Information processing device, monitoring device, and computer program

Abstract

This method suppresses the decrease in the accuracy of estimating the respiratory rate of subjects without requiring the addition of sensors. [Solution] The input interface 111 receives a waveform signal WF corresponding to the waveform of biological information with periodic changes from a sensor 30 attached to the subject 20. The processor 112 visualizes information corresponding to the estimated respiratory rate of the subject 20 on the visualization device 12 based on the waveform signal WF. The processor 112 extracts multiple feature quantities corresponding to the respiratory rate of the subject 20 from the waveform signal WF contained in each of multiple time intervals. The processor 112 obtains multiple preliminary estimates of the respiratory rate by applying frequency analysis processing to each of the multiple feature quantities. The processor 112 determines the estimated value based on the mode of the multiple preliminary estimates.

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus that processes biometric information of a subject. The present disclosure also relates to a monitoring apparatus that monitors biometric information of a subject. The present disclosure also relates to a computer program executable by a processor mounted on the information processing apparatus.

Background Art

[0002] There is known a technique of acquiring a pulse wave waveform, which is an example of a biometric information waveform with periodic changes, from a subject and estimating the respiratory rate of the subject based on the pulse wave waveform. Patent Document 1 discloses a technique of improving the estimation accuracy of the respiratory rate by detecting the body movement of a subject using an acceleration sensor in combination. Patent Document 2 discloses a technique of improving the estimation accuracy of the respiratory rate by increasing the number of sensors for acquiring the pulse wave waveform.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] There is a demand for suppressing a decrease in the estimation accuracy of the respiratory rate of a subject without adding sensors.

Means for Solving the Problems

[0005] One example of an aspect that can be provided by the present disclosure is an information processing apparatus that processes biometric information of a subject, an interface that receives a signal corresponding to a biometric information waveform with periodic changes from a sensor attached to the subject, A processor that causes an output device to output information corresponding to an estimated value of the subject's respiratory rate based on the aforementioned signal, It is equipped with, The aforementioned processor, Multiple feature quantities corresponding to the subject's respiration are extracted from the signals contained in each of the multiple time intervals, By applying frequency analysis processing to each of the aforementioned multiple feature quantities, multiple preliminary estimates of the respiratory rate are obtained. The estimated value is determined based on the mode of the aforementioned multiple preliminary estimates.

[0006] One example of the embodiments that may be provided by this disclosure is a monitoring device for monitoring the biological information of a subject, An information processing device according to the above example embodiment, An output device that outputs information corresponding to the estimated respiratory rate of the subject obtained by the information processing device, It is equipped with.

[0007] One example of the embodiments that may be provided by this disclosure is a computer program that can be executed by a processor installed in an information processing device that processes the biological information of a subject, By being executed, the information processing device will The system receives signals corresponding to biometric waveforms that exhibit periodic changes from sensors attached to the subject. Multiple feature quantities corresponding to the subject's respiration are extracted from the signals contained in each of the multiple time intervals, By applying frequency analysis processing to each of the aforementioned multiple features, multiple preliminary estimates of the subject's respiratory rate are obtained. The estimated respiratory rate is determined based on the mode of the aforementioned multiple preliminary estimates. The output device outputs information corresponding to the aforementioned estimated value.

[0008] The inventor of the present application noted that in a configuration where an estimated value of the respiration rate of a subject is determined based on a plurality of preliminary estimated values, some preliminary estimated values obtained with low accuracy for some reason may affect the final estimated result of the respiration rate. According to the configuration according to the above processing example, since the estimated value of the respiration rate of the subject is determined based on the most frequent value of the plurality of preliminary estimated values, the influence of some preliminary estimated values obtained with low accuracy can be suppressed. Since such processing is executed by a processor mounted on the information processing apparatus, it is possible to suppress a decrease in the estimation accuracy of the respiration rate of the subject without adding a sensor.

Brief Description of the Drawings

[0009] [Figure 1] An example of the functional configuration of a monitoring apparatus according to an embodiment is illustrated. [Figure 2] An example of the processing executed by the processor in FIG. 1 is shown. [Figure 3] The baseline fluctuation of the waveform signal in FIG. 1 is illustrated. [Figure 4] The amplitude fluctuation of the waveform signal in FIG. 1 is illustrated. [Figure 5] The frequency fluctuation of the waveform signal in FIG. 1 is illustrated. [Figure 6] Another example of the processing executed by the processor in FIG. 1 is shown. [Figure 7] Another example of the processing executed by the processor in FIG. 1 is shown. [Figure 8] Another example of the processing executed by the processor in FIG. 1 is shown. [Figure 9] Another example of the processing executed by the processor in FIG. 1 is shown. [Figure 10] Another example of the processing executed by the processor in FIG. 1 is shown.

Modes for Carrying Out the Invention

[0010] Examples of the embodiments will be described in detail while referring to the accompanying drawings.

[0011] FIG. 1 illustrates the functional configuration of a monitoring device 10 according to an exemplary embodiment. The monitoring device 10 is a device that monitors the biological information of a subject 20. The monitoring device 10 includes an information processing device 11 and a visualization device 12.

[0012] The information processing device 11 is a device that processes the biological information of the subject 20. Specifically, the information processing device 11 is configured to obtain information corresponding to an estimated respiration rate of the subject 20 based on a signal corresponding to the waveform of the pulse wave of the subject 20. The pulse wave is an example of biological information accompanied by periodic changes. The term "waveform" used in the present disclosure means the change over time of the value of specific biological information.

[0013] The visualization device 12 is configured to visualize information corresponding to the estimated respiration rate of the subject 20. As an example, the visualization device 12 may be a display device that displays the information. As another example, the visualization device 12 may be a printing device that prints the information on a medium. The visualization device 12 is an example of an output device.

[0014] The information processing device 11 includes an input interface 111. The input interface 111 is configured as a hardware interface that receives a waveform signal WF corresponding to the waveform of the pulse wave from a sensor 30 attached to the subject 20. The waveform signal WF may be an analog signal or a digital signal depending on the specifications of the sensor 30.

[0015] Examples of the sensor 30 for acquiring the pulse wave include a photoelectric pulse wave sensor including a light emitting element and a light receiving element. The light emitted from the light emitting element passes through the biological tissue of the subject 20 and enters the light receiving element. With the pulsation of the subject 20, the intensity of the incident light on the light receiving element changes. Therefore, the change over time of the incident light intensity may correspond to the waveform of the pulse wave.

[0016] If the waveform signal WF is an analog signal, the input interface 111 includes an appropriate conversion circuit, including an A / D converter. This description also applies to other signals that the input interface 111 can accept, as described later.

[0017] The information processing device 11 includes a processor 112. The processor 112 is configured to obtain an estimated respiratory rate of the subject 20 based on the waveform signal WF received by the input interface 111.

[0018] The information processing device 11 is equipped with an output interface 113. The processor 112 is configured to output a control signal CT from the output interface 113, which causes the visualization device 12 to visualize information corresponding to the estimated respiratory rate. The control signal CT may be an analog signal or a digital signal, depending on the specifications of the visualization device 12.

[0019] In other words, the output interface 113 is configured as a hardware interface capable of outputting a control signal CT. When the control signal CT is an analog signal, the output interface 113 includes an appropriate conversion circuit, including a D / A converter. This description also applies to other signals that the output interface 113 can output, as described later.

[0020] Referring to Figure 2, we will now explain the specific processing flow performed by the processor 112 to obtain an estimated respiratory rate for subject 20.

[0021] The pulse wave waveform corresponding to the waveform signal WF received by the input interface 111 has superimposed frequency components corresponding to pulse (0.5~2Hz), frequency components corresponding to respiration (0.1~1Hz), and frequency components of the Mayer wave (0.04~0.4Hz). The frequency components corresponding to respiration overlap with some of the other frequency components in terms of bandwidth, making it difficult to extract them by filtering alone.

[0022] On the other hand, the pulse wave waveform contains several features that reflect the effects of respiration. Figure 3 illustrates baseline fluctuations of the waveform, which are one example of features. Baseline fluctuations can be caused by changes in intrathoracic pressure, vasoconstriction of arteries during inspiration, etc. Figure 4 illustrates amplitude fluctuations of the waveform, which are another example of features. Amplitude fluctuations can be caused by changes in intrathoracic pressure, etc. Figure 5 illustrates frequency fluctuations of the waveform, which are yet another example of features. Frequency fluctuations can be caused by respiratory sinus arrhythmias (RSA), etc.

[0023] Therefore, as illustrated in Figure 2, the processor 112 performs processing to extract baseline fluctuations, amplitude fluctuations, and frequency fluctuations from the waveform signal WF.

[0024] Specifically, the processor 112 extracts the above-mentioned feature quantities from the waveform signal WF contained in the first time interval TS1. The length of the first time interval TS1 is, for example, 30 seconds.

[0025] Next, the processor 112 extracts the above-mentioned feature quantities from the waveform signal WF contained in the second time interval TS2. The length of the second time interval TS2 is equal to the length of the first time interval TS1. The starting point of the second time interval TS2 is later than the starting point of the first time interval TS1. Part of the second time interval TS2 may overlap with the first time interval TS1.

[0026] Processor 112 similarly performs feature extraction for n time intervals. In Figure 2, the nth time interval is represented by the code TSn.

[0027] Next, processor 112 performs a process to obtain a preliminary estimate of the respiratory rate based on each of the multiple features extracted for each of the n time intervals. Specifically, frequency analysis is applied to each feature to identify the frequency with the largest power spectral value. Examples of frequency analysis processes include fast Fourier transform and wavelet transform. Processor 112 then associates the identified frequency with a preliminary estimate of the respiratory rate (number of breaths per minute).

[0028] In this example, the estimated respiratory rate obtained based on baseline fluctuations extracted from the first time interval TS1 is referred to as the first preliminary estimate PE11. Similarly, the estimated respiratory rate obtained based on amplitude fluctuations extracted from the first time interval TS1 is referred to as the second preliminary estimate PE21, and the estimated respiratory rate obtained based on frequency fluctuations extracted from the first time interval TS1 is referred to as the third preliminary estimate PE31.

[0029] Although the feature vectors are not shown in Figure 2, the estimated respiratory rate obtained based on baseline fluctuations extracted from the second time interval TS2 is referred to as the first preliminary estimate PE12. Similarly, the estimated respiratory rate obtained based on amplitude fluctuations extracted from the second time interval TS2 is referred to as the second preliminary estimate PE22, and the estimated respiratory rate obtained based on frequency fluctuations extracted from the second time interval TS2 is referred to as the third preliminary estimate PE32.

[0030] Therefore, the respiratory rate estimate obtained based on baseline fluctuations extracted from the nth time interval TSn is referred to as the first preliminary estimate PE1n. Similarly, the respiratory rate estimate obtained based on amplitude fluctuations extracted from the nth time interval TSn is referred to as the second preliminary estimate PE2n, and the respiratory rate estimate obtained based on frequency fluctuations extracted from the nth time interval TSn is referred to as the third preliminary estimate PE3n.

[0031] Next, processor 112 performs a selection process to identify the mode from the 3n preliminary estimates obtained as described above. Figure 6 shows an example of the selection process flow. Processor 112 determines the identified mode as the estimated respiratory rate ES of subject 20.

[0032] The processor 112 outputs a control signal CT from the output interface 113 to the visualization device 12, which causes the visualization device 12 to visualize information corresponding to the estimated respiratory rate ES of the calculated subject 20. This information may be the estimated value itself, or it may be a color, symbol, or graphic corresponding to the estimated value.

[0033] The inventors of the present invention have focused on the fact that in a configuration in which the estimated respiratory rate of a subject 20 is determined based on multiple preliminary estimates, some preliminary estimates obtained with low accuracy for some reason may adversely affect the final respiratory rate estimation result. According to the configuration of the above processing example, the estimated respiratory rate of the subject 20 is determined based on the mode of the multiple preliminary estimates, so the influence of some preliminary estimates obtained with low accuracy can be suppressed. Since this processing is performed by the processor 112 mounted on the information processing device 11, a decrease in the estimation accuracy of the respiratory rate of the subject 20 can be suppressed without adding sensors.

[0034] Figure 7 shows another example of the selection process performed by processor 112. In this example, the selection process is first performed on multiple preliminary estimates obtained in each of the n time intervals.

[0035] Specifically, the mode is identified from the first preliminary estimate PE11, the second preliminary estimate PE21, and the third preliminary estimate PE31, which are obtained based on multiple features extracted in the first time interval TS1. This mode is treated as the first candidate estimate CE1.

[0036] Similarly, the mode is identified from the first preliminary estimate PE12, the second preliminary estimate PE22, and the third preliminary estimate PE32, which are obtained based on multiple features extracted in the second time interval TS2. This mode is treated as the second candidate estimate CE2.

[0037] This process is performed for all n time intervals, and the mode is identified from the first preliminary estimate PE1n, the second preliminary estimate PE2n, and the third preliminary estimate PE3n, which are obtained based on multiple features extracted in the nth time interval TSn. This mode is treated as the nth candidate estimate CEn.

[0038] Next, the processor 112 performs statistical processing to identify a representative value for the n candidate estimates obtained as described above, and determines this representative value as the estimated respiratory rate ES of the subject 20. Examples of representative values ​​include the mean, median, and mode.

[0039] The accuracy of respiratory rate estimation may decrease due to temporary noise or other factors that occur during the acquisition of the waveform signal WF. According to the processing example above, the mode is identified as a candidate estimate from multiple preliminary estimates acquired in each of the multiple time intervals, and the estimated respiratory rate ES of subject 20 is determined as a representative value of the multiple candidate estimates. Therefore, the impact of temporary causes of decreased estimation accuracy is easily suppressed. Consequently, the decrease in the accuracy of the respiratory rate estimation of subject 20 can be suppressed without adding sensors.

[0040] As illustrated in Figure 8, the above statistical processing may first be performed on multiple preliminary estimates obtained in each of the n time intervals to obtain n candidate estimates CE1 to CEn. In this case, the estimated respiratory rate ES of subject 20 is determined as the mode of the n candidate estimates CE1 to CEn through a selection process.

[0041] Figure 9 shows another example of the selection process performed by processor 112. In this example, the selection process is first performed on n preliminary estimates obtained from n time intervals for each of the three feature quantities extracted from the waveform signal WF.

[0042] Specifically, the mode is identified from n first preliminary estimates, including the first preliminary estimates PE11, PE12, ... PE1n, which are obtained based on baseline fluctuations extracted from the waveform signal WF. This mode is treated as the first candidate estimate CE1.

[0043] Similarly, the mode is identified from n second preliminary estimates, including second preliminary estimates PE21, PE22, ... PE2n, obtained based on amplitude fluctuations extracted from the waveform signal WF. This mode is treated as the second candidate estimate CE2.

[0044] Similarly, the mode is identified from n third preliminary estimates, including third preliminary estimates PE31, PE32, ..., PE3n, obtained based on frequency fluctuations extracted from the waveform signal WF. This mode is treated as the third candidate estimate CE3.

[0045] Next, the processor 112 performs statistical processing to identify a representative value for the three candidate estimates obtained as described above, and determines this representative value as the estimated respiratory rate ES for the subject 20. Examples of representative values ​​include the mean, median, and mode.

[0046] The subject's condition may reduce the accuracy of respiratory rate estimation based on specific features. For example, when the subject's respiratory rate is low (e.g., less than 10 breaths per minute), the accuracy of respiratory rate estimation based on baseline fluctuations tends to decrease. As another example, when the subject is older (especially vascular age), the accuracy of respiratory rate estimation based on frequency fluctuations tends to decrease.

[0047] According to the processing example above, the mode is identified as a candidate estimate from multiple preliminary estimates obtained for each of the multiple features, and the estimated respiratory rate ES of subject 20 is determined as a representative value of the multiple candidate estimates. Therefore, the influence of factors causing a decrease in estimation accuracy for a particular feature is easily suppressed. Consequently, the decrease in estimation accuracy of subject 20's respiratory rate can be suppressed without adding sensors.

[0048] As illustrated in Figure 10, the above statistical processing may first be performed on n preliminary estimates obtained from n time intervals for each of the three features extracted from the waveform signal WF, thereby obtaining three candidate estimates CE1 to CE3. In this case, the estimated respiratory rate ES of subject 20 is determined as the mode of the three candidate estimates CE1 to CE3 through a selection process.

[0049] The processor 112 of the information processing device 11, which has various functions as exemplified above, can be realized by at least one general-purpose microprocessor operating in cooperation with at least one general-purpose memory. Examples of general-purpose microprocessors include CPUs, MPUs, and GPUs. Examples of general-purpose memory include ROMs and RAMs. In this case, the ROM may store computer programs that implement the various functions described above. ROM is an example of a non-temporary computer-readable medium that stores computer programs. The general-purpose microprocessor selects at least a portion of the program stored in the ROM and loads it onto the RAM, and then executes the above-described processes in cooperation with the RAM. The computer program may be pre-installed in the general-purpose memory, or it may be downloaded from an external server device via a communication network and then installed in the general-purpose memory. In this case, the external server device is an example of a non-temporary computer-readable medium that stores computer programs.

[0050] The processor 112 may be implemented by at least one dedicated integrated circuit capable of executing the above-described computer program. Examples of dedicated integrated circuits include microcontrollers, ASICs, FPGAs, etc. In this case, the above-described computer program is pre-installed in a memory element included in the dedicated integrated circuit. This memory element is an example of a computer-readable medium that stores the computer program. The processor 112 can also be implemented by a combination of a general-purpose microprocessor and a dedicated integrated circuit.

[0051] The various configurations described herein are merely examples to facilitate understanding of this disclosure. Each example configuration may be modified or combined with others as appropriate within the scope of the intent of this disclosure.

[0052] In the above embodiment, three feature quantities, including baseline fluctuation, amplitude fluctuation, and frequency fluctuation, are extracted from the waveform signal WF. With this configuration, more preliminary estimates can be obtained based on diverse feature quantities, making it easier to suppress the influence of various factors that degrade the accuracy of respiratory rate estimation. However, if a sufficient number of preliminary estimates can be obtained, then at least one feature quantity may be obtained from the waveform signal WF.

[0053] In the above embodiment, the estimated respiratory rate ES of the subject 20 is visualized by the visualization device 12. However, the estimated value ES may also be output as sound by a speaker, or transmitted as data to another device by a data transmission device. In this case, the speaker or data transmission device can be an example of an output device.

[0054] In the above embodiment, a waveform signal WF corresponding to the pulse wave waveform of subject 20 is acquired. However, a waveform signal WF corresponding to the electrocardiogram waveform of subject 20 may also be acquired. That is, an electrocardiogram can be an example of biological information that involves periodic changes.

[0055] The configurations listed below also constitute part of this disclosure. Item 1: An information processing device for processing the biological information of a subject, An interface that receives signals corresponding to biological information waveforms with periodic changes from sensors attached to the subject, A processor that causes an output device to output information corresponding to an estimated value of the subject's respiratory rate based on the aforementioned signal, It is equipped with, The aforementioned processor, Multiple feature quantities corresponding to the subject's respiration are extracted from the signals contained in each of the multiple time intervals, By applying frequency analysis processing to each of the aforementioned multiple feature quantities, multiple preliminary estimates of the respiratory rate are obtained. The estimated value is determined based on the mode of the plurality of preliminary estimates. Information processing device. Item 2: The aforementioned processor, Multiple candidate estimates are obtained by identifying the mode from the multiple preliminary estimates obtained for each of the multiple feature quantities included in each of the multiple time intervals. The estimated value is determined as a representative value of the aforementioned plurality of candidate estimated values. The information processing device described in item 1. Item 3: The aforementioned processor, For each of the aforementioned feature quantities, a plurality of candidate estimates are obtained by identifying the mode from the plurality of preliminary estimates obtained over the plurality of time intervals, The estimated value is determined as a representative value of the aforementioned plurality of candidate estimated values. The information processing device described in item 1. Item 4: The aforementioned multiple feature quantities include baseline variation, amplitude variation, and frequency variation. An information processing device as described in any one of items 1 to 3. Item 5: The aforementioned biological information waveform is a pulse wave waveform. An information processing device as described in any one of items 1 through 4. [Explanation of Symbols]

[0056] 10: Monitoring device, 11: Information processing device, 111: Input interface, 112: Processor, 12: Visualization device, 20: Subject, 30: Sensor, AG: Age information, CE1: First candidate estimate, CE2: Second candidate estimate, CE3: Third candidate estimate, CEn: nth candidate estimate, ES: Respiratory rate estimate, TS1: First time interval, TS2: Second time interval, TSn: nth time interval, PE11, PE12, PE1n: First preliminary estimate, PE21, PE22, PE2n: Second preliminary estimate, PE31, PE32, PE3n: Third preliminary estimate, WF: Waveform signal

Claims

1. An information processing device for processing the biological information of a subject, An interface that receives signals corresponding to biological information waveforms with periodic changes from sensors attached to the subject, A processor that causes an output device to output information corresponding to an estimated value of the subject's respiratory rate based on the aforementioned signal, It is equipped with, The aforementioned processor, Multiple feature quantities corresponding to the subject's respiration are extracted from the signals contained in each of the multiple time intervals, By applying frequency analysis processing to each of the aforementioned multiple features, multiple preliminary estimates of the respiratory rate are obtained. The estimated value is determined based on the mode of the plurality of preliminary estimates. Information processing device.

2. The aforementioned processor, Multiple candidate estimates are obtained by identifying the mode from the multiple preliminary estimates obtained for each of the multiple feature quantities included in each of the multiple time intervals. The estimated value is determined as a representative value of the aforementioned plurality of candidate estimated values. The information processing apparatus according to claim 1.

3. The aforementioned processor, For each of the aforementioned feature quantities, a plurality of candidate estimates are obtained by identifying the mode from the plurality of preliminary estimates obtained over the plurality of time intervals, The estimated value is determined as a representative value of the aforementioned plurality of candidate estimated values. The information processing apparatus according to claim 1.

4. The aforementioned multiple feature quantities include baseline variation, amplitude variation, and frequency variation. The information processing apparatus according to claim 1.

5. The aforementioned biological information waveform is a pulse wave waveform. The information processing apparatus according to claim 1.

6. A monitoring device that monitors the biological information of a subject, An information processing device according to any one of claims 1 to 5, An output device that outputs information corresponding to the estimated respiratory rate of the subject obtained by the information processing device, It is equipped with Monitoring device.

7. A computer program that can be executed by a processor installed in an information processing device that processes the biological information of a subject, By being executed, the information processing device will The system receives signals corresponding to biometric waveforms that exhibit periodic changes from sensors attached to the subject. Multiple feature quantities corresponding to the subject's respiration are extracted from the signals contained in each of the multiple time intervals, By applying frequency analysis processing to each of the aforementioned multiple features, multiple preliminary estimates of the subject's respiratory rate are obtained. The estimated respiratory rate is determined based on the mode of the aforementioned multiple preliminary estimates. The output device outputs information corresponding to the estimated value. Computer program.

Images