphocyte subpopulation data 42. Specifically, the following processing is performed in Example 3.

[0185] Figure 23 This is a schematic diagram illustrating the evaluation baseline data 41 in Example 3 and the number of granulocytes, single cells, lymphocytes, helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells used for classification data in Example 3. Figure 24 This is a schematic diagram illustrating the proportions of granulocytes, unicellular cells, and lymphocytes in Example 3. Figure 25 This is a schematic diagram illustrating the proportions of helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells in Example 3.

[0186] In Example 3, the same processes as in Example 2 are omitted here. Step S1 in Example 3 is performed using the same method as in Example 1.

[0187] The evaluation data 42 for lymphocyte subpopulation in Example 3 consists of the following data: Figure 23 The data shown are the number of granulocytes, unicellular cells, lymphocytes, helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells used for classification. Figure 24 The proportions of granulocytes, monocytes, and lymphocytes in Example 3 shown; and the proportions of helper T cells, killer T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells.

[0188] The data acquisition unit 210 outputs the evaluation basis data 41 corresponding to the second inspection and the above-mentioned evaluation lymphocyte subpopulation data 42 as classification data to the data classification unit 211.

[0189] Data classification department 211, following step S2 above, refers to Figure 4The classification extraction table 33 shown extracts the corresponding learning type and basic feature 341 from the classification data acquired by the data acquisition unit 210. The data classification unit 211 outputs the acquired classification data, the extracted learning type, and the basic feature 341 as learning data to the learning unit 212.

[0190] Since the subject in Example 3 is the same as the subject in Example 2, the data classification unit 211 is the same as in Example 2, which determines the learning type as colorectal cancer and extracts 34 points of the basic feature 341 as part of the learning data.

[0191] The data classification unit 211 outputs the above-mentioned evaluation lymphocyte subpopulation data 42, evaluation basic data 41, colorectal cancer as a learning type, and 34 points as basic features 341 as learning data to the learning unit 212.

[0192] The learning unit 212 learns the classification model 600 based on the learning data according to the above steps S3 and S4, and makes a judgment on the trained classification model 600.

[0193] In Example 3, the learning unit 212 uses the basic feature 341 as the numerator and the immune dynamic feature 39 as the denominator, and performs adjustments and comparison calculations on the relevant elements so that the ratio (proportion) between the two is 1.

[0194] Learning Unit 212 uses the 34 points contained in the aforementioned learning data as basic features 341. Based on the evaluation baseline data 41 and the evaluation lymphocyte subpopulation data 42 contained in the aforementioned learning data, Learning Unit 212 refers to... Figure 11 The immune dynamic table 35 shown undergoes self-learning to form basic feature weights 381, immune dynamic features 39, and immune dynamic weights 391.

[0195] In Example 3, the number of helper T cells was 78 / μl, corresponding to Level 1 in Immunomodulatory Table 35, which can be evaluated as 0 points. Figure 24 As shown, in Example 3, the ratio of granulocytes, monocytes and lymphocytes was 80:7:13%, corresponding to Level 3 in Immunodynamic Table 35, which can be evaluated as 4 points.

[0196] In Example 3, since the number of helper T cells was 78 / μl, corresponding to Level 1 in Immunodynamic Table 35, and can be evaluated as 0 points, it can be inferred that the Th1 immune strength is less than 1. Therefore, the learning unit 212 judges that the Th1 immune strength is less than 1, corresponding to Level 1 in Immunodynamic Table 35, and can be evaluated as 0 points.

[0197] In Example 3, since the number of helper T cells was 78 / μl, corresponding to Level 1 in the Immunodynamic Table 35, and can be evaluated as 0 points, it can be inferred that the proportion of activated T cells is less than 5%. Therefore, the learning unit 212 judges that the proportion of activated T cells is less than 5%, corresponding to Level 1 in the Immunodynamic Table 35, and can be evaluated as 0 points.

[0198] In Example 3, since the number of helper T cells was 78 / μl, corresponding to Level 1 in Immunomodulatory Table 35, and can be evaluated as 0 points, therefore, as Figure 23 As shown, it can be inferred that the proportions of the core elements in the lymphocyte subpopulation data 42 are in an abnormal state. Therefore, the proportions of helper T cells, αβkiller-T cells, γδT cells, NK cells, NKT cells, and B cells, which are the core elements for evaluating lymphocyte subpopulation data 42, can be determined to correspond to Level 1 of "Other Immune Cell Indicators" in the Immunodynamics Table 35, and are rated as 0 points.

[0199] That is, when helper T cells correspond to Level 1 in the immune dynamics table 35, the evaluation of the proportion of activated T cells, Th1 immune intensity, and other immune cell indicators can be considered as all indicators being at the lowest level and used for learning calculations. However, when actual values ​​for the proportion of activated T cells, Th1 immune intensity, and other immune cell indicators exist, those actual values ​​can also be used.

[0200] Alternatively, when the levels of helper T cells, granulocyte / monocyte / lymphocyte ratio, and activated T cell ratio are all below Level 2, the evaluation of Th1 immune strength and other immune cell indicators can be considered as all indicators being at the lowest level, and used for learning calculations. However, when actual values ​​for activated T cell ratio, Th1 immune strength, and other immune cell indicators exist, these actual values ​​can also be used.

[0201] In Example 3, the detection value of tumor marker CA72-4 (below 6.9 U / ml) was 14.6, and the detection value of anti-P53 antibody (below 1.3 U / ml) was 13.97. Since both were higher than the standard values, they correspond to Level 1 in Immunomodulatory Table 35 and can be evaluated as 0 points.

[0202] That is, in Example 3, since helper T cells correspond to Level 1 in the immune dynamics table 35, the proportions of granulocytes, monocytes and lymphocytes correspond to Level 2 in the immune dynamics table 35, and the total score of the main immune dynamics item is 4 points, it can be considered that the learning unit 212 selects 0 points as the immune dynamics weight 391.

[0203] Learning Unit 212 is based on the actual detection data in Example 3 and Figure 11 The immune dynamic table 35 shown undergoes self-learning to appropriately form the basic feature weights 381 and immune dynamic weights 391. In Example 3, as in Example 2, due to... Figure 11 The sub-projects include detection data for three items: tumor markers, HLA-DR(+)CD38, and HLA-DR(-)CD38. Therefore, the data for these three items can be included in the immunodynamic evaluation. Thus, as the basic feature weight of 381, the sum of the highest scores of the three items—tumor markers, HLA-DR(+)CD38, and HLA-DR(-)CD38—can be selected, which can be considered as selecting 6 points.

[0204] like Figure 26 As shown, in Example 3, the learning unit 212 calculates based on the above-mentioned elements and obtains: the total of basic feature 341 and basic feature weight 381 is 40 points; the immune dynamic feature 39 is 4 points; and the ratio of the two is 10. Next, the learning unit 212 will... Figure 26 The calculation results shown are stored as classification model 600 in classification model box 34. Furthermore, while storing the classification model 600 in classification model box 34, the learning unit 212 can also display it on display unit 5.

[0205] In this case, the learning unit 212 extracts the classification model 600 from Example 2 into a base model 601. Then, the learning unit 212 uses the classification model 600 from Example 3 as an evaluation model 603 and compares it with the base model 601. According to the general comparison rules of the present invention: when the evaluation model 603 is greater than the base model 601, the immune status is judged to be in a deteriorating state. When the evaluation model 603 is equal to the base model 601, the immune status is judged to be unchanged. When the evaluation model 603 is less than the base model 601, the immune status is judged to be in an improving state. The learning unit 212 can store the above judgment results in the classification model box 34 and also display the judgment results on the display unit 5.

[0206] In Example 3, as Figure 26 As shown, since the ratio of evaluation model 603 is greater than the ratio of the basic model 601 (10 > 5), the immune status is judged to be in a deteriorating state.

[0207] Although the present invention has been described above based on preferred embodiments, those skilled in the art will understand that various modifications, alterations, and substitutions can be made to these embodiments without departing from the principles and spirit of the invention. The present invention is not limited to the specific embodiments disclosed herein, and other embodiments included in the claims of this application are also included within the scope of the present invention.

[0208] Symbol Explanation 1. Immune dynamic generation device 2. Immune Dynamic Processing Department 3. Memory Department 4. Information Input Section 5 Display Section 6. External Information Sources 7. Network 21 Learning Institutions 210 Data Acquisition Department 211 Data Classification Department 212 Study Department 22 Immunological Dynamics Evaluation Department 31 Basic Data 32. Dynamic information on immune subpopulations (lymphocyte subpopulation data) 33. Category Extraction Table 34 Classification Model Boxes 341 Basic Features 35. Immune Dynamics Table 38 Basic Elements 600 classification model 601 Basic Model 602 Baseline Model 603 Evaluation Model

Claims

1. An immune dynamic generation device, characterized in that, include: At least one immune dynamic processing unit; At least a memory unit for storing clinical data; And the display section; The immune dynamics processing unit: acquires basic clinical characteristics and dynamic information of immune subpopulations based on the clinical data; forms a classification model by self-learning based on the dynamic information of immune subpopulations and the classification model, and monitors it simultaneously; and evaluates the immune dynamics of the examinee by comparing the overall relationship between the basic clinical characteristics and the dynamic information of immune subpopulations based on the classification model, and controls the display of the evaluation results of the immune dynamics information on the display unit.

2. The immune dynamic generation device according to claim 1, wherein, The clinical data includes at least health checkup data for normal individuals and health checkup data for individuals with abnormal health conditions.

3. The immune dynamic generation device according to claim 1, wherein, The dynamic information of the immune subpopulation includes: the number of B cells, the number of NK cells, the number of NKT cells, and the number and proportion of immune lymphocyte subpopulations, including at least the number of helper T cells, the number of killer T cells, the number of αβkiller-T cells, the number of γδT cells, the number of NK cells, the number of NKT cells, and the number of B cells.

4. The immune dynamic generation device according to claim 2, wherein, The immune dynamic processing unit: acquires the clinical basic characteristics, the dynamic information of the immune subpopulations, and the classification model based on the health examination data of the normal person, and obtains normal immune status information by comparing the overall relationship between the clinical basic characteristics and the dynamic information of the immune subpopulations.

5. The immune dynamic generation device according to claim 2, wherein, The immune dynamic processing unit: acquires the clinical basic characteristics, the dynamic information of the immune subpopulation, and the classification model based on the health examination data of the abnormal person, and obtains abnormal immune status information by comparing the overall relationship between the clinical basic characteristics and the dynamic information of the immune subpopulation.

6. The immune dynamic generation device according to claim 5, wherein, The immune dynamic processing unit: based on the clinical data including cancer cell data, it acquires the basic clinical characteristics, the dynamic information of the immune subpopulation, and the classification model; and based on the abnormal immune state information, it obtains cancer incidence information by comparing the overall relationship between the basic clinical characteristics and the dynamic information of the immune subpopulation.

7. The immune dynamic generation device according to claim 6, wherein, The immune dynamic processing unit determines that cancer cells are in a deteriorating state when the proportion of the cancer incidence information is greater than 1.

8. The immune dynamic generation device according to claim 6, wherein, The immune dynamic processing unit determines that cancer cells are in a prolongation state when the proportion of the cancer incidence information is equal to 1.

9. The immune dynamic generation device according to claim 6, wherein, The immune dynamic processing unit determines that cancer cells are in an improved state when the proportion of the cancer incidence information is less than 1.

10. A method for dynamic immune generation, characterized in that, include: Steps for obtaining clinical data from examinees; The steps for obtaining dynamic information on the immune subpopulations of the subjects; The process includes monitoring the average data of the normal majority and the abnormal data that have changed, calculated using the basic data and lymphocyte subpopulation data stored in the memory unit, and evaluating the immune dynamics of the subject by comparing the overall relationship between the clinical data and the immune subpopulation dynamic information based on the average data and the abnormal data.

Images