JP6656789B2 - Signal processing apparatus and method - Google Patents

Description

  The present invention relates to a signal processing apparatus and method for processing an input signal such as a respiratory sound signal, and a computer program and a recording medium.

  A body sound such as a breathing sound is used as one of indexes when a doctor diagnoses a disease state of a living body. For this reason, in order to suitably assist a medical worker such as a doctor in diagnosing a disease state, a display device that displays characteristics (for example, signal strength) of body sounds (in other words, visualizes body sounds) has been proposed. I have. By displaying the characteristics of the body sound on the display device, the medical staff can recognize the body sound suitably or relatively easily.

  Note that, as prior art documents related to the present invention, Patent Documents 1 to 3 are cited. Patent Literature 1 discloses a monitor that is not a display device that displays a body sound signal indicating a body sound, but that displays a peak level for an arbitrary period (for example, a video scene unit) determined by a user while watching a video. ing. In Patent Document 2, although not a display device for displaying a body sound signal indicating a body sound, when a peak value of an analog input signal is held and displayed for a predetermined peak hold time, the peak hold time is variably set. A display control device for setting is disclosed. Patent Document 3 does not show a body sound signal indicating a body sound, but when the value of the input signal is smaller than the peak level being displayed, the peak level is reduced with time. A display device for displaying an enlarged image is disclosed.

JP 2009-69549 A JP-A-5-157578 JP-A-9-26332

  In order for medical personnel to diagnose a disease state, the peak value (for example, the maximum value) of the characteristic of the body sound can be useful information. Therefore, it is preferable that the display device that displays the characteristic of the body sound displays the peak value of the characteristic of the body sound.

  By the way, body sounds often have periodicity (that is, characteristics change periodically). That is, the body sound often has a periodicity in which the characteristic changes while periodically increasing and decreasing. In this case, it is preferable that the display device displays a peak value for each cycle in order to appropriately assist a medical worker in diagnosing a disease state.

  However, the monitor described in Patent Literature 1 performs peak level display for an arbitrary period designated by the user without considering such periodicity. Therefore, when the cycle of the body sound does not match (or does not synchronize with) the arbitrary period specified by the user, the peak value of the characteristic of the body sound is displayed even if the body sound is stable. May fluctuate significantly. For this reason, the monitor described in Patent Literature 1 has a technical problem that the peak value of the characteristic of the body sound cannot be displayed in such a manner that the medical worker can appropriately assist the diagnosis of the disease state. .

  Similarly, the display control device described in Patent Literature 2 holds and displays a peak value for a predetermined peak hold time without considering such periodicity. Therefore, when the cycle of the body sound does not match the variably settable peak hold time (or does not synchronize), even if the body sound is stable, the peak value of the characteristic of the body sound is obtained. May greatly fluctuate. For this reason, the display control device described in Patent Literature 2 has a technical problem that it is not possible to display the peak value of the characteristic of the body sound in a mode that can appropriately assist the diagnosis of a medical condition by a medical worker. Having.

  Similarly, the display device described in Patent Literature 3 displays the peak level of the input signal for each fixed period without considering such periodicity. Therefore, when the period of the body sound does not coincide with the predetermined period (or is not synchronized), even if the body sound is stable, the display of the peak value of the characteristic of the body sound greatly fluctuates. There is a risk. For this reason, the display device described in Patent Literature 3 has a technical problem that the peak value of the characteristic of the body sound cannot be displayed in such a manner that the medical worker can appropriately assist the diagnosis of the disease state. Have.

  Furthermore, each cycle of the body sound may be further subdivided according to the characteristics of the body sound. For example, focusing on a breathing sound, which is an example of a body sound, each cycle of the breathing sound may be subdivided into a period in which expiration is performed and a period in which inspiration is performed. When such a cycle is subdivided, the display device preferably displays a peak value for each subdivided period. However, as described above, the techniques described in Patent Literatures 1 to 3 do not consider such subdivision of the cycle, and thus can appropriately assist a medical worker in diagnosing a disease state. There is a technical problem that the peak value of the characteristic of the body sound cannot be displayed in a possible mode.

  The present invention has been made in view of, for example, the conventional problems described above, and provides a signal processing apparatus and method capable of displaying characteristics of an input signal in a more easily viewable manner, as well as a computer program and a recording medium. That is the task.

In order to solve the above-mentioned problem, a first signal processing device includes: a detection unit configured to detect a respiratory sound component and an auxiliary noise component included in an input respiratory sound signal having periodicity; A first display object indicating an instantaneous value and a peak value of a respiratory sound component and a second display object indicating an instantaneous value and a peak value of the auxiliary noise component are displayed side by side in different areas of a display unit of a display device. And output means for outputting the instantaneous value, wherein the instantaneous value indicates one value at a certain timing for each respiratory cycle, and the peak value is a maximum value of the instantaneous value for each respiratory cycle. It is shown .

In order to solve the above-mentioned problem, the second signal processing device includes a detecting unit configured to detect a continuous ra component and an intermittent la component included in the input respiratory sound signal having periodicity, and the detecting unit includes: A first display object indicating the detected instantaneous value and peak value of the continuous lane component and a second display object indicating the instantaneous value and peak value of the intermittent lane component are displayed on a display unit of a display device. Output means for outputting so as to be displayed side by side in different areas, wherein the instantaneous value indicates one value at a certain timing for each respiratory cycle, and the peak value is for each respiratory cycle. It shows the maximum value of the instantaneous value .
In order to solve the above-mentioned problem, a third signal processing device includes a detecting means for detecting a snoring sound component and a whistling sound component which are continuous ra sound components contained in an input respiratory sound signal having periodicity. A first display object indicating the instantaneous value and the peak value of the snoring sound component detected by the detection means, and a second display object indicating the instantaneous value and the peak value of the whistling sound component. Output means for outputting so as to be displayed side by side in different areas of the display unit, wherein the instantaneous value indicates one value of a certain timing for each respiratory cycle, and the peak value is It shows the maximum value of the instantaneous value for each cycle.
In order to solve the above-mentioned problem, a fourth signal processing device includes a detecting means for detecting a crunch sound component and a blister sound component, which are intermittent rattle components included in an input respiratory sound signal having periodicity. A first display object indicating an instantaneous value and a peak value of the crunch sound component detected by the detection means, and a second display object indicating an instantaneous value and a peak value of the blister sound component. Output means for outputting so as to be displayed side by side in different areas of the display unit, wherein the instantaneous value indicates one value of a certain timing for each respiratory cycle, and the peak value is It shows the maximum value of the instantaneous value for each cycle.
In order to solve the above problem, a fifth signal processing device includes a respiratory sound component, a snoring sound component, a whistling sound component, a crunch sound component, and a blister sound component included in an input respiratory sound signal having periodicity. Detecting means for detecting an instantaneous value and a peak value of the respiratory sound component detected by the detecting means, and a second display indicating the instantaneous value and the peak value of the snoring sound component An object, a third display object indicating the instantaneous value and the peak value of the whistle sound component, a fourth display object indicating the instantaneous value and the peak value of the crunch sound component, and the instantaneous value of the water bubble sound component. Output means for outputting a fifth display object indicating a peak value so as to be displayed side by side in different areas of the display unit of the display device, wherein the peak value is a maximum value of an instantaneous value for each respiratory cycle. It shows.

In order to solve the above-mentioned problem, the first signal processing method includes a detecting step of detecting a respiratory sound component and an auxiliary noise component included in an input respiratory sound signal having periodicity, and a detecting step of detecting the respiratory sound component and the auxiliary noise component. A first display object indicating the instantaneous value and the peak value of the respiratory sound component and a second display object indicating the instantaneous value and the peak value of the auxiliary noise component are arranged in different areas of the display unit of the display device. An output step of outputting as displayed, wherein the peak value indicates the maximum value of the instantaneous value in each respiratory cycle .

FIG. 2 is a block diagram illustrating a configuration of a signal processing device according to the present embodiment. 4 is a flowchart illustrating a flow of an operation of the signal processing device of the present embodiment. It is the schematic which shows an example of the mounting | wearing mode of the signal acquisition part with respect to the living body, and the graph which shows the waveform of a respiratory sound signal on a time axis. It is a schematic diagram which shows an example of the operation | movement which a signal storage part stores a respiratory sound signal. It is a classification chart which shows the kind of breathing sound. 5 types of breath sounds (alveolar breath sounds, low-pitched continuum sounds (snoring sounds), high-pitched continuity rap sounds (whistle sounds), fine intermittent rap sounds (twisting sounds), and coarse intermittent rap sounds 6 is a graph showing waveforms of five types of signal components corresponding to (water bubble sound)) on a time axis. It is a graph which shows the 1st aspect of a period determination process with the waveform of a respiratory sound signal. It is a graph which shows the 2nd aspect of a period determination process with the waveform of a respiratory sound signal. 6 is a graph showing one mode of a peak hold process together with waveforms of five types of signal components. 9 is a graph showing one mode of the peak hold process together with the waveform of one type of signal component (alveolar respiratory sound component). It is a top view showing an example of a display mode of each characteristic of five kinds of signal components which constitute a respiratory sound signal.

  Hereinafter, embodiments of a signal processing device and method, and a computer program and a recording medium will be sequentially described.

(Embodiment of signal processing device)
<1>
The signal processing device according to the first embodiment includes a recognition unit that recognizes a period of an input signal having periodicity, and n (where n is an integer of 1 or more) n based on the period recognized by the recognition unit. For each continuous period, a detecting unit that detects a characteristic value of the input signal, and an output unit that outputs display data for displaying the characteristic value detected by the detecting unit on a display device to the display device. Prepare.

  The signal processing device according to the first embodiment is a signal processing device for displaying a characteristic of an input signal (that is, an arbitrary index capable of specifying a state of the input signal, for example, a signal strength) on a display device. I do. That is, the signal processing device according to the first embodiment performs signal processing for displaying on the display device any display target capable of visually specifying the characteristics of the input signal.

  An input signal to be subjected to signal processing performed by the signal processing device according to the first embodiment has periodicity. Here, “periodicity” means a property (typically, signal strength) that changes periodically. More specifically, “periodicity” means a property in which characteristics change while periodically increasing and decreasing. In other words, “periodicity” means a property in which characteristics repeatedly change in the same or similar manner or with the same or similar tendency.

  In particular, the signal processing device according to the first embodiment includes at least a recognition unit, a detection unit, and an output unit in order to perform signal processing for displaying characteristics of an input signal on a display device.

  The recognition means recognizes the cycle of the input signal. Here, the “period” means a period required for one change of the repetitive change of the characteristics of the input signal. Therefore, when the change of the characteristics of the input signal in the same manner is repeated a plurality of times, the period required for one change in the same manner is a cycle.

  In the first embodiment, the cycle of the input signal may be always constant. In this case, the recognizing means continuously recognizes a constant cycle (that is, a period) as the cycle of the input signal. For example, after recognizing that the cycle at the first timing is the first cycle, the recognition unit continuously recognizes that the cycle at the second timing subsequent to the first timing is the first cycle. Will be.

  Alternatively, the cycle of the input signal may change over time. In this case, the recognizing means continuously recognizes, as the cycle of the input signal, a cycle (that is, a period) that varies with time. For example, after recognizing that the cycle at the first timing is the first cycle, the recognition unit may determine that the cycle at the second timing following the first timing is different from the first cycle. Will be recognized.

  The detection unit is a characteristic value of the input signal (that is, an arbitrary index that directly or indirectly indicates a characteristic of the characteristic of the input signal, based on the period of the input signal recognized by the recognition unit. Values and peak values such as minimum values and average values) are detected. In particular, in the first embodiment, the detecting unit detects the characteristic value of the input signal at every n consecutive periods (that is, at n consecutive periods continuously recognized by the recognizing unit). In other words, the detecting means detects the characteristic value of the input signal for every n consecutive cycles. In other words, the detecting means detects the characteristic value of the input signal every period corresponding to n consecutive cycles. That is, the detecting means detects the characteristic value of the input signal for each period synchronized with the cycle of the input signal.

  The output means outputs display data for displaying the characteristic value detected by the detection means on the display device to the display device. As a result, the display device displays the characteristic value detected by the detection unit. That is, the display device can display the characteristic of the input signal by displaying a display target that can visually specify the characteristic value detected by the detection unit.

  As described above, in the first embodiment, the period in which the detection unit detects the characteristic value coincides with n consecutive periods. That is, the period in which the detection unit detects the feature value is synchronized with n consecutive periods. Therefore, in the first embodiment, the start point of the period in which the feature value is to be detected coincides with the start point of the cycle recognized by the recognition unit. Similarly, in the first embodiment, the end point of the period in which the feature value is to be detected coincides with the end point of the cycle recognized by the recognition unit. Because the period of the input signal may or may not fluctuate over time, the start of the n consecutive periods coincides with the start of the first of the n consecutive periods and n Is the same as the end point of the last cycle of the n consecutive cycles. As a result, the detecting means can detect the characteristic value of the input signal suitably associated with the cycle of the input signal. In other words, the display device can display the characteristic value of the input signal suitably associated with the cycle of the input signal.

  It is assumed that the detecting means always detects the characteristic value of the input signal at regular intervals. In this case, the period in which the feature value is detected is not always synchronized with the cycle of the input signal. Alternatively, it is assumed that the detection unit detects the characteristic value of the input signal for each period obtained by multiplying the period recognized at a certain timing by n times. Also in this case, the period during which the feature value is detected is not always synchronized with the cycle of the input signal. This is because the cycle of the input signal is not always constant. Therefore, in these cases, a time lag may occur between the period of the input signal and the period in which the feature value is to be detected. As a result, the detection means cannot detect the characteristic value of the input signal suitably associated with the cycle of the input signal. In other words, the detecting means can only detect the characteristic value of the input signal for each divided period regardless of the cycle of the input signal. As a result, even if the input signal is stable, the feature value of the input signal may fluctuate significantly.

  However, in the first embodiment, the detecting unit detects the characteristic value of the input signal every n consecutive periods (that is, every period corresponding to the n consecutive periods). Therefore, no matter how long the cycle of the input signal is, there is little or no possibility that a time lag occurs between the cycle of the input signal and the period for which the feature value is to be detected. . As a result, the detecting means can detect the characteristic value of the input signal suitably associated with the cycle of the input signal. In other words, the detecting means can detect the characteristic value of the input signal for each period synchronized with the cycle of the input signal. In other words, the detecting unit can newly detect (that is, update) the characteristic value of the input signal at a timing synchronized with the cycle of the input signal. As a result, the display device can display the characteristic value of the input signal for each period synchronized with the period of the input signal (that is, a period corresponding to n consecutive periods). In other words, the display device can display the characteristic value of the input signal that is updated in synchronization with the cycle of the input signal. Therefore, the observer who observes the display device (for example, a medical worker such as a doctor) can input the characteristic value of the input signal for each period synchronized with the period of the input signal (that is, a period corresponding to n consecutive periods). Can be suitably recognized. In other words, the observer can appropriately recognize the characteristic value of the input signal that is updated in synchronization with the cycle of the input signal.

  As described above, according to the signal processing device of the first embodiment, the characteristics of the input signal are suitably displayed.

<2>
In another aspect of the signal processing device of the first embodiment, the input signal is a body sound signal corresponding to a sound caused by the activity of the living body, and the cycle is one of a plurality of activities of the living body repeatedly performed. This is the period required for one activity.

  According to this aspect, the signal processing device can perform the above-described signal processing on the body sound signal.

  Note that, as an example of the body sound signal, a breathing sound signal corresponding to a breathing sound caused by an activity of a living body breathing is given as an example. Such a cycle of the respiratory sound signal may be a period required for one respiration of repetition of respiration. Alternatively, as the body sound signal, for example, a heart sound signal corresponding to a heart sound resulting from the activity of the heartbeat of the living body is given as an example. Such a cycle of the heart sound signal may be a period required for one beat of repeated heart beats. Alternatively, as the body sound signal, for example, a pulsation sound signal corresponding to a pulsation sound caused by the activity of the heart of the living body or a blood flow sound corresponding to a blood flow sound caused by the activity of the heart of the living body A signal is an example. The cycle of the pulsation sound signal and the blood flow sound signal may be the same as the cycle of the heart sound signal.

<3>
In another aspect of the signal processing device of the first embodiment, the input signal is a respiratory sound signal corresponding to a respiratory sound of a living body, and the cycle is a period required for one continuous expiration and inspiration.

  According to this aspect, the signal processing device can perform the above-described signal processing on the respiratory sound signal.

<4>
In another aspect of the signal processing device according to the first embodiment, the cycle of the input signal varies with time.

  According to this aspect, even when the cycle of the input signal fluctuates, the detection unit can detect the characteristic value of the input signal suitably associated with the cycle of the input signal. As a result, the display device can display the characteristic value of the input signal for each period synchronized with the cycle of the input signal even when the cycle of the input signal varies.

<5>
In another aspect of the signal processing device of the first embodiment, the recognition unit recognizes the cycle by receiving a notification of the cycle from an external device that detects the cycle by analyzing the input signal. .

  According to this aspect, in order to recognize the period, the recognition unit does not need to analyze the input signal by the recognition unit itself (in other words, the signal processing device itself including the recognition unit). Therefore, the processing load on the signal processing device is reduced. Furthermore, if the performance of the external device is improved, the recognition means compares the signal processing device itself (that is, the recognition means itself), which has limited performance improvement, with the period recognized by analyzing the input signal. Thus, a highly accurate cycle can be recognized.

<6>
In another aspect of the signal processing apparatus according to the first embodiment, the output unit includes a detection unit that replaces the feature value previously detected by the detection unit each time the detection unit newly detects the feature value. Outputs the display data for displaying the newly detected feature value on the display device to the display means.

  According to this aspect, the output unit outputs display data for displaying the latest feature value on the display device to the display unit. Therefore, the latest characteristic value of the input signal is displayed on the display device while being appropriately updated. Therefore, the observer can suitably recognize the characteristic value of the input signal that is updated in synchronization with the cycle of the input signal.

<7>
The signal processing device according to the second embodiment includes a recognition unit that recognizes a subdivision period obtained by subdividing each period of an input signal having periodicity, and a recognition unit that recognizes the subdivision period based on the subdivision period recognized by the recognition unit. , M (where m is an integer of 1 or more) continuous subdivision periods, a detecting unit for detecting a characteristic value of the input signal, and the characteristic value detected by the detecting unit are displayed on a display device. Output means for outputting display data for the display device to the display device.

  The signal processing device of the second embodiment performs signal processing for displaying the characteristics of an input signal on a display device, similarly to the signal processing device of the first embodiment.

  Compared with the signal processing device of the first embodiment, the signal processing device of the second embodiment is obtained by subdividing the period of the input signal instead of recognizing the period of the input signal. The difference is in recognizing the subdivision period to be performed. Further, the signal processing device of the second embodiment is different from the signal processing device of the first embodiment in that the detecting means detects the characteristic value of the input signal at every n consecutive periods, The difference is that the feature value of the input signal is detected every m successive subdivision periods. Other components of the signal processing device of the second embodiment may be the same as the other components of the signal processing device of the first embodiment.

  Here, the “subdivision period” means a period obtained by subdividing the cycle according to the characteristics of the input signal. For example, the subdivision period may mean a period obtained by subdividing the cycle according to the tendency of the change of the input signal (typically, the tendency of the signal strength of the input signal over time). Good. Therefore, in the second embodiment, each cycle recognized by the recognition unit includes one or a plurality of subdivision periods.

  As described above, in the second embodiment, the period in which the detection unit detects the characteristic value coincides with m consecutive subdivision periods. That is, the period in which the detection unit detects the feature value is synchronized with m consecutive subdivision periods. Therefore, in the second embodiment, the start point of the period in which the feature value is to be detected coincides with the start point of the segmentation period recognized by the recognition unit. Similarly, in the second embodiment, the end point of the period in which the feature value is to be detected coincides with the end point of the segmentation period recognized by the recognition unit. Because the period of the input signal may or may not fluctuate over time, the starting point of the m consecutive subdivision periods is the starting point of the first of the m consecutive subdivision periods. And the end point of the m consecutive subdivision periods coincides with the end point of the last subdivision period of the m consecutive subdivision periods. As a result, the detecting means can detect the characteristic value of the input signal suitably associated with the segmentation period obtained by segmenting the cycle of the input signal. In other words, the display device can display the characteristic value of the input signal suitably associated with the segmentation period obtained by segmenting the cycle of the input signal.

  It is assumed that the detecting means always detects the characteristic value of the input signal at regular intervals. In this case, the period for which the feature value is detected is not always synchronized with the segmentation period. Alternatively, it is assumed that the detecting unit detects the characteristic value of the input signal for each period obtained by multiplying the subdivision period recognized at a certain time by m. Also in this case, the period in which the feature value is detected is not necessarily synchronized with the subdivision period. This is because the segmentation period is not always constant. Therefore, in these cases, there is a possibility that a time lag may occur between the segmentation period and the period in which the feature value is to be detected. As a result, the detecting means cannot detect the characteristic value of the input signal suitably associated with the segmentation period. In other words, the detection means can only detect the characteristic value of the input signal for each divided period regardless of the subdivision period.

  However, in the second embodiment, the detecting means detects the characteristic value of the input signal for each of m consecutive subdivision periods. Therefore, no matter how long the subdivision period is, there is little or no possibility that a temporal shift occurs between the subdivision period and the period in which the feature value is to be detected. As a result, the detecting means can detect the characteristic value of the input signal suitably associated with the segmentation period. In other words, the detecting means can detect the characteristic value of the input signal for each period synchronized with the segmentation period. In other words, the detecting means can newly detect (that is, update) the characteristic value of the input signal at a timing synchronized with the segmentation period. As a result, the display device can display the characteristic value of the input signal for each period synchronized with the subdivision period (that is, m continuous subdivision periods). In other words, the display device can display the characteristic value of the input signal that is updated in synchronization with the segmentation period. Therefore, the observer who observes the display device (for example, a medical worker such as a doctor) preferably sets the characteristic value of the input signal for each period synchronized with the segmentation period (that is, m continuous segmentation periods). Can be recognized. In other words, the observer can appropriately recognize the characteristic value of the input signal that is updated in synchronization with the segmentation period.

  As described above, according to the signal processing device of the second embodiment, the characteristics of the input signal are suitably displayed.

<8>
In another aspect of the signal processing device of the second embodiment, the input signal is a body sound signal corresponding to a sound caused by the activity of the living body, and the cycle is one of a plurality of activities of the living body that are repeatedly performed a plurality of times. This is a period required for one activity, and the plurality of subdivision periods are a plurality of subdivision periods obtained by subdividing the one activity according to the mode of the one activity.

  According to this aspect, the signal processing device can perform the above-described signal processing on the body sound signal.

  Note that, as described above, the body sound signal includes a breathing sound signal, a heart sound signal, a pulsation sound signal, and a blood flow sound signal. The subdivision period of the respiratory sound signal is a period obtained by subdividing a period required for one respiration of repetitive respiration according to the mode of the one respiration (for example, a period during which expiration is performed). And a period during which intake is performed). Alternatively, the periods of the heart sound signal, the pulsation sound signal, and the blood flow sound signal are obtained by subdividing a period required for one pulsation among repeated heart beats in accordance with a mode of the one pulsation. (For example, a period during which the ventricle contracts and a period during which the ventricle contracts).

<9>
In another aspect of the signal processing device of the second embodiment, the input signal is a body sound signal corresponding to a breathing sound of a living body, and the cycle is a period required for one continuous expiration and inspiration, The plurality of subdivision periods include a period during which expiration is performed and a period during which inspiration is performed.

  According to this aspect, the signal processing device can perform the above-described signal processing on the respiratory sound signal.

<10>
In another aspect of the signal processing device according to the second embodiment, the cycle of the input signal varies with time.

  According to this aspect, even when the period of the input signal varies, the detection unit can determine the characteristic value of the input signal suitably associated with the subdivision period obtained by subdividing the period of the input signal. Can be detected. As a result, even if the period of the input signal fluctuates, the display device displays the characteristic value of the input signal for each period synchronized with the subdivision period obtained by subdividing the period of the input signal. Can be.

<11>
In another aspect of the signal processing device of the second embodiment, the recognizing unit receives the notification of the subdivision period from an external device that detects the subdivision period by analyzing the input signal. Recognize the subdivision period.

  According to this aspect, the recognition unit does not need to analyze the input signal by the recognition unit itself (in other words, the signal processing device itself including the recognition unit) in order to recognize the subdivision period. Therefore, the processing load on the signal processing device is reduced. Furthermore, if the performance of the external device is improved, the recognition means compares the signal processing device itself (that is, the recognition means itself), which has limited performance improvement, with the period recognized by analyzing the input signal. Thus, a highly accurate cycle can be recognized.

<12>
In another aspect of the signal processing device according to the second embodiment, the output unit includes a detection unit that replaces the feature value previously detected by the detection unit each time the detection unit newly detects the feature value. Outputs the display data for displaying the newly detected feature value on the display device to the display means.

  According to this aspect, the output unit outputs display data for displaying the latest feature value on the display device to the display unit. Therefore, the latest characteristic value of the input signal is displayed on the display device while being appropriately updated. Therefore, the observer can suitably recognize the characteristic value of the input signal that is updated in synchronization with the cycle of the input signal.

(Embodiment of signal processing method)
<13>
The signal processing method according to the first embodiment includes a recognition step of recognizing a cycle of an input signal having periodicity, and n (where n is an integer of 1 or more) pieces based on the cycle recognized by the recognition step. For each successive cycle, a detection step of detecting a characteristic value of the input signal, and an output step of outputting display data for displaying the characteristic value detected by the detection step on a display device to the display device. Prepare.

  According to the signal processing method of the first embodiment, it is possible to suitably receive the same effects as the various effects received by the signal processing device of the first embodiment.

  Incidentally, in response to the various aspects that can be adopted by the signal processing apparatus of the first embodiment, the signal processing method of the first embodiment may also adopt various aspects.

<14>
The signal processing method according to the second embodiment is based on a recognition step of recognizing a subdivision period obtained by subdividing each cycle of an input signal having periodicity, and the subdivision period recognized by the recognition step. , M (where m is an integer of 1 or more) successive subdivision periods, a detecting step of detecting a characteristic value of the input signal, and displaying the characteristic value detected by the detecting step on a display device. Outputting display data for the display device to the display device.

  According to the signal processing method of the second embodiment, it is possible to suitably receive the same effects as the various effects received by the above-described signal processing device of the second embodiment.

  Incidentally, in response to the various aspects that can be adopted by the signal processing device of the second embodiment described above, the signal processing method of the second embodiment may also adopt various aspects.

(Embodiment of computer program)
<15>
The computer program according to the first embodiment is a computer program executed by a computer, and includes a recognition step of recognizing a cycle of an input signal having periodicity, and n (where, , N is an integer of 1 or more) for each of the continuous periods, a detection step of detecting the characteristic value of the input signal, and display data for displaying the characteristic value detected by the detection step on a display device. And causing the computer to execute an output step of outputting to the display device.

  According to the computer program of the first embodiment, it is possible to suitably receive the same effects as the various effects received by the above-described signal processing device of the first embodiment.

  Note that, in response to the various aspects that can be adopted by the signal processing device of the first embodiment, the computer program of the first embodiment may also adopt various aspects.

<16>
The computer program according to the second embodiment is a computer program executed by a computer, and includes: a recognition step of recognizing a subdivision period obtained by subdividing each cycle of a periodic input signal; Detecting a characteristic value of the input signal for each of m (where m is an integer equal to or greater than 1) continuous subdivision periods based on the subdivision periods recognized by Outputting the display data for displaying the characteristic value on a display device to the display device.

  According to the computer program of the second embodiment, it is possible to suitably receive the same effects as the various effects received by the above-described signal processing device of the second embodiment.

  Note that, in response to the various aspects that can be adopted by the signal processing device of the second embodiment, the computer program of the second embodiment may also adopt various aspects.

(Embodiment of recording medium)
<17>
The recording medium according to the first embodiment is a recording medium in which a computer program executed by a computer is recorded, wherein a recognition step of recognizing a cycle of an input signal having a periodicity, and the cycle recognized by the recognition step A detection step of detecting a characteristic value of the input signal for each of n (where n is an integer equal to or greater than 1) continuous cycles, and displaying the characteristic value detected by the detection step on a display device. And a computer program for causing the computer to execute an output step of outputting display data to the display device.

  According to the recording medium of the first embodiment, it is possible to suitably receive the same effects as the various effects received by the above-described signal processing device of the first embodiment.

  Incidentally, the recording medium of the first embodiment may also adopt various aspects in correspondence with various aspects that can be adopted by the signal processing device of the first embodiment described above.

<16>
The recording medium of the second embodiment is a recording medium in which a computer program executed by a computer is recorded, and recognizes a subdivision period obtained by subdividing each cycle of a periodic input signal. A recognition step, and a detection step of detecting a characteristic value of the input signal for each of m (where m is an integer of 1 or more) continuous subdivision periods based on the subdivision periods recognized by the recognition step. And a computer program for causing the computer to execute an output step of outputting display data for displaying the characteristic value detected by the detection step on a display device to the display device.

  According to the recording medium of the second embodiment, it is possible to suitably receive the same effects as the various effects received by the above-described signal processing device of the second embodiment.

  It should be noted that the recording medium of the second embodiment may also adopt various aspects in correspondence with various aspects that can be adopted by the signal processing device of the second embodiment described above.

  The operation and other advantages of the present embodiment will become more apparent from the examples explained below.

  As described above, the signal processing device according to the first or second embodiment includes the recognition unit, the detection unit, and the output unit. The signal processing method according to the first or second embodiment includes a recognition step, a detection step, and an output step. The computer program according to the first or second embodiment causes a computer to execute a recognition step, a detection step, and an output step. The recording medium of the first or second embodiment is a recording medium in which a computer program for causing a computer to execute a recognition step, a detection step, and an output step is recorded. Therefore, the input signal is displayed in a more easily viewable manner.

  Hereinafter, embodiments of a signal processing device and method, a computer program and a recording medium will be described with reference to the drawings. In the following, it is assumed that a respiratory sound signal S indicating a respiratory sound is used as an input signal. However, another biological sound signal different from the respiratory sound signal may be used as the input signal. Examples of other body sound signals include, for example, a heart sound signal indicating a heart sound, a pulsation sound signal indicating a sound caused by a pulsation (in other words, a pulse wave), and a sound caused by an activity state of an internal organ officer (for example, Examples include a visceral sound signal indicating a bowel sound caused by intestinal movement, a blood flow sound signal indicating a sound caused by a blood flow, and the like. Alternatively, any signal other than the body sound signal (however, any signal having periodicity) may be used as the input signal.

(1) Configuration of Signal Processing Device First, the configuration of the signal processing device 10 of the present embodiment will be described with reference to FIG. FIG. 1 is a block diagram illustrating a configuration of a signal processing device 10 according to the present embodiment.

  As illustrated in FIG. 1, the signal processing device 10 of the present embodiment includes a signal acquisition unit 101, a signal storage unit 102, a period determination unit 113 which is a specific example of “recognition unit”, “recognition unit”, It includes a signal analysis unit 103 as a specific example of “detection unit”, a display data generation unit 104 as a specific example of “output unit”, and a display unit 105 as a specific example of “display device”.

  The signal acquisition unit 101 acquires a respiratory sound signal. For example, the signal acquisition unit 101 may directly acquire a respiratory sound signal by directly detecting a respiratory sound of a living body. In this case, the signal acquisition unit 101 may include a respiration sensor attached to the living body and detecting a breathing sound of the living body. Such a respiration sensor is typically a microphone (for example, a coil microphone, a condenser microphone, a piezoelectric microphone, or the like). However, the respiration sensor may be any type of sensor as long as the respiration sensor can detect a respiratory sound.

  FIG. 1 illustrates an example in which the signal processing device 10 includes a signal acquisition unit 101. However, the signal processing device 10 may not include the signal acquisition unit 101. In this case, it is preferable that the signal processing device 10 obtains a respiratory sound signal from the signal obtaining unit 101 disposed outside the signal processing device 10 via a wired or wireless communication line or a wired signal line. .

  The signal storage unit 102 temporarily stores the respiratory sound signal acquired by the signal acquisition unit 101. For this reason, the signal storage unit 102 may include a memory or a buffer. Note that, as described later, it is preferable that the signal storage unit 102 store the respiratory sound signals for a predetermined period among the respiratory sound signals sequentially acquired by the signal acquiring unit 101.

  FIG. 1 illustrates an example in which the signal processing device 10 includes the signal storage unit 102. However, the signal processing device 10 may not include the signal storage unit 102.

  The cycle determination unit 113 performs a predetermined cycle determination process on the respiratory sound signal stored in the signal storage unit 102.

  In the present embodiment, the predetermined cycle determination process may include a process of determining (in other words, specifying or detecting) the cycle of the respiratory sound signal. Specifically, in the predetermined cycle determination process, the amplitude level of the respiratory sound signal is changed by one cycle with respect to the respiratory sound signal in which the amplitude level (that is, the characteristic indicating the signal intensity) changes while periodically increasing and decreasing. It may include a process of determining that the period required to change only is the cycle of the respiratory sound signal. In other words, considering that the respiratory sound signal is a sound signal corresponding to a sound caused by the activity of the living body, the predetermined period determination processing is performed for one respiration (a respiration performed a plurality of times). That is, it may include a process of determining that the period required for continuous expiration and inspiration) is the cycle of the respiratory sound signal.

  Note that the predetermined cycle determination process performed on any body sound signal other than the respiratory sound signal is such that the period required for one activity of the biological activity repeatedly performed a plurality of times is the cycle of the body sound signal. It may include a process of determining that there is. For example, the predetermined cycle determination process is a process of determining that a period required for one beat of a heart beat repeatedly performed a plurality of times is a cycle of a heart sound signal, a pulsation sound signal, or a blood flow sound signal. May be included. The cycle determination process performed on other body sound signals or input signals may include a process of determining a cycle in a similar manner.

  Further, in the present embodiment, the predetermined cycle determination process may include a process of determining a subdivision period obtained by subdividing the cycle of the respiratory sound signal. For example, the predetermined cycle determination process may include a process of determining a subdivision period obtained by subdividing the cycle of the respiratory sound signal in accordance with the tendency of the amplitude level of the respiratory sound signal to change. Specifically, one respiration, which is an activity of a living body, is composed of one expiration and one inspiration, which have different amplitude level changes. Therefore, the predetermined cycle determination process determines the expiration period (that is, the expiration period during which expiration is being performed) and the inspiration period (that is, the period during which inspiration is being performed) in each cycle of the respiratory sound signal. Processing may be included.

  In addition, the predetermined cycle determination process performed on an arbitrary body sound signal other than the respiratory sound signal converts one activity of the biological activity repeatedly performed a plurality of times into the one activity mode. It may include a process of determining a subdivision period obtained by subdividing accordingly. For example, one heart beat, which is an activity of a living body, is composed of one contraction of the ventricle and one dilation of the ventricle in different modes of activity. For this reason, the predetermined cycle determination process performed on the heart sound signal, the pulsation sound signal, or the blood flow sound signal is performed during the period in which the contraction of the ventricle is performed and the expansion of the ventricle in each cycle of the body sound signal. It may include a process of determining a period during which the time period is set. The cycle determination process performed on other body sound signals or input signals may include a process of determining a subdivision period in a similar manner.

  FIG. 1 illustrates an example in which the signal processing device 10 includes the cycle determination unit 113. However, the signal processing device 10 does not have to include the cycle determination unit 113. In this case, the signal processing device 10 is located outside the signal processing device 10 via a wired or wireless communication line or a wired signal line from the period determination unit 113 disposed outside the signal processing device 10. It is preferable to obtain the result of the cycle determination process executed by the cycle determination unit 113.

  The signal analysis unit 103 performs a predetermined analysis process on the respiratory sound signal stored in the signal storage unit 102. In the present embodiment, the predetermined analysis process includes an analysis process of separating a respiratory sound signal into a plurality of types of signal components that can be distinguished from each other. In other words, the predetermined analysis process includes an analysis process of extracting at least one of a plurality of types of signal components that can be distinguished from each other from the respiratory sound signal. In other words, the predetermined analysis processing includes an analysis processing for analyzing what kind of signal component is included in the respiratory sound signal.

  Specifically, for example, the predetermined analysis process may include an analysis process of separating the respiratory sound signal into a signal component corresponding to a normal sound and a signal component corresponding to an abnormal sound different from the normal sound. Good. In other words, for example, the predetermined analysis process may include an analysis process of extracting at least one of a signal component corresponding to a normal sound and a signal component corresponding to an abnormal sound from a respiratory sound signal.

  Alternatively, for example, the predetermined analysis process may include an analysis process of separating a respiratory sound signal into a plurality of types of signal components corresponding to a plurality of types of normal sounds that can be distinguished from each other. In other words, for example, the predetermined analysis process may include an analysis process of extracting at least one of a plurality of types of signal components corresponding to a plurality of types of normal sounds that can be distinguished from each other, from the respiratory sound signal. .

  Alternatively, for example, the predetermined analysis process may include an analysis process of separating a respiratory sound signal into a plurality of types of signal components corresponding to a plurality of types of abnormal sounds that can be distinguished from each other. In other words, for example, the predetermined analysis process may include an analysis process of extracting at least one of a plurality of types of signal components corresponding to a plurality of types of abnormal sounds that can be distinguished from each other, from the respiratory sound signal. .

  In addition, the signal analysis unit 103 determines the peak value of the respiratory sound signal (for example, the amplitude level of the respiratory sound signal (that is, the characteristic indicating the signal intensity)) based on the result of the cycle determination process performed by the cycle determination unit 113. A predetermined peak hold process for detecting the maximum value or the minimum value may be executed. However, instead of the peak value of the respiratory sound signal, the signal analysis unit 103 detects a predetermined peak value that detects an arbitrary characteristic value (for example, an average value) capable of specifying the characteristic of the amplitude level of the respiratory sound signal. Hold processing may be performed.

  In the present embodiment, the predetermined peak hold process may include a process of detecting a peak value of a respiratory sound signal at every n consecutive time periods (where n is an integer of 1 or more). . In other words, the predetermined peak hold process is performed based on the peak value of the respiratory sound signal at every n consecutive periods in time (that is, the peak value of the respiratory sound signal during the period corresponding to n consecutive periods in time). Value) may be included.

  Further, in the embodiment, the predetermined peak hold process includes a process of detecting a peak value of a respiratory sound signal for each of m (where m is an integer of 1 or more) subdivided periods that are temporally continuous. May be. In other words, the predetermined peak hold process performs the peak value of the respiratory sound signal for each of m temporally continuous subdivision periods (that is, the peak value of the respiratory sound signal during the m temporally continuous subdivision periods). Value) may be included.

  The signal analysis unit 103 determines the result of the analysis processing and the peak hold processing (for example, information that can specify a plurality of types of signal components themselves, or n sub-periods or m subdivisions of a plurality of types of signal components, respectively). Any information that can specify the peak value for each period) is output to the display data generation unit 104.

  FIG. 1 illustrates an example in which the signal processing device 10 includes the signal analysis unit 103. However, the signal processing device 10 may not include the signal analysis unit 103. In this case, the signal processing device 10 is located outside the signal processing device 10 via a wired or wireless communication line or a wired signal line from the signal analysis unit 103 disposed outside the signal processing device 10. It is preferable to obtain the results of the analysis processing and the peak hold processing performed by the signal analysis unit 103.

  Based on the results of the analysis process and the peak hold process performed by the signal analysis unit 103, the display data generation unit 104 performs the respective characteristics (for example, the amplitude corresponding to the signal intensity) of the plurality of types of signal components constituting the respiratory sound signal. (For example, an image signal) for displaying the level or the like on the display unit 105. In other words, the display data generation unit 104 displays the display data for displaying on the display unit 105 any display target capable of visually specifying the characteristics of each of a plurality of types of signal components constituting the respiratory sound signal. Generate At this time, the display data generation unit 104 generates display data for displaying at least the respective peak values of a plurality of types of signal components constituting the respiratory sound signal on the display unit 105. The display target may be any display target as long as the characteristics of the respiratory sound signal can be visually specified. For example, the display object may be a display object that includes text (for example, characters or numerical values) that visually specifies the characteristics of the respiratory sound signal. Alternatively, for example, the display object may be a display object including a graphic (for example, a graph or a chart) that visually specifies the characteristics of the respiratory sound signal.

  In addition, the display data generation unit 104 outputs the generated display data to the display unit 105.

  Note that the cycle determination unit 113, the signal analysis unit 103, and the display data generation unit 104 are processing blocks logically realized on a CPU (Central Processing Unit). However, the cycle determination unit 113, the signal analysis unit 103, and the display data generation unit 104 may each be a processing circuit physically implemented by a semiconductor chip or the like.

  The display unit 105 is a display device that performs a display process based on the display data generated by the display data generation unit 104. As a result, the display unit 105 displays, on the display screen of the display unit 105, the characteristics of each of the plurality of types of signal components constituting the respiratory sound signal.

  FIG. 1 illustrates an example in which the signal processing device 10 includes the display unit 105. However, the signal processing device 10 does not have to include the display unit 105. In this case, the signal processing device 10 transmits the display data generated by the display data generation unit 104 to the display unit 105 located outside the signal processing device 10 via a wired or wireless communication line or a wired signal line. Is preferably transferred.

(2) Operation of Signal Processing Apparatus 10 of the Present Embodiment Next, the operation of the signal processing apparatus 10 of the present embodiment will be described with reference to FIGS. FIG. 2 is a flowchart illustrating a flow of the operation of the signal processing device 10 according to the present embodiment. FIG. 3 is a schematic diagram showing an example of a manner of attaching the signal acquisition unit 101 to a living body, and a graph showing a waveform of a respiratory sound signal on a time axis. FIG. 4 is a schematic diagram illustrating an example of an operation in which the signal storage unit 102 stores a respiratory sound signal. FIG. 5 is a classification chart showing types of respiratory sounds. FIG. 6 shows five types of respiratory sounds (alveolar respiratory sounds, low-pitched continuous rales (snoring sounds), high-pitched continuous rales (whistle sounds), fine intermittent rales (twisting sounds), and coarse sounds. It is a graph which shows the waveform of five types of signal components corresponding to intermittent rattle (water bubble sound) on a time axis. FIG. 7 is a graph showing the first mode of the cycle determination process together with the waveform of the respiratory sound signal. FIG. 8 is a graph showing the second mode of the cycle determination process together with the waveform of the respiratory sound signal. FIG. 9 is a graph showing one mode of the peak hold processing together with waveforms of five types of signal components. FIG. 10 is a graph showing one mode of the peak hold process together with the waveform of one type of signal component (alveolar respiratory sound component). FIG. 11 is a plan view showing an example of a display mode of the characteristics of each of the five types of signal components constituting the respiratory sound signal.

  As shown in FIG. 2, first, the signal acquiring unit 101 acquires a respiratory sound signal (Step S101). For example, as shown in FIG. 3A, the signal acquisition unit 101 may be attached to the body surface of a living body (in the example shown in FIG. 3A, the body surface near the left chest). At this time, the mounting position of the signal acquisition unit 101 may be fixed. Alternatively, considering that the position at which the medical staff puts the stethoscope on auscultation is appropriately changed, the mounting position of the signal acquisition unit 101 may be appropriately changed during the acquisition of the respiratory sound signal. . As a result, the signal acquisition unit 101 acquires a respiratory sound signal. At this time, the signal acquisition unit 101 may periodically acquire a respiratory sound signal according to a predetermined sampling frequency. That is, the signal acquisition unit 101 may periodically acquire the sample value of the respiratory sound signal. However, the signal acquisition unit 101 may acquire the respiratory sound signal aperiodically or continuously. As a result, the signal acquisition unit 101 can acquire a respiratory sound signal specified as a waveform on the time axis illustrated in FIG.

  As shown in FIG. 3B, considering that the respiration of the living body is a repetition of inspiration (that is, intake of air into the lungs) and expiration (that is, exhalation of air from the lungs), It can be said that the respiratory sound signal is a signal having periodicity. One cycle of the breathing sound signal is a period obtained by adding an inspiratory period (that is, a period during which inhalation is performed) and an expiration period subsequent to the inspiratory period (that is, a period during which exhalation is performed). It goes without saying that the cycle of the respiratory sound signal may fluctuate over time.

  In parallel with the acquisition of the respiratory sound signal by the signal acquisition unit 101, the signal storage unit 102 temporarily stores the respiration sound signal acquired by the signal acquisition unit 101 (Step S102). At this time, the signal storage unit 102 stores the respiratory sound signal having the larger size of the size required for the analysis process performed by the signal analysis unit 103 described later and the size required for the cycle determination process performed by the cycle determination unit 113. Is preferably temporarily stored. In the present embodiment, it is assumed that the size of the respiratory sound signal required for the analysis process performed by the signal analysis unit 103 is “1 second”. On the other hand, in the present embodiment, the size required for the cycle determination process performed by the cycle determination unit 113 is 20 seconds to 25 seconds (in other words, the size of the respiratory sound signal for 5 cycles, and the length of one cycle. Is about (4 to 5 seconds) × 5 cycle = 20 to 25 seconds). Therefore, in this case, it is preferable that the signal storage unit 102 temporarily store at least the latest (in other words, the latest) respiratory sound signal for 20 seconds. In order to temporarily store such a respiratory sound signal for the last 20 seconds, the signal storage unit 102 preferably includes a ring buffer. However, the size of the respiratory sound signal required for the analysis process performed by the signal analysis unit 103 may be any time other than 1 second. Similarly, the size of the respiratory sound signal required for the cycle determination process performed by the signal analysis unit 103 may be any time other than 20 seconds to 25 seconds.

  When the heart sound signal is used instead of the respiratory sound signal, the size required for the cycle determination process performed by the cycle determination unit 113 is, for example, 4 to 5 seconds (in other words, the heart sound signal for 5 cycles). And, for a heart sound signal whose one cycle is approximately 0.8 seconds to 1 second, it is assumed to be 0.8 seconds to 1 second × 5 cycles = 4 seconds to 5 seconds). However, the size of the respiratory sound signal required for the cycle determination process executed by the signal analyzer 103 may be any time other than 4 to 5 seconds.

  For example, when the signal obtaining unit 101 obtains a respiratory sound signal according to the sampling frequency of 44100 Hz, the signal storage unit 102 temporarily stores the sample values of 44100 × 20 seconds = 882000 respiratory sound signals. Remember. More specifically, as shown in FIG. 4, the signal storage unit 102 acquires the sample value S (t) of the respiratory sound signal acquired at the latest time (t) and the sample value S (t) acquired at the time (t-1). The sample value S (t-1) of the respiratory sound signal obtained, the sample value S (t-2) of the respiratory sound signal obtained at time (t-2),..., At time (t-881999) The sample value S (t-881999) of the acquired respiratory sound signal is temporarily stored.

  In parallel with the acquisition of the respiratory sound signal by the signal acquiring unit 101, the signal analyzing unit 103 performs a predetermined analysis process on the respiratory sound signal stored in the signal storage unit 102 (Step S103a). That is, the signal analysis unit 103 performs an analysis process on the respiration sound signal stored in the signal storage unit 102 to separate the respiration sound signal into a plurality of types of signal components.

  Here, the types of breath sounds will be described with reference to FIG. As shown in FIG. 5, respiratory sounds in a broad sense (that is, lung sounds) are classified into respiratory sounds in a narrow sense and side noises that are examples of abnormal sounds. Breath sounds in a narrow sense are classified into normal breath sounds, which are examples of normal sounds, and abnormal breath sounds, which are examples of abnormal sounds. Normal breath sounds are classified into alveolar breath sounds, bronchial breath sounds, bronchoalveolar breath sounds, and tracheal breath sounds. Abnormal breathing sounds are classified into breathing sounds due to attenuation / loss, breathing sounds due to augmentation, breathing sounds due to prolonged expiration, breathing sounds due to bronchial breathing sounds, and tracheal constriction sounds. Is done. The side noise is classified into a ra sound and other sounds. The rattle is classified into a continuous rattle and an intermittent rattle. The continuous ra sound is classified into a bass continuous ra sound (snoring sound), a high sound continuous ra sound (whistling sound), and a squawk (inspiratory continuous ra sound). The intermittent rattle is classified into a fine intermittent rattle (twisting sound) and a coarse intermittent rattle (water bubble sound). Other sounds are classified as pleural rubs and pulmonary vascular murmurs.

  In the present embodiment, the signal analysis unit 103 converts the respiratory sound signal into a signal component corresponding to an alveolar respiratory sound (hereinafter, referred to as an “alveolar respiratory component”), which is an example of a normal sound, and an abnormal sound. (Hereinafter referred to as "snoring component") and a high tone continuous rasing sound (whistle), which is an example of an abnormal sound. Signal component (hereinafter referred to as a "whistle component"), and a signal component corresponding to a fine intermittent rattle (a crunch sound) which is an example of an abnormal sound (hereinafter referred to as a "crochet component"). An analysis process is performed to separate the signal component into a signal component (hereinafter, referred to as “water bubble sound component”) corresponding to a coarse intermittent rattle (water bubble sound) which is an example of the abnormal sound. In other words, the signal analysis unit 103 performs an analysis process of extracting at least one of an alveolar respiratory component, a snoring component, a whistling component, a crunch component, and a blister component from the respiratory signal. . The first graph in FIG. 6 shows an example of an alveolar respiratory sound component. The second graph in FIG. 6 illustrates an example of the snoring sound component. The third graph in FIG. 6 shows an example of the whistle sound component. The fourth graph in FIG. 6 shows an example of the crunch sound component. The fifth graph in FIG. 6 shows an example of a water bubble sound component.

  The signal analyzer 103 may separate the respiratory sound signal into an alveolar respiratory sound component, a sniffing sound component, a whistling sound component, a crunch sound component, and a blistering sound component from the following viewpoints. .

  More specifically, the frequency spectrum shapes of the alveolar respiratory sound component and the signal components corresponding to the intermittent rattle sound (that is, the crunch sound component and the blister sound component) are relatively broadband and relatively smooth. On the other hand, there is a difference that the frequency spectrum shapes of the signal components corresponding to the continuous ra sound (that is, the snoring sound component and the whistling sound component) are relatively narrow and relatively steep. Therefore, the signal analysis unit 103 focuses attention on such a difference in the frequency spectrum shape, and makes the respiratory sound signal different from the signal component corresponding to the continuous ra sound and the signal component corresponding to the continuous ra sound. It may be separated into a signal component (that is, a signal component corresponding to an alveolar respiratory sound component and an intermittent rattle).

  Further, there is a difference that the frequency of the whistling sound component is relatively high while the frequency of the snoring sound component is relatively low. Therefore, the signal analyzing unit 103 may separate the signal component corresponding to the continuous ra sound into the snoring sound component and the whistling sound component by focusing on such a difference in frequency.

  Also, the difference is that the alveolar respiratory sound component is a temporally continuous (that is, continuous) signal component, while the signal component corresponding to the intermittent rattle is a temporally intermittent pulse-like signal component. There is. Therefore, the signal analysis unit 103 focuses on such a difference in distribution of signal components on the time axis, and converts a signal component different from a signal component corresponding to a continuous ra sound into an alveolar respiratory sound component. The signal may be separated into a signal component corresponding to an intermittent sound.

  Further, there is a difference that the frequency of the whip sound component is relatively high while the frequency of the creasing sound component is relatively low. Therefore, the signal analysis unit 103 may separate the signal component corresponding to the intermittent rattle sound into the water bubble sound component and the crouching sound component by focusing on such a difference in frequency.

  It should be noted that the above-described method relating to the analysis processing for separating the respiratory sound signal into five types of signal components (ie, alveolar respiratory component, snoring component, whistle component, crunch component, and blister component) is merely an example. It is. Therefore, the signal analysis unit 103 may execute an analysis process for separating the respiratory sound signal into five types of signal components by another method. Note that a method of separating a respiratory sound signal into a plurality of types of signal components (or extracting a specific type of signal component from the respiratory sound signal) is described in, for example, JP-T-2004-53309 or JP-A-2005-66045. Japanese Unexamined Patent Publication (Kokai) No. 2001-505085, Japanese Unexamined Patent Publication No. 2007-508899, and "Sparse Representation of Lung Sound Signal and Its Application to Intermittent Sound Separation, Sakai, Satomoto, Kiyasu, Miyahara, Nagasaki University National Studies Report Vol. 41, No. 76 "and the like. Therefore, the signal analysis unit 103 may separate the respiratory sound signal into a plurality of types of signal components using the methods disclosed in these documents.

  Further, the signal analysis unit 103 separates the respiratory sound signal into arbitrary plural types of signal components in addition to or instead of the analysis processing of separating the respiratory sound signal into the above-described five types of signal components (or the respiratory sound). (Any one or more signal components are extracted from the signal).

  Note that, typically, the respiratory sound signal to be analyzed by the signal analyzing unit 103 is likely to always include an alveolar respiratory sound component. Because, as long as the living body is breathing, the respiratory sound should include the alveolar respiratory sound, which is an example of the normal sound, regardless of whether the abnormal sound is included or not. It is. On the other hand, the respiratory sound signal to be analyzed by the signal analyzer 103 may include all of the five types of signal components. Alternatively, the respiratory sound signal to be analyzed by the signal analyzer 103 may include only some of the five types of signal components. Alternatively, the respiratory sound signal to be analyzed by the signal analyzer 103 may not include all of the five types of signal components. In any case, the signal analysis unit 103 can acquire each of the five types of signal components by executing the analysis process of separating the respiratory sound signal into the five types of signal components described above. For example, the signal analysis unit 103 can obtain a signal component included in the respiratory sound signal by performing the analysis process, and set the signal component not included in the respiratory sound signal to have an amplitude level of zero. Can be obtained as the following signal component.

  In parallel with the process of analyzing the respiratory sound signal by the signal analyzing unit 103, the cycle determining unit 113 performs a predetermined cycle determining process on the respiratory sound signal stored in the signal storage unit 102 (Step S113). That is, the signal analyzer 103 determines at least one of the cycle and the subdivision period (that is, the expiration period and the inspiration period) of the respiration sound signal for the respiration sound signal stored in the signal storage unit 102. Execute the cycle determination process.

  The cycle determination unit 113 may determine at least one of the cycle of the respiratory sound signal and the subdivision period (that is, the expiration period and the inspiration period) from the following viewpoints.

  Specifically, as shown in FIG. 7, the cycle determination unit 113 determines that a silent section in which the absolute value of the amplitude level of the respiratory sound signal is equal to or smaller than the predetermined threshold A is at least one boundary between the cycle and the subdivision period By assuming that there is, at least one of the cycle and the subdivision period (that is, the expiration period and the inspiration period) of the respiratory sound signal may be determined. In the example illustrated in FIG. 7, the cycle determination unit 113 determines the subdivision period (that is, the expiration period and the inspiration period) of the respiratory sound signal by regarding the silent part as the boundary of the subdivision period. I have. However, in consideration of the fact that two consecutive subdivided periods correspond to one cycle, in the example shown in FIG. 7, the cycle determination unit 113 determines whether one of the two continuous silent sections is a silent section. Is regarded as the boundary of the cycle, it can be said that the cycle of the respiratory sound signal is determined.

  When the silent section is regarded as the boundary of the period or the subdivision period, the time interval between the silent sections serving as the boundary is preferably equal to or longer than a certain period. That is, it is preferable that the cycle determination unit 113 regards any one of the two silent sections in which the time interval between them is less than a certain period as not a boundary of the cycle or the subdivision period. As a result, even in a case where an abnormally small instantaneous value is mixed in the respiratory sound signal, the cycle determination unit 113 can appropriately determine at least one of the cycle and the subdivision period based on the silent section. it can.

  Alternatively, as illustrated in FIGS. 8A and 8B, the cycle determination unit 113 determines the cycle and the subdivision period of the respiratory sound signal (that is, the expiration period) based on the similarity of the waveform of the respiratory sound signal. And the intake period) may be determined. Specifically, when a waveform having a high similarity (for example, relatively high or higher than a predetermined frequency) is repeated, the period in which each waveform appears is determined by the period and the subdivision. It may be determined that it is at least one of the periods. For example, in the example illustrated in FIG. 8A, the cycle determination unit 113 determines that the waveform A, the waveform A ′, and the waveform A ″ are similar (that is, the similarity is high), and the determination is performed. Based on the result, it is determined that the period in which each of the waveform A, the waveform A ′, and the waveform A ″ appears is one intake period. Similarly, in the example illustrated in FIG. 8A, the cycle determination unit 113 determines that the waveform B and the waveform B ′ are similar (that is, the similarity is high), and based on the determination result. , B, and B ′ are determined to be one expiration period. Alternatively, in the example illustrated in FIG. 8B, the cycle determination unit 113 determines that the waveform a, the waveform a ′, and the waveform a ″ are similar (that is, the similarity is high), and the determination is performed. Based on the result, it is determined that the period during which each of the waveform a, the waveform a ′, and the waveform a ″ appears is one cycle.

  Thereafter, the signal analysis unit 103 performs a predetermined peak hold process for detecting a peak value of the respiratory sound signal based on the result of the cycle determination process executed by the cycle determination unit 113 (Step S103b). In the following, the signal analysis unit 103 calculates the peak value of the respiratory sound signal for each subdivision period (that is, the peak value of the respiratory sound signal for each expiration period and the peak value of the respiratory sound signal for each inspiration period). It shall be detected.

  In addition, the signal analysis unit 103 calculates the peak of each of the five types of signal components (i.e., alveolar respiratory component, snoring component, whistle component, crunch component, and blister component) included in the respiratory signal. Preferably, the value is detected. However, the signal analyzer 103 may detect the peak value of the respiratory sound signal itself.

  As a result, for example, as shown in FIG. 9, the signal analysis unit 103 calculates the peak value P11 of the alveolar respiratory sound component in the inspiratory period # 1, the peak value P21 of the snoring sound component in the inspiratory period # 1, and the inspiratory period # 1, the peak value P31 of the whistling sound component, the peak value P41 of the twisting sound component in the intake period # 1, and the peak value P51 of the water bubble component in the intake period # 1 are detected. Thereafter, the signal analysis unit 103 calculates the peak value P12 of the alveolar respiratory sound component during the expiration period # 1, the peak value P22 of the snoring sound component during the expiration period # 1, the peak value P32 of the whistling sound component during the expiration period # 1, The peak value P42 of the twisting sound component in the expiration period # 1 and the peak value P52 of the water bubble sound component in the expiration period # 1 are detected. Thereafter, the signal analysis unit 103 calculates the peak value P13 of the alveolar respiratory sound component during the inspiration period # 2, the peak value P23 of the snoring sound component during the inspiration period # 2, the peak value P33 of the whistling sound component during the inspiration period # 2, The peak value P43 of the twisting sound component in the intake period # 2 and the peak value P53 of the water bubble sound component in the intake period # 2 are detected. Thereafter, the signal analysis unit 103 calculates the peak value P14 of the alveolar respiratory sound component during the expiration period # 2, the peak value P24 of the snoring sound component during the expiration period # 2, the peak value P34 of the whistling sound component during the expiration period # 2, The peak value P44 of the twisting sound component in the expiration period # 2 and the peak value P54 of the water bubble sound component in the expiration period # 2 are detected.

  In the example illustrated in FIG. 9, the signal analyzer 103 determines the peak value of the respiratory sound signal for each subdivision period (that is, the peak value of the respiratory sound signal for each expiration period and the respiratory sound signal for each inspiration period). Peak value). However, as shown in FIG. 10A, the signal analyzer 103 may detect peak values of five types of signal components (or respiratory sound signals) for each cycle. Alternatively, as shown in FIG. 10B, the signal analysis unit 103 detects peak values of five types of signal components (or respiratory sound signals) every two cycles (or every three or more cycles). May be. Alternatively, as illustrated in FIG. 10C, the signal analysis unit 103 outputs five types of signal components (or respiratory sound signals) in three subdivision periods (or two or more subdivision periods). The peak value may be detected.

  In parallel with the analysis performed by the signal analyzer 103, the display data generator 104 generates five types of signal components (ie, alveolar respiratory sound component, snoring sound component, whistling sound component, crouching sound component, and so on). Display data for displaying the respective characteristics (specifically, the instantaneous value and peak value of the amplitude level) of the water bubble sound component on the display unit 105 is generated (step S104). Note that the display data is data indicating a display object displayed by the display unit 105. Such display data is, for example, an image signal indicating at least one of a text (for example, a character or a numerical value or the like) and a graphic (for example, a graph or a chart) for visually specifying the characteristic of the respiratory sound signal (or , An arbitrary signal having a format different from the image signal).

  At this time, the display data generation unit 104 generates display data based on the result of the analysis performed by the signal analysis unit 103. The display data generation unit 104 outputs the generated display data to the display unit 105 as appropriate.

  In the following description, the description will be given using an example in which the characteristics of the signal components displayed on the display unit 105 are amplitude levels (that is, characteristics indicating signal strength). However, it goes without saying that the characteristic of the signal component may be a characteristic different from the amplitude level (in other words, an index (variable) that can specify the state of the signal component). In this case, it is preferable that the signal analysis unit 103 performs the above-described peak hold processing on a characteristic different from the amplitude level (that is, detects a peak value having a characteristic different from the amplitude level).

  For example, as shown in FIGS. 11A to 11D, the display data generation unit 104 uses a bar graph to display the instantaneous value and peak value of each of the five types of signal components. Display data may be generated. In the example shown in FIGS. 11A to 11D, five bars corresponding to five types of signal components are arranged in order along the horizontal axis. In addition, in the examples shown in FIGS. 11A to 11D, as the amplitude level increases, the length of the bar increases along the vertical axis. In addition, in the examples shown in FIGS. 11A to 11D, in addition to the instantaneous value indicated by the bar, the peak value indicated by the line is also displayed.

  FIGS. 11A to 11D show examples of bar graphs showing the instantaneous value and peak value of the amplitude level of each of the five types of signal components shown in FIG. For example, FIG. 11A shows an example of a bar graph showing the instantaneous value and peak value of the amplitude level of each of the five types of signal components in the intake period # 1. Therefore, when the current time is a time included in the intake period # 1, display data for displaying the bar graph shown in FIG. 11A is generated, and the bar graph shown in FIG. Is displayed on the display unit 105.

  Similarly, FIG. 11B shows an example of a bar graph showing the instantaneous value and peak value of each of the five types of signal components in the expiration period # 1. Therefore, when the current time is a time included in the expiration period # 1, display data for displaying the bar graph shown in FIG. 11B is generated, and the bar graph shown in FIG. Is displayed on the display unit 105. In other words, when the signal analysis unit 103 newly detects the peak value of the amplitude level of each of the five types of signal components in the expiration period # 1, the image data generation unit 104 sets the five types of signal components in the inspiration period # 1. A display for displaying a bar graph showing the instantaneous value and the peak value of each of the five types of signal components in the expiration period # 1 instead of the bar graph showing the instantaneous value and the peak value of each of the signal components. Create new data. As a result, the display unit 105 replaces the bar graph showing the instantaneous value and peak value of each of the five types of signal components in the inspiration period # 1 with the respective amplitudes of the five types of signal components in the expiration period # 1. A bar graph showing the instantaneous value and peak value of the level is displayed.

  Similarly, FIG. 11C shows an example of a bar graph displaying the instantaneous value and the peak value of the amplitude level of each of the five signal components in the intake period # 2. Therefore, when the current time is a time included in the intake period # 2, display data for displaying the bar graph shown in FIG. 11C is generated, and the bar graph shown in FIG. Is displayed on the display unit 105.

  Similarly, FIG. 11D shows an example of a bar graph displaying the instantaneous value and the peak value of each of the five types of signal components in the expiration period # 2. Therefore, when the current time is a time included in the expiration period # 2, display data for displaying the bar graph shown in FIG. 11D is generated, and the bar graph shown in FIG. Is displayed on the display unit 105.

  Of course, the display data generating unit 104 displays the instantaneous value and peak value of each of the five types of signal components in a display format different from the display formats shown in FIGS. 11A to 11D. May be generated. Further, the display data generation unit 104 generates display data for displaying either one of the instantaneous value and the peak value of the amplitude level of each of the five types of signal components while not displaying one of them. You may.

  In parallel with the generation of the display data by the display data generation unit 104, the display unit 105 performs a display process based on the display data generated by the display data generation unit 104 (Step S105). As a result, the display unit 105 displays the instantaneous value and the peak value of the amplitude level of each of the five types of signal components. For example, the display unit 105 displays the instantaneous value and the peak value of the amplitude level of each of the five types of signal components in the format shown in FIGS. 11A to 11D.

  The operations from step S101 to step S105 described above are repeated until the operation of the signal processing device (10) ends (step S106). At this time, when the operations from step S101 to step S105 are repeated, it is preferable that the operations from step S101 to step S105 are performed periodically, aperiodically, or continuously. That is, it is preferable that each operation from step S101 to step S105 is performed in real time or dynamically.

  As described above, according to the signal processing device 10 of the present embodiment, the period in which the peak value is detected by the peak hold processing performed by the signal analysis unit 103 is n continuous cycles or m continuous cycles. Coincides with successive subdivision periods. Therefore, no matter what the cycle of the respiratory sound signal is, a time lag occurs between the cycle or the subdivision period of the respiratory sound signal and the period for which the peak value is to be detected. There is little or no fear. Therefore, in the present embodiment, the signal analysis unit 103 can detect the peak value of the respiratory sound signal appropriately associated with the cycle or the subdivision period (that is, the inspiratory period and the expiratory period) of the respiratory sound signal. . As a result, the display unit 105 can display the peak value of the respiratory sound signal suitably associated with the cycle or the subdivision period (that is, the inspiratory period and the expiratory period) of the incoming breathing sound signal.

  Further, the present invention can be appropriately changed within a scope not contrary to the gist or idea of the present invention which can be read from the claims and the entire specification, and a signal processing device and method with such a change, a computer program and a recording The medium is also included in the technical concept of the present invention.

Reference Signs List 10 signal processing device 101 signal acquisition unit 102 signal storage unit 103 signal analysis unit 104 display data generation unit 105 display unit 113 cycle determination unit

Claims (7)

Detecting means for detecting a respiratory sound component and an auxiliary noise component included in the input respiratory sound signal having periodicity ,
A first display object indicating an instantaneous value and a peak value of the respiratory sound component detected by the detection means, and a second display object indicating an instantaneous value and a peak value of the auxiliary noise component are displayed on a display device. Output means for outputting so that they are displayed side by side in different areas of the
It said peak value to a signal processing apparatus according to claim der Rukoto indicates the maximum value of the instantaneous value for each cycle of breath. Detecting means for detecting a continuous ra component and an intermittent la component contained in the input respiratory sound signal having periodicity ,
A first display object indicating an instantaneous value and a peak value of the continuous lane component detected by the detection means and a second display object indicating an instantaneous value and a peak value of the intermittent lane component are displayed. Output means for outputting so as to be displayed side by side in different areas of the display unit of the device ,
It said peak value to a signal processing apparatus according to claim der Rukoto indicates the maximum value of the instantaneous value for each cycle of breath. Detection means for detecting a snoring sound component and a whistling sound component that are continuous ra sound components included in the input respiratory sound signal having periodicity ,
A first display object indicating the instantaneous value and the peak value of the similar snoring sound component detected by the detection means, and a second display object indicating the instantaneous value and the peak value of the whistling sound component. Output means for outputting so as to be displayed side by side in different areas of the display unit ,
It said peak value to a signal processing apparatus according to claim der Rukoto indicates the maximum value of the instantaneous value for each cycle of breath. Detecting means for detecting a crunch sound component and a water bubble sound component which are intermittent ra sound components included in the input respiratory sound signal having periodicity ,
A first display object indicating the instantaneous value and the peak value of the crunch sound component detected by the detection means, and a second display object indicating the instantaneous value and the peak value of the water bubble sound component, Output means for outputting so as to be displayed side by side in different areas of the display unit ,
It said peak value to a signal processing apparatus according to claim der Rukoto indicates the maximum value of the instantaneous value for each cycle of breath. Detecting means for detecting a breathing sound component, a snoring sound component, a whistling sound component, a crunch sound component, and a blister sound component included in the input breathing sound signal having periodicity ,
A first display object indicating the instantaneous value and peak value of the respiratory sound component detected by the detection means, a second display object indicating the instantaneous value and peak value of the snoring sound component, and a whistle sound component; A third display object indicating the instantaneous value and the peak value, a fourth display object indicating the instantaneous value and the peak value of the crunch sound component, and a fifth display object indicating the instantaneous value and the peak value of the blister sound component. Output means for outputting the display object of the display so as to be displayed side by side in different areas of the display unit of the display device,
It said peak value to a signal processing apparatus according to claim der Rukoto indicates the maximum value of the instantaneous value for each cycle of breath. The instantaneous value indicates one value of a certain timing for each cycle of breathing,
The detecting means detects a respiratory cycle,
The signal according to any one of claims 1 to 5, wherein the output unit outputs the peak value displayed on the display device every time the respiratory cycle changes. Processing equipment. A detection step of detecting a respiratory sound component and an auxiliary noise component included in the input respiratory sound signal having periodicity ,
A first display object indicating an instantaneous value and a peak value of the respiratory sound component detected in the detection step, and a second display object indicating an instantaneous value and a peak value of the auxiliary noise component, And an output step of outputting so as to be displayed side by side in different areas of the display unit.
The peak value, a signal processing method according to claim der Rukoto indicates the maximum value of the instantaneous value for each cycle of breath.

Images