Classification methods

Abstract

This invention provides a method for classifying data into predetermined clusters based on features obtained from time-series data of salivary sIgA. [Solution] The classification method of the present disclosure includes an acquisition step, an extraction step, and a classification step, wherein the acquisition step acquires time-series data of saliva sIgA to be classified, the extraction step extracts features from the time-series data, and the classification step classifies the subject to be classified into one of predetermined clusters based on the features, and each of the steps is performed by a computer.

Description

Technical Field

[0001] This disclosure relates to a classification method.

Background Art

[0002] As described in Non-Patent Document 1, it is known that the concentration of salivary sIgA at a single point in time serves as an indicator of acute stress and the like.

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] On the other hand, since salivary sIgA has large individual differences and further large intra-day and inter-day variations, its usefulness is limited in a single measurement. Therefore, there is a need for a new method for evaluating health risks using salivary sIgA to appropriately grasp health risks.

[0005] Therefore, an object of this disclosure is to provide a classification method for classifying an object to be classified into a predetermined cluster based on feature quantities obtained from time-series data of salivary sIgA.

Means for Solving the Problems

[0006] To achieve the above object, the classification method of this disclosure includes an acquisition step, an extraction step, and a classification step, wherein the acquisition step acquires time-series data of salivary sIgA of an object to be classified, and the extraction step extracts feature quantities from the time-series data, The classification step involves classifying the object to be classified into one of the predetermined clusters based on the feature quantities. This method involves each of the aforementioned steps being performed by a computer. [Effects of the Invention]

[0007] According to this disclosure, it is possible to provide a classification method for classifying objects into predetermined clusters based on features obtained from time-series data of salivary sIgA. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a block diagram showing the configuration of an example of the classification device of this disclosure. [Figure 2] Figure 2 is a block diagram showing an example of the hardware configuration of the classification device of this disclosure. [Figure 3] Figure 3 is a flowchart showing an example of the steps in the classification method of this disclosure. [Figure 4] Figure 4 is a heatmap showing the results of principal component analysis (PCA) as a correlation between "feature quantity × principal component (PC)". [Figure 5] Figure 5 shows the graph of the elbow method used to determine the number of clusters in k-means. [Figure 6] Figure 6 is a scatter plot showing the clustering results. [Figure 7A] Figure 7A shows the mean sIgA concentration between groups (violin plot + statistical significance). [Figure 7B] Figure 7B shows the intergroup comparison of the coefficient of variation (violin plot + statistical significance). [Figure 7C] Figure 7C shows the intergroup comparison of variance instability (violin plot + statistical significance). [Figure 7D] Figure 7D shows the intergroup comparison of entropy (violin plot + statistical significance). [Figure 7E] Figure 7E is a heatmap of the effect size between groups. [Figure 8A] FIG. 8A is a stacked bar graph of the sex ratio by cluster. [Figure 8B] FIG. 8B is a box plot of the sIgA concentration distribution by cluster for each sex. [Figure 9A] FIG. 9A is a bar graph showing the relationship between various symptoms, the number of positive cases, and the positive rate. [Figure 9B] FIG. 9B is a heat map and the average symptom score by cluster. [Figure 10] FIG. 10 is a graph showing the number of cough positive cases by cluster. MODE FOR CARRYING OUT THE INVENTION

[0009] Embodiments of the present disclosure will be described. Note that the present disclosure is not limited to the following embodiments. In the following figures, the same parts are denoted by the same reference numerals. Also, unless otherwise specified, the descriptions of the respective embodiments can be mutually referred to. Furthermore, unless otherwise specified, the configurations of the respective embodiments can be combined. Also, each of the procedures described later in the program of the present disclosure can be read as "processing" instead of "procedure", for example.

[0010] [Embodiment 1] A classification device, a classification method, and a classification program of the present disclosure will be described.

[0011] An example of the classification device of the present disclosure will be described based on FIGS. 1 and 2.

[0012] FIG. 1 is a block diagram showing a configuration of an example of a classification device 10 (this device 10) of the present disclosure. As shown in FIG. 1, the device 10 includes an acquisition unit 11, an extraction unit 12, and a classification unit 13.

[0013] The device 10 may be, for example, a single device including the aforementioned parts, or it may be a system device in which each of the aforementioned parts is a separate device that can be connected via a communication network. The device 10 can also be connected to external devices described later via the communication network. The communication network is not particularly limited and can use any known network, such as a wired or wireless network. Examples of the communication network include the Internet, WWW (World Wide Web), telephone lines, LAN (Local Area Network), SAN (Storage Area Network), DTN (Delay Tolerant Networking), LPWA (Low Power Wide Area), L5G (Local 5G), etc. Examples of wireless communication include Wi-Fi (registered trademark), Bluetooth (registered trademark), Local 5G, LPWA, etc. The wireless communication may be in the form of direct communication between devices (Ad Hoc communication), infrastructure communication, indirect communication via an access point, etc. The device 10 may be, for example, incorporated into a server as a system. Furthermore, the device 10 may be, for example, a personal computer (PC, e.g., desktop or notebook), smartphone, tablet terminal, etc., on which the program of this disclosure is installed. The device 10 may also be in the form of cloud computing or edge computing, for example, in which at least one of the aforementioned parts is on a server and the other aforementioned parts are on a terminal.

[0014] Figure 2 illustrates a block diagram of the hardware configuration of the device 10. The device 10 includes, for example, a central processing unit (CPU, GPU, etc.) 101, memory 102, bus 103, storage device 104, input device 105, output device 106, communication device 107, etc. Each part of the device 10 is interconnected via the bus 103 through its respective interface (I / F).

[0015] The central processing unit 101 operates in coordination with other components via controllers (system controller, I / O controller, etc.) and is responsible for the overall control of the device 10. In the device 10, the central processing unit 101 executes, for example, the program disclosed herein and other programs, and also reads and writes various types of information. Specifically, for example, the central processing unit 101 functions as an acquisition unit 11, an extraction unit 12, and a classification unit 13. The central processing unit 101 may be equipped with arithmetic units such as a CPU, GPU (Graphics Processing Unit), APU (Accelerated Processing Unit), NPU (Neural Processing Unit), or a combination thereof.

[0016] Bus 103 can also be connected to external devices, for example. Examples of such external devices include external storage devices such as external databases, printers, external input devices, external display devices, and external imaging devices. The device 10 can be connected to an external network (the aforementioned communication network) by a communication device 107 connected to bus 103, for example, and can also be connected to other devices via the external network.

[0017] Memory 102 may be, for example, main memory. When the central processing unit 101 performs processing, memory 102 reads various operational programs, such as the program of this disclosure, stored in the storage device 104 (described later), and the central processing unit 101 receives data from memory 102 and executes the program. The main memory may be, for example, RAM (random access memory). Alternatively, memory 102 may be, for example, ROM (read-only memory).

[0018] The storage device 104 is also called an auxiliary storage device, for example, in relation to the main memory (primary memory). As described above, the storage device 104 stores an operating program including the program of this disclosure. The storage device 104 may be, for example, a combination of a recording medium and a drive for reading and writing to the recording medium. The recording medium is not particularly limited and may be internal or external, for example, an HD (hard disk), CD-ROM, CD-R, CD-RW, MO, DVD, flash memory, memory card, etc. The storage device 104 may be, for example, a hard disk drive (HDD) in which the recording medium and the drive are integrated, or a solid state drive (SSD).

[0019] In this device 10, the memory 102 and storage device 104 can also store various types of information, such as log information, information obtained from an external database (not shown) or external devices, information generated by this device 10, and information used by this device 10 when executing processing. In this case, the memory 102 and storage device 104 may store, for example, time-series data, features, and clusters, as described later. At least some of the information may be stored on an external server other than the memory 102 and storage device 104, or it may be stored in a distributed manner across multiple terminals using blockchain technology or the like.

[0020] The device 10 further includes, for example, an input device 105 and an output device 106. The input device 105 may include, for example, a pointing device such as a touch panel, trackpad, or mouse; a keyboard; imaging means such as a camera or scanner; a card reader such as an IC card reader or magnetic card reader; an audio input means such as a microphone; and so on. The output device 106 may include, for example, a display device such as an LED display or liquid crystal display; an audio output device such as a speaker; a printer; and so on. In this disclosure, the input device 105 and the output device 106 are configured separately, but the input device 105 and the output device 106 may be configured as an integrated unit, such as a touch panel display.

[0021] Next, an example of the classification method of this disclosure will be explained based on the flowchart in Figure 3. The classification method of this disclosure is carried out as follows, for example, using the apparatus 10 in Figure 1 or Figure 2. However, the classification method of this disclosure is not limited to the use of the apparatus 10 in Figure 1 or Figure 2. In the classification method of this disclosure, each step described later is performed by a computer, but for example, each step described later may be performed by something other than a computer. That is, the entity performing each step described later in the classification method of this disclosure may be a computer or a human. Furthermore, some of the steps described later may be performed by a computer, and some of the steps described later may be performed by a human.

[0022] The acquisition unit 11 acquires time-series data of saliva sIgA to be classified (S11, acquisition step). The subject to be classified is, for example, a human or a non-human animal. For example, the time-series data is, for example, continuous data of the saliva sIgA concentration. The time-series data may be, for example, continuous data of 2 days or more, 5 days or more, 10 days or more, 30 days or more, 50 days or more, or 55 days or more. The upper limit of the number of days for the time-series data is not particularly limited, but may be, for example, continuous data of 1000 days or less, 500 days or less, 365 days or less, 300 days or less, 200 days or less, 100 days or less, or 60 days or less.

[0023] The extraction unit 12 extracts features from the time series data (S12, extraction step). Examples of these features include the mean, coefficient of variation, temporal stability, standard deviation, autocorrelation, trend, periodicity, entropy, and trend-removed variation analysis index of the time series data.

[0024] The classification unit 13 classifies the object to be classified into one of the predetermined clusters based on the features (S13, classification step). The classification unit 13 may also perform cluster analysis on the object to be classified based on the features and classify it into a predetermined cluster. The clusters can also be called, for example, phenotypes. The number of clusters can be, for example, 2 or more, 3 or more, 4 or more, 5 or more, 10 or more, or 15 or more, and 100 or less, 50 or less, 40 or less, 30 or less, or 20 or less. The classification unit 13 may, for example, perform cluster analysis on the object to be classified based on the features and classify it into 3 clusters. The 3 clusters can be, for example, a stable / balanced type, a high-immunity / high-variability type, and a low-immunity / unstable type. Here, the stable / balanced type is, for example, a cluster that shows a moderate salivary sIgA concentration and is characterized by low variability. The high-immunity / high-variability type is a cluster that shows a high salivary sIgA concentration and is characterized by high variability. The aforementioned hypoimmune / unstable cluster is characterized by low salivary sIgA concentrations and high dispersion instability.

[0025] Here, the classification unit 13 inputs the features into a pre-trained classification model, for example, to classify the objects to be classified into predetermined clusters. The classification model is, for example, a model trained to classify input data into predetermined clusters based on training time-series data.

[0026] The predetermined cluster may, for example, be a cluster generated in advance by cluster analysis. Alternatively, the predetermined cluster may be a cluster generated by applying the k-means method based on the number of clusters determined by the elbow method. Furthermore, the predetermined cluster may be a cluster generated based on a reset number of clusters, for example, when the number of samples in each cluster in the clustering results is less than a predetermined lower limit and the number of clusters is deemed inappropriate.

[0027] The programs of this disclosure are programs that cause a computer to execute each step of this disclosure as a procedure. Specifically, the classification program of this disclosure is a program that causes a computer to execute an acquisition procedure, an extraction procedure, and a classification procedure. The classification program of this disclosure can also be described as a program that causes a computer to function as an acquisition procedure, an extraction procedure, and a classification procedure. The classification program of this disclosure can be described by reference to the descriptions of the classification apparatus and classification method of this disclosure.

[0028] According to this disclosure, time-series data of salivary sIgA to be classified is acquired in the acquisition step, features are extracted from the time-series data in the extraction step, and the classification is classified into one of the predetermined clusters based on the features in the classification step. This makes it possible to classify the subjects into predetermined clusters based on the features obtained from the time-series data of salivary sIgA. Furthermore, for example, by classifying the subjects into predetermined clusters, it is possible to visualize the state differences and trends among multiple subjects, making it easy to perform grouping, state evaluation, anomaly detection, or understanding of intervention effects based on the fluctuation patterns of salivary sIgA. This makes it possible to perform analysis or judgment according to the time-series fluctuation characteristics of salivary sIgA flexibly and efficiently.

[0029] [Embodiment 2] The program of this disclosure may be recorded on, for example, a computer-readable storage medium. The storage medium is, for example, a non-transitory computer-readable storage medium. The storage medium is not particularly limited and includes, for example, random access memory (RAM), read-only memory (ROM), hard disk (HD), flash memory (e.g., USB flash memory, SD / SDHC card, etc.), optical disc (e.g., CD-R / CD-RW, DVD-R / DVD-RW, BD-R / BD-RE, etc.), magneto-optical disk (MO), floppy disk (FD), etc. The program of this disclosure (for example, also referred to as a programming product or program product) may also be delivered, for example, from an external computer. The “delivery” may be, for example, delivered via a communication network or delivered via a wired device. The program of this disclosure may be installed and executed on the delivered device, or it may be executed without being installed. [Examples]

[0030] This section explains the results of cluster analysis based on features obtained from time-series data of salivary sIgA.

[0031] Comparison of basic participant characteristics and salivary sIgA concentration across different attributes. Fifty participants took part in this analysis, and 46 of them (23 males and 23 females) provided sufficient data for the analysis. The participants' ages ranged from 22 to 68 years (mean 45.2 ± 12.7 years).

[0032] Inter-attributable comparisons using the median salivary sIgA concentration for each participant over 56 days revealed a significant difference between sexes (males: 116.4±49.3 μg / mL, females: 97.2±100.7 μg / mL, Mann-Whitney U test, p = 0.012). However, no significant differences were found in age-based comparisons (20s: 86.4±36.8 μg / mL, 30s: 93.0±44.7 μg / mL, 40s: 105.8±57.6 μg / mL, 50s and over: 121.9±107.6 μg / mL, Kruskal-Wallis test, p = 0.896).

[0033] Identification of salivary sIgA phenotypes by time-series feature extraction Longitudinal data of salivary sIgA concentration measured continuously for 56 days from 46 participants were analyzed. To characterize individual variability patterns, 23 time-series features were extracted from each participant's data. These include statistical indicators (mean, standard deviation, coefficient of variation), temporal characteristics (autocorrelation, trend, periodicity), and complexity indicators (entropy, trend-removed variability analysis index, variance stability). Figure 4 is a heatmap showing the results of principal component analysis (PCA) as correlations between "feature × principal component (PC)". Principal component analysis (PCA) of these features showed that the first two principal components explained 81.48% of the total variance (PC1: 54.24%, PC2: 27.24%), demonstrating that individual variability patterns can be effectively captured in a low-dimensional space (Figure 4).

[0034] Multiple statistical indicators and practical constraints were considered in determining the number of clusters. Figure 5 is a graph of the elbow method used to determine the number of clusters in k-means. In Figure 5, the horizontal axis represents the number of clusters, and the vertical axis represents inertia. Statistical evaluation of clusters from 2 to 8 using the k-means method showed that for k≧4, clusters containing only a single individual appeared, making statistical evaluation impossible. The silhouette coefficient showed a maximum value (0.465) at k=2, but maintained an acceptable value (0.277) even at k=3. 3 was adopted as the maximum number of clusters that could ensure a sufficient sample size (minimum 7 individuals) for all clusters to be statistically compliant. Figure 6 is a scatter plot of the clustering results. In Figure 6, the horizontal axis represents PC1, and the vertical axis represents PC2. The three obtained phenotypes were clearly separated in the PCA space and each showed different salivary sIgA dynamics (Figures 5 and 6).

[0035] Characteristics of three salivary sIgA phenotypes Phenotype 1 (stable / balanced, n=26) was the largest group, characterized by moderate salivary sIgA levels (mean: 99.4 ± 41.6 μg / mL) and low variability (CV: 0.41 ± 0.15). This phenotype showed the highest temporal stability, with 72% of individuals maintaining a consistent pattern throughout the first and second halves of the study period.

[0036] Phenotype 2 (highly immune and highly variable, n=7) showed significantly higher salivary sIgA concentrations (mean: 272.2 ± 170.7 μg / mL), but also exhibited high variability (CV: 0.66 ± 0.21). Despite the high variability, 71.4% of individuals in this group maintained their phenotypic characteristics throughout the study period.

[0037] Phenotype 3 (hypoimmune and unstable, n=13) was characterized by low salivary sIgA levels (mean: 76.8 ± 38.6 μg / mL) and the highest dispersion instability (0.72 ± 0.18). Only 30.8% of individuals in this phenotype showed temporal stability, suggesting a more dynamic and unpredictable immune state.

[0038] Statistical verification of phenotypic differences Figure 7 visualizes the differences in cluster features. Figures 7A to 7D show the intergroup comparison (violin plot + statistical significance) of each feature (mean sIgA concentration, coefficient of variation, variance instability, entropy). Figure 7E is a heatmap of the intergroup effect size. In Figures 7A to 7D (violin plot + statistical significance), the horizontal axis of each graph represents the cluster type, and the vertical axis represents the feature (mean sIgA concentration, coefficient of variation, variance instability, entropy). Analysis of variance revealed significant differences in multiple time-series features among the three phenotypes. The most discriminative features were range (F = 69.15, p < 0.0001), maximum value (F = 66.46, p < 0.0001), standard deviation (F = 51.43, p < 0.0001), and peak power in frequency analysis (F = 43.09, p < 0.0001). Post-hoc testing (Tukey HSD) confirmed that pairwise comparisons of all phenotypes for these major features were statistically significant (p < 0.01) (Figures 7A-7E). Interestingly, the phenotypes showed clear salivary sIgA dynamics.

[0039] Figure 8 shows the relationship between sex differences and phenotype. Figure 8A is a stacked bar graph of sex ratios by cluster. Figure 8B is a box plot of sIgA concentration distribution by sex and cluster. In Figure 8A, the horizontal axis is phenotype (cluster) and the vertical axis is proportion. In Figure 8B, the horizontal axis is phenotype (cluster) and the vertical axis is salivary sIgA concentration. As shown in Figures 8A and 8B, no sex bias was observed in the distribution of phenotypes (χ²). 2 (Statistical test, p = 0.830). However, within each phenotype, sex comparisons showed that men had significantly higher sIgA levels than women in the stable phenotype (121.0 ± 35.6 vs 80.9 ± 45.1 μg / mL, p = 0.013), suggesting that the overall observed sex differences were mainly attributable to this phenotype. No significant sex differences were observed in the highly volatile phenotype (p = 0.629) and the unstable phenotype (p = 0.073) (Figures 8A and 8B).

[0040] Clinical significance of phenotypic stratification Phenotypic classification demonstrated potential clinical utility. ROC analysis showed that phenotypic grouping moderately predicted symptom onset (AUC = 0.636), outperforming predictions based solely on mean salivary sIgA levels. Each phenotype exhibited different temporal stability (72%, 71.4%, 30.8%) and abnormality detection rates (4.4%, 5.1%, 3.7%), suggesting the need for individualized monitoring strategies.

[0041] In this example, time-series analysis of salivary sIgA concentration obtained through continuous measurement over 56 days identified three distinct temporal phenotypes. These phenotypes provide a new framework for characterizing individual differences in salivary sIgA dynamics and suggest potential applications of mucosal immunity in personalized medicine.

[0042] Biological significance of the temporal phenotype of salivary sIgA The three identified phenotypes are thought to reflect different mechanisms of mucosal immune homeostasis. The stable / balanced phenotype (phenotype 1) accounted for 56.5% of the total and was characterized by moderate sIgA levels and low variability. The high temporal stability of this phenotype (72%) indicates that sIgA production and its regulatory mechanisms are relatively stable.

[0043] On the other hand, the highly immune, highly volatile type (phenotype 2) exhibits high sIgA production capacity, but its regulatory mechanism is more dynamic. Interestingly, the fact that 71.4% of individuals maintained their phenotypic characteristics despite high variability suggests that this variability pattern itself is stable as an individual's immune characteristic.

[0044] The hypoimmune / unstable phenotype (phenotype 3) shows the lowest temporal stability (30.8%) and may be more strongly influenced by environmental factors and lifestyle. Since decreased salivary sIgA levels are generally reported to be associated with an increased risk of upper respiratory tract infections, more frequent monitoring may be useful for individuals belonging to this phenotype.

[0045] Relationship between sex differences and phenotypes In this study, men generally showed significantly higher salivary sIgA levels than women, but this sex difference was mainly due to the stable / balanced type (phenotype 1). The absence of a sex difference in the highly volatile and unstable types suggests that other factors (stress, lifestyle, etc.) may have a stronger influence on these phenotypes.

[0046] The difficulty in predicting outcomes in this case highlights the limitations of clinical judgment based on single-point salivary sIgA measurements, suggesting that risk assessment and intervention strategies based on individual phenotypic classification may be more effective. Indeed, the prediction of symptom onset based on phenotypic classification (AUC=0.636) outperformed predictions using only mean sIgA values, supporting the clinical utility of this approach.

[0047] Implications for clinical application The findings of this embodiment provide important insights for the individualization of mucosal immune monitoring. Possible management strategies based on the characteristics of each phenotype include: Stable and balanced type: Regular monitoring considering high temporal stability. High Immunity / High Variability Type: Evaluation through multiple measurements considering high variability. Hypoimmune / Unstable Type: Consider more frequent monitoring and early intervention.

[0048] Relationship with symptom onset Figure 9 shows the relationship between each phenotype and various symptoms. Figure 9A is a bar graph showing the relationship between various symptoms, the number of positive cases, and the positivity rate. Figure 9B shows the average symptom score by cluster and a heat map. In Figure 9A, the horizontal axis represents symptoms, and the vertical axis represents the number of positive cases. When the relationship between each phenotype and various symptoms was investigated, the most common symptom was cough, experienced by 8 out of 46 people (Figures 9A and 9B). Figure 10 is a graph showing the number of cough-positive cases by cluster. In Figure 10, the horizontal axis represents the cluster, and the vertical axis represents the number of positive cases. Note that for cough, χ² 2 The values ​​were 6.111, p = 0.0471, and df = 2, confirming a significant association between each phenotype and the presence or absence of cough (p < 0.05) (Figure 10).

[0049] conclusion As described above, we successfully characterized individual differences in salivary sIgA dynamics as three temporal phenotypes. These phenotypes are relatively stable over time and offer a new perspective for evaluating and managing an individual's mucosal immune status. While predicting future values ​​at the individual level is difficult, a phenotypic-based stratification approach has shown potential to be clinically more useful than single-time point-in-time measurements. This analysis provides an important foundation for realizing personalized medicine for mucosal immunity.

[0050] Although the present disclosure has been described above with reference to embodiments and examples, the present disclosure is not limited to the embodiments and examples described above. The configuration and conditions of the present disclosure can be modified in various ways that can be understood by those skilled in the art within the scope of the present disclosure.

[0051] <Note> Some or all of the above embodiments may be described as follows, but are not limited to the following: <Classification method> (Note 1) This includes acquisition, extraction, and classification processes. The acquisition step involves acquiring time-series data of saliva sIgA to be classified, The extraction step involves extracting features from the time-series data, The classification step involves classifying the object to be classified into one of the predetermined clusters based on the feature quantities. A classification method in which each of the above steps is performed by a computer. (Note 2) The aforementioned features include the mean, coefficient of variation, and temporal stability of the time series data. The classification method described in Appendix 1. (Note 3) The classification step involves performing cluster analysis on the items to be classified based on the features and classifying them into three clusters. The classification method described in Appendix 1 or 2. (Note 4) The three clusters mentioned above are the stable / balanced type, the highly immune / highly variable type, and the low immune / unstable type. The classification method described in Appendix 3. (Note 5) The aforementioned time series data consists of continuous data spanning 50 days or more. The classification method described in any of the appendices 1 to 4. (Note 6) The classification step involves inputting the features into a pre-trained classification model to classify the object to be classified into a predetermined cluster. The classification model is a model trained to classify input data into predetermined clusters based on training time-series data. The classification method described in any of the appendices 1 to 5. <Classification device> (Note 7) Including an acquisition unit, an extraction unit, and a classification unit, The acquisition unit acquires time-series data of saliva sIgA to be classified, The extraction unit extracts features from the time-series data, The classification unit classifies the object to be classified into one of the predetermined clusters based on the feature quantities. Classification device. (Note 8) The aforementioned features include the mean, coefficient of variation, and temporal stability of the time series data. The classification device described in Appendix 7. (Note 9) The classification unit performs cluster analysis on the items to be classified based on the features and classifies them into three clusters. A classification device as described in Appendix 7 or 8. (Note 10) The three clusters mentioned above are the stable / balanced type, the highly immune / highly variable type, and the low immune / unstable type. The classification device described in Appendix 9. (Note 11) The aforementioned time series data consists of continuous data spanning 50 days or more. A classification device as described in any of the appendices 7 to 10. (Note 12) The classification unit inputs the features into a pre-trained classification model and classifies the object to be classified into a predetermined cluster. The classification model is a model trained to classify input data into predetermined clusters based on training time-series data. A classification device as described in any of the appendices 7 to 11. <Classification Program> (Note 13) This includes acquisition procedures, extraction procedures, and classification procedures. The acquisition procedure described above involves acquiring time-series data of salivary sIgA to be classified, The extraction procedure involves extracting features from the time-series data, The classification procedure classifies the object to be classified into one of the predetermined clusters based on the features. A classification program that causes a computer to perform each of the above steps. (Note 14) The aforementioned features include the mean, coefficient of variation, and temporal stability of the time series data. The classification program described in Appendix 13. (Note 15) The classification procedure involves performing cluster analysis on the items to be classified based on the features and classifying them into three clusters. The classification program described in Appendix 13 or 14. (Note 16) The three clusters mentioned above are the stable / balanced type, the highly immune / highly variable type, and the low immune / unstable type. The classification program described in Appendix 15. (Note 17) The aforementioned time series data consists of continuous data spanning 50 days or more. A classification program as described in any of the appendices 13 to 16. (Note 18) The classification procedure involves inputting the features into a pre-trained classification model to classify the object to be classified into a predetermined cluster. The classification model is a model trained to classify input data into predetermined clusters based on training time-series data. A classification program as described in any of the appendices 13 to 17. <Program recording medium> (Note 19) This includes acquisition procedures, extraction procedures, and classification procedures. The acquisition procedure described above involves acquiring time-series data of salivary sIgA to be classified, The extraction procedure involves extracting features from the time-series data, The classification procedure classifies the object to be classified into one of the predetermined clusters based on the features. A computer-readable recording medium containing a classification program for causing a computer to perform each of the aforementioned steps. (Note 20) The aforementioned features include the mean, coefficient of variation, and temporal stability of the time series data. Recording medium as described in Appendix 19. (Note 21) The classification procedure involves performing cluster analysis on the items to be classified based on the features and classifying them into three clusters. Recording media as described in Appendix 19 or 20. (Note 22) The three clusters mentioned above are the stable / balanced type, the highly immune / highly variable type, and the low immune / unstable type. Recording medium as described in Appendix 21. (Note 23) The aforementioned time series data consists of continuous data spanning 50 days or more. A recording medium as described in any of the appendices 19 to 22. (Note 24) The classification procedure involves inputting the features into a pre-trained classification model to classify the object to be classified into a predetermined cluster. The classification model is a model trained to classify input data into predetermined clusters based on training time-series data. A recording medium as described in any of the appendices 19 to 23. [Industrial applicability]

[0052] According to this disclosure, it is possible to provide a classification method for classifying objects into predetermined clusters based on features obtained from time-series data of salivary sIgA. The fields to which this disclosure can be applied are not limited, and it is useful in various fields using the program of this disclosure. [Explanation of Symbols]

[0053] 10 Classifier 11 Acquisition Department 12 Extraction part 13 Classification section 101 Central Processing Unit 102 memory 103 Bus 104 Storage device 105 Input device 106 Output device 107 Communication devices

Claims

1. This includes acquisition, extraction, and classification processes. The acquisition step involves acquiring time-series data of saliva sIgA to be classified, The extraction step involves extracting features from the time-series data, The classification step involves classifying the object to be classified into one of three clusters by cluster analysis based on the features, and assigning it to one of the predetermined clusters. The three clusters mentioned above are the stable / balanced type, the highly immune / highly variable type, and the low immune / unstable type. A classification method in which each of the above steps is performed by a computer.

2. Includes an acquisition step, an extraction step, and a classification step, The acquisition step involves acquiring time-series data of saliva sIgA to be classified, The aforementioned time-series data consists of continuous data spanning 50 days or more. The extraction step involves extracting features from the time-series data, The classification step involves classifying the object to be classified into one of the predetermined clusters based on the feature quantities. A classification method in which each of the above steps is performed by a computer.

3. The aforementioned features include the mean, coefficient of variation, and temporal stability of the time series data. The classification method according to claim 1 or 2.

4. The classification step involves inputting the features into a pre-trained classification model to classify the object to be classified into a predetermined cluster. The classification model is a model trained to classify input data into predetermined clusters based on training time-series data. The classification method according to claim 1 or 2.

5. Including an acquisition unit, an extraction unit, and a classification unit, The acquisition unit acquires time-series data of saliva sIgA to be classified, The extraction unit extracts features from the time-series data, The classification unit performs cluster analysis on the object to be classified based on the features and classifies it into one of three predetermined clusters. The three clusters mentioned above are the stable / balanced type, the highly immune / highly variable type, and the low immune / unstable type. Classification device.

6. Includes an acquisition unit, an extraction unit, and a classification unit, The acquisition unit acquires time-series data of saliva sIgA to be classified, The aforementioned time-series data consists of continuous data spanning 50 days or more. The extraction unit extracts features from the time-series data, The classification unit classifies the object to be classified into one of the predetermined clusters based on the feature quantities. Classification device.

7. The aforementioned features include the mean, coefficient of variation, and temporal stability of the time series data. The sorting device according to claim 5 or 6.

8. This includes acquisition procedures, extraction procedures, and classification procedures. The acquisition procedure described above involves acquiring time-series data of saliva sIgA to be classified, The extraction procedure involves extracting features from the time-series data, The classification procedure involves classifying the object to be classified into one of three clusters by performing cluster analysis based on the features, and assigning it to one of the predetermined clusters. The three clusters mentioned above are the stable / balanced type, the highly immune / highly variable type, and the low immune / unstable type. A classification program that causes a computer to perform each of the above steps.

9. Includes acquisition procedure, extraction procedure, and classification procedure, The acquisition procedure described above involves acquiring time-series data of saliva sIgA to be classified, The aforementioned time-series data consists of continuous data spanning 50 days or more. The extraction procedure involves extracting features from the time-series data, The classification procedure classifies the object to be classified into one of the predetermined clusters based on the features. A classification program that causes a computer to perform each of the above steps.

10. This includes acquisition procedures, extraction procedures, and classification procedures. The acquisition procedure described above involves acquiring time-series data of saliva sIgA to be classified, The extraction procedure involves extracting features from the time-series data, The classification procedure involves classifying the object to be classified into one of three clusters by performing cluster analysis based on the features, and assigning it to one of the predetermined clusters. The three clusters mentioned above are the stable / balanced type, the highly immune / highly variable type, and the low immune / unstable type. A computer-readable recording medium containing a classification program for causing a computer to perform each of the aforementioned steps.

Images