JP7774305B2 - Method and device for predicting livestock parturition time - Google Patents

Description

The present invention relates to a method for predicting the timing of parturition in livestock and a device for predicting the timing of parturition in livestock.

In recent years, with the decline in the number of livestock workers and the increase in the number of livestock per farm household, the number of livestock workers per farm is decreasing. When livestock give birth, livestock workers often need to be present and provide support. Because the timing of birth cannot be predicted, observation may be required day and night for several days. However, in a situation where the number of livestock workers per livestock is decreasing, this places a heavy burden on farms.
As a means for solving such problems, a method for predicting the time of parturition using information on the heart rate of livestock as an index has been used (Patent Document 1 and Non-Patent Document 1). Patent Document 1 describes that a transient decrease in heart rate occurs that is specific only to parturition. Non-Patent Document 1 also describes that a high percentage of livestock have a heart rate ratio of 1.3 or higher within two days of parturition compared to the heart rate 30 days before parturition.

JP 2008-11916 A

Fujimoto et al., Bulletin of the Japanese Society of Animal Science, Vol. 59, No. 4, pp. 301-305, 1988

However, the above-mentioned conventional technology has the problem that it can only predict parturition immediately before it occurs because it detects a transient decrease in heart rate that is unique only to parturition. Furthermore, since parturition is predicted based on the heart rate ratio compared to the heart rate 30 days before parturition, it is necessary to determine approximately 30 days before parturition, but this determination may be difficult due to individual differences. If the time of parturition could be predicted at the appropriate time, it would be possible to prepare a sufficient support system at the time of parturition, which would reduce the burden on livestock workers and be useful.
One aspect of the present invention has been made to solve the above-mentioned problems, and an object of the present invention is to realize a technology for predicting the time of parturition in livestock.

In order to solve the above-mentioned problems, one embodiment of the present invention provides a method for predicting the timing of parturition in livestock, which includes an analysis step of decomposing electrocardiogram data representing the temporal progression of heart rate information over a predetermined period obtained from an electrocardiogram measured in a test livestock into a diurnal fluctuation component, a trend component, and a residual component, and a prediction step of predicting the timing of parturition in the test livestock based on the inflection point of the trend component obtained in the analysis step.

In one embodiment of the method for predicting the timing of parturition in livestock, the prediction step may predict the parturition time of the test livestock to be 30 to 80 hours after the inflection point of the trend component.

In a method for predicting the timing of livestock parturition according to one aspect of the present invention, the inflection point of the trend component may be the point at which the difference between the trend component and the moving average of the trend component reaches a minimum value greater than or equal to 0.

In one aspect of the method for predicting the timing of parturition in livestock, the specified period may be one week or more before the expected date of parturition of the test livestock.

In one aspect of the method for predicting the timing of livestock parturition according to the present invention, the moving average may be the average value for the 6 to 18 hours immediately preceding the calculation time.

A method for predicting the timing of parturition in livestock according to another aspect of the present invention includes an analysis step of comparing a first average value of heart rate information obtained from an electrocardiogram measured in a test livestock over a predetermined time period with a second average value of the heart rate information over the predetermined time period on at least one of the previous day and the day before that, and a prediction step of predicting the timing of parturition in the test livestock based on the results of the comparison in the analysis step.

In another aspect of the method for predicting the timing of livestock parturition according to the present invention, the specified time span may be either a continuous period of 4 to 5 hours between 1:00 and 8:00, or a continuous period of 1 to 3 hours between 1:00 and 5:00.

In another aspect of the method for predicting the time of parturition in livestock according to the present invention, when the first average value is lower than the second average value in the prediction step, the parturition time of the test livestock may be predicted to be within 20 to 30 hours from the date of calculation of the first average value.

In another aspect of the method for predicting the time of parturition in livestock, when the first average value is lower than the second average value by 5% or more in the prediction step, the parturition time of the test livestock may be predicted to be within 20 to 30 hours from the date of calculation of the first average value.

A livestock parturition timing prediction device according to one aspect of the present invention includes an analysis unit that decomposes electrocardiogram data representing the temporal progression of heart rate information over a predetermined period obtained from an electrocardiogram measured on a test livestock into a diurnal fluctuation component, a trend component, and a residual component, and a prediction unit that predicts the parturition timing of the test livestock based on the inflection point of the trend component obtained by the analysis unit.

Another aspect of the present invention provides a livestock parturition timing prediction device that includes an analysis unit that compares a first average value of heart rate information obtained from an electrocardiogram measured on a test livestock over a predetermined time span with a second average value of the heart rate information over the predetermined time span on at least one of the previous day and the day before that, and a prediction unit that predicts the parturition timing of the test livestock based on the results of the comparison made by the analysis unit.

One aspect of the present invention provides technology for predicting the timing of livestock parturition.

1 is a block diagram showing the configuration of a main part of a delivery timing prediction device according to one aspect of the present invention. FIG. 1 is a diagram illustrating an outline of trend analysis. 10 is a flowchart illustrating an example of a delivery timing prediction process executed by a delivery timing prediction device according to an aspect of the present invention. 10 is a flowchart illustrating another example of the delivery timing prediction process executed by the delivery timing prediction device according to an aspect of the present invention. FIG. 10 is a diagram showing the results of trend analysis in an example. FIG. 10 is a diagram showing the results of trend analysis in an example. FIG. 10 is a diagram showing the extraction results of inflection points in an example. FIG. 10 is a diagram showing the extraction results of inflection points in an example. FIG. 10 is a diagram showing the extraction results of inflection points in an example. FIG. 10 is a diagram showing the extraction results of inflection points in an example. FIG. 10 is a diagram showing the results of parturition time prediction in an example. FIG. 10 is a diagram showing the results of parturition time prediction in an example. FIG. 10 is a diagram showing the results of parturition time prediction in an example. FIG. 10 is a diagram showing the results of parturition time prediction in an example.

[Method for predicting delivery time]
One aspect of the present invention provides a calving timing prediction method for predicting the calving timing of livestock. In the calving timing prediction method according to one aspect of the present invention, examples of livestock for which the calving timing is predicted include cows, horses, pigs, goats, sheep, etc. In the calving timing prediction method, one example of the livestock for which the calving timing is predicted is cows.

A method for predicting the timing of parturition in livestock according to one aspect of the present invention predicts the timing of parturition in livestock using heartbeat information of the livestock as an index. Here, heartbeat information is information that represents the pulsation, which is the electrical activity of the heart. Examples include the heart rate and heartbeat interval obtained from an electrocardiogram, which represents pulsation as a waveform signal, as well as the pulse wave, pulse rate, and pulse interval, which represent changes in the flow rate in peripheral blood vessels transmitted in response to pulsation. In this embodiment, we will mainly explain the prediction of the timing of parturition in livestock using the RR interval, which represents the interval between heartbeats (R waves), out of the heartbeat information obtained from an electrocardiogram.

A method for predicting the timing of parturition in livestock according to one embodiment of the present invention includes an analysis step of decomposing electrocardiogram data representing the temporal progression of heart rate information over a predetermined period obtained from an electrocardiogram measured in a test livestock into a diurnal fluctuation component, a trend component, and a residual component, and a prediction step of predicting the timing of parturition in the test livestock based on the inflection point of the trend component obtained in the analysis step.

One example of a livestock calving timing prediction method according to one aspect of the present invention is a livestock calving timing prediction process executed in the livestock calving timing prediction device 10 shown in Figure 1. Therefore, in this embodiment, a livestock calving timing prediction method according to one aspect of the present invention will be described as a livestock calving timing prediction process executed in the livestock calving timing prediction device 10.

(Parturition timing prediction device 10)
Fig. 1 is a block diagram showing the main components of a delivery timing prediction device 10 according to one embodiment of the present invention. As shown in Fig. 1, the delivery timing prediction device 10 is connected to a measurement device 20, a storage device 30, and an output device 40 so as to be able to send and receive data.

The measuring device 20 measures the electrocardiogram of livestock. One example of the measuring device 20 is a wearable device attached to the body of the livestock. One example of the measuring device 20 is equipped with multiple electrodes, and measures the electrocardiogram by detecting the current flowing between the electrodes attached to the livestock. Another example of the measuring device 20 is equipped with a light source (LED or laser) and a photodiode, and measures a pulse wave from which an RR interval approximating the electrocardiogram can be obtained by detecting light emitted from the light source and reflected from the blood vessels of the livestock using the photodiode. The RR interval is the interval between heartbeats (R waves) on an electrocardiogram, and is a value corresponding to the heart rate within a certain period of time.

The measuring device 20 outputs the measured electrocardiogram to the storage device 30 and stores the heart rate information obtained from electrocardiograms measured continuously within a predetermined period in chronological order. This makes it possible to obtain changes in heart rate information over time. The measuring device 20 may also output the measured electrocardiogram to the delivery timing prediction device 10. The measuring device 20 may be configured to measure the electrocardiogram continuously, or may be configured to measure it periodically at predetermined intervals.

The storage device 30 stores the programs and data used by the delivery timing prediction device 10. The storage device 30 stores electrocardiograms measured by the measurement device 20. As an example, the storage device 30 stores, in chronological order, heart rate information obtained from electrocardiograms measured continuously within a predetermined period. The storage device 30 may have a database on the cloud or a server that stores electrocardiograms and heart rate information.

The storage device 30 preferably stores heart rate information obtained from the electrocardiogram measured by the measurement device 20 in association with the time of measurement. The heart rate information may be waveform data representing the electrocardiogram over a predetermined period of time.

The output device 40 outputs information regarding the delivery timing predicted by the delivery timing prediction device 10. The manner of output by the output device 40 is not particularly limited. The output device 40 may be, for example, a display device that displays the information as an image, a printing device that prints the information, or an alarm device that outputs the information as sound.

The delivery timing prediction device 10 includes a control unit 11. The control unit 11 controls all of the components of the delivery timing prediction device 10, and is realized, for example, by a processor and memory. In this example, the processor accesses storage (not shown), loads a program (not shown) stored in the storage into memory, and executes a series of instructions contained in the program. This configures the components of the control unit 11. As these components, the control unit 11 includes a data acquisition unit 12, a data conversion unit 13, an analysis unit 14, and a prediction unit 15.

The data acquisition unit 12 acquires heart rate information obtained from an electrocardiogram measured by the measurement device 20. The data acquisition unit 12 acquires the heart rate information based on an input signal representing an instruction to start prediction from an input device (not shown). As an example, the data acquisition unit 12 may read the heart rate information from the storage device 30 or acquire the heart rate information directly from the measurement device 20. The data acquisition unit 12 outputs the acquired heart rate information to the data conversion unit 13.

The data conversion unit 13 acquires heartbeat information and converts it into data representing the RR interval, which is the interval between heartbeats (R waves). As an example, the data conversion unit 13 acquires waveform data representing an electrocardiogram over a predetermined period of time, and generates RR interval data (electrocardiogram data) from the waveform data that represents the changes in the RR interval over time. The data conversion unit outputs the generated RR interval data to the analysis unit 14.

Here, the waveform data may be waveform data representing the electrocardiogram of the test livestock one week or more before the expected calving date, and more preferably waveform data representing the electrocardiogram of the test livestock two weeks or more before the expected calving date. One example of the waveform data is waveform data representing electrocardiograms measured continuously from one week before the expected calving date until the expected calving date, more preferably from two weeks before the expected calving date until the expected calving date. The expected calving date is set based on the insemination date of the livestock, and, for example, in the case of dairy cows, may be 280 days after the insemination date.

The analysis unit 14 decomposes the RR interval data, which represents the temporal changes over a predetermined period of time in the RR interval, which is the interval between heartbeats (R waves) obtained from an electrocardiogram measured in the test livestock, into a diurnal fluctuation component, a trend component, and a residual component. As shown in Figure 2, the analysis unit 14 performs trend analysis on the RR interval data to decompose the RR interval data into a diurnal fluctuation component, a trend component, and a residual component. Figure 2 is a diagram illustrating an overview of trend analysis. The diurnal fluctuation component is a data component that represents the daily fluctuation of the RR interval. The trend component is a data component that represents the general tendency of the temporal changes in the RR interval. The residual component is a data component that represents a numerical difference that is neither a diurnal fluctuation nor a trend. The analysis unit 14 outputs the trend component to the prediction unit 15.

The trend analysis of the RR interval data in the analysis unit 14 can be performed using a conventionally known trend analysis algorithm. One example of a trend analysis algorithm is STL (Seasonal Decomposition of Time Series by Loess) decomposition (R. B. Cleveland et al., Journal of Official Statistics, 6, 3-73, 1990).

The prediction unit 15 predicts the time of delivery of the test livestock based on the inflection points of the trend component. The prediction unit 15 extracts the inflection points of the trend component. As an example, the inflection points of the trend component may be transient peak values or inflection points seen in a trend graph representing the trend component. The method for extracting the inflection points of the trend component in the prediction unit 15 is not particularly limited, and the inflection points may be extracted by visually checking the trend graph. Examples of inflection points extracted in the trend graph are shown in Figures 5 and 6, which show the results of the examples described below.

The prediction unit 15 may also extract, as an inflection point of the trend component, the point at which the difference between the trend component and the moving average of the trend component becomes a minimum value greater than or equal to 0. The prediction unit 15 extracts, as an inflection point, the point at which the difference between the trend component and the moving average becomes 0 or a minimum value greater than or equal to 0.

The moving average used to extract the inflection point of the trend component may be calculated from the measurement values of the previous day, or may be calculated from the measurement values of a specified time immediately prior to the calculation point. As an example, the moving average is the average value for the 6 to 18 hours immediately prior to the calculation point. Furthermore, the moving average is preferably the average value for the 8 to 15 hours immediately prior to the calculation point, and more preferably the average value for the 10 to 13 hours immediately prior to the calculation point. For example, it is the average value for the 12 hours immediately prior to the calculation point.

The calculation of the moving average in the prediction unit 15 can be performed using a conventionally known moving average calculation algorithm. Examples of moving average calculation algorithms include DPO (Detrended Price Oscillator) and MAD (Moving Average Deviation Rate). The prediction unit 15 can modify these moving average calculation algorithms to calculate the moving average using measured values from a predetermined time immediately prior to the calculation time, and use this to calculate the moving average. Examples of inflection points extracted based on the difference between the trend component and the moving average of the trend component are shown in Figures 7 to 10, which illustrate the results of the examples described below.

The prediction unit 15 predicts the parturition time of the test livestock to be 30 to 80 hours after the inflection point of the trend component. As an example, the prediction unit 15 predicts the parturition time of the test livestock to be 35 to 75 hours, preferably 36 to 72 hours, after the inflection point of the trend component. In other words, the prediction unit 15 predicts the parturition time of the test livestock to be 2 to 3 days after the inflection point of the trend component. The prediction unit 15 outputs information indicating the predicted parturition time of the test livestock to the output device 40. The prediction unit 15 may store the information indicating the predicted parturition time of the test livestock in the storage device 30.

(Flow of delivery prediction process)
FIG. 3 is a flowchart illustrating an example of a delivery timing prediction process executed by a delivery timing prediction device according to an aspect of the present invention.

First, the data acquisition unit 12 acquires heart rate information obtained from an electrocardiogram measured by the measurement device 20 (step S11). Next, the data conversion unit 13 acquires the heart rate information and generates RR interval data representing the temporal progression of the RR interval (step S12). The analysis unit 14 then performs trend analysis on the RR interval data and decomposes it into a diurnal fluctuation component, a trend component, and a residual component (step S13). The prediction unit 15 extracts the inflection point of the trend component and predicts the time of delivery of the test livestock (step S14), thereby completing the delivery time prediction process.

The livestock parturition timing prediction method and parturition timing prediction device according to one aspect of the present invention makes it possible to predict the parturition timing of a livestock subject based on the electrocardiogram of the subject livestock. Furthermore, the livestock parturition timing prediction method and parturition timing prediction device according to one aspect of the present invention predicts the parturition timing based on electrocardiograms taken a predetermined period before the expected parturition date, thereby shortening the observation period for the subject livestock. Furthermore, the livestock parturition timing prediction method and parturition timing prediction device according to one aspect of the present invention makes it possible to predict the expected parturition timing two to three days before parturition, ensuring sufficient time to prepare for parturition.

When livestock give birth, livestock workers often need to be present and provide support. If the time of birth cannot be predicted, it may be necessary to observe the animal day and night for several days. Furthermore, if the time of birth is predicted at the last minute, there may not be enough time to prepare before the birth and sufficient support may not be possible. According to a method and device for predicting the time of livestock birth according to one aspect of the invention, the time of birth can be predicted several days in advance, eliminating the need for long-term observation and ensuring sufficient time for preparation.

(Modification of the method for predicting the time of delivery)
A method for predicting the time of parturition in livestock according to one embodiment of the present invention may predict the time of parturition in livestock by analyzing an electrocardiogram as described below.

A modified example of a method for predicting the timing of parturition in livestock according to one aspect of the present invention includes an analysis step of comparing a first average value of heart rate information obtained from an electrocardiogram measured in a test livestock over a predetermined time period with a second average value of the heart rate information over the predetermined time period on at least one of the previous day and the day before that, and a prediction step of predicting the timing of parturition in the test livestock based on the results of the comparison in the analysis step.

An example of a livestock parturition timing prediction method according to an aspect of the present invention is a modified version of the livestock parturition timing prediction process executed in the livestock parturition timing prediction device 10 shown in FIG. 1. Therefore, in this embodiment, a modified version of the livestock parturition timing prediction method according to an aspect of the present invention will be described as a modified version of the livestock parturition timing prediction process executed in the livestock parturition timing prediction device 10. Note that in this modified version, only the differences from the livestock parturition timing prediction method and parturition timing prediction device according to an aspect of the present invention described above will be described, and a description of the common points will be omitted.

In a modified example, the analysis unit 14 compares a first average value of RR intervals obtained from an electrocardiogram measured in the test livestock over a predetermined time span with a second average value of the RR intervals over the predetermined time span on at least one of the previous day and the day before that. The analysis unit 14 calculates the first average value over the predetermined time span using the RR interval data generated by the data conversion unit 13. Then, the analysis unit 14 calculates a second average value over the same predetermined time span as the first average value calculated on at least one of the day before and the day before that on which the measurement values used to calculate the first average value were measured.

As an example, the specified time span is either a four to five consecutive hour period between 1:00 and 8:00, or a one to three consecutive hour period between 1:00 and 17:00. The analysis unit 14 calculates the average value of the RR intervals over a four to five consecutive hour period between 1:00 and 8:00 on a certain day as the first average value, and calculates the average value of the RR intervals over a four to five consecutive hour period between 1:00 and 8:00 on the previous day or the day before that as the second average value. As an example, the specified time span is 2:00 to 7:30, preferably 2:30 to 7:00.

As an example, the analysis unit 14 compares the first average value with the second average value and extracts points where the first average value decreases. As an example, the analysis unit 14 extracts points where the first average value decreases by a predetermined percentage relative to the second average value. Here, the predetermined percentage may be 5% or more, 8% or more, 10% or more, 14% or more, 16% or more, or 20% or more. The analysis unit 14 outputs the comparison result between the first average value and the second average value to the prediction unit 15.

When the first average value is lower than the second average value, the prediction unit 15 predicts that the test livestock's parturition time will be 20 to 30 hours after the calculation date of the first average value. As an example, the prediction unit 15 predicts that the test livestock's parturition time will be within 24 hours after the calculation date of the first average value. In other words, the prediction unit 15 predicts that the test livestock's parturition time will be the day after the calculation date of the first average value. The prediction unit 15 outputs information indicating the predicted parturition time of the test livestock to the output device 40. The prediction unit 15 may store the information indicating the predicted parturition time of the test livestock in the storage device 30.

As an example, when the first average value is 5% or more lower than the second average value, the prediction unit 15 predicts that the test livestock will calve within 20 to 30 hours from the date on which the first average value was calculated. As an example, when the first average value is lower than the second average value by 5% or more, 8% or more, 10% or more, 14% or more, 16% or more, or 20% or more, the prediction unit 15 predicts that the test livestock will calve within 20 to 30 hours from the date on which the first average value was calculated.

As an example, when the first average value from 2:30 to 7:00 is 10% or more lower than the second average value from 2:30 to 7:00 on the previous day, the prediction unit 15 predicts that the delivery period will be within 24 hours. Also, as an example, when the first average value from 2:30 to 7:00 is 8% or more lower than the second average value from 2:30 to 7:00 on the day before that, the prediction unit 15 predicts that the delivery period will be within 24 hours. Note that if the second average value is calculated from measurements taken on the day before that, and the first average value is lower than the second average value for two consecutive days, the day before that and the day before that, and the first average value is 8% or less lower than the second average value on the day before that, the delivery period may be predicted to be within 24 hours.

Furthermore, as an example, when the first average value from 2:00 PM to 4:00 PM is lower than the second average value from 2:00 PM to 4:00 PM on the day before that by 14% or more, the prediction unit 15 predicts that the delivery time will be within 24 hours.

The prediction unit 15 outputs information representing the predicted parturition time of the test livestock to the output device 40. The prediction unit 15 may store the information representing the predicted parturition time of the test livestock in the storage device 30.

(Flow of modified example of delivery time prediction process)
FIG. 4 is a flowchart illustrating another example of the delivery timing prediction process executed by the delivery timing prediction device according to one aspect of the present invention.

First, the data acquisition unit 12 acquires heart rate information obtained from an electrocardiogram measured by the measurement device 20 (step S21). Next, the data conversion unit 13 acquires the heart rate information and generates RR interval data that represents the time progression of the RR interval, which is the interval between heart rates (R waves) (step S22). Then, the analysis unit 14 compares a first average value of the RR interval data over a predetermined time width with a second average value over a predetermined time width on the previous day or the day before that (step S23). The prediction unit 15 predicts the time of parturition of the test livestock based on the prediction results of the first average value and the second average value (step S24), and the parturition time prediction process ends.

[Software implementation example]
The functions of the delivery timing prediction device 10 (hereinafter referred to as the "device") can be realized by a program that causes a computer to function as the device, and a program that causes a computer to function as each control block of the device (particularly each part included in the control unit 11).

In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory) as hardware for executing the program. The functions described in each of the above embodiments are realized by executing the program using this control device and storage device.

The above program may be stored non-transitory on one or more computer-readable storage media. These storage media may or may not be included in the device. In the latter case, the program may be supplied to the device via any wired or wireless transmission medium.

Furthermore, some or all of the functions of each of the above control blocks can be realized by logic circuits. For example, integrated circuits incorporating logic circuits that function as each of the above control blocks are also included in the scope of the present invention. In addition, the functions of each of the above control blocks can also be realized by, for example, a quantum computer.

Furthermore, each process described in each of the above embodiments may be executed by AI (Artificial Intelligence). In this case, the AI may run on the control device, or on another device (for example, an edge computer or a cloud server).

The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention.

Trend analysis was performed using RR interval data obtained by measuring the heart rates of livestock. Holter electrocardiogram monitors were attached to the bodies of seven dairy cows starting one day or two days before heart rate measurements began, and heart rate measurements began seven days before the expected calving date of 280 days after conception, until the day of calving. As a general rule, feeding was carried out between 9:30 and 10:00 and 15:30 and 16:00 each day. Before the morning feeding, the electrocardiogram monitors were reattached and batteries were replaced as necessary, and recording resumed from 10:00. Electrocardiogram data was collected until calving.

The RR intervals obtained from the data acquired by the electrocardiograph were calculated as averages every 30 minutes, counting backward from the time of delivery. These 30-minute averages were used to perform a time series analysis and calculate averages by time period. Time series analysis was performed on six dairy cows with few missing values, using the calculation software R. Hourly analysis was performed on seven dairy cows, using Microsoft Excel. The times shown in the examples include an error of less than ±15 minutes. For example, when calculating the average value from 2:30 to 7:00, this also includes periods such as 2:35 to 7:05.

The results of the time series analysis are shown in Figures 5 and 6. Figures 5 and 6 show the results of the trend analysis of the example. Figures 5 and 6 show graphs of the analysis results for each dairy cow, and from top to bottom, the graphs show the RR interval data, diurnal fluctuation component, trend component, and residual component. In Figures 5 and 6, as indicated by the arrows on the trend component graphs, an inflection point or transient peak value was observed in the trend component, and parturition began 36 to 72 hours after that.

Figures 7 to 10 show the results of extracting inflection points of the trend component based on the difference between the trend component and its moving average. Figures 7 to 10 show the results of extracting inflection points in the example. Figures 7 to 10 show graphs of the analysis results for each dairy cow, with the upper graph showing the trend component of the RR interval data and its moving average, and the lower graph showing the difference between the trend component and the moving average. Figures 7 and 8 show the results using the moving average calculated by DPO, with the points indicated by arrows where the difference is the smallest value exceeding 0 being the inflection points. Figures 9 and 10 show the results using the moving average calculated by MAD, with the points indicated by circles where the difference is the smallest value exceeding 0 being the inflection points. As shown in Figures 7 to 10, inflection points of the trend component were extracted based on the difference between the trend component and the moving average, and calving began 36 to 72 hours after that.

The results of the analysis by time period are shown in Figures 11 and 12. Figures 11 and 12 are diagrams showing the results of predicting the timing of delivery using the analysis by time period in the example. As shown in Figures 11 and 12, the RR interval within 24 hours tends to decrease, but to examine the trend in more detail, further analysis was performed by narrowing down to specific time periods. The average RR interval from around 2:30 to around 7:00 is shown in Figure 13, and the average RR interval from around 14:00 to around 16:00 is shown in Figure 14.

As shown in Figures 13 and 14, when the average RR interval between approximately 2:30 and 7:00 decreased by 10% or more compared to the previous day, delivery occurred within 24 hours. Also, when the average RR interval between approximately 2:30 and 7:00 decreased for two consecutive days and by 8% or more compared to the day before last, delivery occurred within 24 hours. Furthermore, as shown in Figure 14, when the average RR interval between approximately 2:00 and 4:00 PM decreased by 14% or more compared to the day before last, delivery occurred within 24 hours.

10 Delivery time prediction device 14 Analysis unit 15 Prediction unit

Claims (7)

an analysis step of decomposing electrocardiogram data representing time transitions of heart rate information obtained from electrocardiograms measured on test livestock over a predetermined period into a diurnal fluctuation component, a trend component, and a residual component;
A prediction step of predicting the delivery time of the test livestock based on the inflection point of the trend component obtained in the analysis step;
It encompasses
The inflection point of the trend component is a time point at which the difference between the trend component and the moving average of the trend component becomes a minimum value equal to or greater than 0.
A method for predicting the time of calving in livestock. an analysis step of decomposing electrocardiogram data representing a time transition of heart rate information obtained from an electrocardiogram measured on a test livestock over a predetermined period into a diurnal fluctuation component, a trend component, and a residual component;
A prediction step of predicting the delivery time of the test livestock based on the inflection point of the trend component obtained in the analysis step;
It encompasses
The inflection point of the trend component is a transient peak value or inflection point seen in a trend graph representing the trend component,
An inflection point seen in the trend graph is a point at which the trend graph changes from upward to downward.
A method for predicting the time of calving in livestock. 3. The method for predicting the time of parturition of livestock according to claim 1 or 2 , wherein in the prediction step, the parturition time of the test livestock is predicted to be 30 to 80 hours after the inflection point of the trend component. The method for predicting the timing of livestock parturition described in claim 1 or 2, wherein the specified period is at least one week before the expected parturition date of the test livestock. 2. The method for predicting the time of livestock parturition described in claim 1 , wherein the moving average is an average value for the 6 to 18 hours immediately prior to the calculation time. an analysis unit that decomposes electrocardiogram data representing time transitions of heart rate information obtained from electrocardiograms measured on test livestock over a predetermined period into a diurnal fluctuation component, a trend component, and a residual component;
A prediction unit that predicts the delivery time of the test livestock based on the inflection point of the trend component obtained in the analysis unit;
Equipped with
The inflection point of the trend component is a time point at which the difference between the trend component and the moving average of the trend component becomes a minimum value equal to or greater than 0.
A device for predicting livestock birth timing. an analysis unit that decomposes electrocardiogram data representing time transitions of heart rate information obtained from electrocardiograms measured on test livestock over a predetermined period into a diurnal fluctuation component, a trend component, and a residual component;
A prediction unit that predicts the delivery time of the test livestock based on the inflection point of the trend component obtained in the analysis unit;
Equipped with
The inflection point of the trend component is a transient peak value or inflection point seen in a trend graph representing the trend component,
An inflection point seen in the trend graph is a point at which the trend graph changes from upward to downward.
A device for predicting livestock birth timing.