Method for assisting diagnosis of mental stress

Abstract

This method for assisting the diagnosis of mental stress comprises: a step for measuring the concentrations of a plurality of components contained in a biological sample collected from a subject; classifying the detected persons based on the personality features; and generating stress diagnosis assistance information for diagnosing the degree of mental stress of the subject on the basis of the determined concentrations of the components including betaine, creatine, GABA, creatinine, serotonin, kynurenic acid, 3-hydroxybutyric acid, and citric acid and the classification of the personality characteristics. According to the method, information serving as a support for accurately diagnosing the presence or absence or degree of mental stress of a subject can be provided.

Description

Technical Field

[0001] This invention relates to an auxiliary method for diagnosing mental stress. Background Technology

[0002] In recent years, the proportion of workers experiencing work-related mental stress, intense anxiety, and unease has increased, leading to a rise in cases of mental disorders caused by stress and other factors, and an increasing trend in the number of workers recognized as having suffered work-related injuries. In this specification, "stress" refers to "mental stress." Working under stress can potentially lead to serious accidents, depending on the nature of the work. Based on this situation, the Japanese Ministry of Health, Labour and Welfare has formulated the "Guidelines for the Maintenance and Promotion of Workers' Mental Health" to promote the implementation of measures for maintaining and promoting the mental health of workers in the workplace (mental health care) (Non-Patent Literature 1).

[0003] The aforementioned guidelines define "poor mental health" as "not only mental disorders classified as mental and behavioral disorders and suicide, but also mental and behavioral problems that may affect the physical and mental health, social life and quality of life of workers, such as stress, intense anxiety and unease." This implements countermeasures to prevent the occurrence of poor mental health or to detect it early.

[0004] For example, the "Partial Amendment Bill of the Occupational Safety and Health Law" (Law No. 82 of 2014), promulgated in Japan on June 25, 2014, established a "stress check system," which includes checks to assess the level of psychological stress (stress checks) and conduct interviews and guidance based on the results. This system mandates the active implementation of workplace mental health care, including stress check systems. Here, the "stress check system" refers to a series of comprehensive measures in the workplace covered by Article 66-10 of the Occupational Safety and Health Law.

[0005] The main methods for assessing stress are medical history taking and blood tests. Typical medical history taking methods include using occupational stress questionnaires and the PHQ9 (Patient Health Questionnaire-9). Typical blood tests include methods that use liquid chromatography, mass spectrometry, or similar instruments to measure the concentration of metabolites (biomarkers) in the blood that indicate stress levels, and then determine the degree of stress based on the results.

[0006] Existing technical documents

[0007] Non-patent literature

[0008] Non-patent document 1: Ministry of Health, Labor and Welfare Independent Administrative Agency Workers' Health and Safety Organization "Workplace Health and Safety Guidelines for Workplace Health and Safety, Maintenance and Improvement Guidelines~", March 2017, [Retrieved on May 17, 2023], Internet

[0009] <URL:https: / / www.mhlw.go.jp / content / 000560416.pdf> Summary of the Invention

[0010] The problem the invention aims to solve

[0011] Medical history taking largely relies on the subject's subjective account and attitude, making it difficult to establish absolute benchmarks for determining the presence and degree of stress based on the results. On the other hand, while blood tests determine the presence and degree of stress based on objective values ​​(concentrations) of biomarkers, they do not consider the influence of personality traits such as temperament on these values. For example, an outgoing and resilient subject may perceive stress differently than a timid and sensitive subject, resulting in differences in biomarker values ​​even under the same stressful environment. Therefore, problems exist such as overlooking the severe stress experienced by an outgoing and resilient subject, or conversely, overestimating the stress experienced by a timid and sensitive subject.

[0012] Furthermore, this section primarily addresses stress assessment in work and professional life, but the same issues exist in other areas such as family, living environment, and school life.

[0013] The problem this invention aims to solve is to provide information that can be used to accurately diagnose the presence and degree of mental stress in a subject.

[0014] Solution for solving the problem

[0015] The diagnostic aid method for mental stress according to the present invention, which addresses the aforementioned issues, includes the following steps: measuring the concentration of multiple components contained in a biological sample collected from a subject; classifying the subject based on personality traits; and generating stress diagnostic aid information based on the measured concentration of components and the classification of personality traits. This stress diagnostic aid information is used to diagnose the degree of stress in the subject. The multiple components include betaine, creatine, GABA (γ-aminobutyric acid), creatine anhydride, serotonin, kynurenic acid, 3-hydroxybutyric acid, and citric acid.

[0016] Invention Effects

[0017] In the diagnostic aid method for mental stress involved in this invention, stress diagnostic aid information is generated based on two factors: the concentration of specific metabolites in a biological sample, which is an objective indicator of mental stress, and the classification based on the individual characteristics of the examinee. Therefore, by using this information, the diagnostician can appropriately diagnose the presence and degree of mental stress in the examinee. Attached Figure Description

[0018] Figure 1 This is a process diagram illustrating one embodiment of the diagnostic aid method for mental stress involved in the present invention.

[0019] Figure 2 This is a graph showing the relationship between the measured and predicted scores of PHQ9, obtained from a stress assessment model (Model 1) constructed using the consultation score and the concentration of 11 components.

[0020] Figure 3 This is a graph showing the relationship between the measured and predicted scores of PHQ9, obtained from a stress assessment model (Model 2) constructed using the consultation score and the concentration of eight components.

[0021] Figure 4 This is a graph showing the relationship between the measured and predicted PHQ9 scores obtained from a stress assessment model (Model 3) constructed using consultation scores and the concentrations of 10 components.

[0022] Figure 5 This is a graph showing the relationship between the measured and predicted PHQ9 scores obtained from a stress assessment model (Model 4) constructed using the consultation score and the concentration of 10 components.

[0023] Figure 6 This is a graph showing the relationship between the measured and predicted scores of PHQ9, obtained from a stress assessment model (Model 5) constructed using consultation scores and the concentrations of 10 components. Detailed Implementation

[0024] Figure 1 This is a process diagram illustrating one embodiment of the diagnostic aid method for mental stress according to the present invention. The diagnostic aid method for mental stress of this embodiment includes the following steps: determining the concentration of multiple components contained in a biological sample collected from a subject; classifying the subject based on personality traits; and generating stress diagnostic aid information based on the determined component concentrations and the personality trait-based classification. The stress diagnostic aid information is information used to diagnose the degree of mental stress in the subject. The multiple components contained in the biological sample include betaine, creatine, GABA (γ-aminobutyric acid), creatine anhydride, serotonin, kynurenic acid, 3-hydroxybutyric acid, and citric acid.

[0025] Traditionally, assessments of a subject's mental stress rely heavily on subjective statements and attitudes, which can be problematic as blood tests may fail to reflect personality traits such as temperament. In contrast, the mental stress diagnostic aid method of this embodiment generates stress diagnostic aid information based on both the results of classifying subjects according to their personality traits and the concentrations of various specified components in biological samples collected from the subjects. This information allows for a more appropriate assessment of the subject's mental stress.

[0026] In the diagnostic aid method for mental stress of this embodiment, either the step of measuring the concentration of various components contained in a biological sample collected from the subject (hereinafter referred to as the "measurement step") or the step of classifying the subject based on personality characteristics (hereinafter referred to as the "classification step") can be performed. Furthermore, the diagnostic aid method for mental stress of this embodiment does not include a step in which the diagnostician judges the presence or degree of mental stress in the subject. Therefore, it is not a method for diagnosing humans.

[0027] In the testing process, regarding the "biological sample collected from the subject," there are no particular restrictions as long as the concentration of the components contained in the biological sample can be determined. Examples include blood, biological tissue, feces, urine, sweat, and saliva, with blood being preferred. Blood can be not only blood collected from the subject (whole blood), but also serum, plasma, etc., obtained by processing blood. For example, serum and plasma can be obtained by allowing blood to stand or by centrifugation.

[0028] Biological samples can be used directly to determine the concentration of components, but they can also be used for concentration determination after appropriate pretreatment as needed. Examples of pretreatment include, for instance, stopping enzyme reactions in the biological sample, removing lipid-soluble substances, and removing proteins. These pretreatments can be performed using known methods. Alternatively, biological samples can also be used for concentration determination after appropriate dilution or concentration.

[0029] As a method for determining the concentration of components in a biological sample, a known method can be appropriately selected based on the type of component. For example, the concentration of a component can be determined by selecting a quantitative method appropriate to the component being measured, from methods such as nuclear magnetic resonance (NMR)-based quantification, acid-base titration-based quantification, amino acid analyzer-based quantification, enzymatic quantification, quantification using aptamers such as nucleic acid aptamers and peptide aptamers, and colorimetric quantification. Alternatively, commercially available quantitative kits appropriate to the component can be used to determine the concentration. Furthermore, the concentration of a component can be determined, for example, by using capillary electrophoresis, liquid chromatography, gas chromatography, mass spectrometry, etc., alone or in appropriate combinations. These methods are particularly suitable for the simultaneous determination of multiple components.

[0030] Based on the advantages of obtaining analytical results from trace amounts of biological samples containing multiple components in a short time and the excellent separation capability of metabolites, liquid chromatography-mass spectrometry (LC-MS) is preferred for the determination process. In LC-MS, for example, the biological sample is appropriately pretreated, and the obtained peptide fragments are separated according to their respective retention times by liquid chromatography (LC), resulting in multiple peak outputs for each component. The output of this liquid chromatography is ionized by a mass spectrometer, and separation and detection are performed based on the mass-to-charge ratio (m / z). In addition to general single-type mass spectrometers, triple quadrupole mass spectrometers, Q-TOF mass spectrometers, TOF-TOF mass spectrometers, ion trap mass spectrometers, and ion trap time-of-flight mass spectrometers can also be appropriately used.

[0031] Among the components of the biological sample used in the testing process, biomarkers of mental stress include betaine, creatine, GABA, creatine anhydride, serotonin, kynurenic acid, 3-hydroxybutyric acid, and citric acid, which are known metabolites and have been shown, as described later, to improve the diagnostic performance of mental stress when combined with personality-based classification. In addition to these eight components, at least one of the three components consisting of kynurenine, tryptophan, and 3-hydroxykynurenine may also be included in the components used in the testing process. The choice between using eight components as biomarkers of mental stress, or using nine to eleven components obtained by adding at least one of the three components to the eight components, can be determined based on the individual characteristics of the subject obtained during the classification process. In this case, the classification process is performed before the testing process.

[0032] As a method of classification based on personality traits, the TACS-22 assessment scale (TACS22), HQ-25 questionnaire (HQ25), and Tachikawa Psychological Resilience Scale (TRS) are preferred, but other well-known and appropriate methods can also be used.

[0033] The TACS22 is one of the known methods for assessing modern depressive temperament (Modern Depressive Temperament Assessment Scale), consisting of 22 items. These 22 items are divided into three groups: one related to "avoidance of social roles," one related to "low self-esteem," and one related to "resentment." In the TACS22, each item is assigned a score from 0 to 4 based on the degree to which the respondent conforms to or does not conform to each item. The respondent is then categorized based on personality traits using the total score of all 22 items and the total score of each of the three groups.

[0034] The HQ25 is one of the known methods for assessing the degree of social withdrawal, consisting of 25 questions. These 25 questions are divided into five groups related to "socialization," "isolation," "lack of emotional support," and "lack of motivation." Similar to the TACS22, the HQ25 assigns scores from 0 to 4 to each item based on the degree to which the respondent conforms to or does not conform to them. The respondent is then categorized based on personality traits using the total scores of all 25 items and the total scores of each of the five group.

[0035] The Tachikawa Psychological Resilience Scale is one of the known methods for assessing resilience to mental stress. It consists of 10 items. Each item is scored from 1 to 7 based on the responses. The higher the total score across all 10 items, the greater the resilience. In other words, a higher total score indicates greater resilience to mental stress and a greater ability to maintain health while under stress.

[0036] The present invention will now be illustrated in more detail with examples of experiments, but the present invention is not limited to the following examples.

[0037] [Experimental Example 1]

[0038] [1] The examinee

[0039] The subjects were 259 Japanese adults (age range = average age 34 years (16-72 years), 132 men and 127 women).

[0040] [2] Recruitment of examinees

[0041] The subjects were patients confirmed to have depressive symptoms through diagnostic interviews and healthy individuals confirmed to have no depressive symptoms.

[0042] Based on the explanations provided to the individuals in the supplementary explanatory documents, patients and healthy individuals who consented to participate in this study were registered as subjects, and the following clinical information was obtained through medical records and consultations.

[0043] [Clinical Information]

[0044] Information related to age, sex, height, weight, and medical history (diagnosis, onset period, treatment course, medication history).

[0045] The consultation involves interviews conducted by psychiatrists and psychologists with both patients and healthy individuals. These interviews may be based on either the SCID (Structured Clinical Interview for DSM-IV) method, a simplified structured interview method for mental illness based on the MINI (Mini-International Neuropsychiatric Interview) method, or a general psychiatric interview that includes information on parenting history, lifestyle history, and present illness. When the SCID-based or MINI-based simplified structured interview method was used during treatment, this information is obtained from the patient's medical records.

[0046] In addition, 25 ml of blood was drawn to measure metabolites in the blood. All participants signed informed consent forms.

[0047] [2] Classification based on personality traits

[0048] All participants were interviewed using questionnaires corresponding to the Tachikawa Psychological Resilience Scale (TRS), TACS22, and HQ25, respectively. The personality traits of each participant were evaluated based on the results. The following eight items were used to assess the personality traits.

[0049] (1) Total score of all 10 items in the Tachikawa Psychological Resilience Scale

[0050] (2) Total score of questions related to “Avoidance of social roles” in TACS22

[0051] (3) Total score of the question items associated with “LowSE” in TACS22

[0052] (4) Total score of questions related to “Complaint” in TACS22

[0053] (5) Total score of questions related to “Socialization” in HQ25

[0054] (6) Total score of questions related to “Isolation” in HQ25

[0055] (7) Total score of questions related to “lack of emotional support” in HQ25

[0056] (8) Total score of questions related to “lack of motivation” in HQ25.

[0057] [3] Preparation of plasma samples

[0058] Plasma samples were collected as biological specimens via peripheral blood collection via venipuncture. For extracts of water-soluble metabolites, 100 μL (4 Vol) of ice-cold methanol was added to 25 μL of plasma, followed by vortexing, sonication, and centrifugation (14,000 × g, 4 °C, 15 min). The supernatant was collected and stored in 1.5 mL Eppendorf microcentrifuge tubes. For amino acid extracts, 100 μL (4 Vol) of 0.1 M perchloric acid was added to 25 μL of plasma, followed by vortexing, sonication, and centrifugation (14,000 × g, 4 °C, 15 min). The supernatant was collected in 1.5 mL Eppendorf microcentrifuge tubes.

[0059] For LC-MS analysis, the collected solution was diluted 10-fold with each mobile phase, and 5 μL of sample solution (equivalent to 0.1 μL of plasma) was used. Additionally, an internal standard corresponding to the metabolite being measured was added to the sample solution.

[0060] [4] LC-MS determination

[0061] The sample solutions were analyzed by LC-MS using an LCMS-8060 (manufactured by Shimadzu Corporation) as an ultra-high-speed triple quadrupole LC / MS / MS system. To determine a wide variety of water-soluble metabolites, the extracted solutions were separated using a Shim-pack GIST PFPP (2.1 mm I.D x 150 mm L., 3.0 μm (manufactured by Shimadzu Corporation)). A mobile phase consisting of solvent A (water + 0.06% formic acid) and solvent B (acetonitrile + 0.1% formic acid) was used. A gradient elution program was employed at a flow rate of 0.3 mL / min and an oven temperature of 40 °C.

[0062] 0~2 minutes: 0% Solvent B

[0063] 5 minutes: 25% Solvent B

[0064] 10 minutes: 35% Solvent B

[0065] 11-15 minutes: 95% Solvent B

[0066] 15.1-20 minutes: 0% Solvent B

[0067] In addition, the parameters for the positive / negative electrospray ionization modes are as follows.

[0068] Ionization mode: ESI positive / negative

[0069] Atomizer gas flow rate: 3.0 L / min

[0070] Heating gas flow rate: 10L / min

[0071] Drying gas flow rate: 10 L / min

[0072] Interface temperature: 400℃

[0073] DL temperature: 300℃

[0074] Heating block temperature: 500℃

[0075] CID gas: Tuning file (270 kPa)

[0076] [5] Data Analysis

[0077] Multiple response monitoring (MRM) data were processed using LabSolutions software (manufactured by Shimadzu Corporation) for peak acquisition and adjustment. Data on the individual characteristics of eight items and the composition of plasma samples were analyzed using Python (version 3.10.9). Data were automatically scaled, and k-means clustering analysis and chart plotting were performed using sklearn (machine learning library), XGboost (machine learning algorithm), and matplotlib (plotting library) to create a stress diagnostic model.

[0078] [6] Construction of the pressure determination model

[0079] In constructing the stress assessment model, a dataset consisting of the results (metabolite concentrations) obtained from LC-MS measurements of the aforementioned sample solutions and scores for eight items representing a classification based on individual characteristics was used as training data. Furthermore, for the metabolite concentration data used to construct the stress assessment model, the value obtained by dividing the ionic strength of the metabolite measured by LC-MS by the ionic strength of the corresponding internal standard was used. Additionally, a stress assessment was conducted on all subjects using the PHQ9 questionnaire, and the results were used as the positive result data for the training dataset. In this experimental example, the PHQ9 scores (0–27) were binned into 5-point intervals (0–4, 5–9, 10–14, 15–19, and 20 and above), with each interval designated as a score of 1–5 (Score 1: 0–4, Score 2: 5–9, Score 3: 0–14, Score 4: 15–19, Score 5: 20 and above). The detailed breakdown of the PHQ9 scores for the 259 subjects is as follows: score 1: 109 people, score 2: 49 people, score 3: 34 people, score 4: 36 people, and score 5: 31 people.

[0080] Next, the prescribed machine learning processing was performed using the aforementioned learning data to construct a learning model for stress assessment (stress assessment model). In this experimental example, when the stress assessment model was input into a dataset consisting of scores for eight items representing metabolite concentrations and personality-based classifications, it output a PHQ9 score as an indicator of the degree of mental stress. In other words, a stress assessment model capable of predicting PHQ9 scores was constructed based on scores for eight items representing the subject's metabolite concentrations and personality-based classifications.

[0081] The gradient boosting decision tree (regression model) algorithm was used when constructing the stress determination model. However, the learning algorithm is not limited to this; for example, it can use logistic regression, support vector machine (SVM), random forest, or deep learning algorithms that use neural networks.

[0082] In addition, the learning data of all subjects was split into training data (80%) and test data (20%). The training data was used as the model building data to adjust hyperparameters. Specifically, the model building data was split into 10 parts to construct a cross-validation model, and the hyperparameters were optimized based on the positive resolution rate, RMSE (root mean square error), and MAE (mean absolute error) of each model. As a result, 11 components (betaine, creatine, GABA, creatinine, serotonin, kynurenic acid, 3-hydroxybutyric acid, citric acid, kynurenine, tryptophan, and 3-hydroxykynurenine) were extracted as effective indicators (biomarkers) for judging whether subjects have mental stress.

[0083] Therefore, the stress judgment model was reconstructed using a dataset consisting of the concentrations of the aforementioned 11 components and scores representing 8 items in a personality-based classification. Evaluation of the constructed stress judgment models using test data showed that the model constructed with the 11 components, the model constructed with 10 components (removing any one of kynurenine, tryptophan, and 3-hydroxykynurenine) from the 11 components, and the model constructed with 8 components (removing kynurenine, tryptophan, and 3-hydroxykynurenine) all had high positive solution rates.

[0084] The following are the pressure determination models constructed using 11 components (betaine, creatine, GABA, creatine anhydride, serotonin, kynurenic acid, 3-hydroxybutyric acid, citric acid, kynurenine, tryptophan, and 3-hydroxykynurenine) as Model 1, the pressure determination model constructed using 8 components (betaine, creatine, GABA, creatine anhydride, serotonin, kynurenic acid, 3-hydroxybutyric acid, and citric acid) as Model 2, and the pressure determination models constructed using 10 components obtained by removing 3-hydroxykynurenine, tryptophan, and kynurenine from the above 11 components as Models 3-5. The pressure determination accuracy of each model was evaluated.

[0085] Figures 2-6 The evaluation results for models 1 through 5 are shown in the figure. Figures 2-6 The evaluation results are plotted using both test data and training data. Figures 2-6 In the original text, black dots of varying shades are used to plot the evaluation results of test and training data. However, in the actual graph, red dots represent the evaluation results using the test data, and blue dots represent the evaluation results using the training data. Furthermore, darker dots indicate a larger number of data points. Figures 2-6 In the diagram, the horizontal axis represents the PHQ9 score (1~5) value (measured value) of each training data, and the vertical axis represents the output value (predicted value) when each training data is input into the stress judgment model.

[0086] In addition, the coefficient of determination (R²) is calculated based on the output and measured values ​​of the PHQ9 scores when test and training data are input into models 1-5. 2 The results are shown in Table 1 below.

[0087] [Table 1]

[0088]

[0089] As shown in Table 1, models 1 through 4 all display the determination coefficients (R²) of the training data. 2The coefficient of determination (R²) for the test data is a high value above 0.70. 2 The coefficient of determination for the training data is lower than that for the test data, but both show values ​​above 0.50. Additionally, Model 5 shows the coefficients of determination (R²) for both the training and test data. 2 The scores were 0.56 and 0.47, respectively, and lower than other models, but all around 0.50. Based on these results, models 1-5 are considered useful as models for predicting PHQ9 scores using a dataset consisting of scores for eight items representing categories based on personality traits, with specified component concentrations.

[0090] The PHQ9 score can supplement information used to diagnose the presence and degree of stress (relative to the stress diagnostic aid information of this invention). Therefore, it has been confirmed that 11 components obtained by adding 3 components (kynurenine, tryptophan, and 3-hydroxykynurenine) to these 8 components (betaine, creatine, GABA, creatinine, serotonin, kynurenine, 3-hydroxykynurenine), or 10 components obtained by removing any one of kynurenine, tryptophan, and 3-hydroxykynurenine from these 11 components, can serve as biomarkers of mental stress through combination with personality-based classifications. Diagnostic aid information that improves the accuracy of stress diagnosis can be provided based on the concentration of these biomarkers and the personality-based classifications.

[0091] Furthermore, in Experiment 1 above, a dataset consisting of metabolite concentrations in the blood collected from the subjects and scores on eight items representing a personality-based classification was used as input data. The results of a stress assessment conducted using the PHQ9 were used as training data for positive responses. However, alternatively, the results of an occupational stress questionnaire could be used instead of the PHQ9 assessment results. Similar to the PHQ9 scores, the results of the occupational stress questionnaire can supplement information used to diagnose the presence and severity of stress.

[0092] [Way]

[0093] The exemplary embodiments described above are specific examples of the following methods, which will be obvious to those skilled in the art.

[0094] (First item) One aspect of the present invention relates to a diagnostic aid method for mental stress, comprising the following steps:

[0095] Determine the concentration of multiple components contained in biological samples collected from subjects;

[0096] The subjects were classified based on their individual characteristics; and

[0097] Stress diagnostic auxiliary information is generated based on the measured concentrations of the components and a classification based on personality traits. This stress diagnostic auxiliary information is used to diagnose the degree of mental stress in the subject.

[0098] The various components include betaine, creatine, GABA, creatine anhydride, serotonin, kynurenic acid, 3-hydroxybutyric acid, and citric acid.

[0099] According to the diagnostic aid method for mental stress involved in the first item, information useful for diagnosing whether the subject is under mental stress and for diagnosing the degree of mental stress can be generated based on the concentration of various components contained in the biological sample collected from the subject and the subject's personality characteristics. Therefore, by providing this information to the diagnostician, the accuracy of the diagnostician's stress diagnosis can be improved.

[0100] (Second item) Based on the diagnostic aids for mental stress involved in the first item, the multiple components of the diagnostic aids for mental stress involved in the second item also include kynurenine, tryptophan and 3-hydroxykynurenine.

[0101] Based on the diagnostic aids for mental stress involved in the second item, the accuracy of stress diagnosis by the diagnostician can be further improved.

[0102] (Third item) Based on the diagnostic aids for mental stress involved in the first or second item, the diagnostic aids for mental stress involved in the third item determine the concentration of the components contained in the biological sample by liquid chromatography-mass spectrometry analysis.

[0103] According to the diagnostic aids for mental stress involved in the third item, it is possible to determine the concentration of multiple components contained in trace biological samples in a short time.

[0104] (Fourth item) Based on the diagnostic aids for mental stress involved in any of the first to third items, the fourth item's diagnostic aids for mental stress involves a classification based on the aforementioned personality traits that utilizes one or more of the classifications selected from TACS-22, HQ-25, and Tachikawa Psychological Resilience Scale.

[0105] Based on the diagnostic aids for mental stress involved in the fourth item, it is possible to provide diagnostic aids for stress that can take into account the subject's personality and temperament to determine the presence and degree of stress.

[0106] (Fifth item) Based on the diagnostic aids for mental stress involved in any of the first to fourth items, the diagnostic aids for mental stress involved in the fifth item further includes the process of providing a report, which includes the generated diagnostic aids for stress and the results of the occupational stress questionnaire.

[0107] (Sixth item) Based on the diagnostic aids for mental stress involved in any of the first to fourth items, the diagnostic aids for mental stress involved in the sixth item further includes the step of providing a report, the report containing the generated diagnostic aids for stress and a PHQ9 score.

[0108] According to the diagnostic aids for mental stress involved in item 5 or 6, a report containing diagnostic aids for stress, as well as information equivalent to the results of the occupational stress questionnaire and the PHQ9 score, can be provided to the diagnostician. The occupational stress questionnaire is effective in diagnosing the presence and degree of mental stress in the examinee's work and professional life.

[0109] (Seventh item) Based on the diagnostic aids for mental stress involved in any of the first to fourth items, the diagnostic aids for mental stress involved in the seventh item are a machine learning model that takes the concentration of the measured components and the classification based on personality traits as input data and information equivalent to the results of the interview used for stress examination as output data.

[0110] According to the diagnostic aids for mental stress involved in item seven, information can be provided that is effective in diagnosing the presence and degree of mental stress in subjects based on both subjective and objective indicators.

[0111] (Eighth item) Another aspect of the present invention relates to a method for generating a judgment model for determining the degree of mental stress of a subject, the method comprising the following steps:

[0112] Acquire a dataset in which the concentrations of various components contained in biological samples collected from the subject and the subject’s personality traits are used as input data, and information equivalent to the results of the Occupational Stress Questionnaire or the PHQ9 score is used as positive data.

[0113] Machine learning is performed based on the dataset, thereby enabling the decision model to learn; and

[0114] The appropriateness of the judgment model is evaluated using the prescribed evaluation dataset.

[0115] In the method for generating the aforementioned judgment model, the aforementioned components include, for example, betaine, creatine, GABA, creatine anhydride, serotonin, kynurenic acid, 3-hydroxybutyric acid, and citric acid. According to the above generation method, by inputting the concentrations of various components contained in a biological sample collected from the subject and the subject's personality characteristics, a judgment model can be generated that outputs information useful for diagnosing whether the subject is experiencing psychological stress, information useful for diagnosing the degree of psychological stress experienced, and the appropriateness of the generated judgment model can be evaluated.

[0116] (Ninth item) Based on the method for generating the judgment model involved in the eighth item, in the evaluation dataset of the method for generating the judgment model involved in the ninth item, it is possible to use data that is partially or completely different from the input data contained in the dataset used in the process of making the judgment model learn, and to use information equivalent to the results of the occupational stress questionnaire or the PHQ9 score as the correct answer data.

[0117] Based on the method for generating the judgment model involved in item nine, it is possible to correctly evaluate whether the judgment model obtained through machine learning is appropriate as a model that provides effective information for diagnosing the presence and degree of mental stress in the examinee.

Claims

1. A diagnostic aid for mental stress, comprising the following steps: Determine the concentration of multiple components contained in biological samples collected from subjects; The subjects were categorized based on their individual characteristics; as well as Stress diagnostic auxiliary information is generated based on the measured concentrations of the components and a classification based on personality traits. This stress diagnostic auxiliary information is used to diagnose the degree of mental stress in the subject. The various components include betaine, creatine, γ-aminobutyric acid, creatine anhydride, serotonin, kynurenic acid, 3-hydroxybutyric acid, and citric acid.

2. The diagnostic aid method for mental stress according to claim 1, wherein, The various components also include kynurenine, tryptophan, and 3-hydroxykynurenine.

3. The diagnostic aid method for mental stress according to claim 1 or 2, wherein, The concentration of components contained in the biological sample was determined by liquid chromatography-mass spectrometry analysis.

4. The diagnostic aid method for mental stress according to claim 1 or 2, wherein, The personality-based classification utilizes one or more of the following: the TACS-22 assessment scale, the HQ-25 questionnaire, and the Tachikawa psychological resilience score.

5. The diagnostic aid method for mental stress according to claim 1 or 2, wherein, It also includes the process of providing a report, which contains the generated stress diagnostic support information and the results of the occupational stress questionnaire.

6. The diagnostic aid method for mental stress according to claim 1 or 2, wherein, It also includes the process of providing a report containing the generated stress diagnostic aid information and the PHQ9 questionnaire score.

7. The diagnostic aid method for mental stress according to claim 1 or 2, wherein, The stress diagnostic auxiliary information is a machine learning model that takes the concentration of the measured components and the classification based on personality traits as input data, and information equivalent to the results of a consultation used for stress examination as output data.

8. A generation method for generating a judgment model to determine the degree of mental stress in a subject, wherein, The generation method includes the following steps: Acquire a dataset in which the concentrations of various components contained in biological samples collected from the subject and the subject’s personality traits are used as input data, and information equivalent to the results of the occupational stress questionnaire or the score of the PHQ9 questionnaire is used as positive data. Machine learning is performed based on the dataset, thereby enabling the decision model to learn; as well as The appropriateness of the judgment model is evaluated using the prescribed evaluation dataset.

9. The generation method according to claim 8, wherein, In the evaluation dataset, data that is partially or completely different from the input data contained in the dataset used in the process of making the judgment model learn is used as input data, and information equivalent to the results of the occupational stress questionnaire or the scores of the PHQ9 questionnaire are used as positive answer data.

Images