Vital data evaluation method, vital data evaluation system, and program

Abstract

This vital data evaluation method includes: acquiring a first vital data set corresponding to a first vital sign of a first type from a patient in a measurement period; acquiring a second vital data set corresponding to a second vital sign of a second type different from the first type; classifying the first vital data set on the basis of a first classification condition; classifying the second vital data set on the basis of a second classification condition; generating first time integrated value information at each time point in the measurement period on the basis of first vital data in an integration interval corresponding to each time point and shorter than the measurement period; generating second time integrated value information at each time point in the measurement period on the basis of second vital data in the integration interval corresponding to each time point; and generating patient state information for estimating a state of the patient on the basis of a relationship between the first time integrated value information and the second time integrated value information.

Description

Vital data evaluation method, vital data evaluation system, and program

[0001] The present invention relates to a vital data evaluation method, a vital data evaluation system, and a program.

[0002] Vital (sign) data obtained by digitizing "vital signs" indicating that a person is alive is widely used in medical fields and daily health management. For example, in medical fields, data such as pulse, blood pressure, body temperature, and respiratory rate are recorded as vital data.

[0003] Further, Patent Document 1 discloses an example of calculating a cardiac output by acquiring electrocardiogram data and pulse wave data and performing arithmetic processing.

[0004] Japanese Patent Application Laid-Open No. 2013-220175

[0005] In medical fields, vital data may be measured over time. At this time, medical staff grasp the patient's condition from the changes and trends of multiple vital data and perform medical acts. However, it is difficult for medical staff to always monitor the same subject and continuously grasp the patient's condition.

[0006] One of the objects of the present invention is to easily make a judgment close to that of a skilled medical professional regarding the patient's condition.

[0007] According to one embodiment of the present invention, a vital data evaluation method is provided, which includes a computer acquiring a first vital data set corresponding to a first type of first vital sign of a patient over time during a measurement period, acquiring a second vital data set corresponding to a second type of second vital sign different from the first type during the measurement period over time, classifying the first vital data set based on a first classification criterion, classifying the second vital data set based on a second classification criterion, generating first time integral value information at each time point based on the first vital data in an integration interval shorter than the measurement period corresponding to each time point in the measurement period from the classified first vital data set, generating second time integral value information at each time point based on the second vital data in the integration interval corresponding to each time point in the measurement period from the classified second vital data set, and generating patient state information for estimating the patient's condition based on the relationship between the first time integral value information and the second time integral value information.

[0008] In the vital data evaluation method described above, a first sub-patient status information may be generated based on the first time integral value information, a second sub-patient status information may be generated based on the second time integral value information, and the patient status information may be generated based on the relationship between the first sub-patient status information and the second sub-patient status information.

[0009] In the vital data evaluation method described above, when the first time integral value information satisfies a first cutoff condition, which is greater than or equal to a first cutoff value for defining the state of the first vital sign, the first sub-patient state information that conforms to the classification of the first classification condition may be generated. When the second time integral value information satisfies a second cutoff condition, which is greater than or equal to a second cutoff value for defining the state of the second vital sign, the second sub-patient state information that conforms to the classification of the second classification condition may be generated.

[0010] In the above vital data evaluation method, response information may be generated based on a response information dataset associated with the patient's condition information and relating to the initial response to the patient.

[0011] In the vital data evaluation method described above, when the third cutoff condition is met, where the first time integral value information is greater than or equal to the third cutoff value which exceeds the first cutoff value, the corresponding information may be generated based on a second corresponding information dataset, which is modified from the corresponding information dataset and relates to the initial and secondary responses to the patient.

[0012] In the vital data evaluation method described above, the first classification condition includes a 1-1 category corresponding to the 1-1 state of the patient's first vital sign and a 1-2 category corresponding to a 1-2 state different from the 1-1 state, and the second classification condition includes a 2-1 category corresponding to the 2-1 state of the patient's second vital sign and a 2-2 category corresponding to a 2-2 state different from the 2-1 state, and among the first vital data set classified into at least one of the 1-1 category and the 1-2 category, the first time integral value information at the first time point within the measurement period satisfies the first cutoff condition, and the predetermined measurement period after the first time point When the third cutoff condition is met, such that the first time integral value information at a second time point within a change period threshold shorter than the first cutoff value is greater than or equal to a third cutoff value exceeding the first cutoff value, at least one of the first cutoff value, the third cutoff value, the second cutoff value, and the fourth cutoff value exceeding the second cutoff value may be changed for the first time integral value information generated based on a first vital dataset classified into at least one of the 1-1 and 1-2 categories, the second cutoff value, and the fourth cutoff value exceeding the second cutoff value.

[0013] In the above vital data evaluation method, a first preprocessing step may be performed on the first vital dataset based on predetermined conditions before classifying based on the first classification condition, and a second preprocessing step may be performed on the second vital dataset based on predetermined conditions before classifying based on the second classification condition.

[0014] In the vital data evaluation method described above, the first preprocessing may include excluding first vital data from the first vital dataset that exceeds a predetermined numerical range and performing a smoothing process on the first vital dataset, and the second preprocessing may include excluding second vital data from the second vital dataset that exceeds a predetermined numerical range and performing a smoothing process on the second vital dataset.

[0015] In the above vital data evaluation method, at least one of the patient status information and the corresponding information may be displayed on a display device as visually recognizable information.

[0016] In the above vital data evaluation method, predictive patient status information may be generated based on the changes in the patient status information during the measurement period to predict the patient's status.

[0017] In the above vital data evaluation method, the first vital data set and the second vital data set may be applied to a trained model to generate the patient status information.

[0018] In the above vital data evaluation method, the trained model may correspond to the relationship between the first vital data and the second vital data at each point in time and the patient's condition information.

[0019] According to one embodiment of the present invention, a non-temporary computer-readable medium is provided which stores a program that causes a computer to execute the above-described vital data evaluation method.

[0020] According to one embodiment of the present invention, a vital data evaluation system is provided which includes a control unit that acquires a first vital data set corresponding to a first type of first vital sign of a patient over time during a measurement period, acquires a second vital data set corresponding to a second type of second vital sign different from the first type during the measurement period over time, classifies the first vital data set based on a first classification condition, classifies the second vital data set based on a second classification condition, generates first time integral value information for each time point based on the first vital data in an integration interval shorter than the measurement period corresponding to each time point in the measurement period from the classified first vital data set, generates second time integral value information for each time point based on the second vital data in the integration interval corresponding to each time point in the measurement period from the classified second vital data set, and generates patient state information that estimates the patient's state based on the relationship between the first time integral value information and the second time integral value information.

[0021] In the vital data evaluation system described above, a first sub-patient status information may be generated based on the first time integral value information, a second sub-patient status information may be generated based on the second time integral value information, and the patient status information may be generated based on the relationship between the first sub-patient status information and the second sub-patient status information.

[0022] In the vital data evaluation system described above, the control unit may generate the first sub-patient status information that conforms to the classification of the first classification condition when the first time integral value information satisfies a first cutoff condition in which the first time integral value information is greater than or equal to a first cutoff value for defining the state of the first vital sign, and generate the second sub-patient status information that conforms to the classification of the second classification condition when the second time integral value information satisfies a second cutoff condition in which the second time integral value information is greater than or equal to a second cutoff value for defining the state of the second vital sign.

[0023] In the vital data evaluation system described above, the control unit may generate response information based on a response information dataset associated with the patient's condition information and relating to the initial response to the patient.

[0024] In the vital data evaluation system described above, the control unit may, when the third cutoff condition is met, where the first time integral value information is greater than or equal to a third cutoff value that exceeds the first cutoff value, generate the corresponding information based on a second corresponding information dataset that is modified from the corresponding information dataset and relates to the initial and secondary responses to the patient.

[0025] In the vital data evaluation system described above, the first classification condition includes a 1-1 category corresponding to the 1-1 state of the patient's first vital sign and a 1-2 category corresponding to a 1-2 state different from the 1-1 state, and the second classification condition includes a 2-1 category corresponding to the 2-1 state of the patient's second vital sign and a 2-2 category corresponding to a 2-2 state different from the 2-1 state, and among the first vital data set classified into at least one of the 1-1 category and the 1-2 category, the first time integral value information at the first time point within the measurement period satisfies the first cutoff condition, and the period after the first time point is set to a predetermined value for the measurement period. When the third cutoff condition is met, which is that the first time integral value information at a second time point within a short change period threshold is greater than or equal to a third cutoff value that exceeds the first cutoff value, the control unit may change at least one of the first cutoff value, the third cutoff value, the second cutoff value, and the fourth cutoff value that exceeds the second cutoff value for the first time integral value information generated based on a first vital dataset classified into at least one of the 1-1 and 1-2 categories, the second cutoff value for the second time integral value information generated based on a second vital dataset classified into at least one of the 2-1 and 2-2 categories.

[0026] In the vital data evaluation system described above, the control unit may perform a first preprocessing on the first vital dataset based on predetermined conditions before classifying based on the first classification condition, and may perform a second preprocessing on the second vital dataset based on predetermined conditions before classifying based on the second classification condition.

[0027] In the vital data evaluation system described above, the first preprocessing may include excluding first vital data from the first vital dataset that exceeds a predetermined numerical range and performing a smoothing process on the first vital dataset, and the second preprocessing may include excluding second vital data from the second vital dataset that exceeds a predetermined numerical range and performing a smoothing process on the second vital dataset.

[0028] In the vital data evaluation system described above, the control unit may display at least one of the patient status information and the corresponding information on a display device as visible information.

[0029] In the vital data evaluation system described above, the control unit may generate predicted patient status information that predicts the patient's condition based on the changes in the patient's condition information during the measurement period.

[0030] In the vital data evaluation system described above, the control unit may apply the first vital data set and the second vital data set to a trained model to generate the patient status information.

[0031] In the vital data evaluation system described above, the trained model may correspond to the relationship between the first vital data and the second vital data at each point in time and the patient's condition information.

[0032] According to the present invention, it is possible to easily make judgments about a patient's condition that are similar to those of a skilled medical professional.

[0033] This is a diagram illustrating a vital data evaluation system in one embodiment of the present invention. This is a diagram illustrating the hardware configuration of each device constituting the vital data evaluation system in one embodiment of the present invention. This is a diagram illustrating the software configuration of the control unit of the vital data evaluation server in one embodiment of the present invention. This is a diagram of the first classification information data table in one embodiment of the present invention. This is a diagram of the second classification information data table in one embodiment of the present invention. This is a diagram of the patient status information data table in one embodiment of the present invention. This is a diagram of the correspondence information data table in one embodiment of the present invention. This is a flowchart showing the vital evaluation process of the vital data evaluation system in one embodiment of the present invention. This is a flowchart showing the vital evaluation process of the vital data evaluation system in one embodiment of the present invention. This is a flowchart showing the vital evaluation process of the vital data evaluation system in one embodiment of the present invention. This is a flowchart showing the vital evaluation process of the vital data evaluation system in one embodiment of the present invention. This is a flowchart showing the vital evaluation process of the vital data evaluation system in one embodiment of the present invention. This is an example of the acquired first vital dataset in one embodiment of the present invention. This is an example of the acquired second vital dataset in one embodiment of the present invention. This is an example of the first vital dataset from which error values ​​have been removed in one embodiment of the present invention. This is an example of the smoothed first vital dataset in one embodiment of the present invention. This is an example of a second vital dataset from which error values ​​have been removed in one embodiment of the present invention. This is an example of a smoothed second vital dataset from one embodiment of the present invention. This is an example of a classified first vital dataset from one embodiment of the present invention. This is an example of a classified second vital dataset from one embodiment of the present invention. This is an example of a dataset of time integral values ​​of the first vital data from one embodiment of the present invention. This is an example of a dataset of time integral values ​​of the first vital data from one embodiment of the present invention. This is an example of a dataset of time integral values ​​of the first vital data from one embodiment of the present invention.This is an example of a dataset of time-integrated information of the first vital data in one embodiment of the present invention. This is a flowchart showing the vital evaluation process of the vital data evaluation system in one embodiment of the present invention. This is an example of a dataset of time-integrated information of the second vital data in one embodiment of the present invention. This is an example of a dataset of time-integrated information of the second vital data in one embodiment of the present invention. This is a flowchart showing the vital evaluation process of the vital data evaluation system in one embodiment of the present invention. This is an example of a dataset of time-integrated information of the first vital data second vital data in one embodiment of the present invention. This is a diagram showing a correspondence information data table in one embodiment of the present invention. This is a flowchart showing the vital evaluation process of the vital data evaluation system in one embodiment of the present invention. This is an example of a dataset of time-integrated information of the first vital data in one embodiment of the present invention. This is an example of a dataset of time-integrated information of the second vital data in one embodiment of the present invention. This is an example of how to display arrows for patient status information in one embodiment of the present invention. This is an example of how to display arrows for patient status information in one embodiment of the present invention. This is an example of how to display arrows for patient status information in one embodiment of the present invention. This is an example of how to display arrows for patient status information in one embodiment of the present invention. This is an example of how to display arrows for patient status information in one embodiment of the present invention. This is a flowchart showing the vital evaluation process of the vital data evaluation system in one embodiment of the present invention. This is a flowchart showing the vital evaluation process of the vital data evaluation system in one embodiment of the present invention.

[0034] The embodiments described below are examples of embodiments of the present invention, and the present invention is not limited to these embodiments. In the drawings referenced in these embodiments, the same or similar reference numerals are used to denote identical parts or parts having similar functions, and repeated descriptions thereof may be omitted.

[0035] <First Embodiment> The vital data evaluation system in this embodiment will be described in detail with reference to the drawings.

[0036] (1-1. Overall Configuration of the Vital Data Evaluation System) Figure 1 is a diagram illustrating the vital data evaluation system 1 in this embodiment. This vital data evaluation system 1 includes a vital data evaluation device 10, a first vital data measuring device 20-1, a second vital data measuring device 20-2, and a terminal device (external device) 40. When it is not necessary to distinguish between the first vital data measuring device 20-1 and the second vital data measuring device 20-2, they will be described as vital data measuring device 20. In this example, one first vital data measuring device 20-1 and one second vital data measuring device 20-2 are provided, but the present invention is not limited to this. As will be described later, multiple vital data measuring devices may be used when measuring one type of vital data.

[0037] A database 50 is connected to the vital data evaluation device 10. In Figure 1, the vital data evaluation device 10, the first vital data measurement device 20-1, the second vital data measurement device 20-2, and the terminal device 40 are connected to a network NW. The network NW is a communication network such as the Internet or an intranet, and an appropriate network is used depending on the communication environment.

[0038] According to the vital data evaluation system 1, two types of vital data can be measured using two types of vital data measuring devices 20, and time integral information of each vital data for a given period of time can be generated. In this embodiment, the two types of vital data used are heart rate data and mean blood pressure data measured during surgery under general anesthesia. In this embodiment of the vital data, not only basic information such as heart rate, pulse, blood pressure, pulse pressure, body temperature, respiration, and oxygen saturation can be used, but also detailed biological information obtained from electrocardiograms, pacemakers, ventilators, arterial blood pressure monitors, wearable devices, etc., as long as it is biological information measured over time (arrhythmia, electrical resistance, heart sounds, electrocardiogram fluctuations, respiratory fluctuations of blood pressure and pulse waves, pressure measurements of various parts [venous pressure, central venous pressure, pulmonary artery pressure, pulmonary artery wedge pressure, pulmonary venous pressure, atrial pressure, ventricular pressure, arterial pressure, etc.], blood flow velocity, breath sounds, breathing pattern, breathing rhythm, apnea, ventilation, expiratory and inspiratory flow rates, airway pressure, gas analysis [Partial pressure of carbon dioxide, partial pressure of oxygen, concentration of inhaled anesthetic, etc.], cardiac output, shock index, exercise volume, exercise mode, tissue pressure, etc.), state of consciousness, cognitive function, electroencephalogram, sleep, response to external stimuli, intracranial pressure, cerebral blood flow, pupil diameter, pain, itching, nausea, vomiting, speech, laboratory values ​​(blood glucose, blood cell count, hemoglobin level, inflammation, enzyme levels, renal function, hormones, electrolytes, etc.), muscle tone, muscle relaxation, peripheral coldness, cyanosis, skin condition, weight, food intake, water intake, urine output, stool volume, blood loss, fetal heart rate, fetal growth and condition, amniotic fluid characteristics, uterine contractions, cervical dilation, etc. are also acceptable. Furthermore, information based on the body's electrical activity, physical, chemical, optical, and acoustic changes, and multimodal information combining these are also included. Furthermore, the data may not be limited to continuous values, discrete values, images, waveforms, audio, or video, but may also include features extracted from these, as well as indices generated by AI (Artificial Intelligence). Moreover, the vital data in this invention is not limited to biological information commonly used at present, but is interpreted to include all information that may be recognized in the future as indicators of physiological or pathological conditions in medical settings or research fields.Furthermore, the vital data evaluation device 10 generates patient status information that estimates the patient's condition based on the relationship between the time integral values ​​of two types of vital data, and also generates response information for treatment of the patient. The generated time integral value information, patient status information, and response information are displayed on the vital data evaluation device 10 (or terminal device 40). The configuration of the vital data evaluation system 1 for realizing such processing will be described below.

[0039] (1-2. Hardware Configuration) Figure 2 is a diagram illustrating the hardware configuration of each device that makes up the vital data evaluation system 1.

[0040] (1-2-1. Vital Data Evaluation Device 10) The vital data evaluation device 10 includes a control unit 11, a storage unit 12, a communication unit 13, a display unit 14, and an operation unit 15. The vital data evaluation device 10 may be an on-premise server, a cloud server, or other processing device. The control unit 11 is an example of a computer that includes arithmetic processing circuits such as a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), and an FPGA (Field Programmable Gate Array). The control unit 11 executes a program stored in the storage unit 12 to realize various functions in the vital data evaluation device 10. This enables the vital data evaluation system 1 to perform processing by the vital data evaluation device 10 (vital data evaluation processing).

[0041] The storage unit 12 includes a storage device and stores a control program. Further, the storage unit 12 also stores various data used when the control program is executed. Note that this program may be provided in a state recorded on a computer-readable recording medium such as a magnetic recording medium, an optical recording medium, a magneto-optical recording medium, or a semiconductor memory. In this case, the vital data evaluation device 10 only needs to include an interface for connecting the recording medium. Here, the recording medium (computer-readable recording medium) may be defined as a medium different from the storage unit 12 included in the vital data evaluation device 10, or may be the medium used for the storage unit 12.

[0042] The communication unit 13 includes a communication module and connects to the network NW under the control of the control unit 11 to transmit and receive information with other devices connected to the network NW. In this example, the communication unit 13 further connects to the database 50 to transmit and receive information. Information registered in the database 50 will be described later. Note that the communication unit 13 may connect to the database 50 via the network NW. Also, the information registered in the database 50 may be stored in the storage unit 12. In this case, the database 50 may not exist.

[0043] The display unit 14 includes a display device whose display content is controlled under the control of the control unit 11. The operation unit 15 can include a keyboard, a switch, a handle, a microphone, etc., and outputs information corresponding to the operation to the control unit 11.

[0044] Note that, in addition to the above configuration, the vital data evaluation device 10 may have other configurations such as an audio output unit and a light emitting unit.

[0045] (1-2-2. Vital Data Measuring Device 20) The vital data measuring device 20 includes a control unit 21, a storage unit 22, a communication unit 23, a display unit 24, an operation unit 25, and a measurement unit 26. The control unit 21 has basically the same configuration as the control unit 11 described above, and realizes various functions in the vital data measuring device 20. Thereby, the processing in the vital data measuring device 20 in the vital data evaluation system 1 is executed. The measured vital data may be stored in the storage unit 22 via the control unit 21.

[0046] The storage unit 22 has basically the same configuration as the storage unit 12, and the content of the program instructions stored is different to some extent, so the description thereof is omitted. The communication unit 23 has basically the same configuration as the communication unit 13, and the connectable network is different to some extent, so the description thereof is omitted.

[0047] The display unit 24 has basically the same configuration as the display unit 14, and can display the acquired vital data. The operation unit 25 has the same configuration as the operation unit 15, and the content of the operation is different to some extent, so the description thereof is omitted.

[0048] The measurement unit 26 is used to measure the vital data of the patient. The measurement unit 26 may measure the vital data invasively or non-invasively.

[0049] (1-2-4. Terminal Device 40) The terminal device 40 is composed of at least one of a desktop PC (personal computer), a notebook PC, a mobile phone, a smartphone, a tablet terminal, and other electronic application mechanical devices. The terminal device 40 can function as a client. The terminal device includes a control unit 41, a storage unit 42, a communication unit 43, a display unit 44, and an operation unit 45. The control unit 41 has basically the same configuration as the control unit 11 described above, and realizes various functions in the terminal device 40. The processing in the terminal device 40 in the vital data evaluation system 1 is executed.

[0050] The memory unit 42 has a configuration that is basically the same as the memory unit 12, and the only difference is the content of the program instructions stored in it, so its explanation will be omitted. The communication unit 43 has a configuration that is basically the same as the communication unit 13, and the only difference is the network it can connect to, so its explanation will be omitted.

[0051] The display unit 44 includes a display device whose display content is controlled by the control unit 41. The operation unit 45, in this example, includes a touch sensor and outputs information corresponding to the position operated by the user (e.g., a medical professional) to the control unit 41. This touch sensor is provided on the display area of ​​the display unit 44. In other words, the display unit 44 and the operation unit 45 constitute a touch panel. Note that touch panels may also be provided in the vital data evaluation device 10 and the vital data measurement device 20.

[0052] (1-3. Software Configuration of Vital Data Evaluation Device) Figure 3 shows the software configuration of the control unit 11 of the vital data evaluation device 10. The control unit 11 includes an acquisition unit 11a, a data preprocessing unit 11b, a classification unit 11c, an analysis unit 11d, a generation unit 11e, and an output instruction unit 11f.

[0053] The acquisition unit 11a has the function of acquiring various types of information from each device. In this example, vital data is stored in the vital database (DB) 50a of the database 50. Time integral value information is stored in the time integral value information database (DB) 50b. Patient status information is stored in the patient status information database (DB) 50c. Correspondence information is stored in the correspondence information database (DB) 50d.

[0054] The data preprocessing unit 11b has the function of performing preprocessing on vital data (vital data set) which is a time-series arrangement of vital data over a certain period. In this example, the data preprocessing unit 11b excludes data (also called error values) that meet predetermined conditions (in this example, exceeding a pre-set range) from the vital data set, and performs smoothing on the vital data set. By performing preprocessing, the effects of measurement errors can be mitigated.

[0055] The classification unit 11c has the function of classifying vital data sets based on classification conditions. The classification conditions are included in the classification condition data table described later.

[0056] The analysis unit 11d has the function of calculating the time integral value of vital data and analyzing the time integral value information. In this example, it calculates the time integral value at each point in time in a vital data set in which vital data over a certain period is arranged in time series, and analyzes the time integral value information.

[0057] The generation unit 11e has the function of generating various types of information. In this embodiment, the generation unit 11e generates patient status information based on the relationship between the time integral values ​​of two types of vital data. Furthermore, the generation unit 11e generates response information to propose a response method based on the patient status information.

[0058] The output instruction unit 11f has the function of instructing the output of patient status information and response information. In this example, the output instruction unit 11f instructs the display unit 14 to display the patient status information and response information, or to output it to the terminal device 40.

[0059] (1-4. Various Tables) Next, we will explain the various data tables used in the vital data evaluation system 1.

[0060] (1-4-1. First Classification Information Data Table) Figure 4A is the first classification information data table 100. The first classification information data table 100 is stored in the vital database (DB) 50a. In this example, the first classification information data table 100 is used to divide the heart rate state into several categories (or classifications). The first classification information data table 100 includes heart rate numerical range information 101 and classification categories (heart rate state information) 103.

[0061] (1-4-2. Second Classification Information Data Table) Figure 4B is the second classification information data table 110. The second classification information data table 110 is stored in the vital database (DB) 50a. In this example, the second classification information data table 110 is used to divide blood pressure conditions into several categories (or groups). The second classification information data table 110 includes blood pressure numerical range information 111 and classification categories (blood pressure condition information) 113.

[0062] (1-4-3. Patient Status Data Table) Figure 5 is the patient status data table 200. The patient status data table 200 is stored in the vital database (DB) 50c. The patient status data table 200 includes a first sub-patient status information 201 determined by the time integral value information of the first vital data, a second sub-patient status 203 determined by the time integral value information of the second vital data, and patient status information 205. In the patient status data table 200, patient status information is shown based on the relationship between the first sub-patient status information determined based on the time integral value information of the first vital data and the second sub-patient status information determined based on the time integral value information of the second vital data. For example, if the first sub-patient status information is "mild tachycardia" and the second sub-patient status information is "mild hypotension", then the patient status information "mild circulating plasma volume deficit" is associated.

[0063] (1-4-4. Correspondence Information Data Table) Figure 6 is the correspondence information data table 300. The correspondence information data table 300 is stored in the vital database (DB) 50d. The correspondence information data table 300 includes the first sub-patient status information 301 determined by the time integral value information of the first vital data, the second sub-patient status information 303 determined by the time integral value information of the second vital data, and correspondence information 305. The correspondence information data table 300 shows the correspondence information associated with the patient status information. In this example, the correspondence information "infusion or blood transfusion" is associated with the patient status information "mild circulating plasma volume deficit". Note that the first sub-patient status information 301 and the second sub-patient status information 303 are not necessarily included in the correspondence information data table 300.

[0064] (1-5. Vital Data Evaluation Processing) Next, the vital data evaluation processing implemented in the vital data evaluation system of this embodiment will be described. Figures 7 to 12 are flowcharts showing the vital data evaluation processing of the vital data evaluation system 1 of this embodiment.

[0065] (1-5-1. Acquisition of First and Second Vital Data) First, in Figure 7, the vital data evaluation device 10 generates vital data measurement instruction information (step S101). For example, when it receives a vital data measurement request from the terminal device 40, or when it receives any other vital data measurement request, the vital data evaluation device 10 may generate vital data measurement instruction information. The vital data evaluation device 10 transmits the vital measurement instruction information to the first vital data measuring device 20-1 and the second vital data measuring device 20-2 (steps S103, S105).

[0066] Based on vital data measurement instruction information, the first vital data measuring device 20-1 measures one type (first type) of first vital data. In this example, the first vital data measuring device 20-1 measures heart rate (bpm) (step S107). At this time, the first vital data measuring device may measure over a predetermined period, or it may continuously acquire data until it receives an instruction to stop measurement. Alternatively, it may measure based on instructions (information) directly input to the individual vital data measuring device, without relying on vital data measurement instruction information. The measured first vital data is transmitted to the vital data evaluation device 10 as a dataset arranged chronologically over a certain period (step S109), and the vital data evaluation device 10 acquires the first vital data over time (step S111). The acquired first vital data (dataset) is stored in the vital DB 50a. Figure 13A is an example of the acquired first vital dataset (more specifically, a graph visualizing the first vital dataset). As shown in Figure 13A, the acquired first vital data (dataset) may be a discrete dataset with time (minutes) on the horizontal axis and heart rate on the vertical axis. In this example, the first vital dataset can be data acquired every minute for a predetermined period (also called the "measurement period," which is 400 minutes in this example) for a patient undergoing surgery under general anesthesia (for space limitations, all data from 0 minutes to 400 minutes is shown as already acquired, but in this embodiment, it is assumed that measurements are taken continuously from 0 minutes and calculations are performed each time). The measurement period may include past measurement periods, the current time, and the period scheduled to end.

[0067] Similarly, based on the vital data measurement instruction information, the second vital data measuring device 20-2 measures second vital data of a different type (second type) from the first vital data. In this example, the second vital data measuring device 20-2 measures mean arterial pressure (mmHg) (step S113). At this time, the second vital data measuring device 20-2 may measure over a predetermined period, or it may continuously acquire data until it receives an instruction to stop measurement. Alternatively, it may measure based on instructions (information) directly input to individual vital data measuring devices, rather than based on vital data measurement instruction information. The measured second vital data is transmitted to the vital data evaluation device 10 as a dataset arranged in chronological order (step S115), and the vital data evaluation device 10 acquires the second vital data (dataset) (step S117). Figure 13B is an example of the acquired second vital dataset. As shown in Figure 13B, the acquired first vital data (dataset) is a discrete dataset with time (minutes) on the horizontal axis and mean blood pressure on the vertical axis. The acquired first vital dataset and the acquired second vital dataset correspond in time. In this example, the second vital dataset can be data acquired every minute over a predetermined period (400 minutes in this example) for a patient undergoing surgery under general anesthesia (for space limitations, all data from time 0 to 400 minutes is shown as already acquired, but in this embodiment, it is assumed that measurements are taken continuously from time 0 and calculations are performed each time).

[0068] (1-5-2. Preprocessing of Vital Data) Next, as shown in Figure 8, the vital data evaluation device 10 performs preprocessing of vital data based on predetermined conditions. First, the vital data evaluation device 10 performs preprocessing of the acquired first vital data (step S201). As shown in Figure 9, data (error values) that exceed a predetermined numerical range are excluded (cleaned) from the first vital data set (step S2011). Specifically, in the case of heart rate, the upper cutoff value can be set to 220 bpm and the lower cutoff value to 10 bpm. As a result, as shown in Figure 14A, a first vital data set from which error values ​​have been removed is generated.

[0069] Let's return to Figure 9 for explanation. After the cleaning process, the vital data evaluation device 10 then performs a smoothing process on the first vital dataset from which data exceeding a predetermined numerical range has been excluded (step S2013). As a smoothing method, one can appropriately select from methods such as functional data analysis, autoregression, and moving average. For example, when performing functional data analysis, B-splines, Fourier models, polygonal bases, polynomial bases, and constant bases can be used. As a result, a smoothed first vital dataset (graph) is generated as shown in Figure 14B. If there is insufficient computational power, the smoothing process may be omitted, and missing values ​​may be simply imputed using line graphs or step graphs.

[0070] Since the first vital data may be measured at predetermined time intervals, the first vital dataset will be discrete time-series data. When generating time-integrated value information, directly applying the acquired first vital dataset will be strongly affected by outliers and missing values. However, in this embodiment, by performing the preprocessing described above, the time-integrated value information described later can be generated more accurately. The preprocessed first vital dataset may be stored in the vital DB 50a. At this time, the first vital dataset before preprocessing may be deleted from the vital DB 50a.

[0071] Next, the vital data evaluation device 10 preprocesses the acquired second vital data. In this embodiment, similar to the first vital data, the second vital data that exceeds a preset range is excluded (cleaned) from the acquired second vital data set and smoothed (Figures 15A and 15B). The preprocessed second vital data set may be stored in the vital DB 50a. At this time, the second vital data set before preprocessing may be deleted from the vital DB 50a.

[0072] (1-5-3. Classification Processing of Vital Data) Next, as shown in Figure 10, the vital data evaluation device 10 classifies the first vital data set into corresponding groups (categories) based on a pre-set first classification condition (first classification information data table 100) (step S301). As shown in Figure 4A, in this embodiment, the first classification information data table 100 used for the first classification condition includes five classification categories according to the cutoff value. Specifically, the first classification information data table 100 includes severe bradycardia (heart rate ≤ 45 (bpm)), mild bradycardia (45 < heart rate ≤ 60), normal (60 < heart rate < 90), mild tachycardia (90 ≤ heart rate < 120), and severe tachycardia (120 ≤ heart rate)). As a result, as shown in Figure 16A, a first vital data set (graph) classified based on the first classification condition is generated.

[0073] Furthermore, in the first classification criterion, the classification categories "mild bradycardia" and "mild tachycardia," which correspond to mild heart rate conditions (also called "1-1 condition"), may be collectively defined as "1-1 category." Similarly, the classification categories "severe hypotension" and "severe hypertension," which correspond to severe heart rate conditions (also called "1-2 condition"), may be defined as "1-2 category." Also, as shown in Figure 16A, in this embodiment, due to the calculation of the time integral value as described later, if a certain point in time is classified as a severe heart rate condition, it may also be classified as a mild heart rate condition. For example, the heart rate is 140 bpm at 150 minutes. In this case, the first vital data (heart rate) at 150 minutes is classified as either "severe tachycardia" or "mild tachycardia."

[0074] Furthermore, the cutoff value (or threshold) used in the first classification condition may be changed as appropriate depending on the purpose and target individuals. Also, in this embodiment, the cutoff value is not used to determine the patient's condition or issue an alarm based on whether or not the cutoff value is exceeded, but rather to classify categories for generating time integral value information. For this reason, the range of "normal" in this embodiment may be set narrower than the range of "normal" in normal clinical practice. As will be described later, since alarms and other outputs are generated based on the time integral value based on this classification, even if the range is set narrowly, unnecessary alarms are less likely to be output, and clinical practice is less likely to be interrupted.

[0075] Similarly, as shown in Figure 10, the vital data evaluation device 10 classifies the second vital data at each time point into the corresponding category based on a pre-set second classification condition (second classification information data table 110) (step S303). As shown in Figure 4B, the second classification information data table 100 used for the second classification condition includes five categories according to the cutoff value. Specifically, the second classification condition data table 110 includes severe hypotension (mean arterial pressure ≤ 45 (bpm)), mild hypotension (45 < mean arterial pressure ≤ 70), normal (70 < mean arterial pressure < 90), mild hypertension (90 ≤ mean arterial pressure < 110), and severe hypertension (110 ≤ mean arterial pressure)). As a result, as shown in Figure 16B, a second vital data set classified based on the second classification condition is generated. At this time, in the second classification condition, the classification categories "mild hypotension" and "mild hypertension" corresponding to mild blood pressure states (also called "second-1 state") may be defined as "second-1 category". Similarly, in the second classification criteria, the classification categories "severe hypotension" and "severe hypertension," which correspond to severe blood pressure conditions (also called "second-second condition"), may be defined as "second-second category." The cutoff values ​​used in the second classification criteria may be changed as appropriate depending on the purpose and target group. Alternatively, they may be automatically selected or updated using machine learning or other methods to suit individual patients and situations.

[0076] (1-5-4. Generation of Time Integral Values ​​of Vital Data) Next, as shown in Figure 11, the vital data evaluation device 10 calculates time integral values ​​(bpm·min) using the first vital data set classified into each category (generates time integral value information) (step S401). The method for calculating the time integral value can be appropriately selected according to the purpose. In this embodiment, the time integral values ​​are calculated continuously in the interval corresponding to each time point within the predetermined period (400 minutes) described above (also called the "integration interval," specifically the interval from each time point to 15 minutes prior) (Figure 17A). Note that for the first 15 minutes from the start of measurement, the calculation may be based on the period from the start of measurement (0 minutes) to the corresponding time point. Also, in this embodiment, for simplification, the time integral value is calculated as the sum of strips (rectangles) for each minute. Furthermore, in this embodiment, in order to prevent the time integral values ​​for mild tachycardia and severe tachycardia, and mild bradycardia and severe bradycardia from being calculated twice, the time integral values ​​may be calculated separately above and below these cutoff values. In this example, as shown in Figure 16A, the heart rate is 140 bpm at 150 minutes. In this case, the vital data evaluation device 10 calculates the time integral values ​​of mild tachycardia and severe tachycardia, using the cutoff value between severe tachycardia and mild tachycardia (120 bpm) as the boundary. As a result, the time integral value of severe tachycardia for a 1-minute interval at 150 minutes is calculated as 20 bpm·min (140 - 120 (cutoff value between severe tachycardia and mild tachycardia)), and the time integral value of mild tachycardia is calculated as 30 bpm·min (120 - 90 (cutoff value between mild tachycardia and normal)). Note that for the first vital data classified as "normal" in the classification process described above, it is not necessary to generate a time integral value.

[0077] Similarly, the vital data evaluation device 10 calculates time integral values ​​using the second vital data set classified as shown in Figure 16B (generates time integral value information) (step S403). The method for calculating time integral values ​​can be the same as in the case of the first vital data, so a detailed explanation is omitted. In addition, as in this embodiment, the time integral values ​​for mild hypertension and severe hypertension, and for mild hypotension and severe hypotension may be calculated separately above and below these cutoff values ​​so that they are not calculated twice.

[0078] (1-5-5. Analysis of time-integrated vital data information, generation of patient status information and corresponding information) Next, in Figure 12, the vital data evaluation device 10 analyzes the time-integrated information of the first vital data and the time-integrated information of the second vital data (step S501).

[0079] Figure 17A is a data (graph) showing the change in the time integral value for severe tachycardia at each time point. Figure 17B is a data (graph) showing the change in the time integral value for mild tachycardia at each time point, calculated based on the above. Figure 18A is a data (graph) showing the change in the time integral value for severe bradycardia at each time point. Figure 18B is a data (graph) showing the change in the time integral value for mild bradycardia at each time point. The vital data evaluation device 10 generates first sub-patient status information at each time point based on the first vital data time integral value information (step S503).

[0080] Figure 19 is a flowchart of the process for generating the first sub-patient status information. Since the method for determining tachycardia and bradycardia is the same, it will be explained as tachycardia (or bradycardia). As shown in Figure 19, the vital data evaluation device 10 first determines whether the time integral value in the classification category "severe tachycardia (or severe bradycardia)" is greater than or equal to the cutoff value (step S5031) if the vital classification category at that time is "severe tachycardia (or severe bradycardia)". In this embodiment, as shown in Figures 17A to 18B, it is determined whether the time integral value at each time point calculated for the first vital data classified into each category is greater than or equal to a pre-set cutoff value (also called the "first cutoff value," in this example, 50 bpm·min). The first cutoff value is an index for determining whether the sub-patient status information at each time point conforms to the state corresponding to the category classified by the classification conditions (it defines whether the state has clinical significance, the state of vital signs, and is also called "state definition").

[0081] If the time integral value at that time for the classification category "severe tachycardia (or severe bradycardia)" is greater than or equal to the cutoff value (step S5031; Yes), then "severe tachycardia (or severe bradycardia)" is set as the first sub-patient status information at that time (step S5033). If the time integral value at that time for the classification category "severe tachycardia (or severe bradycardia)" is less than the cutoff value (step S5031; No), then the vital data evaluation device 10 determines whether the time integral value for the classification category "mild tachycardia (or mild bradycardia)" is greater than or equal to the cutoff value (step S5035: Yes). If the time integral value at that time for the classification category "mild tachycardia (or mild bradycardia)" is greater than or equal to the cutoff value (step S5035: Yes), then "mild tachycardia (or mild bradycardia)" is set as the first sub-patient status information at that time (step S5037). If the time integral value at that point in time for the classification category "mild tachycardia (or mild bradycardia)" is less than the cutoff value (step S5031; No), then "normal" is set as the first sub-patient status information at that point in time (step S5039). This clarifies clinically meaningful and non-clinical points in time, allowing for accurate understanding of the patient's condition in a single vital data point. Note that if the vital classification category at that point in time is classified as "normal," the first sub-patient status information may be set to "normal" without making a determination based on the time integral value. The calculation of the time integral value and the setting of sub-patient status information as described above may be performed in real time.

[0082] Next, the vital data evaluation device 10 generates second sub-patient status information at each time point based on the time integral value information of the second vital data (step S505). Figure 20A is data (graph) showing the trend of the time integral value of severe hypotension at each time point. Figure 20B is data (graph) showing the trend of the time integral value of mild hypotension at each time point. Note that in the second vital data set shown in Figure 16B, there are no time points corresponding to mild hypertension and severe hypertension, so the graphs showing the trends of the time integral values ​​of mild hypertension and severe hypertension are omitted. The method for setting the second sub-patient status information is the same as in the case of the first sub-patient status information described above. As a result, in this embodiment, one of the following is set as the second sub-patient status information at each time point: "severe hypotension," "mild hypotension," and "normal" (although not present in this example, "mild hypertension" and "severe hypertension" would also be options if they existed).

[0083] Next, the vital data evaluation device 10 generates patient status information based on the relationship between the first sub-patient status information and the second sub-patient status information (step S507). Patient status information is information that estimates the patient's condition at each point in time. In this example, the patient status information is selected from the patient status data table 200 (also called the "patient status data set") shown in Figure 5. Specifically, when the first sub-patient status information is "mild tachycardia" and the second sub-patient status information is "mild hypotension," the vital data evaluation device 10 generates patient status information "mild circulating plasma volume deficit." The vital data evaluation device 10 displays the generated patient status information on the display unit 14. This allows healthcare professionals to understand the patient's condition at each point in time. The generation and display of patient status information may be performed in real time.

[0084] Next, the vital data evaluation device 10 generates response information based on the patient's condition information (step S509). In this example, the response information is selected from the response information data table 300 (also called the "response information dataset") shown in Figure 6. The response information is associated with the patient's condition information. In this example, the device generates response information "infusion or blood transfusion" regarding the initial response to a patient associated with the patient's condition information "mild circulating plasma volume deficit". The vital data evaluation device 10 displays the generated response information as visible information on the display unit 14 (step S511). This allows healthcare professionals to understand the appropriate response to the patient at each point in time. This concludes the vital data evaluation method in this embodiment.

[0085] According to this embodiment, by generating time-integrated values ​​(vital data time-integrated value information) from two types of vital data measured over a predetermined period, it is possible to accurately grasp the changes in vital data over time. Furthermore, patient condition information and corresponding information can be easily generated based on the relationship between the time-integrated value information of the two types of vital data. In other words, by using this embodiment, it is possible to easily make judgments about the patient's condition that are similar to those of a skilled medical professional.

[0086] <Second Embodiment> In this embodiment, a method for generating patient status information and response information that differs from that of the first embodiment will be described. Specifically, an example of optimizing the response information according to the situation will be described.

[0087] Figure 21 is a flowchart of the vital data evaluation method. As shown in Figure 21, when analyzing the relationship between the time integral information of the first vital data and the time integral information of the second vital data, it may be determined whether the time integral information of the first vital data or the time integral information of the second vital data satisfies predetermined conditions (step S5061).

[0088] Figure 22A shows the time integral values ​​for severe tachycardia at each time point. Figure 22B shows the time integral values ​​for mild tachycardia at each time point. Figure 23A shows the time integral values ​​for severe bradycardia at each time point. Figure 23B shows the time integral values ​​for mild bradycardia at each time point. Figure 24A shows the time integral values ​​for severe hypotension at each time point. Figure 24B shows the time integral values ​​for mild hypotension at each time point. Note that in this example, a case in which neither mild hypertension nor severe hypertension was observed is used, so these figures may be omitted.

[0089] As shown in Figures 22A to 24B, in this example, the vital data evaluation device 10 determines whether the time integral value of the first vital data "heart rate" is greater than or equal to the corresponding change cutoff value (also called the "third cutoff value," 200 bpm / min in this example) which exceeds the normal (state-defining) cutoff value (first cutoff value, 50 bpm / min) (whether the third cutoff condition is met), and whether the time integral value of the second vital data "mean blood pressure" is greater than or equal to a cutoff value (also called the "fourth cutoff value," 150 mmHg / min in this example) which is higher than the normal (state-defining) cutoff value (second cutoff value, 50 mmHg / min) (whether the fourth cutoff condition is met) (step S5061). If the time integral value of the first vital data "heart rate" is 50 bpm·min or more and less than 200 bpm·min, and the time integral value of the second vital data "mean blood pressure" is 50 mmHg·min or more and less than 150 mmHg·min (Step S5061; No), proceed to Step S507. If the time integral value of "heart rate" is 200 bpm·min or more, or the time integral value of the second vital data "mean blood pressure" is 150 mmHg·min or more (Step S5061; Yes), modify (change) the corresponding information data set in the corresponding information data table 300 to the second corresponding information data table 300A shown in Figure 25 (Step S5063).

[0090] The second correspondence information data table 300A includes secondary correspondence information in addition to the initial correspondence information as correspondence information 305A (in Figure 25, the initial correspondence information is placed before the parentheses, and the secondary correspondence information is placed inside the parentheses). The second correspondence information data table 300A may be generated in advance. In this case, after generating patient status information (step S507), correspondence information is generated from the patient status information based on the second correspondence information dataset 300A (step S510).

[0091] By using this embodiment, in addition to the cutoff value that defines the state (state of vital signs), a cutoff value for changing the corresponding information can be set as a cutoff value (threshold) for the time integral value. This allows the severity to be changed for the same patient state information, and the corresponding information to be changed flexibly. These changes can also be performed in real time. Therefore, by using this embodiment, it becomes possible to perform sequential responses similar to those of a skilled physician.

[0092] In this embodiment, an example of adding one cutoff value is shown, but the present invention is not limited to this. For example, two or more cutoff values ​​may be added, changed, or deleted depending on patient attributes or environmental information. Furthermore, the added cutoff values ​​may be used not to change the corresponding information, but to change patient status information, etc.

[0093] Furthermore, in cases where patient status information is estimated by combining severe first sub-patient status information (or second sub-patient status information) and mild second sub-patient status information (or first sub-patient status information), such as severe tachycardia and mild hypotension, the determination may be made based on whether the severe first sub-patient status information (or second sub-patient status information) exceeds the added (third) cutoff value. For example, if severe circulating plasma volume deficiency is estimated with severe tachycardia and mild hypotension, the determination of whether to generate (display) correspondence information based on the second correspondence information data table may be made based on whether the severe tachycardia exceeds the third cutoff value.

[0094] <Third Embodiment> This embodiment describes a method for generating patient status information and response information that differs from the first embodiment. Specifically, an example of quickly determining the patient's status and its response in response to a rapid change in the time integral of vital data will be described.

[0095] Figure 26 is a flowchart for generating patient status information. As shown in Figure 26, the vital data evaluation device 10 analyzes the time integral information of the first vital data and the time integral information of the second vital data (step S501). When analyzing the relationship between the time integral information of the first vital data and the time integral information of the second vital data, the vital data evaluation device 10 determines whether the time integral value of the first vital data "heart rate" is greater than or equal to the corresponding change cutoff value (third cutoff value, 200 bpm·min), which is higher than the normal cutoff value for state definition (first cutoff value, 50 bpm·min), and whether the time integral value of the second vital data "mean blood pressure" is greater than or equal to the corresponding change cutoff value (fourth cutoff value, 150 mmHg·min), which is higher than the normal cutoff value for state definition (second cutoff value, 50 mmHg·min) (S5061). When the time integral value of the first vital data "heart rate" is 200 bpm·min or more, or the time integral value of the second vital data "mean blood pressure" is 150 mmHg·min or more (step S5061; Yes), the vital data evaluation device 10 modifies the corresponding information data table 300 and generates the second corresponding information data table 300A shown in Figure 26 (step S5063).

[0096] Furthermore, the vital data evaluation device 10 determines whether the period from the point in time when the time integral value of the first (or second) vital data becomes equal to or greater than the cutoff value for state definition to the point in time when it becomes equal to or greater than the cutoff value for corresponding change (also called the "change period") is less than or equal to a predetermined period (also called the "change period threshold") (step S5065). If the change period is greater than the change period threshold (step S5065; No), the process proceeds directly to step S507. On the other hand, if the change period is less than or equal to the change period threshold (step S5065; Yes), at least one of the cutoff value for state definition and the cutoff value for corresponding change is reduced. For example, in this embodiment, if the change period threshold is set to "5 minutes," and the vital data evaluation device 10 continuously calculates the integrated values ​​of heart rate and mean blood pressure from 0 minutes, as shown in Figure 26B, the time integrated value of mild hypotension becomes 60 mmHg·min (≧50 mmHg·min) at 274 minutes, which is above the cutoff value for state definition, and becomes 158 mmHg·min (≧150 mmHg·min) at 278 minutes, which is above the corresponding change cutoff value. In this case, since the change period is less than or equal to the change period threshold of 5 minutes, the vital data evaluation device 10 changes to "rapid change mode" from 278 minutes onward.

[0097] In this embodiment, as shown in Figure 27A, the vital data evaluation device 10 reduces the state definition cutoff value of the time integral value of the first vital data "heart rate" classified into each category in the "rapid change mode" from "50 bpm·min" to "25 bpm·min" (corrected to the modified cutoff condition), and reduces (changes) the cutoff value for corresponding information change from "200 bpm·min" to "100 bpm·min". Similarly, as shown in Figure 27B, the vital data evaluation device 10 reduces (changes) the state definition cutoff value of the time integral value of the second vital data "mean blood pressure" from "50 mmHg·min" to "25 mmHg·min", and the cutoff value for corresponding information change from "150 mmHg·min" to "75 mmHg·min".

[0098] In this embodiment, by changing the cutoff value for the period corresponding to the sudden change mode, as shown in Figure 27A, the time integral value of severe bradycardia is greater than or equal to the cutoff value for state definition from 326 minutes to 339 minutes (if the sudden change mode was not applied, the sub-patient state information during this period would have been mild bradycardia). Also, as shown in Figure 27B, in the case of the time integral value of severe hypotension, before the sudden change mode is activated, it is greater than or equal to the cutoff value for state definition at 284 minutes, but by changing to the sudden change mode, it is greater than or equal to the cutoff value at 279 minutes, so that the sub-patient state information changes from mild hypotension to severe hypotension 5 minutes earlier, allowing for a more rapid assessment of the severity of the patient's condition. Furthermore, in the case of the time integral value of severe hypotension, by changing to the sudden change mode, it is greater than or equal to the cutoff value for changing the response information from 289 minutes to 298 minutes. Therefore, at that point, the response information can include a secondary response in addition to the initial response.

[0099] Therefore, by using this embodiment, the cutoff value can be flexibly changed according to the degree of change in the time integral value of vital data, making it possible to make judgments similar to those of a skilled physician when the patient's condition changes rapidly.

[0100] Furthermore, the reduced cutoff value may be changed back to its original value when the subsequent sub-patient status information returns to "normal." More specifically, since the patient status information becomes "normal" from point 363 onwards, the emergency mode may be deactivated.

[0101] The rapid change mode determination process and rapid change mode modification process of this embodiment can be applied to time integral values ​​generated based on first vital data classified into at least one of mild heart rate states (Category 1-1) and severe heart rate states (Category 1-2), or to time integral values ​​generated based on second vital data classified into at least one of mild blood pressure states (Category 2-1) and severe blood pressure states (Category 2-2). In this embodiment, the first to fourth cutoff values ​​for the time integral values ​​of severe (mild) tachycardia, severe (mild) bradycardia, severe (mild) hypertension, and severe (mild) hypotension were changed, but the present invention is not limited thereto. For example, at least one of the first to fourth cutoff values ​​may be changed. Alternatively, the cutoff values ​​of other numerical values ​​may be changed, or only some of these cutoff values ​​may be changed. The values ​​to be changed may also be the cutoff values ​​of other normal alarms other than time integral values. The cutoff value doesn't have to be limited to just two types: normal mode and abrupt mode. There could be three or more types, or intermediate states could be defined using fuzzy logic, and the degree of urgency could be treated as a continuous value (for example, the higher the numerical value, the higher the degree of urgency, and the cutoff value decreases accordingly).

[0102] In this embodiment, an example was shown where the cutoff value was changed due to rapid changes over time, but the present invention is not limited to this. In addition to changes over time, other meaningful circumstances may be defined, and the cutoff value may be similarly changed for those periods. Furthermore, the cutoff value mode may be changed for any period based on information input by the user. For example, if fine adjustment of the cutoff value is required, the cutoff value may be lowered, and if fine control is not necessary, the cutoff value may be raised.

[0103] <Fourth Embodiment> In this embodiment, a display method for showing the changes in patient status information and response information, which differs from that of the first embodiment, will be described.

[0104] In this embodiment, the patient status information data table (and corresponding information data table) may be used as a map to map and display the progression of the patient's status in real time. In this case, the progression of the patient's status information (what path it follows) may be displayed with arrows, and the direction in which it is likely to progress in the future may be predicted (predicted patient status information may be generated). An example of how to display the arrows at each point in time is given below.

[0105] Figure 28A is a schematic diagram showing the display of arrows at 68 minutes. As shown in Figure 28A, the patient's condition information at 68 minutes is "normal". In this case, directions corresponding to the time integral values ​​of mild tachycardia, mild bradycardia, mild hypertension, and mild hypotension are provided. In this example, the time integral value of the classification category "mild hypotension" at 68 minutes is 36 mmHg·min, the time integral values ​​of the classification categories "mild tachycardia" and "mild bradycardia" are 0 bpm·min, and the time integral value of "mild hypertension" is 0 mmHg·min. In this case, the arrow is displayed extending to the right, corresponding to the direction of the time integral value of mild hypotension. Furthermore, the length of the arrow may be adjusted to correspond to a maximum value of 50 mmHg·min, which is the first cutoff value for the time integral value of mild hypotension. This makes it easy to grasp the patient's condition at 68 minutes.

[0106] Figure 28B is a schematic diagram showing the display of the arrow at 118 minutes. As shown in Figure 28B, the patient status information at 118 minutes is "Mild hypotension of unknown cause (first sub-patient status information is set to "Normal (heart rate)", and second sub-patient status information is set to "Mild hypotension")". Also, the time integral value of the classification category "Mild tachycardia" at 118 minutes is 20 bpm·min, the time integral value of the classification category "Mild bradycardia" is 0 bpm·min, and the time integral value of "Severe hypotension" is 0 mmHg·min. At this time, the arrow is displayed extending upward, corresponding to the direction of the time integral value of mild tachycardia.

[0107] Figure 29A is a schematic diagram showing the display of the arrow at 135 minutes. As shown in Figure 29A, the patient status information at 135 minutes is "mild circulating plasma volume deficit (first sub-patient status information: "mild tachycardia," second sub-patient status information: "mild hypotension")." The time integral value for classification "severe tachycardia" is 22 bpm·min, and the time integral value for "severe hypotension" is 8 mmHg·min. At this time, the arrow is displayed extending upwards and to the right, combining the time integral values ​​of "severe tachycardia" and "severe hypotension."

[0108] Figure 29B is a schematic diagram showing the display of the arrow at 143 minutes. As shown in Figure 29B, the patient status information at 143 minutes is "severe circulating plasma volume deficiency (first sub-patient status information: "severe tachycardia," second sub-patient status information: "mild hypotension")." The time integral value of the classification "severe hypotension" is 34 bpm·min. At this time, the arrow is displayed extending to the right, corresponding to the direction of the time integral value of "severe hypotension."

[0109] Figure 30 shows an example of patient status information being highlighted based on the progression of patient status information. In this embodiment, the progression of past patient status information may be shown by a dashed arrow. Current patient status information may be shown by a solid, thick arrow. The vital data evaluation device 10 can highlight (generate predicted patient status information) the patient status information (in this example, "severe circulating plasma volume deficit" and "fatal circulating plasma volume deficit") corresponding to the direction the solid arrow is pointing (in this example, the upper right direction). The method of highlighting is not particularly limited, and the corresponding portion of the patient status information may be colored, written in bold, or displayed as lit or flashing.

[0110] Thus, by using this embodiment, it is possible to generate predictive patient status information that predicts the patient's condition based on the changes in patient status information. This allows users to quickly grasp the changes in the patient's condition visually by displaying not only the current patient status information and response information, but also the past and future progress (changes in patient status). Furthermore, by modifying the current patient status information and response information while considering the changes in the patient's condition, it becomes possible to estimate patient status information and suggest response information to users in a manner similar to that of a skilled medical professional.

[0111] In this embodiment, the magnitude and speed of changes may be converted into visually recognizable information such as the thickness of the arrow. Also, in this embodiment, heart rate and mean blood pressure are used to display the arrow (marker) in two dimensions, but three or more may be combined to create a multidimensional display, or if there is no need to combine them, it may be one-dimensional, or these may be combined as needed.

[0112] Furthermore, in this embodiment, the progression of past state information is shown as a straight line passing through the center of each state information, but the present invention is not limited to this. Information such as whether the progression is normal but slightly tachycardia-oriented may be incorporated from time integral value information, etc., into the progression of past state information, and movement within the same "normal" state may be represented by a curve. In addition, the line may be made thicker if the patient stayed at the same point in time for a long time, or thinner if the state changed quickly, and time-related information may be added to the progression of past patient state information (for example, by displaying a linearly moving marker and changing its speed). Regarding the progression of future patient state information, an arrow may be emphasized to reflect a rapid increase in integral value information, etc. Note that the marker does not have to be an arrow, and different markers may be used, or audio information may be used as long as the change is recognizable. By using this embodiment, not only is the current patient state information displayed, but the path taken (past patient state information) and subsequent changes (future patient state information) can also be visualized.

[0113] Furthermore, in this embodiment, in addition to the pathways and pathway predictions described above, arrows may be added as needed to indicate directions in which the patient's condition is improving. In addition, the actual actions taken may be input, and the current patient status information and response information may be modified. Combining these, for example, if the patient's status information, which is derived from the relationship between the sub-patient status information "severe bradycardia" of heart rate and the sub-patient status information "severe hypotension" of mean blood pressure, is rapidly recovering to "normal", the response information may be changed to treat the patient status information as close to "normal" as possible. In this case, if the "rapid change mode" shown in the third embodiment is activated, the "rapid change mode" may be deactivated. Furthermore, in this case, "responding to treatment" may be displayed to indicate that the treatment is progressing smoothly, so as not to cause unnecessary anxiety to the medical staff involved. If the response information is determined solely from the current patient status information, there is a possibility that additional medications may be administered unnecessarily even though the patient is improving, potentially worsening the situation. However, by adding such a function, it becomes possible to prevent such excessive medical intervention.

[0114] <Fifth Embodiment> In this embodiment, the progression of patient status information for multiple patients based on patterns or combinations thereof of time integral values ​​of first vital data and second vital data measured in the past is used as the ground truth label. A trained model is constructed by machine learning using this ground truth label and actually measured vital signs as training data, and an example of predicting future patient status information is described. Note that explanations of parts that overlap with the first to fourth embodiments of the present invention will be omitted as appropriate.

[0115] Figure 31 is a flowchart showing the vital data evaluation process in this embodiment. First, the vital data evaluation device 10 acquires first vital data (first vital data set) and second vital data (second vital data set) arranged in time series in real time (step S601).

[0116] Next, the acquired first vital data (first vital dataset) and second vital data (second vital dataset) are applied to a trained model for predicting the patient's condition (step S603). The trained model for predicting the patient's condition corresponds to the relationship between the first vital data at each time point (more specifically, the time integral information of the first vital data at each time point), the second vital data (more specifically, the time integral information of the first vital data at each time point), and the patient's condition information.

[0117] Figure 32 is a flowchart of the generation (construction) of a trained model for predicting patient status in this embodiment. As shown in Figure 32, in order to generate (construct) a trained model for predicting patient status, the transitions of patient status information (and / or combinations of time integral value information patterns) for multiple patients and the corresponding vital data in time are acquired in advance (step S6031). At this time, the transitions of patient status information (or combinations of time integral value information patterns) can be used as the ground truth label, and the corresponding vital data in time (in particular, vital data prior to the point in time to be predicted) can be used as features. In this embodiment, the transitions of patient status information and corresponding vital data for multiple patients are acquired in advance. The transitions of patient status information consist of a dataset of patient status information generated based on the relationship between the time integral value information of the first vital data and the time integral value information of the second vital data, but the first vital data and the second vital data, or the time integral value information of the first vital data and the time integral value information of the second vital data may be used instead or included. Furthermore, new ground truth data may be defined by using other patient information and environmental information in combination with these.

[0118] Next, training data is generated based on the changes in patient status information and corresponding vital data for multiple individuals that have been acquired in advance (step S6033). At this time, information may be arbitrarily added to the training data by user input. Specifically, training data may be generated by inputting information corresponding to the actual patient's condition and the changes in vital data. Furthermore, patient information or environmental information may be used as features in the generation of training data. In addition, the time integral values ​​of the first vital data and the time integral values ​​of the second vital data may be included as the correct labels or features of the training data.

[0119] Next, machine learning is performed using the generated training data (step S6035). Known learning methods such as backpropagation and genetic algorithms (GA) may be used for machine learning, or other prediction methods such as logistic regression analysis may be applied. By repeatedly performing machine learning, a trained model for predicting patient conditions is generated (constructed) (step S6037).

[0120] Let's return to Figure 31 for explanation. Next, the vital data evaluation device 10 applies the first vital data and the second vital data as input information to a trained model for predicting the patient's condition. As a result, the vital data evaluation device 10 generates (outputs) predicted patient condition information (step S605). Finally, the vital data evaluation device 10 generates corresponding information based on the generated patient condition information (step S607). This concludes the vital data evaluation method in this embodiment. When applying this embodiment to an individual patient, the difference between the predicted patient condition information and the actual patient condition information may be used to optimize, update, or improve the trained model for the individual patient or the user of the model. In this case, retraining may be performed with real-time feedback. This allows for the construction of a model tailored to the user, enabling more precise prediction of patient condition information.

[0121] By using this embodiment, it becomes possible not only to predict, for example, that blood pressure will simply drop, but also to distinguish whether the drop in blood pressure is due to insufficient circulating plasma volume or parasympathetic nervous system dominance, and to what extent these factors are present. Furthermore, it is possible to apply training data with high-quality, ground truth labels that are robust to error values. As a result, highly accurate machine learning becomes possible, and the patient's condition can be predicted more accurately. In other words, by using this embodiment, it becomes easier to make judgments about the patient's condition that are similar to those of a skilled medical professional.

[0122] In this embodiment, as in other embodiments, two or more types of vital data may be used during model construction, when predicting patient status information for patients after model construction (when generating predicted patient status information), and when optimizing patient status information for patients, or one type of vital data may be used if it is not necessary. In each case, information other than vital data may also be used (basic information (e.g., age, gender), test results, medication information, environmental information (location information, temperature, humidity)). For example, when the input information "When an elderly person goes outdoors, the outside temperature is 30 degrees or higher" is added, it is conceivable that the patient status information will be predicted (output) as "heatstroke" by using it simultaneously with the already input vital data such as heart rate and body temperature. This makes it easier to manage the health of all kinds of people, including workers, the elderly, and astronauts. Furthermore, this information may or may not be chronological. Smoothing and other processes may also be performed on information other than vital data. In this embodiment, we have shown an example where time integral values ​​are calculated for both the ground truth data and the features used for prediction, but time integral information may be used for only one of them. For example, if the event to be predicted is "seizures," time integral information may be used as part of the features for prediction, without using integration in the definition of the ground truth data for "seizures." Conversely, time integral information may be used to create the ground truth data for the event to be predicted, but not for the features used for prediction. Furthermore, although this embodiment describes a supervised machine learning model, time integral information may also be used to create an unsupervised model.

[0123] Furthermore, in this embodiment, since predicted patient state information is generated, it is conceivable that the patient will not necessarily actually reach that state. Therefore, different response information may be used compared to the response information in other embodiments. For example, response information may be generated based on a preventative response information dataset to avoid predicting the patient state. Alternatively, the original prediction may be modified or the response to the response may be learned based on whether or not the preventative response was taken.

[0124] <Modifications> This disclosure is not limited to the embodiments described above, but includes various other modifications. For example, the embodiments described above are described in detail for the purpose of explaining this disclosure clearly, and are not necessarily limited to those having all the configurations described. Also, it is possible to replace parts of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add configurations from other embodiments to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace parts of the configuration of each embodiment with other configurations. Some modifications are described below. Note that examples of modifications of each embodiment can also be applied as examples of modifications of other embodiments.

[0125] (1) In one embodiment of the present invention, an example was shown in which five classification categories were included as the first classification condition, but the present invention is not limited thereto. The number of classification categories can be fewer or more than five, and both can contribute to solving the problems of the present invention.

[0126] (2) In one embodiment of the present invention, an example was shown in which a smoothing process is performed after a cleaning process, but the present invention is not limited thereto. The order of the cleaning process and the smoothing process may be reversed. Alternatively, either the cleaning process or the smoothing process may be performed. The smoothing process may be represented by a line graph or a step graph. Furthermore, in order to obtain a more intuitively understandable cutoff value, the integral value may be recalculated by dividing it by the length of the interval to obtain the average value or moving average (curve) over a certain interval.

[0127] (3) In one embodiment of the present invention, an example was shown in which vital data is classified into five categories by adding "mild" or "severe," but the present invention is not limited thereto. For example, in addition to "mild" or "severe," "moderate" may be added to make the number of categories more than five, or the number of categories may be less than five, such as "normal," "tachycardia," and "bradycardia," without adding "mild" or "severe."

[0128] (4) In one embodiment of the present invention, an example was shown in which the cutoff values ​​for classifying into each category are set to be clearly distinct, but the present invention is not limited thereto. Cutoff values ​​may overlap between categories. Conversely, there may be intervals between categories in which the integral value is not calculated. In a special case, for example, if the main interest is in the time when the mean blood pressure is below the cutoff value (e.g., 60 mmHg), the classification condition may be set to 59 mmHg ≤ mean blood pressure ≤ 60 mmHg. Performing interval integration within this range will yield a numerical value that is numerically equivalent to the value of the time when the value is below the cutoff value within the integration interval (although the units will be different as is) (if the corresponding vital sign is an integer). This is useful when there is no need to distinguish between a case where 59 mmHg lasts for 2 minutes and a case where 58 mmHg lasts for 2 minutes. Here, if the body temperature is recorded including the first decimal place, the same can be done by setting the classification condition to 37°C ≤ body temperature ≤ 37.1°C and then multiplying the obtained integral value by 10. If the decimal part is long, the same result can be achieved by rounding the decimal part appropriately. Alternatively, a simpler method to achieve this is to calculate the length of time that is below the cutoff value within a unit of time and use that as a substitute for the integral value.

[0129] (5) In one embodiment of the present invention, the data tables for patient status information and response methods were shown to each contain 5 x 5 = 25 pieces of information, but the present invention is not limited thereto. The data tables for patient status information and response methods may contain more or less than 25 pieces of information, and an appropriate number of patient status information and response methods can be set as needed. Furthermore, in cases where a medical professional skilled in interpreting vital signs is monitoring, it is not always necessary to display patient status information or response information, and only sub-patient status information or their respective time integral values ​​may be displayed. In this way, it is possible to display only simple information that supports the medical professional's judgment without interfering with their thinking. In addition, if a simpler representation is more appropriate even for a less skilled medical professional, patient response information and response information may be displayed in the form of simplified diagrams or videos. Furthermore, although only current patient status information was used in this example, past patient status information and current patient status information may be combined. For example, when respiratory status deteriorates, the respiratory rate increases compensatoryly, but if the deterioration of respiratory status continues thereafter, the respiratory rate may decrease due to respiratory muscle fatigue, and a decrease in respiratory rate does not necessarily mean recovery of respiratory status. In this case, by combining information on the patient's past severe tachypnea status with information on the apparent normalization of respiratory rate afterward, it is possible to detect further deterioration of the respiratory condition.

[0130] (6) In one embodiment of the present invention, an example was shown in which the patient status information data table is determined based on the relationship between the first sub-patient status information and the second sub-patient status information, but the present invention is not limited thereto. For example, intermediate states may be defined using fuzzy logic or the like, and the degree of circulating plasma volume deficiency may be expressed as a continuous value or a percentage.

[0131] (7) In one embodiment of the present invention, an example was shown in which 15 minutes of vital data is used when calculating the time integral value (generating time integral value information), but the present invention is not limited thereto. For example, when calculating the time integral value using 1 hour (60 minutes) of vital data, it is possible to generate patient status information and corresponding information that reflect longer-term trends. Furthermore, integral values ​​with different time intervals may be used together for the same vital sign or different vital signs. In addition, instead of defining intervals by time, intervals may be set according to the purpose, such as defining an interval from when the value exceeds a cutoff value until it returns to normal, and calculations may be performed by setting intervals accordingly.

[0132] (8) In one embodiment of the present invention, an example was shown in which "heart rate" is used as the first vital data and "mean blood pressure" is used as the second vital data, but the present invention is not limited thereto. The types of vital signs used as vital data may be changed, and patient status information may be generated not only by combining the time integral values ​​of two types of vital data, but also by combining the time integral values ​​of three or more types of vital data, or if there is little need to combine the time integral values ​​of the first vital data and the time integral values ​​of the second vital data, patient status information may be generated from only the time integral value of one type of vital sign data. Furthermore, patient status information may be generated not only by combining the time integral values ​​of two types of vital sign data, but also by using other information such as patient information other than vital sign data, medication information, environmental information, and actually measured vital sign data. Patient information includes at least one piece of information such as gender, age, weight, occupation, and laboratory values. Environmental information includes patient-related information such as time information, spatial information such as an operating room or hospital room, and information on drug administration. This patient information and environmental information may be information that changes over time. Furthermore, similar time-series data that is recorded continuously, such as vital signs, may be treated in the same way as vital signs. This allows for the generation of more accurate patient status information. The above information may also be used when generating correspondence information. This improves the accuracy of both patient status information and correspondence information.

[0133] (9) In one embodiment of the present invention, an example was shown in which the cutoff value (threshold) used in determining each time integral value information is set uniformly, but the present invention is not limited thereto. The cutoff value used in determining each time integral value information may differ depending on the purpose and the characteristics of the target patient.

[0134] (10) In one embodiment of the present invention, an example was shown in which the time integral value is calculated after preprocessing (cleaning, smoothing) to reduce the influence of measurement errors, but the present invention is not limited thereto. When vital signs that are less prone to measurement errors are used, or when there are limitations to the computational power, it is not always necessary to perform smoothing on the vital data, and the time integral value may be calculated using the acquired (measured) data. In this case, for missing values ​​between measurement points, for example, imputation of missing values ​​using a line graph may be selected.

[0135] (11) In one embodiment of the present invention, it was assumed that patient status information and response information would be displayed and that medical professionals would respond accordingly. However, for events that can be handled without the intervention of medical professionals, the system may directly respond without the intervention of medical professionals. In this case, the vital data evaluation system may include medical devices, air conditioning equipment, and other electronic devices, and instruction information may be transmitted to each device. For example, various processes such as adjusting the dosage of medication by medical devices or adjusting the room temperature by air conditioning equipment may be performed based on the instruction information generated by the control unit of the vital data evaluation device. AI may also be used in conjunction with such responses (i.e., the vital data evaluation device may store a trained model that performs various processes based on input data and perform inference processing that performs inference of input data based on the model).

[0136] (12) When the same vital data is measured simultaneously by multiple devices, preprocessing of the vital data may be performed to take into account measurement errors specific to vital data. For example, when measuring the second vital data "heart rate" using two second vital data measuring devices, if the "heart rate" measured by the electrocardiogram is approximately twice the "heart rate" measured by the pulse oximeter, the "heart rate" from the electrocardiogram may be considered a double count and excluded, or it may be divided by two. This makes it less susceptible to measurement errors, making it easier to generate vital data function information and to grasp the trends in vital data more accurately.

[0137] (13) In one embodiment of the present invention, when measuring the first vital data, the measurement may be performed not only with one first vital data measuring device but also with multiple first vital data measuring devices. For example, blood pressure measurement using a cuff is non-invasive and can be widely used from daily life to hospitals, but there are limitations to shortening the measurement interval. Also, frequent and repeated measurements may lead to skin damage at the measurement site. On the other hand, blood pressure measurement using an invasive arterial pressure measuring device is invasive but can continuously monitor blood pressure, so it is used intensively when there are large changes in the subject's blood pressure. For this reason, the period during which invasive arterial pressure measurement is performed is often short. Also, when blood pressure measurement using an invasive arterial pressure measuring device is performed, blood pressure measurement using a cuff is often not performed or performed less frequently. For this reason, if data measured by only one of the first vital data measuring devices is used, missing values ​​may occur during a certain period. By combining data measured by multiple primary vital data measuring devices, it is possible to supplement primary vital data when the measurement values ​​of primary vital data are interrupted (i.e., primary vital data is missing for a predetermined period) in one primary vital data measuring device, thereby generating a continuous (stable) time-series dataset over a predetermined period.

[0138] (14) In one embodiment of the present invention, the generated patient status information and corresponding information may be subjected to analysis processing based on the information input to the terminal device 40. For example, if the patient status information and corresponding information need to be corrected, the patient status information DB 50c and corresponding information DB 50d may be appropriately corrected based on the information input to the terminal device 40.

[0139] (15) In one embodiment of the present invention, the vital data evaluation device 10 may generate warning information when patient status information or sub-patient status information satisfies predetermined conditions (for example, when it is severe sub-patient status information or patient status information derived from severe sub-patient status information, such as "severe tachycardia," "severe bradycardia," "severe hypertension," or "severe hypotension"). These alarms may coexist with or be combined with alarms from a normal monitoring system.

[0140] (16) In addition, in one embodiment of the present invention, the measurement of the first vital data and the second vital data may be performed with a single measuring device.

[0141] (17) In one embodiment of the present invention, patient status information or response information may be appropriately changed based on information such as environmental information and patient information (age, sex, weight, complications, etc.). This makes it possible to provide patient status information and response information that are tailored to the patient.

[0142] (18) In one embodiment of the present invention, an example was shown in which the integral value is calculated continuously and patient status information and response information are displayed in real time. However, this information may also be used for other purposes as training data for a machine learning model. For example, this information may be used not as ground truth data, but as features to predict other events that you want to predict. Furthermore, training may be performed using vital sign data, smoothed vital sign data and its derivatives and integrals, and other data as training data.

[0143] (19) In one embodiment of the present invention, an example of mapping and displaying patient status information has been shown, but the present invention is not limited thereto. For example, a table showing patient status information may be displayed to show the history of movement. More abstract diagrams or videos may be used to display the history of vital changes as an image.

[0144] (20) In one embodiment of the present invention, the subject may be healthy individuals as well as patients. Furthermore, it may be applied to animals as well as humans. Vital data evaluation may be performed outside the operating room or outside the hospital. Furthermore, information on the subject's condition may be provided not only to medical personnel but also to the subject themselves or non-medical personnel. The recipient of the information may be an AI or the like. In addition, the patient status information may be displayed not only to the person being measured or medical personnel, but also to security systems, medical institutions, emergency centers, etc., and transmitted to remote locations via communication systems. Regarding response information, in addition to displaying response information, the device equipped with the present invention may make its own decisions and take action (e.g., adjusting medication, adjusting air conditioning).

[0145] (21) In one embodiment of the present invention, the vital data evaluation device 10 is shown to display the generated patient status information and corresponding information on the display unit 14, but the present invention is not limited thereto. For example, the vital data evaluation device 10 may output the patient status information and / or corresponding information as audio, or output it to an external device (terminal device 40).

[0146] (22) In one embodiment of the present invention, an example was shown in which patient status information and response information are displayed as visible information, but at least one of the patient status information and response information may be displayed as visible information.

[0147] 1...Vital data evaluation system, 10...Vital data evaluation device, 11...Control unit, 11a...Acquisition unit, 11b...Data preprocessing unit, 11c...Classification unit, 11d...Analysis unit, 11e...Generation unit, 11f...Output instruction unit, 12...Storage unit, 13...Communication unit, 14...Display unit, 15...Operation unit, 20...Vital data measurement device, 20-1...First vital data measurement device, 20-2...Second vital data measurement device, 21...Control unit, 22...Storage unit, 23...Communication unit, 24...Display unit, 25...Operation unit, 26...Measurement unit, 40...Terminal device (external device), 41...Control unit, 42...Storage unit, 43...Communication unit, 44...Display unit, 45...Operation unit, 50...Database, 50 a...Vitals Database (DB), 50b...Time Integral Value Information Database (DB), 50c...Patient Status Information Database (DB), 50d...Correspondence Information Database (DB), 100...First Classification Information Data Table, 101...Numerical Range Information, 103...Classification Category, 110...Second Classification Information Data Table, 111...Numerical Range Information, 113...Classification Category, 200...Patient Status Data Table, 201...First Sub-Patient Status Information, 203...Second Sub-Patient Status Information, 205...Patient Status Information, 300...Correspondence Information Data Table, 300A...Second Correspondence Information Data Table, 301...First Sub-Patient Status Information, 303...Second Sub-Patient Status Information, 305...Correspondence Information, 305A...Correspondence Information

Claims

1. A vital data evaluation method comprising: a computer acquiring a first vital data set over time corresponding to a first type of first vital sign of a patient during a measurement period; acquiring a second vital data set over time corresponding to a second type of second vital sign different from the first type during the measurement period; classifying the first vital data set based on a first classification criterion; classifying the second vital data set based on a second classification criterion; generating first time integral value information at each time point based on the first vital data in an integration interval shorter than the measurement period corresponding to each time point in the measurement period from the classified first vital data set; generating second time integral value information at each time point based on the second vital data in the integration interval corresponding to each time point in the measurement period from the classified second vital data set; and generating patient state information for estimating the patient's state based on the relationship between the first time integral value information and the second time integral value information.

2. A vital data evaluation method according to claim 1, comprising generating first sub-patient status information based on first time integral value information, generating second sub-patient status information based on second time integral value information, and generating patient status information based on the relationship between the first sub-patient status information and the second sub-patient status information.

3. A vital data evaluation method according to claim 2, wherein when the first time integral value information satisfies a first cutoff condition that is greater than or equal to a first cutoff value for defining the state of the first vital sign, the first sub-patient state information that conforms to the classification of the first classification condition is generated, and when the second time integral value information satisfies a second cutoff condition that is greater than or equal to a second cutoff value for defining the state of the second vital sign, the second sub-patient state information that conforms to the classification of the second classification condition is generated.

4. A vital data evaluation method according to claim 3, which generates response information based on a response information dataset associated with the patient's condition information and relating to the initial response to the patient.

5. The vital data evaluation method according to claim 4, wherein when the first time integral value information satisfies a third cutoff condition where it is greater than or equal to a third cutoff value that exceeds the first cutoff value, the corresponding information is generated based on a second corresponding information dataset which is modified from the corresponding information dataset and relates to the initial and secondary responses to the patient.

6. The first classification criterion includes a 1-1 category corresponding to the 1-1 state of the patient's first vital sign and a 1-2 category corresponding to a 1-2 state different from the 1-1 state; the second classification criterion includes a 2-1 category corresponding to the 2-1 state of the patient's second vital sign and a 2-2 category corresponding to a 2-2 state different from the 2-1 state; and when, among the first vital data set classified into at least one of the 1-1 category and the 1-2 category, the first time integral value information at a first time point within the measurement period satisfies the first cutoff condition, and the first time integral value information at a second time point within a preset change period threshold shorter than the measurement period after the first time point satisfies the third cutoff condition, where the first cutoff value is greater than or equal to a third cutoff value exceeding the first cutoff value, A vital data evaluation method according to claim 4, comprising changing at least one of the first cutoff value, the third cutoff value, the second cutoff value, and the fourth cutoff value exceeding the second cutoff value for the second time integral value information generated based on a first vital data set classified into at least one of the first-1 and first-2 categories.

7. A vital data evaluation method according to claim 1, wherein a first preprocessing is performed on the first vital dataset based on predetermined conditions before classifying based on the first classification condition, and a second preprocessing is performed on the second vital dataset based on predetermined conditions before classifying based on the second classification condition.

8. The vital data evaluation method according to claim 7, wherein the first preprocessing includes excluding first vital data from the first vital dataset that exceeds a predetermined numerical range and performing a smoothing process on the first vital dataset, and the second preprocessing includes excluding second vital data from the second vital dataset that exceeds a predetermined numerical range and performing a smoothing process on the second vital dataset.

9. The vital data evaluation method according to claim 4, wherein at least one of the patient status information and the corresponding information is displayed on a display device as visible information.

10. A vital data evaluation method according to claim 1, comprising generating predictive patient status information to predict the patient's condition based on the changes in the patient's condition information during the measurement period.

11. The vital data evaluation method according to claim 1, wherein the first vital data set and the second vital data set are applied to a trained model to generate the patient status information.

12. The vital data evaluation method according to claim 11, wherein the trained model corresponds to the relationship between first vital data and second vital data at each time point and patient status information.

13. A non-temporary computer-readable medium storing a program that causes a computer to execute the vital data evaluation method described in claim 1.

14. A vital data evaluation system comprising a control unit that acquires a first vital data set over time corresponding to a first type of first vital sign of a patient during a measurement period; acquires a second vital data set over time corresponding to a second type of second vital sign different from the first type during the measurement period; classifies the first vital data set based on a first classification criterion; classifies the second vital data set based on a second classification criterion; generates first time integral value information for each time point based on the first vital data in an integration interval shorter than the measurement period corresponding to each time point in the measurement period from the classified first vital data set; generates second time integral value information for each time point based on the second vital data in the integration interval corresponding to each time point in the measurement period from the classified second vital data set; and generates patient state information for estimating the patient's state based on the relationship between the first time integral value information and the second time integral value information.

15. The vital data evaluation system according to claim 14, wherein the control unit generates first sub-patient status information based on the first time integral value information, generates second sub-patient status information based on the second time integral value information, and generates patient status information based on the relationship between the first sub-patient status information and the second sub-patient status information.

16. The vital data evaluation system according to claim 15, wherein the control unit generates first sub-patient status information that conforms to the classification of the first classification condition when the first time integral value information satisfies a first cutoff condition in which the first time integral value information is greater than or equal to a first cutoff value for defining the state of the first vital sign, and generates second sub-patient status information that conforms to the classification of the second classification condition when the second time integral value information satisfies a second cutoff condition in which the second time integral value information is greater than or equal to a second cutoff value for defining the state of the second vital sign.

17. The vital data evaluation system according to claim 16, wherein the control unit generates response information based on a response information dataset associated with the patient status information and relating to the initial response to the patient.

18. The vital data evaluation system according to claim 17, wherein the control unit generates the corresponding information based on a second corresponding information dataset which is modified from the corresponding information dataset and relates to the initial and secondary responses to the patient, when the third cutoff condition is met such that the first time integral value information is greater than or equal to a third cutoff value which exceeds the first cutoff value.

19. The first classification condition includes a 1-1 category corresponding to the 1-1 state of the patient's first vital sign and a 1-2 category corresponding to a 1-2 state different from the 1-1 state; the second classification condition includes a 2-1 category corresponding to the 2-1 state of the patient's second vital sign and a 2-2 category corresponding to a 2-2 state different from the 2-1 state; and when, among the first vital data set classified into at least one of the 1-1 category and the 1-2 category, the first time integral value information at a first time point within the measurement period satisfies the first cutoff condition, and the first time integral value information at a second time point within a preset change period threshold shorter than the measurement period after the first time point satisfies the third cutoff condition, where the first cutoff value is greater than or equal to a third cutoff value exceeding the first cutoff value, the control unit shall: A vital data evaluation system according to claim 17, which modifies at least one of the first cutoff value, the third cutoff value, the second cutoff value, and the fourth cutoff value exceeding the second cutoff value, for the first time integral value information generated based on a first vital data set classified into at least one of the 1-1 and 1-2 categories, based on a second vital data set classified into at least one of the 2-1 and 2-2 categories.

20. The vital data evaluation system according to claim 14, wherein the control unit performs a first preprocessing on the first vital dataset based on predetermined conditions before classifying it based on the first classification conditions, and performs a second preprocessing on the second vital dataset based on predetermined conditions before classifying it based on the second classification conditions.

21. The vital data evaluation system according to claim 20, wherein the first preprocessing includes excluding first vital data from the first vital dataset that exceeds a predetermined numerical range and performing a smoothing process on the first vital dataset, and the second preprocessing includes excluding second vital data from the second vital dataset that exceeds a predetermined numerical range and performing a smoothing process on the second vital dataset.

22. The vital data evaluation system according to claim 18, wherein the control unit displays at least one of the patient status information and the corresponding information as visible information on a display device.

23. The vital data evaluation system according to claim 14, wherein the control unit generates predicted patient status information that predicts the patient's condition based on the changes in the patient status information during the measurement period.

24. The vital data evaluation system according to claim 14, wherein the control unit applies the first vital data set and the second vital data set to a trained model to generate the patient status information.

25. The vital data evaluation system according to claim 24, wherein the trained model corresponds to the relationship between first vital data and second vital data at each time point and patient status information.

Images