WO2016132529A1 - 情報処理装置、情報処理方法及び情報処理プログラム - Google Patents
情報処理装置、情報処理方法及び情報処理プログラム Download PDFInfo
- Publication number
- WO2016132529A1 WO2016132529A1 PCT/JP2015/054790 JP2015054790W WO2016132529A1 WO 2016132529 A1 WO2016132529 A1 WO 2016132529A1 JP 2015054790 W JP2015054790 W JP 2015054790W WO 2016132529 A1 WO2016132529 A1 WO 2016132529A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- days
- moving average
- term
- menstrual
- cycle
- Prior art date
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7278—Artificial waveform generation or derivation, e.g. synthesising signals from measured signals
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B10/00—Other methods or instruments for diagnosis, e.g. instruments for taking a cell sample, for biopsy, for vaccination diagnosis; Sex determination; Ovulation-period determination; Throat striking implements
- A61B10/0012—Ovulation-period determination
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/01—Measuring temperature of body parts ; Diagnostic temperature sensing, e.g. for malignant or inflamed tissue
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/43—Detecting, measuring or recording for evaluating the reproductive systems
- A61B5/4306—Detecting, measuring or recording for evaluating the reproductive systems for evaluating the female reproductive systems, e.g. gynaecological evaluations
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K7/00—Measuring temperature based on the use of electric or magnetic elements directly sensitive to heat ; Power supply therefor, e.g. using thermoelectric elements
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
- G16H40/63—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for local operation
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
- G16H40/67—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for remote operation
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B10/00—Other methods or instruments for diagnosis, e.g. instruments for taking a cell sample, for biopsy, for vaccination diagnosis; Sex determination; Ovulation-period determination; Throat striking implements
- A61B10/0012—Ovulation-period determination
- A61B2010/0019—Ovulation-period determination based on measurement of temperature
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K13/00—Thermometers specially adapted for specific purposes
- G01K13/20—Clinical contact thermometers for use with humans or animals
Definitions
- the present invention relates to a technical field for estimating an ovulation day or predicting a physiological day based on a measured body temperature.
- the menstrual cycle is composed of a low temperature period and a high temperature period, and the number of days in the high temperature period is considered to vary less than the number of days in the low temperature period. Therefore, techniques using such properties are conventionally known.
- Patent Document 1 when the day when the body temperature is equal to or higher than the reference body temperature continues for three days, the day two days after that day is set as the start of the high temperature phase, and the day before the start date of the high temperature phase is set as the ovulation day And predicting the next menstrual day by adding the average high-temperature phase period to the date of ovulation.
- body temperature may fluctuate greatly up and down before and after the timing of the transition from the low temperature period to the high temperature period. Therefore, if the timing when the body temperature becomes high is determined to be the start of the high temperature period, the estimation will be erroneous if it is actually still in the low temperature period.
- Patent Document 1 if it is determined that the high temperature period has started when the number of days in which the body temperature has become high body temperature has continued for a predetermined number of days, it has already shifted to the high temperature period. Therefore, it is impossible to estimate the menstrual date and predict the menstrual date.
- the present invention has been made in view of the above points, and even when the body temperature fluctuates up and down before and after the timing of actually shifting from the low temperature period to the high temperature period, the date of ovulation is suppressed. It is an object of the present invention to provide an information processing apparatus, an information processing method, and the like that enable at least one of the estimation of menstruation and the prediction of menstrual days.
- the invention according to claim 1 is characterized in that, for body temperatures measured during one or more previous menstrual cycles, the moving average of short-term days is more than the moving average of long-term days.
- the specifying means for specifying the timing at which the moving average of the short-term days is higher than the moving average of the long-term days determined by the determining means, and the target based on the timing specified by the specifying means
- An estimation means for predicting the next menstruation day of the menstrual cycle or estimating the arrival of an ovulation day during the target menstrual cycle.
- the information processing apparatus has a short period of days and a long period of time so that the timing at which the short-term moving average of the body temperature becomes lower than the long-term moving average overlaps on the physiological day when the low temperature period is considered to start.
- the information processing device uses the determined number of days to calculate the current menstrual cycle temperature based on the short-term moving average and the long-term moving average, and the timing at which the short-term moving average becomes higher than the long-term moving average. Is identified. Since the time series of body temperature is smoothed by the moving average, even if the body temperature fluctuates up and down before and after the timing of actually shifting from the low temperature period to the high temperature period by specifying this timing, Timing can be specified appropriately.
- the estimating means is configured such that the moving average of the short-term days is greater than the moving average of the long-term days in the one or more previous menstrual cycles.
- a representative value of the number of days from the timing when the moving average of the short-term days becomes lower than the moving average of the long-term days is added to the timing specified by the specifying means, and the next physiological day It is characterized by predicting.
- the information processing apparatus has a short-term moving average lower than the long-term moving average from the timing at which the short-term moving average is higher than the long-term moving average corresponding to the number of days in the high temperature period of the subject. Specify the number of days until a certain timing. Therefore, at least one of the estimation of the ovulation date and the prediction of the physiological date can be performed more appropriately according to the tendency of the number of days in the high temperature period of the subject.
- the determining means is configured such that a short-term moving average of body temperature measured in the one or more previous menstrual cycles is a long-term body temperature. So that the time difference between the timing lower than the moving average of the period and the menstrual day becomes small, and the integral value of the difference between the short-term moving average and the long-term moving average in the past one or more menstrual cycles becomes large. The short-term days and the long-term days are determined.
- the timing at which the short-term moving average becomes higher than the long-term moving average can be appropriately identified as the integral of the difference between the short-term moving average and the long-term moving average of body temperature is larger. According to the present invention, based on this integration and the time difference between the period when the short-term moving average of the body temperature is lower than the long-term moving average and the menstrual day, more appropriate short-term days and long-term days are determined. be able to.
- the timing at which the moving average of the short-term days is higher than the moving average of the long-term days is determined by the specifying unit. If not specified, it further comprises a predicting means for predicting the next menstrual day based on the number of days in the past one or more menstrual cycles.
- the menstrual date can be predicted even when the temperature has not yet shifted from the low temperature period to the high temperature period.
- the invention according to claim 5 is an information processing method executed by a computer, wherein a moving average of short-term days is a long-term number of days for body temperature measured during one or more previous menstrual cycles.
- the determination step of determining the short-term days and the long-term days so that the timing lower than the moving average overlaps one or more menstrual days in the past, and the determination of the body temperature measured in the target menstrual cycle Based on the specific step of specifying the timing when the moving average of the short-term days determined by the step is higher than the moving average of the long-term days determined by the determining step, and the timing specified by the specific step,
- the timing when the moving average of the short-term days is lower than the moving average of the long-term days for the body temperature for a plurality of days measured during one or more menstrual cycles in the past Determination means for determining the short-term days and the long-term days so as to overlap with one or more menstrual days in the past, movement of the short-term days determined by the determination means for the body temperature measured in the target menstrual cycle Specifying means for specifying a timing at which the average is higher than the moving average of the long-term days determined by the determining means; and based on the timing specified by the specifying means, the next of the target menstrual cycle It is made to function as an estimation means for predicting a menstrual day or estimating the arrival of an ovulation day during the target menstrual cycle.
- the information processing apparatus has a short period of days and a long period of time so that the timing at which the short-term moving average of the body temperature is lower than the long-term moving average overlaps on the physiological day when the low temperature period is considered to start.
- the information processing device uses the determined number of days to calculate the current menstrual cycle temperature based on the short-term moving average and the long-term moving average, and the timing at which the short-term moving average becomes higher than the long-term moving average. Is identified.
- Timing can be specified appropriately. Moreover, it becomes a menstrual day at the timing of shifting from the high temperature period to the low temperature period. Therefore, it is possible to appropriately determine the short-term days and the long-term days. Therefore, at least one of the estimation of the ovulation date and the prediction of the menstrual date can be performed while suppressing the decrease in accuracy.
- FIG. 1 It is a figure showing an example of outline composition of information processing system S concerning one embodiment. It is a block diagram which shows an example of schematic structure of the information processing server 1 which concerns on one Embodiment. It is a figure which shows an example of the functional block of the system control part 14 which concerns on one Embodiment. It is a figure which shows an example of the content registered into the database constructed
- FIG. It is a figure which shows an example of the candidate of a parameter. It is a figure which shows the example of three candidate combinations among the produced
- FIG. 1 is a diagram illustrating an example of a schematic configuration of an information processing system S according to the present embodiment.
- the information processing system S includes an information processing server 1, a plurality of user terminals 2, and a plurality of thermometers 3.
- the information processing server 1 and each user terminal 2 can exchange data with each other via the network NW, for example, using TCP / IP as a communication protocol.
- the network NW is constructed by, for example, the Internet, a dedicated communication line (for example, a CATV (CommunityCommunAntenna Television) line), a mobile communication network (including a base station, etc.), a gateway, and the like.
- the information processing server 1 is a server device that distributes information about female health to the user terminal 2. Further, the information processing server 1 acquires information such as the user's basal body temperature and menstrual date from the user terminal 2. And the information processing server 1 estimates a user's ovulation day based on the acquired information, or predicts the next menstrual day.
- the user terminal 2 is a terminal device of a user who uses the information processing system S.
- the user terminal 2 may be, for example, a smartphone, a tablet computer, a PDA (Personal Digital Assistant), a mobile phone, a personal computer, or the like.
- the user terminal 2 transmits the basal body temperature measured by the thermometer 3 to the information processing server 1. Further, the user terminal 2 transmits the actual menstrual date input by the user to the information processing server 1. Then, the user terminal 2 displays information such as the ovulation date and the menstrual date estimated by the information processing server 1.
- the thermometer 3 is an electronic thermometer that measures the user's basal body temperature.
- the thermometer 3 transmits the measured body temperature to the user terminal 2 by, for example, short-range wireless communication. Note that the user may manually input the body temperature measured by the thermometer into the user terminal 2.
- the user measures the body temperature using the thermometer 3, for example, every day (once a day).
- the actually measured body temperature is referred to as an actual measurement value.
- the menstrual cycle is fixed.
- the menstrual cycle is, for example, a period from the previous menstruation day to the day before the next menstruation day.
- the information processing server 1 determines the first day of the consecutive menstrual days as the menstrual date used for determining the menstrual cycle.
- the information processing server 1 predicts the next menstrual day using the period-based prediction method and the body temperature-based prediction method, respectively.
- the cycle-based prediction method is a method using the number of days in a plurality of past menstrual cycles.
- the body temperature-based prediction method is a method using measured values of body temperature.
- the information processing server 1 may predict the next menstrual day and may estimate the ovulation day in the current menstrual cycle.
- the current menstrual cycle is a menstrual cycle that starts from the last menstrual day of the past menstrual days and ends on the day before the next menstrual day. That is, the current menstrual cycle is the menstrual cycle currently in progress.
- FIG. 2A is a block diagram illustrating an example of a schematic configuration of the information processing server 1 according to the present embodiment.
- the information processing server 1 includes a communication unit 11, a storage unit 12, an input / output interface 13, and a system control unit 14.
- the system control unit 14 and the input / output interface 13 are connected via a system bus 15.
- the communication unit 11 is connected to the network NW and controls the communication state with the user terminal 2 and the like.
- the storage unit 12 is composed of, for example, a hard disk drive.
- the storage unit 12 is an example of a storage unit.
- databases such as a member information DB 12a, a body temperature DB 12b, a menstrual day DB 12c, and a parameter DB 12d are constructed.
- DB is an abbreviation for database.
- FIG. 3 is a diagram illustrating an example of contents registered in the database constructed in the storage unit 12b of the information processing server 1.
- user information related to users who use the information processing system S is registered for each user.
- user attributes such as user ID, password, nickname, name, date of birth, sex, postal code, address, telephone number, and e-mail address are registered in association with each other as user information. Is done.
- the user ID is user identification information.
- Information related to actual measurement values is registered in the body temperature DB 12b. Specifically, a user ID, a measurement date, and an actual measurement value are registered in association with each other in the body temperature DB 12b.
- the user ID indicates the user who measured the body temperature.
- a measurement day shows the day when body temperature was measured.
- the system control unit 14 receives from the user terminal 2 a user ID that uses the user terminal 2, a measurement date, and an actual measurement value. Then, the system control unit 14 registers the received information in the body temperature DB 12b.
- Information on the menstrual day is registered in the menstrual day DB 12c.
- a user ID and a menstrual date are registered in association with the menstrual date DB 12c.
- the user ID indicates the user who entered the menstrual date.
- the system control unit 14 receives from the user terminal 2 a user ID that uses the user terminal 2 and a menstruation date input by the user. Then, the system control unit 14 registers the received information in the menstrual day DB 12c.
- parameter information including parameters used for menstrual day prediction is registered.
- the system control unit 14 registers parameter information in the parameter DB 12d for each user basically every time a new menstrual cycle starts.
- parameter information a user ID, a registration date, a parameter of the cycle-based prediction method, a predicted physiological date of the cycle-based prediction method, a body temperature-based parameter, and an estimated high-temperature period days are registered in association with each other.
- the user ID indicates a user whose menstrual date is predicted using a parameter.
- the registration date indicates the date when the parameter information is registered in the parameter DB 12d.
- the parameters of the period-based prediction method include, for example, an abnormal value removal threshold value, a winsolization threshold value, and a representative value type.
- the abnormal value removal is to remove data that exceeds a threshold value or is located outside the threshold value as an outlier from a plurality of menstrual cycles. This outlier is called an abnormal value.
- the abnormal value is an example of a first outlier in the present invention.
- the abnormal value removal threshold indicates the threshold of the menstrual cycle that is removed as an abnormal value. Specifically, for example, a percentage may be stored as the abnormal value removal threshold.
- the abnormal value removal threshold is an example of the first threshold in the present invention.
- the winsolization is to change the value of data that exceeds or exceeds the threshold as an outlier in the distribution of statistical data to the threshold.
- the threshold of the winsolization of the present embodiment indicates the threshold of the menstrual cycle in which the number of days is changed from among a plurality of menstrual cycles, and indicates the number of days after the change. Specifically, for example, a percentage may be stored in the threshold of winsolization.
- the threshold of windization is an example of the second threshold in the present invention.
- the representative value type indicates a method for acquiring representative values of a plurality of menstrual cycles. Specifically, the representative value type indicates whether an average value is acquired or a median value is acquired.
- the predicted physiological date of the cycle-based prediction method is a physiological date predicted by the cycle-based prediction method.
- the parameters of the body temperature-based prediction method include, for example, short-term days and long-term days.
- Short-term days are days used when calculating a short-term moving average of body temperature.
- Long-term days are days used when calculating a long-term moving average of body temperature.
- Long days are longer than short days.
- the estimated number of days in the high temperature period is the number of days in the high temperature period estimated by the body temperature-based prediction method.
- the menstrual cycle is divided into a low temperature period and a high temperature period.
- the low temperature period is a period in which the body temperature is relatively low within the menstrual cycle
- the high temperature period is a period in which the body temperature is relatively high within the menstrual cycle.
- the low temperature period begins with the start of the menstrual cycle, and then transitions to the high temperature period.
- the storage unit 12 stores various data such as an HTML document, an XML (Extensible Markup Language) document, image data, text data, and an electronic document for displaying a web page.
- the storage unit 12 stores various set values, threshold values, constants, and the like.
- the storage unit 12 stores various programs such as an operating system, a WWW (World Wide Web) server program, a DBMS (Database Management System), and a prediction processing program.
- the prediction processing program is a program for estimating an ovulation date and predicting a menstrual date.
- the prediction processing program is an example of an information processing program in the present invention.
- the various programs may be acquired from, for example, another server device or the like via the network NW, or may be recorded on a recording medium such as a magnetic tape, an optical disk, or a memory card and read via the drive device. You may be made to do.
- the prediction processing program may be a program product.
- the storage unit 12 stores a terminal application program.
- the terminal application program is a program executed by the user terminal 2.
- the terminal application program is a program for transmitting information such as measured values and menstrual dates to the information processing server 1 and displaying information such as estimated ovulation dates and predicted menstrual dates.
- the user terminal 2 downloads a terminal application program from the information processing server 1.
- the input / output interface 13 performs interface processing between the communication unit 11 and the storage unit 12 and the system control unit 14.
- the system control unit 14 includes a CPU 14a, a ROM (Read Only Memory) 14b, a RAM (Random Access Memory) 14c, and the like.
- the CPU 14 is an example of a processor.
- the present invention can also be applied to various processors different from the CPU.
- Each of the storage unit 12, the ROM 14b, and the RAM 14c is an example of a memory.
- the present invention can also be applied to various memories different from the hard disk, ROM, and RAM.
- the information processing server 1 may be configured by a plurality of server devices.
- a server device that acquires information from the user terminal 2 a server device that estimates ovulation dates and menstrual days, a server device that provides information to the user terminal 2, a server device that manages a database, and the like are mutually connected. You may connect by LAN etc.
- an example of the information processing apparatus according to the present invention may be a server device that performs ovulation day estimation and menstrual day prediction.
- the information processing apparatus according to the present invention may be implemented in one server apparatus, or may be implemented by a coordinated process of a plurality of server apparatuses.
- FIG. 2B is a diagram illustrating an example of functional blocks of the system control unit 14 according to the present embodiment.
- the system control unit 14 determines whether to use either the cycle-based prediction method or the body temperature-based prediction method based on a predetermined condition. Then, the system control unit 14 predicts the next menstrual day using the determined method. Therefore, as shown in FIG.
- the system control unit 14 reads and executes a program such as a prediction processing program by the CPU 14a, so that a parameter candidate acquisition unit 141, a cycle set acquisition unit 142, a modified reference cycle acquisition unit 143, It functions as a representative value acquisition unit 144, a parameter determination unit 145, a period base prediction unit 146, a period number of days determination unit 147, a timing identification unit 148, a body temperature base prediction unit 149, and the like.
- the period base prediction unit 146 is an example of a prediction unit in the present invention.
- the period days determination unit 147 is an example of a determination unit in the present invention.
- the timing specifying unit 148 is an example of specifying means in the present invention.
- the body temperature base prediction unit 149 is an example of an estimation unit in the present invention.
- the system control unit 14 acquires a plurality of past menstrual cycles. Next, the system control unit 14 performs a predetermined process on the outlier specified by the threshold among a plurality of past menstrual cycles, and acquires a plurality of menstrual cycles listed.
- the predetermined process may be, for example, removing an outlier specified by the threshold value, or may perform winsolization on the outlier specified by the threshold value.
- the system control unit 14 may perform both of the abnormal value removal and the winsolization as the predetermined process. In this case, the system control unit 14 may perform the abnormal value removal first and then perform the winsolization.
- the variation in the number of days in the reference cycle may be large. If the variation remains large, the accuracy of the prediction may decrease. However, even if the outliers are identified, the outliers can be identified and the outliers can be simply removed. Prediction accuracy may be reduced. Therefore, the decrease in prediction accuracy can be suppressed by changing the outlier so as to leave information indicating that the outlier is located outside the distribution of the menstrual cycle.
- the system control unit 14 calculates representative values of the corrected menstrual cycles. Then, the system control unit 14 predicts the next menstrual day based on the representative value of the menstrual cycle.
- the system control unit 14 determines, for each user, a parameter used in the cycle-based prediction method in order to predict a physiological date using the cycle-based prediction method.
- the parameter to be determined is at least one of determining an abnormal value removal threshold, a winsolization threshold, and a representative value type.
- the abnormal value removal threshold value, the winsolization threshold value, and the representative value type may be determined in advance for the entire information processing system S, for example.
- a threshold for winsolization is not necessary.
- an abnormal value removal threshold is not necessary when only the winsolization is used to correct a plurality of menstrual cycles.
- the system control unit 14 predicts a menstrual cycle by a cycle-based prediction method using a plurality of past menstrual cycles and parameter candidates. Then, the system control unit 14 compares the predicted menstrual cycle with another past menstrual cycle, and determines one of the parameter candidates as a parameter used for actual prediction. By using the determined parameters by the system control unit 14, even if there are variations in the number of days in the past menstrual cycles, the next menstrual day can be accurately predicted.
- the parameter candidate acquisition unit 141 acquires a plurality of candidates for outlier thresholds. Specifically, the parameter candidate acquisition unit 141 acquires at least one of a plurality of candidates for abnormal value removal thresholds and a plurality of candidates for threshold values for winsolization. For example, a table that stores a plurality of candidates for abnormal value removal thresholds may be stored in the storage unit 12. In addition, for example, tables each storing a plurality of candidates for the threshold of winsolization may be stored. The parameter candidate acquisition unit 141 may acquire a plurality of threshold candidates from the table stored in the storage unit 12. The number of candidates for each threshold is 2 or more. The number of outlier removal threshold candidates and the number of winsolization thresholds may or may not be the same.
- the parameter candidate acquisition unit 141 may acquire “average value” and “median value” as representative value type candidates.
- the prediction accuracy may be improved by using an average value.
- the prediction accuracy may be improved by using the median.
- FIG. 4A is a diagram illustrating an example of parameter candidates. In the example of FIG. 4A, the number of candidates for abnormal value removal threshold is three, and the number of candidates for threshold for windization is three. Further, there are “average value” and “median value” as representative value type candidates.
- the parameter candidate acquisition unit 141 When two or more parameters are determined among the abnormal value removal threshold value, the winsolization threshold value, and the representative value type, the parameter candidate acquisition unit 141 generates a plurality of combinations of two or more parameter candidates. This combination is called a candidate combination. For example, when three parameters are determined, the parameter candidate acquisition unit 141 selects one of a plurality of candidates for the abnormal value removal threshold, one of the plurality of candidates for the winsolation threshold, and the representative value. A plurality of candidate combinations composed of any of a plurality of types of candidates are generated.
- FIG. 4B is a diagram illustrating an example of three candidate combinations among a plurality of generated candidate combinations. In the example of FIG. 4A, 18 candidate combinations can be generated.
- the cycle set acquisition unit 142 acquires a cycle set composed of one reference cycle and a reference cycle from a plurality of past menstrual cycles of menstrual day prediction subjects.
- the plurality of reference cycles are menstrual cycles used for provisional prediction of the menstrual cycle.
- the reference cycle is a menstrual cycle that is compared with the tentatively predicted menstrual cycle.
- the reference period may be any of a plurality of reference periods, and the reference period may be a menstrual period different from the plurality of reference periods.
- the plurality of reference cycles may be continuous or not continuous.
- the reference period and the plurality of reference periods may or may not be continuous.
- the timing position of the reference period may be before or after a plurality of reference periods. Further, the reference period may be sandwiched between a plurality of reference periods.
- the cycle set acquisition unit 142 may acquire only one cycle set or may acquire a plurality of cycle sets.
- the number of period sets may be set in advance by the administrator of the information processing system S, or may be determined by the period set acquisition unit 142.
- the cycle set acquisition unit 142 is configured such that, for example, a plurality of cycle sets in which at least one of the timing position of the reference cycle, some or all of the plurality of reference cycles, and the number of reference cycles are different from each other. May be obtained.
- FIG. 4C is a diagram illustrating an example of acquiring a periodic set. In FIG. 4C, the numbers displayed above the past menstrual cycles C1 to C9 indicate the time position of the menstrual cycle.
- a number indicating a temporal position is attached as n of the symbol Cn.
- the first menstrual cycle C1 is the latest menstrual cycle among the past menstrual cycles
- the second menstrual cycle C2 is the second new menstrual cycle.
- the number of reference periods in each period set is six. Further, the six reference periods are continuous. The reference period is continuous with six reference periods and is newer than these reference periods.
- three period sets are acquired.
- the first cycle set includes a reference cycle C1 and reference cycles C2 to C7.
- the second period set includes a reference period C2 and reference periods C3 to C8.
- the third period set includes a reference period C3 and reference periods C4 to C9.
- the modified reference cycle acquisition unit 143 corrects the plurality of reference cycles acquired by the cycle set acquisition unit 142 for each threshold candidate acquired by the parameter candidate acquisition unit 141 using the threshold candidates. Then, the cycle set acquisition unit 142 acquires a plurality of corrected reference cycles. When the candidate combination including at least the abnormal value removal threshold value and the winsolization threshold value is acquired by the parameter candidate acquisition unit 141, the corrected reference cycle acquisition unit 143 corrects a plurality of reference cycles for each candidate combination. , Obtain a plurality of corrected reference periods.
- the parameter candidate acquisition unit 141 acquires at least an abnormal value removal threshold candidate OF as a threshold candidate.
- the cycle set acquisition unit 142 calculates the appearance probability of the reference cycle for each number of days.
- the period set acquisition part 142 removes the reference period of the number of days whose appearance probability is less than OF% from a plurality of reference periods as an abnormal value.
- the period set acquisition part 142 acquires the some reference period from which the abnormal value was removed as a corrected reference period.
- the parameter candidate acquisition unit 141 acquires at least a candidate WN for a winsolation threshold as a threshold candidate.
- the cycle set acquisition unit 142 changes the number of reference cycles having a number of days less than the WN percentile to a WN percentile from among a plurality of reference cycles arranged in ascending order of days.
- the period set acquisition unit 142 changes the number of days of the reference period longer than the (100-WN) percentile from the plurality of reference periods to the (100-WN) percentile.
- a reference cycle at a position that completely matches the position indicated by the threshold value WN may not exist among the plurality of reference cycles before correction.
- the corrected reference period acquisition unit 143 may calculate the WN percentile by interpolation based on the number of days of the reference period at the positions immediately before and after the position indicated by the threshold value WN and its position.
- the modified reference cycle acquisition unit 143 may change, for example, the number of days in the reference cycle having a number of days less than the WN percentile to the number of days in the latest reference cycle among the reference cycles inside the position indicated by the threshold value WN. . “Inside” refers to the range from the WN percentile to the (100 ⁇ WN) percentile. Further, when there is no reference cycle of a position that completely matches the position indicated by the threshold (100-WN), the same as the WN percentile may be used.
- the correction reference cycle acquisition unit 143 For example, when the parameter candidate acquisition unit 141 acquires a candidate combination that includes both the candidate OF for the abnormal value removal threshold and the candidate WN for the winsolization threshold as threshold candidates, the correction reference cycle acquisition unit 143 For example, the threshold value may be removed first, and then the winsolization may be performed. 5A to 6C show an example of a process in which threshold removal and winsolization are performed. For example, it is assumed that the first cycle set shown in FIG. 4C is acquired by the cycle set acquisition unit 142. The number of days in the reference period C1 is 30. The days of the reference cycles C2 to C7 are 35, 24, 34, 29, 29, and 29, respectively. FIG.
- FIG. 5A is a graph showing an example of the frequency distribution of the reference period acquired by the modified reference period acquisition unit 143 and an example of the positional relationship between the frequency distribution and the threshold value OF.
- FIG. 5A shows an example of the frequency distribution of the reference cycle for the sake of explanation, and does not necessarily match the actual frequency distribution of the reference cycles C2 to C7.
- FIG. 5B is a diagram illustrating an example of a plurality of reference periods from which abnormal values are removed. For example, it is assumed that the appearance probability of the reference period of 35 days is less than OF percent, and the appearance probability of each of the reference periods of the 24th, 29th, and 34th reference cycles is more than OF percent.
- the corrected reference cycle acquisition unit 143 removes the reference cycle C2 having 35 days as an abnormal value from the reference cycles C2 to C7. Thereby, the corrected reference period acquisition unit 143 acquires the corrected reference periods C3 to C7 as shown in FIG. 5B.
- FIG. 5C is a graph showing an example of the frequency distribution of the reference period after the abnormal value is removed.
- FIG. 5C shows that reference periods with appearance probabilities less than OF percent have been removed.
- FIG. 6A is a graph showing an example of the frequency distribution of the reference period from which abnormal values have been removed and an example of the positional relationship between the frequency distribution and the threshold value WN.
- the corrected reference period acquisition unit 143 determines the OF percentile in the abnormal value removal as the 0th percentile in the winsolization.
- the modified reference period acquisition unit 143 determines the (100-OF) percentile in the abnormal value removal as the 100th percentile in the winsolization.
- the modified reference period acquisition unit 143 determines the WN percentile and the (100-WN) percentile within the determined range.
- FIG. 6B is a diagram illustrating an example of a plurality of reference periods that have been subjected to winsolization.
- FIG. 6C is a graph showing an example of the frequency distribution of the reference period after winsolization.
- the reference period for days less than the WN percentile is changed to the WN percentile
- the reference period for days longer than the (100-WN) percentile is changed to the (100-WN) percentile.
- the representative value acquisition unit 144 calculates, for each threshold candidate acquired by the parameter candidate acquisition unit 141, representative values of a plurality of corrected reference cycles acquired by the corrected reference cycle acquisition unit 143.
- the corrected reference cycle acquisition unit 143 includes a plurality of corrected reference cycles for each candidate combination.
- the representative value of is calculated.
- the representative value acquisition unit 144 acquires only the predetermined one of the corrected average value or median of the reference periods.
- the representative value acquisition unit 144 acquires both the average value and the median value of the plurality of corrected reference periods.
- the representative value acquisition unit 144 calculates an average value when the representative value type included in the acquired candidate combination is “average value”, and when the representative value type is “median”, Determine the value.
- FIG. 7A is a graph showing an example of the frequency distribution of the reference period after the winsolization and an example of the positional relationship between the frequency distribution and the representative value.
- FIG. 7B is a diagram illustrating an example of how the representative value is determined and how the difference between the representative value and the number of days in the reference period is calculated.
- the representative value acquisition unit 144 calculates an average value 28.8 as shown in FIG. 7B for the reference period shown in FIG. 6C.
- the representative value acquisition unit 144 determines the median value 29.
- the acquired representative value corresponds to a provisional predicted value of the menstrual cycle.
- the parameter determination unit 145 compares the representative value acquired by the representative value acquisition unit 144 with the number of days in the reference cycle for each threshold candidate acquired by the parameter candidate acquisition unit 141 or for each candidate combination. Then, the parameter determination unit 145 determines a parameter or combination used for prediction based on the comparison result. For example, the parameter determination unit 145 may calculate the difference between the representative value and the reference period. For example, as shown in FIG. 7B, the parameter determination unit 145 calculates a difference 1.2 between the number of days 30 of the reference period C1 and the average value 28.8. The parameter determination unit 145 may determine a parameter or combination having the smallest calculated difference as a parameter or combination used for prediction. The difference between the representative value and the number of days in the reference cycle is the difference between the provisional predicted value of the menstrual cycle and the actual menstrual cycle. Therefore, it is considered that the smaller the difference, the higher the menstrual cycle prediction accuracy.
- the corrected reference cycle acquisition unit 143 corrects each of the plurality of reference periods of the plurality of period sets for each threshold candidate or for each candidate combination, and acquires each of the corrected reference periods.
- the representative value acquisition unit 144 acquires representative values of a plurality of reference periods corrected in each of a plurality of period sets for each threshold candidate or for each candidate combination.
- the parameter determination unit 145 compares the representative value acquired for each of the plurality of cycle sets with the number of days in each reference cycle of the plurality of cycle sets for each threshold candidate or for each candidate combination.
- the parameter determination unit 145 may calculate the sum of differences calculated for each of a plurality of periodic sets, for example, for each parameter candidate or for each candidate combination. Then, the parameter determination unit 145 may determine a parameter candidate or candidate combination having the smallest difference as a parameter or combination used for prediction. In the following, when the term “candidate” is simply used, it indicates a parameter candidate or a candidate combination. The parameter determination unit 145 may calculate, for example, an average value or a median value instead of the total.
- the parameter determination unit 145 determines the representative value and the reference cycle based on the temporal position of the reference cycle included in the cycle set.
- the difference may or may not be weighted.
- the parameter determination unit 145 may decrease the weight as the timing position of the reference period is older.
- the parameter determination unit 145 determines a parameter for predicting the number of days in the current menstrual cycle. The older the reference period, the farther from the current menstrual cycle. Therefore, the older the timing position of the reference cycle, the lower the similarity between the reference cycle and the current menstrual cycle may be. Therefore, by using a parameter determined by reducing the weight as the time position of the reference cycle is older, the prediction accuracy of the number of days in the menstrual cycle can be increased.
- FIG. 8A is a diagram showing a calculation example of the sum of the difference between the representative value and the reference period.
- the parameter determination unit 145 calculates the weighted sum of the difference between the representative value and the reference period for each candidate combination.
- the parameter determination unit 145 multiplies the difference between the representative value of the first period set and the reference period by the weighting factor 1, and sets the weighting factor 0.8 to the difference between the representative value of the second period set and the reference period.
- the difference between the representative value of the third period set and the reference period may be multiplied by a weighting factor of 0.5.
- the weighting coefficient of each period set may be freely set.
- a difference similar to the difference of the candidate having the smallest difference between the representative value and the reference period may be calculated for one or more other candidates. That is, there may be a plurality of candidates with the same high accuracy of provisional prediction.
- a candidate having the smallest difference between the representative value and the reference period is referred to as a first candidate.
- a candidate for which a difference within a predetermined range is acquired from the difference acquired for the first candidate is referred to as a second candidate.
- the parameter determination unit 145 may determine the first candidate as a parameter or combination used for prediction.
- the parameter determination unit 145 selects the first candidate and the second candidate based on a predetermined condition. You may determine the parameter or combination used for prediction.
- the parameter determination unit 145 may specify only a predetermined number or less of these candidates as second candidates, or may specify all candidates as second candidates.
- the parameter determination unit 145 includes, for example, the number of users for whom the first candidate is determined as a parameter or combination used for prediction among users different from the target person of the current menstrual day prediction, and a second parameter or combination used for prediction.
- the number of users whose candidates are determined may be acquired.
- the parameter determination unit 145 can acquire the number of such users based on the parameter information registered in the parameter DB 12d.
- the parameter determination unit 145 may determine, for example, a parameter or combination to be used for prediction of a candidate having the largest number of acquired people among the first candidate and one or a plurality of second candidates. The reason is that the parameters used for many users are likely to be parameters that can be predicted suitable for various users.
- FIG. 8B is a graph showing an example of the number of users for which the first candidate is determined and the number of users for which the second candidate is determined. It is assumed that candidate combination CC2 is specified as the first candidate and candidate combination CC8 is specified as the second candidate. Here, as shown in FIG. 8B, the number of users whose candidate combination CC8 is determined is larger than the number of users whose candidate combination CC2 is determined. Therefore, the parameter determination unit 145 may determine the candidate combination CC8 as a combination of parameters used for prediction.
- the parameter determination unit 145 may count the number of people using, for example, parameter information with the latest registration date. Or the parameter determination part 145 may each calculate the frequency determined as a parameter or combination which a 1st candidate and a 2nd candidate use for prediction based on several parameter information, for example. Then, the parameter determination unit 145 may determine that, for example, a candidate having the highest frequency among the first candidate and the second candidate has determined a parameter or combination used for prediction.
- the parameter determination unit 145 may acquire the number of users having characteristics that are the same as the characteristics of the target person among the users whose first candidates are determined. In addition, the parameter determination unit 145 may acquire the number of users who have characteristics that are the same as the characteristics of the target person among the users whose second candidates are determined. And the parameter determination part 145 may determine the parameter or combination used for prediction based on the acquired number of persons. The reason is that the tendency of the menstruation or menstrual cycle of the user having the same characteristics as the characteristics of the subject has a probability of being similar to the tendency of the menstruation or menstrual cycle of the subject.
- the parameter determination unit 145 can acquire age or age based on, for example, the member information DB 12a. Moreover, the parameter determination part 145 can acquire the season or month in which the parameter or combination used for prediction was determined based on the parameter information registered into parameter DB12d, for example. Moreover, the parameter determination part 145 can acquire the transition tendency of the body temperature, the days in the low temperature period, and the days in the high temperature period based on the body temperature DB 12b and the menstrual day DB 12c.
- the feature having the same feature as the subject's feature may be, for example, the same feature as the subject's feature, a feature within a predetermined range from the subject's feature, It may be a feature similar to the feature.
- the cycle-based prediction unit 146 predicts the number of days in the current menstrual cycle of the subject using the parameter or combination determined by the parameter determination unit 145. Specifically, the cycle-based prediction unit 146 acquires a plurality of past menstrual cycles of the target person for menstrual day prediction. Next, the cycle-based prediction unit 146 acquires a plurality of corrected menstrual cycles by performing a predetermined process on an outlier specified by a threshold determined as a parameter among the plurality of acquired menstrual cycles. More specifically, the cycle-based prediction unit 146 removes abnormal values from a plurality of menstrual cycles or performs a winsolization on the plurality of menstrual cycles.
- the period base prediction unit 146 further performs winsolization after removing the abnormal value.
- the period-based prediction unit 146 removes the abnormal value using the threshold.
- the period-based prediction unit 146 performs windization using the threshold value.
- the period base prediction unit 146 acquires the corrected representative values of a plurality of menstrual cycles as predicted values of the number of days in the current menstrual cycle.
- the cycle-based prediction unit 146 calculates the average value.
- the cycle-based prediction unit 146 Determines the median.
- the cycle-based prediction unit 146 calculates the predicted physiological date by adding the acquired representative value to the first day of the current menstrual cycle.
- the cycle base prediction unit 146 may determine only one representative value for the entire frequency distribution, for example.
- the cycle-based prediction unit 146 may divide the menstrual cycle distribution, for example, at a cycle position at which the frequency is a minimum value.
- the period base prediction unit 146 may acquire a representative value for each divided distribution, for example.
- the cycle-based prediction unit 146 may calculate the predicted physiological date by adding the representative value to the first day of the current menstrual cycle for each representative value. That is, the cycle-based prediction unit 146 may predict two next menstrual days.
- the cycle base prediction unit 146 may cause the user terminal 2 to display, for example, “the next menstrual scheduled date when the cycle is short is X days and the next menstrual scheduled date when the cycle is long is Y days”.
- X is the predicted menstrual date closest to today among the two predicted menstrual days
- Y is the predicted menstrual date far from today.
- the cycle-based prediction unit 146 may predict only one next menstrual day without dividing the distribution of menstrual cycles, for example.
- the parameter may be stored in the storage unit 12.
- the threshold candidate acquisition unit 141, the cycle set acquisition unit 142, the modified reference cycle acquisition unit 143, the representative value acquisition unit 144, and the parameter determination unit 145 are unnecessary.
- the system control unit 14 calculates a short-term moving average and a long-term moving average of measured values of the body temperature of the subject whose physiological days are predicted, for example, every day. And the system control part 14 specifies the timing when a short-term moving average becomes larger than a long-term moving average in the present menstrual cycle.
- the timing at which the short-term moving average becomes larger than the long-term moving average is, for example, the short-term moving average immediately after the timing at which the moving average line intersects among the timing at which the short-term moving average line and the long-term moving average line intersect Is higher than the long-term moving average.
- P crossover positive crossover
- FIG. 9 is a diagram showing an example of a graph of actual measured values of body temperature in the current menstrual cycle, a short-term moving average line, and a long-term moving average line of body temperature.
- the system control unit 14 predicts the next menstrual day or estimates the arrival of the ovulation day based on the specified timing. Alternatively, the system control unit 14 may perform both prediction of the next menstrual day and estimation of arrival of the ovulation day.
- the menstrual cycle generally consists of a low temperature period and a high temperature period.
- the egg is discharged from the follicle.
- the corpus luteum is formed from the follicle.
- Progesterone is secreted from this corpus luteum.
- This luteinizing hormone raises the body temperature and shifts from the low temperature phase to the high temperature phase.
- the corpus luteum is atrophied, the secretion of luteinizing hormone ends. Then, the endometrium cannot be maintained, and menstruation occurs.
- body temperature decreases. Therefore, it shifts to the low temperature period of the next menstrual cycle.
- the life span of luteinizing hormone is fairly stable. For example, lutein hormone life is typically 14 days ⁇ 2 days.
- the day of transition from the low temperature period to the high temperature period or a day close to that day is the day of ovulation.
- the actual daily body temperature may not change stably.
- the body temperature may greatly fluctuate up and down before and after the timing that should shift from the low temperature period to the high temperature period. Therefore, it is difficult to determine whether or not the temperature has shifted to the high temperature period even if only the actual measured values of daily body temperature are observed.
- the body temperature-based prediction method of the present embodiment since a moving average is used, the time series of body temperature is smoothed. Moreover, a short-term moving average shows the tendency of the body temperature in the latest short period, and a long-term moving average shows the tendency of the body temperature in the long term. Therefore, it is possible to appropriately specify the timing of transition from the low temperature period to the high temperature period based on the timing at which the short-term moving average becomes larger than the long-term moving average. Therefore, the arrival of the ovulation day can be accurately estimated. In the example of FIG. 9, there is a high probability that an ovulation day has arrived on January 12 or a day around it.
- the number of days in the low temperature period is relatively easy to change depending on the menstrual cycle.
- the days in the high temperature period are relatively stable. The reason is that the number of days in the high temperature period corresponds to the life of the corpus luteum. Therefore, if the day when the ovulation day arrives can be estimated, the next menstrual cycle can be predicted.
- the calculated moving average may be any kind of moving average.
- a simple moving average, a weighted moving average, or an exponential moving average may be calculated.
- Any known formula can be used as a formula for calculating the exponential moving average.
- the exponential moving average EMA p (t) of the day t days after the day on which the calculation of the exponential moving average is started may be calculated by, for example, the following Equation 1.
- e (t) is an actual measurement value of the body temperature of the day after t days.
- d is an attenuation coefficient.
- d may be calculated by the following equation 2.
- the period days determination unit 147 determines parameters used in the body temperature-based prediction method in order to estimate the arrival of the ovulation day or the physiological date using the body temperature-based prediction method.
- the parameters to be determined are the short and long days used for calculating the moving average.
- the short-term days and the long-term days indicate the number of actually measured body temperature values that are used as a basis for calculating the average value.
- the short-term moving average is an average value of actually measured values of body temperature in the latest s days.
- the long-term moving average is an average value of the actual measured values of body temperature in the most recent l days.
- the short and long days are one of the parameters for calculating the average. Each time the exponential moving average is calculated, the weight of the actual measured body temperature used for the calculation decreases exponentially.
- the short-term days and the long-term days may be, for example, the number of days required until the weight of the actual measurement value of the body temperature falls below a predetermined value.
- the short-term days and the long-term days are the days required until the weight of the actual measurement value of the body temperature becomes half or less.
- the period days determination unit 147 may determine the parameters used in the body temperature-based prediction method using the measured values of the body temperature of one or more past menstrual cycles of the subject.
- the past one or more menstrual cycles may be a menstrual cycle that is continuous with the current menstrual cycle, for example.
- these menstrual cycles may be a continuous menstrual cycle, for example.
- the period days determination unit 147 may calculate a short-term moving average and a long-term moving average in one or more past menstrual cycles using, for example, the respective short-term days and long-term days used as parameters.
- FIG. 10 is a diagram illustrating an example of a graph of body temperature in the past three menstrual cycles, a short-term moving average line, and a long-term moving average line of body temperature.
- the period days determination unit 147 makes the timing at which the short-term moving average of the body temperature in the past one or more menstrual cycles becomes smaller than the long-term moving average overlap with the menstrual day that is the first day of the past one or more menstrual cycles.
- the short-term days and long-term days used as parameters are determined.
- the timing at which the short-term moving average becomes smaller than the long-term moving average is, for example, the short-term moving average immediately after the timing at which the moving average line intersects among the timing at which the short-term moving average line and the long-term moving average line intersect Is lower than the long-term moving average.
- Such crossing of the moving average lines is referred to as N crossover (negative crossover).
- FIG. 11 is a diagram illustrating an example of a relationship between a short-term moving average line and a long-term moving average line of body temperature in the past three menstrual cycles. In the example of FIG. 11, the timing of N crossover roughly overlaps each of the three menstrual days.
- the period days determining unit 147 determines the short-term days and the long-term days so that the timing of the N crossover overlaps with the actual menstrual days, so that the estimation accuracy of the arrival of the ovulation day using the moving average or the prediction of the menstrual days Accuracy can be increased.
- the period days determination unit 147 may determine the short-term days and the long-term days so that the function err indicating the difference between the menstrual day and the N crossover timing is minimized, for example.
- the parameters of the function err are the short-term days s and the long-term days l.
- the transposed matrix of the matrix (s, l) is x
- the parameter of err is the matrix x.
- err may be represented by the following Expression 3, for example.
- menday (i) is the first day of the i-th menstrual cycle out of the past n menstrual cycles, that is, a menstrual day.
- ncodey (x, i) is the day closest to menday (i) among the days when the N crossover specified based on the short-term moving average and the long-term moving average calculated by x occurs.
- the period days determination unit 147 may solve the convex optimization problem for obtaining x that minimizes err (x), for example. This problem may be shown, for example, by Equation 4 below.
- the period days determining unit 147 may use any algorithm.
- the period days determination unit 147 may use a simplex method or the like.
- the period days determination unit 147 calculates err for all combinations of short-term days and long-term days, for example, and calculates the short-term days and long-term days for which the smallest err is calculated. May be determined as short-term days and long-term days used for estimating the arrival of the ovulation day or predicting the next period.
- the period days determination unit 147 determines the short-term days and the long-term days so that the difference between the period when the short-term moving average of the body temperature in the past one or more menstrual cycles becomes smaller than the long-term moving average and the menstrual day becomes small.
- the short-term days and the long-term days may be determined so that the integrated value of the difference between the short-term moving average and the long-term moving average in one or more previous menstrual cycles becomes large. That is, for example, in FIG. 10, the period days determining unit 147 may determine the short-term days and the long-term days so that the area of the region surrounded by the short-term moving average line and the long-term moving average line becomes as large as possible. .
- the reason is to clearly identify the N crossover. This is because when the difference between the short-term moving average and the long-term moving average is small as a whole, it may be difficult to specify the N crossover or a plurality of N crossovers may occur within one menstrual cycle.
- the definite integral area (x) of the difference between the short-term moving average and the long-term moving average is expressed by the following Equation 5.
- TE is the number of days that have passed from the first day of the first menstrual cycle to the last day of the last menstrual cycle of one or more previous menstrual cycles.
- EMA s is a short-term moving average in which the number of days in the period is s
- EMA l is a long-term moving average in which the number of days in the period is l.
- f (x) may be expressed by Equation 6 below.
- Expression 6 is an example of an expression for the function f (x).
- the function f (x) may be calculated using an expression different from Expression 6.
- the period days determination unit 147 may determine the short-term days and the long-term days so that, for example, f is maximized. For example, the period days determination unit 147 may solve the convex optimization problem for obtaining x that maximizes f. This problem may be shown, for example, by Equation 7 below.
- the period days determining unit 147 may use any algorithm.
- the period days determination unit 147 may use a simplex method or the like.
- the period days determination unit 147 calculates f for all combinations of short-term days and long-term days, for example, and calculates the short-term days and long-term days for which the largest f is calculated. May be determined as short-term days and long-term days used for estimating the arrival of the ovulation day or predicting the next period.
- the period days determination unit 147 may determine the short-term days and the long-term days so that g (x) is minimized.
- the function changes to an optimization problem for obtaining a short-term day and a long-term day in which ⁇ f (x) is minimized. Therefore, the optimization problem of the function f (x) and the function g (x ) Optimization problem is substantially the same.
- the timing specifying unit 148 calculates a short-term moving average and a long-term moving average in the current menstrual cycle using the short-term days and long-term days determined by the period days determining unit 147.
- the timing specifying unit 148 may start calculating the exponential moving average from any day within one or more past menstrual cycles, for example.
- the timing specifying unit 148 specifies the timing at which the short-term moving average becomes larger than the long-term moving average in the current menstrual cycle, that is, the day when the P crossover occurs.
- the body temperature base prediction unit 149 estimates the arrival of the ovulation day or predicts the next menstrual day based on the timing specified by the timing specifying unit 148.
- the body temperature base prediction unit 149 may estimate the day when the P crossover has occurred, for example, as the ovulation day. Alternatively, the body temperature base prediction unit 149 may estimate a day after a predetermined date or a predetermined date before the day when the P crossover has occurred as an ovulation day, for example. Alternatively, when the ovulation date in the past menstrual cycle of the subject is known by the ovulation day test, the body temperature-based prediction unit 149 may, for example, calculate the ovulation date and the date on which the P crossover occurred in the past menstrual cycle. The difference may be calculated. Then, the body temperature base prediction unit 149 may calculate the estimated ovulation date by adding the calculated difference to the day when the P crossover occurs in the current menstrual cycle.
- the body temperature-based predicting unit 149 may calculate the predicted menstrual day by adding a predetermined number of days, for example, on the day when the P crossover occurs, or may calculate the predetermined number of days on the estimated ovulation day.
- the predicted menstrual date may be calculated by adding.
- the number of days to be added may be predetermined within a range of 14 days ⁇ 2 days, for example.
- the body temperature-based prediction unit 149 may perform N crossovers from the day when a P crossover in one or more past menstrual cycles occurs based on, for example, a short-term moving average and a long-term moving average of one or more previous menstrual cycles. A representative value of the number of days until the day of occurrence may be calculated.
- the body temperature base prediction unit 149 may calculate, for example, an average value or a median value. Then, the body temperature base prediction unit 149 may calculate the predicted physiological date by adding the calculated representative value to the day when the P crossover occurs in the current menstrual cycle, for example.
- FIG. 12A is a diagram illustrating an example of how a predicted physiological date is calculated. For example, in the example of FIG. 11, the days from the day when the P crossover occurred to the day when the N crossover occurred in the menstrual cycle are 11, 13 and 15, respectively. Therefore, the average value is 13. As shown in FIG. 12A, the day when the P crossover occurred in the current menstrual cycle is January 14. Therefore, the body temperature base prediction unit 149 predicts the next menstruation date as January 27.
- the menstrual cycle subject to estimation of the ovulation date may be the current menstrual cycle or a past menstrual cycle.
- the timing specifying unit 148 calculates a short-term moving average and a long-term moving average in a past menstrual cycle using the short-term days and long-term days determined by the period days determining unit 147, and in the menstrual cycle, The timing when the short-term moving average becomes larger than the long-term moving average may be specified.
- body temperature base prediction part 149 may estimate the ovulation day in the menstrual cycle based on the specified timing.
- the short-term days and long-term days used in the body temperature-based prediction method may be determined in advance.
- the short-term days and the long-term days may be stored in the storage unit 12 in advance. In this case, the period days determining unit 147 is not necessary.
- the cycle-based prediction method if the number of days in the past three or more menstrual cycles is specified, the next menstrual day is predicted regardless of whether it is in the low temperature period or the high temperature period. Can do.
- the next menstrual day is accurately predicted based on the timing when the short-term moving average of the body temperature becomes larger than the long-term moving average, that is, the timing when the P crossover occurs. Can do.
- the body temperature-based prediction method cannot be used.
- the cycle-based predicting unit 146 may predict the next menstrual day. Good.
- the timing specifying unit 148 specifies the timing when the short-term moving average becomes larger than the long-term moving average
- the body temperature base prediction unit 149 may predict the next menstrual date.
- FIG. 12B is a diagram showing a determination example of a method used for prediction of the next menstrual day.
- the short-term moving average is lower than the long-term moving average. Therefore, during this period, the period-based prediction unit 146 predicts the next menstrual day using the period-based prediction method.
- the body temperature base prediction unit 149 predicts the next menstrual day using the body temperature base prediction method.
- the cycle-based prediction unit 146 may always predict the next menstrual day. In this case, the body temperature base prediction unit 149 is unnecessary. Also, for example, if the timing at which the P crossover occurred in the current menstrual cycle has not yet been specified, the system control unit 14 uses the method different from the cycle-based prediction method described in Section 3-1 to A menstrual day may be predicted. For example, the system control unit 14 may use any method as long as the method predicts the next menstrual day by adding the representative value of the number of days in the past one or more cycles to the menstrual day of the first menstrual cycle. Good. In this case, for example, the number of days in at least one of the past one or more cycles may be corrected or changed.
- the body temperature base prediction unit 149 may predict the next menstrual day.
- the system control unit 14 may not predict the next menstrual date.
- the body temperature base prediction unit 149 may predict the next menstrual day. In this case, the period base prediction unit 146 is not necessary.
- FIG. 13 is a flowchart showing an example of a menstrual day prediction process of the system control unit 14 of the information processing server 1 according to the present embodiment.
- the user terminal 2 activates a terminal application program. Then, the user measures today's body temperature using the thermometer 3. Then, the user terminal 2 receives the measured body temperature from the thermometer 3 as an actual measurement value. Then, the user terminal 2 transmits the measured body temperature value to the information processing server 1 together with the user ID of the user who uses the user terminal 2.
- the system control unit 14 executes the menstrual day prediction process.
- the user indicated by the user ID is a target person whose next menstruation is predicted.
- the body temperature base prediction unit 149 records the received actual measurement value (step S1). Specifically, the body temperature base prediction unit 149 acquires today's date as the measurement date. Then, the body temperature base prediction unit 149 registers the received actual measurement value and user ID in association with the measurement date in the body temperature DB 12b.
- the cycle-based prediction unit 146 determines whether or not the number of past menstrual cycles recorded for the user indicated by the user ID of the target person is greater than the threshold value KN stored in advance in the storage unit 12 (step). S2). For example, the cycle base prediction unit 146 searches the menstrual date DB 12c for the menstrual date corresponding to the user ID of the subject. The cycle-based prediction unit 146 calculates the number of past menstrual cycles by subtracting 1 from the number of menstrual days found by this search.
- the threshold value KN may indicate any of two or more numbers, for example.
- step S2 determines that the number of past menstrual cycles is larger than the threshold value KN (step S2: YES)
- the cycle base prediction unit 146 proceeds to step S3.
- step S2: NO the cycle-based prediction unit 146 ends the menstrual day prediction process.
- the body temperature base prediction unit 149 determines whether or not the number of past menstrual cycles in which the measured values of body temperature of each day are recorded is greater than the threshold value KN. For example, the body temperature base prediction unit 149 sorts the menstrual dates found in step S2 in order of date. The body temperature base prediction unit 149 specifies the menstrual cycle for every two adjacent menstrual days in the sorted menstrual days. A period from one menstrual day to the day before the next menstrual day is one menstrual cycle. The body temperature base prediction unit 149 searches the body temperature DB 12b for each menstrual cycle for the actual measured values from the first day to the last day of the menstrual cycle, among the actual measured values of the body temperature corresponding to the user ID of the subject.
- the body temperature base prediction unit 149 increases the number of past menstrual cycles in which the actual temperature measurement values of each day are recorded by one. If the body temperature base prediction unit 149 determines that the number of menstrual cycles counted in this way is larger than the threshold value KN (step S3: YES), the process proceeds to step S4. On the other hand, when the body temperature base prediction unit 149 determines that the number of past menstrual cycles in which the measured values of the body temperature of each day are recorded is not larger than the threshold KN (step S3: NO), the physiological date prediction End the process.
- the body temperature base prediction unit 149 uses the actual measurement value of the body temperature such as days before and after the part of the day, The actual measured value may be interpolated. Accordingly, the body temperature base prediction unit 149 may increase the number of menstrual cycles in which the measured values of the body temperature of each day are recorded. Alternatively, the body temperature base prediction unit 149 may transmit a message that prompts the user terminal 2 to input a body temperature on a day for which an actual measurement value of body temperature is not recorded, for example. And the body temperature base prediction part 149 may acquire the body temperature manually input by the user from the user terminal 2 as a measured value, for example.
- step S4 the system control unit 14 determines whether the parameter information for the current menstrual cycle is registered. For example, the system control unit 14 specifies the latest menstrual day among the menstrual days found in step S2 as the first day of the current menstrual cycle. Next, the system control unit 14 searches the parameter DB 12d for parameter information whose registration date is after the first day of the current menstrual cycle among the parameter information corresponding to the user ID of the subject. As a result of the search, if the system control unit 14 determines that the parameter information is registered in the parameter DB 12d (step S4: YES), the system control unit 14 proceeds to step S6.
- step S5 the system control unit 14 executes parameter determination processing.
- the system control unit 14 generates parameters for the cycle-based prediction method and the body temperature-based prediction method. Further, the system control unit 14 predicts the next menstruation date using a cycle-based prediction method. In addition, the system control unit 14 estimates the number of days in the high temperature period of the current menstrual cycle by a body temperature-based prediction method. Then, the system control unit 14 generates parameter information for the current menstrual cycle. Details of the parameter determination process will be described later. Next, the system control unit 14 proceeds to step S6.
- step S6 the body temperature base prediction unit 149 determines whether or not an actual measurement value of the body temperature of each day of the current menstrual cycle is recorded. For example, the body temperature base prediction unit 149 searches the body temperature DB 12b for the actually measured values from the first day of the current menstrual cycle to today, among the measured body temperature values corresponding to the user ID of the subject. As a result of the search, when the actual measurement values are found on all the days from the first day to today, the body temperature base prediction unit 149 determines that the actual temperature measurement values for each day of the current menstrual cycle are recorded (step S6: YES) In this case, the body temperature base prediction unit 149 proceeds to step S7.
- the body temperature base prediction unit 149 determines that the actual temperature measurement value of each day of the current menstrual cycle is not recorded (step S6: NO). In this case, the body temperature base prediction unit 149 proceeds to step S10. Note that the body temperature base prediction unit 149 may determine, for example, the body temperature of the day on which the actual measurement value of the body temperature is not recorded, or may cause the user to input the body temperature.
- step S7 the timing specifying unit 148 acquires the short-term days and the long-term days from the parameter information for the current menstrual cycle.
- the timing specifying unit 148 determines the short-term moving average and the long-term average of each day from the first day of the current menstrual cycle to the present day based on the actual measurement value of each day of the current menstrual cycle and the short-term and long-term days.
- the moving average is calculated (step S8).
- step S9 the timing specifying unit 148 determines whether or not a P crossover is specified in the current menstrual cycle.
- the timing specifying unit 148 determines that a P crossover has occurred. You may judge. In this case, the timing specifying unit 148 specifies, for example, one of these two days as the day when the P crossover has occurred. For example, the timing specifying unit 148 may specify the day with the smaller difference between the short-term moving average and the long-term moving average as the day when the P crossover has occurred.
- the timing specifying unit 148 may determine that the P crossover has occurred. In this case, for example, the timing specifying unit 148 may specify the central day of these three days as the day when the P crossover has occurred.
- the timing specifying unit 148 proceeds to step S11.
- the timing specifying unit 148 proceeds to step S10.
- step S10 the cycle-based prediction unit 146 transmits the predicted menstrual date included in the parameter information for the current menstrual cycle to the user terminal 2, and ends the menstrual date prediction process.
- the user terminal 2 displays the predicted physiological date received from the information processing server 1 on the screen.
- the body temperature base prediction unit 149 determines the estimated ovulation date based on the day when the P crossover occurs in the current menstrual cycle. For example, the body temperature base prediction unit 149 may determine the day when the P crossover has occurred and the estimated ovulation day. Next, the body temperature base prediction unit 149 calculates a predicted physiological date by adding the estimated high temperature period days included in the parameter information for the current menstrual cycle to the day when the P crossover occurs (step S12). Next, the body temperature base prediction unit 149 transmits the estimated ovulation date and the predicted physiological date to the user terminal 2 (step S13). The menstrual day prediction process is terminated. The user terminal 2 displays the estimated ovulation date and the predicted menstrual date received from the information processing server 1 on the screen.
- FIG. 14 is a flowchart showing an example of parameter determination processing of the system control unit 14 of the information processing server 1 according to the present embodiment.
- the system control unit 14 performs a cycle-based parameter determination process (step S21).
- the cycle-based parameter determination process the system control unit 14 determines parameters used in the cycle-based prediction method. Details of the cycle-based parameter determination process will be described later.
- the cycle base prediction unit 146 performs a cycle base prediction process (step S22). In the cycle-based prediction process, the cycle-based prediction unit 146 determines a predicted physiological date using a cycle-based prediction method. Details of the cycle-based prediction process will be described later.
- the period days determination unit 147 performs a body temperature base parameter determination process (step S23).
- the body temperature-based parameter determination process determines parameters used in the body temperature-based prediction method. Details of the body temperature-based parameter determination process will be described later.
- the body temperature-based prediction unit 149 determines one or more past ones based on the short-term moving average and the long-term moving average calculated based on the short-term days and the long-term days determined as the parameters of the body temperature-based prediction method. The day when the P crossover occurs in the menstrual cycle and the day when the N crossover occurs are specified. Then, the body temperature base prediction unit 149 calculates the average value of the days from the day when the P crossover occurs until the day when the N crossover occurs as the estimated high temperature period days (step S24).
- the system control unit 14 acquires today's date as the registration date.
- the system control unit 14 includes the user ID of the subject, the registration date, the parameters determined by the cycle-based parameter determination process and the body temperature-based parameter determination process, the predicted physiological date determined by the cycle-based prediction method, and the estimated high temperature period date. Generate parameter information including numbers.
- the system control unit 14 registers the generated parameter information in the parameter DB 12d (step S25), and ends the parameter determination process.
- FIG. 15 is a flowchart showing an example of the cycle-based parameter determination process of the system control unit 14 of the information processing server 1 according to the present embodiment.
- the parameter candidate acquisition unit 141 determines the number RN of reference periods (step S31). For example, the parameter candidate acquisition unit 141 acquires the initial value of the number of reference cycles stored in advance in the storage unit 12. Next, the parameter candidate acquisition unit 141 determines whether the number of past menstrual cycles is larger than the initial value. When the number of past menstrual cycles is larger than the initial value, the parameter candidate acquisition unit 141 determines the initial value as RN. On the other hand, when the number of past menstrual cycles is equal to or less than the initial value, the parameter candidate acquisition unit 141 calculates RN by subtracting 1 from the number of past menstrual cycles.
- the parameter candidate acquisition unit 141 sets the number i to 1 (step S32). Then, the parameter candidate acquisition unit 141 acquires from the storage unit 12 a candidate OF (i) for removing an abnormal value threshold. Next, the parameter candidate acquisition unit 141 sets the number j to 1 (step S33). Then, the parameter candidate acquisition unit 141 acquires the candidate WN (j) for the winsolation threshold from the storage unit 12. Next, the parameter candidate acquisition unit 141 sets the representative value type to “average value” (step S34). Next, the system control unit 14 executes a day difference calculation process (step S35). In the day difference calculation process, the system control unit 14 corrects a plurality of reference periods based on the parameter candidate combinations determined in steps S32 to S34. Then, the system control unit 14 calculates the difference between the corrected representative value of the reference period and the reference period as the total number of days difference. Details of the day difference calculation process will be described later.
- the parameter candidate acquisition unit 141 determines whether or not the representative value type is “average value” (step S36). At this time, if the parameter candidate acquisition unit 141 determines that the representative value type is “average value” (step S36: YES), the parameter candidate acquisition unit 141 proceeds to step S37. In step S37, the parameter candidate acquisition unit 141 changes the representative value type to “median”, and the process proceeds to step S35. On the other hand, if the parameter candidate obtaining unit 141 determines that the representative value type is not “average value” (step S36: NO), the parameter candidate obtaining unit 141 proceeds to step S38. In step S38, the parameter candidate acquisition unit 141 determines whether the number j is less than the number of candidates for the winsolization threshold.
- step S38 determines that the parameter number j is less than the number of candidates for the threshold for winsolization.
- step S38 determines that the parameter candidate acquisition unit 141 determines that the parameter number j is less than the number of candidates for the threshold for winsolization.
- step S40 the parameter candidate acquisition unit 141 determines whether the number i is less than the number of abnormal value removal threshold candidates.
- step S40 determines that the parameter number i is less than the number of abnormal value removal threshold candidates.
- step S40 determines that the parameter number i is not less than the number of abnormal value removal threshold candidates.
- step S42 the parameter determination unit 145 rearranges all candidate combinations in ascending order of the total number of days.
- the parameter determination unit 145 identifies the candidate combination with the smallest total number of days difference as the first candidate.
- the parameter determination unit 145 determines whether there is another candidate combination for which the total number of days difference within a predetermined range is calculated from the total number of days difference of the first candidate (step S43). For example, the parameter determination unit 145 calculates a reference value for the total number of days difference by multiplying the total number of days difference of the first candidate by a coefficient stored in advance in the storage unit 12. The value of this coefficient is larger than 1, for example.
- the parameter determination unit 145 determines whether or not there is one or more candidate combinations whose total days difference is equal to or less than a reference value except for the first candidate. At this time, when there is one or more candidate combinations whose total number of days difference is equal to or less than the reference value, the parameter determination unit 145 calculates another candidate combination in which the total number of days difference within a predetermined range is calculated from the total number of days difference of the first candidate (Step S43: YES). In this case, the parameter determination unit 145 specifies a candidate combination whose total number of days difference is equal to or less than the reference value as the second candidate. Then, the parameter determination unit 145 proceeds to step S45.
- Step S43 NO
- the parameter determination unit 145 proceeds to step S44.
- step S44 the parameter determination unit 145 determines the first candidate as a parameter used in the cycle-based prediction method, and ends the cycle-based parameter determination process.
- step S45 the parameter determination unit 145 executes the number acquisition process.
- the parameter determination unit 145 acquires the number of users for which the first candidate is determined as a parameter for the period-based prediction method, and the number of users for which the second candidate is determined as a parameter for the period-based prediction method. To get. Details of the number acquisition process will be described later.
- the parameter determining unit 145 determines the candidate having the largest number of users determined as a parameter from the first candidate and the second candidate as a parameter to be used in the period-based prediction method (step S46). The parameter determination process is terminated.
- FIG. 16 is a flowchart illustrating an example of the day difference calculation process of the system control unit 14 of the information processing server 1 according to the present embodiment.
- the parameter determination unit 145 sets the total number of days difference TD (i, j, k) to 0 (step S51).
- the cycle set acquisition unit 142 sets the number m of the menstrual cycle to 1 (step S52).
- the cycle set acquisition unit 142 acquires the number of days in the menstrual cycle (m) among the plurality of past menstrual cycles identified in the menstrual day prediction process as the number of days in the reference cycle (step S53).
- the menstrual cycle (m) is the m-th latest menstrual cycle among the past menstrual cycles.
- the cycle set acquisition unit 142 sets the number of days from the menstrual cycle (m + 1) to the menstrual cycle (m + RN + 1) among the past menstrual cycles identified in the menstrual day prediction process as the number of reference cycle days. Obtain (step S54). Then, the cycle set acquisition unit 142 rearranges the acquired days of the reference cycle in ascending order.
- the modified reference period acquisition unit 143 counts the number of appearances of the reference period for each number of days. And the correction period group acquisition part 143 calculates the appearance probability of a reference period for every number of days by dividing each appearance number by RN (step S55). Next, the corrected reference period acquisition unit 143 removes, from the acquired reference period, the reference period of days whose appearance probability is less than OF (i) percent as an abnormal value (step S56).
- the modified reference cycle acquisition unit 143 determines the WN (j) percentile and the (100-WN (j)) percentile from the remaining number of days of the reference cycle from which the abnormal value has been removed. Then, the modified reference cycle acquisition unit 143 changes the number of days in the reference cycle that is less than the WN (j) percentile among the remaining reference cycles to the WN (j) percentile (step S57). Further, the modified reference cycle acquisition unit 143 changes the number of days of the reference cycle that exceeds the (100-WN (j)) percentile among the remaining reference cycles to the (100-WN (j)) percentile (step S58). ).
- step S59 determines whether or not the representative value type k is “average value” (step S59). At this time, if the representative value acquisition unit 144 determines that the representative value type k is “average value” (step S59: YES), the process proceeds to step S60. In step S60, the representative value acquisition unit 144 calculates the average value of the corrected number of days in the reference period as the representative value P, and proceeds to step S62. On the other hand, if the representative value acquisition unit 144 determines that the representative value type k is not “average value” (step S59: NO), the process proceeds to step S61. In step S61, the representative value acquisition unit 144 determines the corrected median number of days of the reference cycle as the representative value P, and proceeds to step S62.
- step S62 the parameter determination unit 145 calculates the absolute value D of the difference between the number of days in the reference period and the representative value P.
- the parameter determination unit 145 acquires the weighting coefficient W (m) from the storage unit 12. The larger the value of m, the smaller the value of W (m).
- the parameter determination unit 145 calculates a weighted difference by multiplying the absolute value D of the difference by W (m). Then, the parameter determination unit 145 adds T to the total number of days difference T (i, j, k) and updates T (i, j, k) (step S63).
- the cycle set acquisition unit 142 determines whether the sum of the number m and the number of reference cycles RN is less than the number of past menstrual cycles (step S64). At this time, when the cycle set acquisition unit 142 determines that the sum of the number m and the number of reference cycles RN is less than the number of past menstrual cycles (step S64: YES), the cycle set acquisition unit 142 proceeds to step S65. In step S65, the period set acquisition unit 142 adds 1 to the number m, and proceeds to step S53. On the other hand, when it is determined that the sum of the number m and the number of reference cycles RN is not less than the number of past menstrual cycles (step S64: NO), the cycle set acquisition unit 142 ends the day difference calculation process. .
- FIG. 17 is a flowchart showing an example of the number acquisition process of the system control unit 14 of the information processing server 1 according to the present embodiment.
- the parameter determination unit 145 acquires the age corresponding to the user ID of the target person from the member information DB 12a. And the parameter determination part 145 acquires a subject's age based on the acquired age (step S71). Next, the parameter determination unit 145 selects one from the first candidate and one or a plurality of second candidates (step S72).
- the parameter determination unit 145 searches the parameter DB 12d for parameter information including all of the selected candidate abnormal value removal threshold value, the winsolization threshold value, and the representative value type. Then, the parameter determining unit 145 generates a list of parameter information found by the search (step S73).
- the parameter determining unit 145 deletes parameter information that is not the latest from the list (step S74). For example, the parameter determination unit 145 acquires the user ID from the parameter information in the list. Next, the parameter determination unit 145 searches the parameter DB 12d for parameter information with the latest registration date among the parameter information corresponding to the acquired user ID. Next, when the registration date of the parameter information in the list is older than the latest registration date, the parameter determination unit 145 deletes the parameter information. The parameter determination unit 145 performs these processes for each parameter information in the list.
- the parameter determination unit 145 counts the number of users that match the age of the target person (step S75). For example, the parameter determination unit 145 acquires the age corresponding to the user ID included in the parameter information in the list from the member information DB 12a. And the parameter determination part 145 acquires the user's age which user ID shows based on the acquired age. When the acquired age matches the age of the target person, the parameter determination unit 145 increases the number of users by one. The parameter determining unit 145 executes these processes for each parameter information in the list (step S75).
- the parameter determination unit 145 determines whether or not all of the first candidate and one or more second candidates have been selected (step S76). At this time, if the parameter determination unit 145 determines that there is a candidate that has not yet been selected (step S76: NO), the parameter determination unit 145 proceeds to step S77. In step S77, the parameter determination unit 145 selects one of the candidates that have not been selected, and proceeds to step S73. On the other hand, if the parameter determination unit 145 determines that all candidates have been selected (step S76: YES), the parameter acquisition process ends.
- FIG. 18 is a flowchart illustrating an example of the cycle-based prediction process of the system control unit 14 of the information processing server 1 according to the present embodiment.
- the period-based prediction unit 146 acquires the abnormal value removal threshold OF, the windization threshold WN, and the representative value type PR determined as parameters (step S81).
- the cycle-based prediction unit 146 acquires the number of days in each of a plurality of past menstrual cycles of the subject (step S82).
- the cycle-based prediction unit 146 counts the number of appearances of past menstrual cycles for each number of days. Then, the cycle-based prediction unit 146 calculates the appearance probability of the past menstrual cycle for each number of days by dividing the number of appearances by the number of past menstrual cycles (step S83). Next, the cycle-based prediction unit 146 removes, from the acquired menstrual cycle, the reference cycle having the number of days whose appearance probability is less than OF percent as an abnormal value (step S84). Next, the cycle-based prediction unit 146 determines a WN percentile and a (100-WN) percentile from the number of days in the remaining menstrual cycle from which the abnormal value has been removed.
- the cycle-based prediction unit 146 changes the number of days of the menstrual cycle that is less than the WN percentile among the remaining menstrual cycles to the WN percentile (step S85). Further, the cycle-based prediction unit 146 changes the number of days of the menstrual cycle that exceeds the (100-WN) percentile among the remaining menstrual cycles to the (100-WN) percentile (step S86).
- step S87 determines whether or not the representative value type PR is “average value” (step S87). At this time, if the cycle-based prediction unit 146 determines that the representative value type PR is “average value” (step S87: YES), the cycle-based prediction unit 146 proceeds to step S88. In step S88, the cycle base prediction unit 146 calculates the average value of the number of days in the corrected menstrual cycle as the representative value P, and proceeds to step S90. On the other hand, if the cycle-based prediction unit 146 determines that the representative value type PR is not “average value” (step S87: NO), the cycle-based prediction unit 146 proceeds to step S89. In step S89, the cycle base prediction unit 146 determines the median value of the number of days in the corrected menstrual cycle as the representative value P, and proceeds to step S90.
- step S90 the cycle-based prediction unit 146 calculates the predicted physiological date by adding the representative value P to the date of the first day of the current menstrual cycle. Then, the cycle base prediction unit 146 ends the cycle base prediction process.
- FIG. 19 is a flowchart showing an example of the body temperature-based parameter determination process of the system control unit 14 of the information processing server 1 according to the present embodiment.
- the period days determination unit 147 acquires one combination of the short-term days and the long-term days (step S ⁇ b> 101).
- the period days determination unit 147 determines a short-term moving average and a long-term moving average of the body temperature of the past one or more menstrual cycles based on the acquired combination and the actual measurement value of the past one or more menstrual cycles of the subject.
- the moving average is calculated (step S102).
- the period days determination unit 147 identifies the day on which the N crossover has occurred based on the calculated moving average (step S103).
- the method for identifying the day when the N crossover occurs is the same as the method for identifying the day when the P crossover occurs, except that the high-low relationship between the short-term moving average and the long-term moving average is reversed.
- the method for specifying the day when the P crossover has occurred is described in the description of step S9 of the menstrual day prediction process shown in FIG.
- the period days determination unit 147 calculates the average value of the difference of the menstrual day and the day when P crossover occurred (step S104). That is, the period days determination unit 147 calculates Equation 3.
- the period days determination unit 147 calculates a definite integral of the difference between the short-term moving average and the long-term moving average in the past one or more menstrual cycles (step S105). That is, the period days determination unit 147 calculates Equation 5.
- the period days determination unit 147 calculates the function f shown in Expression 6 using the calculation results of steps S104 and S105 (step S106).
- the period days determination unit 147 determines whether or not all combinations of short-term days and long-term days have been acquired (step S107). At this time, if it is determined that there is a combination that has not yet been acquired (step S107: NO), the period days determination unit 147 proceeds to step S108. In step S108, the period days determination unit 147 acquires one of the combinations that have not yet been acquired, and proceeds to step S102. On the other hand, if the period days determining unit 147 determines that all combinations have been acquired (step S107: YES), the process proceeds to step S109.
- step S109 the period days determination unit 147 determines the combination having the largest function f among the plurality of combinations of the short-term days and the long-term days as a parameter of the body temperature-based prediction method. Then, the body temperature base prediction unit 149 ends the body temperature base parameter determination process.
- the system control unit 14 includes a plurality of candidates for at least one of a plurality of candidates for the threshold value for abnormal value removal and a plurality of candidates for the threshold value for windization. May be obtained. Further, the system control unit 14 may acquire a reference period and a plurality of reference periods from a plurality of past menstrual periods of the subject. Further, the system control unit 14 obtains a plurality of reference periods corrected by performing predetermined processing on outliers specified by the threshold candidates among a plurality of reference circumferences for each threshold candidate. Also good. The system control unit 14 may acquire representative values of a plurality of corrected reference periods for each threshold candidate. Further, the system control unit 14 may compare the representative value acquired for each threshold candidate and the reference period.
- the system control unit 14 may determine a threshold value used for predicting the number of days in the menstrual cycle of the target person from a plurality of threshold value candidates based on the comparison result. In this case, it is possible to determine a threshold value that suppresses a decrease in prediction accuracy even when the menstrual cycle varies greatly. Therefore, based on this threshold, it is possible to predict the menstrual cycle while suppressing a decrease in prediction accuracy, and it is possible to predict a menstrual day using this menstrual cycle.
- the system control unit 14 may acquire a plurality of candidates for abnormal value removal thresholds and a plurality of candidates for threshold values for windization. Also, the reference period specified as an abnormal value by the abnormal value removal threshold from a plurality of reference periods for each candidate combination including the abnormal value removal threshold candidate and the winsolization threshold candidate. And the number of days of the reference period specified as an outlier by the threshold of the winsolization is changed to the number of days corresponding to the threshold of the winsolization among the plurality of reference periods from which the abnormal value is removed A plurality of corrected reference periods may be acquired. Further, the system control unit 14 may acquire representative values of a plurality of corrected reference periods for each candidate combination. Further, the system control unit 14 may compare the representative value and the reference period for each candidate combination.
- system control unit 14 may determine a candidate combination used for prediction from among a plurality of candidate combinations based on the comparison result. In this case, even when the menstrual cycle has a large variation, it is possible to determine an abnormal value removal threshold and a winsolization threshold that suppress a decrease in prediction accuracy.
- the system control unit 14 may acquire an abnormal value removal threshold value and a winsolization threshold value. Further, the system control unit 14 may remove a menstrual cycle specified as an abnormal value by an abnormal value removal threshold from a plurality of menstrual cycles. In addition, the system control unit 14 changes the number of days of the menstrual cycle identified as an outlier by the threshold of the winsolization among a plurality of menstrual cycles from which abnormal values have been removed to the number of days corresponding to the threshold of the winsolization May be. Then, the system control unit 14 may predict the number of days in the menstrual cycle of the subject based on a plurality of menstrual cycles in which the number of days in the menstrual cycle has been changed. In this case, even when the menstrual cycle has a large variation, it is possible to predict the menstrual cycle by suppressing the decrease in prediction accuracy, and it is possible to predict the menstrual day using this menstrual cycle. is there.
- system control unit 14 may acquire a plurality of periodic sets. Further, the system control unit 14 may acquire a plurality of corrected reference periods for each combination of threshold candidates and period sets. Further, the system control unit 14 may acquire representative values of a plurality of corrected reference periods for each combination of threshold candidates and period sets. Further, the system control unit 14 may compare the representative value and the reference period for each combination of the threshold candidate and the period set. Then, the system control unit 14 may determine a threshold used for prediction based on a plurality of comparison results obtained for each threshold candidate. In this case, a more appropriate threshold value can be determined.
- system control unit 14 may acquire a plurality of continuous reference cycles and a reference cycle that is continuous with the plurality of reference cycles and is newer than the plurality of reference cycles. Further, the system control unit 14 may determine a threshold value used for prediction by weighting each of a plurality of comparison results obtained for each threshold candidate. At this time, the system control unit 14 may decrease the weight of the comparison result between the reference period and the representative value as the reference period is older. In this case, a more appropriate threshold can be determined for the prediction of the menstrual cycle.
- the system control unit includes a second candidate in which a difference within a predetermined range is acquired from the difference of the first candidate having the smallest difference between the representative value and the reference period among the plurality of threshold candidates. Based on the number of users whose first candidate is determined as a threshold used for prediction among users different from the target person, and the number of users whose second candidate is determined as a threshold used for prediction A threshold value used for prediction may be determined from the second candidates. In this case, a more appropriate threshold can be determined for the prediction of the menstrual cycle.
- the system control unit 14 sets the number of users having characteristics identical to the characteristics of the target person among the users whose first candidates are determined as threshold values used for prediction, and the threshold values used for prediction.
- the threshold used for prediction may be determined based on the number of users having features that are identical to the features of the subject among the users for which two candidates are determined. In this case, a threshold value that is more appropriate for the subject can be determined.
- system control unit 14 may acquire the average value and the median value of a plurality of corrected reference periods for each threshold candidate. Further, the system control unit 14 may compare the average value and the median value with the reference period for each threshold candidate. Then, the system control unit 14 may determine a candidate combination to be used for prediction from among a plurality of candidate combinations including threshold candidates and representative value types. In this case, a representative value suitable for the distribution status of a plurality of menstrual cycles in the past can be acquired.
- the system control unit 14 determines that the timing at which the short-term moving average of the body temperature measured in the current menstrual cycle of the subject is larger than the long-term moving average has not arrived within the current menstrual cycle.
- the number of days in the subject's current menstrual cycle may be predicted based on the threshold value and the subject's past menstrual cycles, and the next menstrual day may be predicted based on the predicted number of days .
- the system control unit 14 may predict the next menstrual day based on that timing. Good. In this case, after the transition from the low temperature period to the high temperature period, a more accurate menstrual day can be predicted.
- the timing at which the moving average of short-term days is lower than the moving average of long-term days is one or more times in the past.
- the short-term days and the long-term days may be determined so as to overlap with the menstrual days.
- the system control unit 14 may specify the timing at which the moving average of the short-term days is higher than the moving average of the long-term days for the moving average of the body temperature measured in the target menstrual cycle. Then, the system control unit 14 may predict the next menstrual day of the target menstrual cycle based on the identified timing, or may estimate the arrival of the ovulation day during the target menstrual cycle.
- the time series of the body temperature is smoothed by the moving average, and by specifying this timing, the body temperature fluctuates up and down before and after the timing of actually shifting from the low temperature period to the high temperature period.
- the timing can be specified appropriately.
- the system control unit 14 determines that the moving average of the short-term days is lower than the moving average of the long-term days from the timing when the moving average of the short-term days is higher than the moving average of the long-term days in one or more menstrual cycles in the past.
- the next physiological date may be predicted by adding the representative value of the number of days until the timing to the timing when the moving average of the short-term days is higher than the moving average of the long-term days in the current menstrual cycle. In this case, at least one of the estimation of the ovulation date and the prediction of the physiological date can be performed more appropriately according to the tendency of the subject in the high temperature period.
- the system control unit 14 may reduce the time difference between the timing when the short-term moving average of the body temperature measured in the past one or more menstrual cycles is lower than the long-term moving average of the body temperature and the physiological date.
- the short-term days and the long-term days may be determined so that the integrated value of the difference between the short-term moving average and the long-term moving average in one or more past menstrual cycles becomes large. In this case, based on the integrated value of the difference between the short-term moving average of the body temperature and the long-term moving average, and the time difference between the period when the short-term moving average of the body temperature is lower than the long-term moving average and the physiological date, Appropriate short and long days can be determined.
- the system control unit 14 determines the next menstrual day based on the number of days in the past one or more menstrual cycles. May be predicted. In this case, the menstrual date can be predicted even if the transition from the low temperature period to the high temperature period has not yet occurred.
- the information processing apparatus of the present invention is applied to a server apparatus in a client server system.
- the information processing apparatus of the present invention may be applied to an information processing apparatus other than the server apparatus.
- the information processing apparatus of the present invention may be applied to the user terminal 2 or the like.
- the information processing apparatus may accept an input of an actual measurement value and a physiological date from the user's body temperature, estimate the ovulation date of the user, or predict the next physiological date.
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Pathology (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Veterinary Medicine (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- Heart & Thoracic Surgery (AREA)
- Biophysics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- General Physics & Mathematics (AREA)
- Reproductive Health (AREA)
- Gynecology & Obstetrics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Medical Treatment And Welfare Office Work (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
Description
先ず、本実施形態に係る情報処理システムSの構成及び機能概要について、図1を用いて説明する。図1は、本実施形態に係る情報処理システムSの概要構成の一例を示す図である。
次に、情報処理サーバ1の構成について、図2A乃至図3を用いて説明する。
次に、図2B、図4A乃至図12Bを用いて、システム制御部14の機能概要について説明する。図2Bは、本実施形態に係るシステム制御部14の機能ブロックの一例を示す図である。システム制御部14は、所定の条件に基づいて、周期ベース予測手法及び体温ベース予測手法の何れか用いるかを決定する。そして、システム制御部14は、決定した手法を用いて次回の生理日を予測する。そのため、システム制御部14は、CPU14aが、予測処理プログラム等のプログラムを読み出し実行することにより、図2Bに示すように、パラメータ候補取得部141、周期セット取得部142、修正参照周期取得部143、代表値取得部144、パラメータ決定部145、周期ベース予測部146、期間日数決定部147、タイミング特定部148、体温ベース予測部149等として機能する。周期ベース予測部146は、本発明における予測手段の一例である。期間日数決定部147は本発明における決定手段の一例である。タイミング特定部148は、本発明における特定手段の一例である。体温ベース予測部149は、本発明における推定手段の一例である。
周期ベース予測手法を用いる場合、システム制御部14は、過去の複数の月経周期を取得する。次いで、システム制御部14は、過去の複数の月経周期のうち閾値により特定される外れ値に所定の処理を施して、収載された複数の月経周期を取得する。所定の処理は、例えば、閾値により特定される外れ値を除去することであってもよいし、閾値により特定される外れ値にウインソリゼーションを施すことであってもよい。或いは、システム制御部14は、所定の処理は、異常値除去とウインソリゼーションの両方を行うことであってもよい。この場合、システム制御部14は、先に異常値除去を行い、その後でウインソリゼーションを行ってもよい。複数の月経周期から異常値が除去された後であっても、参照周期の日数のばらつきが大きい場合がある。ばらつきが大きいままであると、予測の精度が低下する可能性がある。しかしながら、異常値が除去された状態で、更に外れ値を特定して、この外れ値を単に除去してしまっても。予測の精度が低下する可能性がある。そこで、ウインソリゼーションにより、外れ値を、月経周期の分布の外側に位置するという情報を残すように変更することで、予測精度の低下を抑制することができる。複数の月経周期が修正されると、システム制御部14は、修正された複数の月経周期の代表値を計算する。そして、システム制御部14は、月経周期の代表値に基づいて、次回の生理日を予測する。
システム制御部14は、周期ベース予測手法を用いて生理日を予測するために、周期ベース予測手法で用いられるパラメータを、ユーザごとに決定する。決定されるパラメータは、異常値除去の閾値、ウインソリゼーションの閾値及び代表値種別を決定のうち少なくとも何れか1つである。なお、異常値除去の閾値、ウインソリゼーションの閾値及び代表値種別の少なくとも1つは、例えば情報処理システムS全体で予め定められていてもよい。また、例えば複数の月経周期の修正に異常値除去のみが用いられる場合、ウインソリゼーションの閾値は不要である。また、例えば複数の月経周期の修正にウインソリゼーションのみが用いられる場合、異常値除去の閾値は不要である。
周期ベース予測部146は、パラメータ決定部145により決定されたパラメータ又は組み合わせを用いて、対象者の現月経周期の日数を予測する。具体的に、周期ベース予測部146は、生理日の予測の対象者の過去の複数の月経周期を取得する。次いで、周期ベース予測部146は、取得された複数の月経周期のうちパラメータとして決定された閾値により特定される外れ値に所定の処理を施すことにより、修正された複数の月経周期を取得する。より詳細に、周期ベース予測部146は、複数の月経周期から異常値を除去し、又は複数の月経周期に対してウインソリゼーションを施す。或いは、周期ベース予測部146は、異常値の除去を行った後に、更にウインソリゼーションを行う。パラメータ決定部145により異常値除去の閾値が決定された場合、周期ベース予測部146は、この閾値を用いて異常値を除去する。パラメータ決定部145によりウインソリゼーションの閾値が決定された場合、周期ベース予測部146は、この閾値を用いてウインソリゼーションを行う。周期ベース予測部146は、修正された複数の月経周期の代表値を、現月経周期の日数の予測値として取得する。パラメータ決定部145により代表値種別として「平均値」が決定された場合、周期ベース予測部146は平均値を計算し、代表値種別として「中央値」が決定された場合、周期ベース予測部146は中央値を決定する。次いで、周期ベース予測部146は、現月経周期の初日に、取得した代表値を加算して、予測生理日を計算する。
体温ベース予測手法を用いる場合、システム制御部14は、生理日が予測される対象者の体温の実測値の短期の移動平均と長期の移動平均とを、例えば日ごとに計算する。そして、システム制御部14は、現月経周期において、短期の移動平均が長期の移動平均よりも大きくなるタイミングを特定する。短期の移動平均が長期の移動平均よりも大きくなるタイミングとは、例えば短期の移動平均線と長期の移動平均線が交差するタイミングのうち、移動平均線が交差するタイミングの直後の短期の移動平均が長期の移動平均よりも高いタイミングである。このように移動平均線が交差することを、Pクロスオーバー(ポジティブクロスオーバー)という。短期の移動平均が長期の移動平均よりも大きくなるとは、Pクロスオーバーが起こることである。図9は、現在の月経周期における体温の実測値のグラフと、体温の短期の移動平均線と長期の移動平均線との例を示す図である。図9に示すように、1月1日から現在の月経周期が始まった後、1月12日でPクロスオーバーが発生している。システム制御部14は、特定されたタイミングに基づいて、次の生理日を予測し又は排卵日の到来を推定する。或いは、システム制御部14は、次の生理日の予測と排卵日の到来の推定の両方を行ってもよい。
期間日数決定部147は、体温ベース予測手法を用いた排卵日の到来の推定又は生理日の予測を行うために、体温ベース予測手法で用いられるパラメータを決定する。決定されるパラメータは、移動平均の計算に用いられる短期日数と長期日数である。
タイミング特定部148は、期間日数決定部147により決定された短期日数と長期日数を用いて、現月経周期における短期の移動平均と長期の移動平均を計算する。指数移動平均を計算する場合、タイミング特定部148は、例えば過去の1以上の月経周期内の何れかの日から指数移動平均の計算を開始してもよい。タイミング特定部148は、現月経周期において、短期の移動平均が長期の移動平均よりも大きくなるタイミング、すなわちPクロスオーバーが起きた日を特定する。
周期ベース予測手法によれば、過去の3以上の月経周期のそれぞれの日数が特定されていれば、現在が低温期であるか高温期であるかに関わらず、次の生理日を予測することができる。一方、体温ベース予測手法によれば、体温の短期の移動平均が長期の移動平均よりも大きくなったタイミング、すなわちPクロスオーバーが起きたタイミングに基づいて、次の生理日を精度よく予測することができる。しかしながら、現月経周期においてPクロスオーバーがまだ起きていない場合、すなわち現在が低温期である蓋然性が高い場合、体温ベース予測手法を用いることができない。
次に、情報処理システムSの動作について、図13乃至図19を用いて説明する。
2 ユーザ端末
11 通信部
12 記憶部
12a 会員情報DB
12b 体温DB
12c 生理日DB
12d パラメータDB
13 入出力インターフェース
14 システム制御部
14a CPU
14b ROM
14c RAM
15 システムバス
141 パラメータ候補取得部
142 周期セット取得部
143 修正参照周期取得部
144 代表値取得部
145 パラメータ決定部
146 周期ベース予測部
147 期間日数決定部
148 タイミング特定部
149 体温ベース予測部
NW ネットワーク
S 情報処理システム
Claims (6)
- 過去の1以上の月経周期の間に測定された複数日分の体温について、短期日数の移動平均が長期日数の移動平均よりも低くなるタイミングが、過去の1以上の生理日と重なるように、前記短期日数と前記長期日数とを決定する決定手段と、
対象とする月経周期で測定された体温について、前記決定手段により決定された前記短期日数の移動平均が、前記決定手段により決定された前記長期日数の移動平均よりも高くなるタイミングを特定する特定手段と、
前記特定手段により特定されたタイミングに基づいて、前記対象とする月経周期の次の生理日を予測し、又は前記対象とする月経周期中の排卵日の到来を推定する推定手段と、
を備えることを特徴とする情報処理装置。 - 請求項1に記載の情報処理装置において、
前記推定手段は、前記過去の1以上の月経周期において、前記短期日数の移動平均が前記長期日数の移動平均よりも高くなるタイミングから、前記短期日数の移動平均が前記長期日数の移動平均よりも低くなるタイミングまでの日数の代表値を、前記特定手段により特定された前記タイミングに加算して、前記次の生理日を予測することを特徴とする情報処理装置。 - 請求項1又は2に記載の情報処理装置において、
前記決定手段は、前記過去の1以上の月経周期で測定された体温の短期の移動平均が該体温の長期の移動平均よりも低くなるタイミングと生理日との時間差が小さくなり、且つ、前記過去の1以上の月経周期における短期の移動平均と長期の移動平均との差の積分値大きくなるなるように、前記短期日数と前記長期日数とを決定することを特徴とする情報処理装置。 - 請求項1乃至3の何れか1項に記載の情報処理装置において、
前記短期日数の移動平均が前記長期日数の移動平均よりも高くなるタイミングが前記特定手段により特定されなかった場合、前記過去の1以上の月経周期の日数に基づいて、次の生理日を予測する予測手段を更に備えることを特徴とする情報処理装置。 - コンピュータにより実行される情報処理方法であって、
過去の1以上の月経周期の間に測定された複数日分の体温について、短期日数の移動平均が長期日数の移動平均よりも低くなるタイミングが、過去の1以上の生理日と重なるように、前記短期日数と前記長期日数とを決定する決定ステップと、
対象とする月経周期で測定された体温について、前記決定ステップにより決定された前記短期日数の移動平均が、前記決定ステップにより決定された前記長期日数の移動平均よりも高くなるタイミングを特定する特定ステップと、
前記特定ステップにより特定されたタイミングに基づいて、前記対象とする月経周期の次の生理日を予測し、又は前記対象とする月経周期中の排卵日の到来を推定する推定ステップと、
を含むことを特徴とする情報処理方法。 - コンピュータを、
過去の1以上の月経周期の間に測定された複数日分の体温について、短期日数の移動平均が長期日数の移動平均よりも低くなるタイミングが、過去の1以上の生理日と重なるように、前記短期日数と前記長期日数とを決定する決定手段、
対象とする月経周期で測定された体温について、前記決定手段により決定された前記短期日数の移動平均が、前記決定手段により決定された前記長期日数の移動平均よりも高くなるタイミングを特定する特定手段、及び、
前記特定手段により特定されたタイミングに基づいて、前記対象とする月経周期の次の生理日を予測し、又は前記対象とする月経周期中の排卵日の到来を推定する推定手段、
として機能させることを特徴とする情報処理プログラム。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2015/054790 WO2016132529A1 (ja) | 2015-02-20 | 2015-02-20 | 情報処理装置、情報処理方法及び情報処理プログラム |
US15/552,311 US10485487B2 (en) | 2015-02-20 | 2015-02-20 | Information processing device, information processing method, and information processing program |
JP2015520745A JP5775242B1 (ja) | 2015-02-20 | 2015-02-20 | 情報処理装置、情報処理方法及び情報処理プログラム |
TW104122785A TWI540997B (zh) | 2015-02-20 | 2015-07-14 | 資訊處理裝置、資訊處理方法及資訊處理程式產品 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2015/054790 WO2016132529A1 (ja) | 2015-02-20 | 2015-02-20 | 情報処理装置、情報処理方法及び情報処理プログラム |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016132529A1 true WO2016132529A1 (ja) | 2016-08-25 |
Family
ID=54192557
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2015/054790 WO2016132529A1 (ja) | 2015-02-20 | 2015-02-20 | 情報処理装置、情報処理方法及び情報処理プログラム |
Country Status (4)
Country | Link |
---|---|
US (1) | US10485487B2 (ja) |
JP (1) | JP5775242B1 (ja) |
TW (1) | TWI540997B (ja) |
WO (1) | WO2016132529A1 (ja) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6259147B1 (ja) * | 2017-05-16 | 2018-01-10 | 佐々木 修 | 体調予測システム |
JP2020008922A (ja) * | 2018-07-03 | 2020-01-16 | グラフテック株式会社 | 計測データ管理装置、計測データ管理方法、および計測データ管理プログラム |
JP2020014665A (ja) * | 2018-07-25 | 2020-01-30 | 日本光電工業株式会社 | 生体情報表示装置、および生体情報表示用データの出力方法 |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016132528A1 (ja) | 2015-02-20 | 2016-08-25 | 楽天株式会社 | 情報処理装置、情報処理方法及び情報処理プログラム |
JP6987534B2 (ja) * | 2017-05-26 | 2022-01-05 | 京セラ株式会社 | 測定装置、測定器具及び測定システム |
AU2018431742A1 (en) * | 2018-07-12 | 2021-03-11 | Richter Gedeon Nyrt. | Vaginal temperature sensing apparatus and methods |
JP2024515933A (ja) * | 2021-03-12 | 2024-04-11 | オーラ ヘルス オサケユキチュア | 月経周期追跡 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS63191345A (ja) * | 1987-02-04 | 1988-08-08 | Fujitsu Ten Ltd | 磁気テ−プの無音部の検出装置 |
JP2000230866A (ja) * | 1999-02-10 | 2000-08-22 | Matsushita Electric Ind Co Ltd | 婦人体温計 |
JP2002063360A (ja) * | 2000-08-23 | 2002-02-28 | Kentex Kk | 投資情報提供装置及び方法並びにシステム |
JP2014176599A (ja) * | 2013-02-12 | 2014-09-25 | Qol Kk | ホルモンバランス推定装置およびホルモンバランス推定方法 |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4151831A (en) * | 1976-11-15 | 1979-05-01 | Safetime Monitors, Inc. | Fertility indicator |
DE3474076D1 (en) * | 1983-02-24 | 1988-10-20 | Bioself Int Inc | Apparatus indicating the present fertility conditions of a person |
JP5179799B2 (ja) | 2007-08-16 | 2013-04-10 | テルモ株式会社 | 婦人体温計及び生理周期予測プログラム |
GB201311580D0 (en) * | 2013-06-27 | 2013-08-14 | Fertility Focus Ltd | Data analysis system and method |
AU2015234868A1 (en) * | 2014-03-28 | 2016-10-20 | Mti Ltd. | Ovulation day prediction program and ovulation day prediction method |
-
2015
- 2015-02-20 WO PCT/JP2015/054790 patent/WO2016132529A1/ja active Application Filing
- 2015-02-20 JP JP2015520745A patent/JP5775242B1/ja active Active
- 2015-02-20 US US15/552,311 patent/US10485487B2/en active Active
- 2015-07-14 TW TW104122785A patent/TWI540997B/zh active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS63191345A (ja) * | 1987-02-04 | 1988-08-08 | Fujitsu Ten Ltd | 磁気テ−プの無音部の検出装置 |
JP2000230866A (ja) * | 1999-02-10 | 2000-08-22 | Matsushita Electric Ind Co Ltd | 婦人体温計 |
JP2002063360A (ja) * | 2000-08-23 | 2002-02-28 | Kentex Kk | 投資情報提供装置及び方法並びにシステム |
JP2014176599A (ja) * | 2013-02-12 | 2014-09-25 | Qol Kk | ホルモンバランス推定装置およびホルモンバランス推定方法 |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6259147B1 (ja) * | 2017-05-16 | 2018-01-10 | 佐々木 修 | 体調予測システム |
WO2018211768A1 (ja) * | 2017-05-16 | 2018-11-22 | 佐々木 修 | 体調予測システム |
JP2018195000A (ja) * | 2017-05-16 | 2018-12-06 | 佐々木 修 | 体調予測システム |
JP2020008922A (ja) * | 2018-07-03 | 2020-01-16 | グラフテック株式会社 | 計測データ管理装置、計測データ管理方法、および計測データ管理プログラム |
JP2020014665A (ja) * | 2018-07-25 | 2020-01-30 | 日本光電工業株式会社 | 生体情報表示装置、および生体情報表示用データの出力方法 |
Also Published As
Publication number | Publication date |
---|---|
JP5775242B1 (ja) | 2015-09-09 |
US10485487B2 (en) | 2019-11-26 |
TWI540997B (zh) | 2016-07-11 |
US20180035954A1 (en) | 2018-02-08 |
TW201609042A (zh) | 2016-03-16 |
JPWO2016132529A1 (ja) | 2017-04-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5775242B1 (ja) | 情報処理装置、情報処理方法及び情報処理プログラム | |
JP5508610B1 (ja) | 情報処理装置、情報処理方法及び情報処理プログラム | |
US11004551B2 (en) | Sleep improvement system, and sleep improvement method using said system | |
US10983671B2 (en) | Terminal device, information processing method, and information processing program | |
CN105228531B (zh) | 月经开始预定日计算装置和月经开始预定日计算方法 | |
CN106874416A (zh) | 排行榜单生成方法及排行榜单生成装置 | |
JP5775243B1 (ja) | 情報処理装置、情報処理方法及び情報処理プログラム | |
CN113228196B (zh) | 用于优化用于计算睡眠得分的睡眠相关参数的系统和方法 | |
WO2017038177A1 (ja) | 情報提供装置、端末装置、情報提供方法、プログラム | |
US20090144069A1 (en) | System and method for providing personalized service | |
CN108784748A (zh) | 一种预测排卵日期的方法、装置和电子设备 | |
WO2021167081A1 (ja) | 睡眠評価方法、睡眠評価システム、端末装置、及び、プログラム | |
US20150250457A1 (en) | Ovulation day estimation apparatus, ovulation day estimation method and storage medium in which ovulation day estimation program is recorded | |
WO2022045082A1 (ja) | 情報処理方法、情報処理装置及び情報処理プログラム | |
KR20130103086A (ko) | 대기시간을 포인트화하는 시스템 및 방법 | |
JP6571046B2 (ja) | サーバ装置、情報処理方法及びプログラム | |
JP2011054101A (ja) | 評判検索サーバ、評判検索システム及び評判検索方法 | |
CN108282760B (zh) | 一种基于用户移动特征规律的d2d设备发现方法及装置 | |
WO2024210097A1 (ja) | レコメンド配信サーバ | |
CN117076916A (zh) | 模型选择方法、装置及网络侧设备 | |
CN117241095A (zh) | 满意下滑模型的训练方法、装置及电子设备 | |
JP6572116B2 (ja) | 繁忙情報提供方法、繁忙情報提供装置、及びプログラム | |
CN116932869A (zh) | 信息推荐方法及装置 | |
Ag Daud | Least congested channel recommendation for uncoordinated WiFi access point | |
CN116776989A (zh) | 模型的准确度确定方法、装置及网络侧设备 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
ENP | Entry into the national phase |
Ref document number: 2015520745 Country of ref document: JP Kind code of ref document: A |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 15882631 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15552311 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 15882631 Country of ref document: EP Kind code of ref document: A1 |