KR101800739B1 - Apparatus and method for detecting respiration rate - Google Patents

Abstract

The present invention relates to a device and a method for detecting a breathing rate, which remove invalid peak and valley pairs by considering a difference of a peak value and a valley value of the peak and the valley pairs detected from a measured waveform and a size and an interval of the waveform of the peak and the valley pairs, and detect a breathing rate, so that an accurate breathing rate can be calculated even when a shape of the measured waveform for detecting breathing is irregular. The device for detecting a breathing rate comprises: a first filter part; a second filter part; and a breathing rate calculation part.

Description

[0001] The present invention relates to an apparatus and method for detecting respiration,

Field of the Invention [0002] The present invention relates to a breath detection technique, and more particularly, to a respiration rate detection apparatus and method.

Korean Patent Laid-Open Publication No. 10-2012-0045664 (2012.05.09) detects a respiration signal that has passed through a low-pass filter in a predetermined frequency band, and detects that the time difference between the closest valley among the peaks of the detected respiration signal is a predetermined Time is selected and the number of the selected peaks per unit time is measured by the number of breaths.

As shown in FIG. 1, when the waveform of the measurement waveform for respiration detection is regular, the measurement by the respiration rate is relatively easy and accurate. However, as shown in FIG. 2, the shape of the measured waveform for the breath detection is often not regular even if the measured person maintains a stable state. Therefore, even when the waveform of the measured waveform for breath detection is irregular, Is required.

Korean Patent Publication No. 10-2012-0045664 (2012.05.09)

It is an object of the present invention to provide a respiratory rate detecting apparatus and method capable of calculating an accurate respiration rate even when the shape of a measurement waveform for respiration detection is irregular.

According to one aspect of the present invention for achieving the above object, there is provided a respiration detection apparatus for detecting a peak value of a pair of peaks and valleys detected from a measurement waveform for respiration detection and an apparatus for removing peak and valley pairs having a valley value difference lower than a threshold value 1 filter unit; A second filter portion for removing peak and valley pairs located in the non-dense region in consideration of the waveform size and spacing for the remaining pairs of peaks and valleys excluding the pairs of peaks and valleys removed by the first filter portion; A respiration rate calculation unit for calculating a respiration rate using the remaining pairs of peaks and valleys excluding the pairs of peaks and valleys removed by the second filter unit; And the like.

According to a further aspect of the present invention, the waveform size of the remaining pairs of peaks and valleys excluding the pairs of peaks and valleys removed by the first filter section by the second filter section is calculated by interquartile range (IQR) A size quadrature range arithmetic unit for removing peak and valley pairs having an outlier out of normal distribution; .

According to a further aspect of the present invention, the second filter unit calculates the waveform interval of the peak and valley pairs from which peak and valley pairs having an outlier whose waveform size is out of the normal distribution are removed by the magnitude quadrature range arithmetic unit (IQR: Interquartile Range) to remove peak and valley pairs having an outlier whose waveform interval is out of the normal distribution; And further comprising:

According to a further aspect of the present invention, the quadrature range (IQR) for the waveform sizes of the peak and valley pairs until the size quadrature range arithmetic operation part finds outliers whose waveform size is out of the normal distribution, And the operation is repeated.

According to a further aspect of the present invention, the quadrature range (IQR) for the waveform intervals of the peak and valley pairs is measured until the interval quadrature range arithmetic operation section detects an outlier whose waveform interval is out of the normal distribution, And the operation is repeated.

According to a further aspect of the present invention, the first filter section normalizes the difference between the peak value and the valley value, and removes the peak and valley pairs whose difference between the normalized peak value and the valley value is less than the threshold value.

According to a further aspect of the present invention, when the number of peak and valley pairs in which the difference between the peak value and the valley value is less than the threshold value is greater than or equal to a certain percentage, the threshold value is adjusted and filtered again.

According to a further aspect of the present invention, when the number of peak and valley pairs in which the difference between the peak value and the valley value is less than the adjusted threshold value is greater than or equal to a predetermined percentage, the first filter section determines the measurement error.

According to a further aspect of the present invention, there is provided an alarm apparatus, comprising: an error warning unit for performing a warning when the breathing rate detection apparatus generates a measurement error; And further comprising:

According to a further aspect of the present invention, there is provided a breathing apparatus comprising: a breathing number display unit for displaying breathing counts calculated by the breathing amount calculation unit; And further comprising:

According to a further aspect of the present invention, there is provided a respiration detection apparatus comprising: a peak and valley pair detection section for detecting a peak and a pair of valleys by differentiating a measured waveform for breath detection; A peak / valley difference calculation unit for calculating a difference between a peak value and a valley value included in pairs of peaks and valleys detected by the peak and valley pair detection unit; And further comprising:

According to a further aspect of the present invention, there is provided a measurement waveform obtaining unit in which the breathing rate detecting apparatus obtains a measured waveform for breath detection; And further comprising:

According to a further aspect of the present invention, the respiratory rate detection apparatus includes a resampler section for sampling the measured waveform acquired by the measured waveform acquisition section in a specific time unit; And further comprising:

According to a further aspect of the present invention, the resampling unit performs a moving average calculation on the resampled measured waveform data.

According to a further aspect of the present invention, the measured waveform for the breath detection is obtained by detecting a change in the biometric information.

According to a further aspect of the present invention, there is provided a method for controlling a living body, wherein the biological information change is a change in temperature during respiration or a change in body volume caused by air flow or pulse or blood pressure or blood volume or respiration due to respiration or sound change due to bodily injury, snoring, And is at least one.

According to another aspect of the present invention, there is provided a method of detecting a respiratory tract comprising: detecting a pair of peaks and valleys for detecting pairs of peaks and valleys from a measurement waveform for respiration detection; A peak / valley difference calculation step of calculating a difference between a peak value and a valley value included in the peak and valley pairs detected by the peak and valley pair detection step; A first filtering step of removing peak and valley pairs whose difference between the peak value and the valley value calculated by the peak / valley difference calculation step is less than the threshold; A second filtering step for removing peak and valley pairs located in the non-dense region taking into account the waveform size and spacing for the remaining peak and valley pairs excluding the peak and valley pairs removed by the first filtering step; Calculating a number of breaths using the remaining pairs of peaks and valleys excluding the pairs of peaks and valleys removed by the second filtering step; And the like.

According to a further aspect of the present invention, the second filtering step comprises calculating the waveform sizes of the remaining pairs of peaks and valleys excluding the pairs of peaks and valleys removed by the first filtering step, in a quadrature range (IQR: Interquartile Range) A magnitude quadrature range computation step of removing peak and valley pairs with size outliers out of normal distribution; Calculating a quadrature range (IQR: Interquartile Range) of the waveform intervals of the peak and valley pairs from which peak and valley pairs having an outlier whose waveform size deviates from the normal distribution by the size quadrature range calculation step, An interval interval quadrature range calculation step of removing peak and valley pairs having an outlier whose interval is out of normal distribution; .

According to a further aspect of the present invention, the quadrature range (IQR) for the waveform sizes of the peak and valley pairs until the waveform size is out of the normal distribution, ) Operation is repeated (iteration).

According to a further aspect of the present invention, in the calculating of the interval quartile range, the quadrature range (IQR) of the waveform intervals of the peak and valley pairs until an outlier whose waveform interval deviates from the normal distribution is not detected ) Operation is repeated (iteration).

The present invention has an effect of accurately calculating the breathing number even when the waveform of the measurement waveform for breath detection is irregular.

Further, the present invention has an effect applicable to various types of respiratory waveforms that may appear in actual clinical practice.

FIG. 1 is a waveform diagram showing an example of a case where a waveform of a measurement waveform for respiration detection is regular.
FIG. 2 is a waveform diagram showing an example of a case in which the waveform of the measurement waveform for respiration detection is irregular.
3 is a block diagram showing an embodiment of a breathing apparatus according to the present invention.
FIG. 4 is a flowchart illustrating an embodiment of a breathing detection method according to the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the like elements throughout.

In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

The terms used throughout the specification of the present invention have been defined in consideration of the functions of the embodiments of the present invention and can be sufficiently modified according to the intentions and customs of the user or operator. It should be based on the contents of.

3 is a block diagram showing an embodiment of a breathing apparatus according to the present invention. 3, the respiratory rate detection apparatus 100 according to this embodiment includes a peak and valley pair detection unit 110, a peak / valley difference calculation unit 120, a first filter unit 130, A second filter unit 140, and a respiratory rate calculation unit 150.

The peak and valley pair detection unit 110 detects peak and valley pairs from a measurement waveform for breath detection. For example, the peak and valley pair detecting section 110 may be implemented to differentiate the measured waveform to detect peak and valley pairs.

The peak points in the measured waveform are points located in the section where the gradient of the measured waveform is differentiated from positive to negative and the valley points are points located in the section where the gradient changes from negative to positive when the measured waveform is differentiated. The peak and valley points can be obtained by differentiating the measured waveform. Then, peak and valley pairs can be detected by determining the peak point and the valley point neighboring in time with a pair of peak and valley pairs.

On the other hand, the measured waveform for the breath detection can be obtained by detecting a change in the biometric information. The biological information change may be at least one of a temperature change during respiration or an air flow caused by respiration or a change in body volume caused by a pulse or blood pressure, blood volume, or respiration, or a change in sound due to breathing, snoring, or discoloration.

The peak / valley difference calculation unit 120 calculates the difference between the peak value and the valley value included in the peak and valley pairs detected by the peak and valley pair detection unit 110. The reason for calculating the difference between the peak value and the valley value included in the peak and valley pairs is that the peak point and the valley point included in the peak and valley pair correspond to one breath.

The first filter unit 130 removes peak and valley pairs whose difference between the peak value and the valley value calculated by the peak / valley difference calculation unit 120 is less than the threshold value. The peak and valley pair detecting section 110 removes the peak and valley pairs having a difference between the peak value and the valley value less than the threshold value in order to remove the noise signal which is not correlated with normal breathing when detecting the peak and valley pair through the first differential.

In this case, the first filter unit 130 may normalize the difference between the peak value and the valley value, and remove the peak and valley pair having the difference between the normalized peak value and the valley value less than the threshold value. Normalizing the difference between the peak value and the valley value means that the difference between the peak value and the valley value is converted into a value within a predetermined normalization interval. For example, the difference between the peak value and the valley value can be normalized to a range from -1 to 1.

On the other hand, if the number of peak and valley pairs in which the difference between the peak value and the valley value is less than the threshold value is greater than or equal to a predetermined percentage, the first filter unit 130 may be configured to adjust and refine the threshold value.

For example, if the detected peak and valley pairs are 100, the difference between the peak value and the valley value is normalized to a section from -1 to 1, the initial threshold is set to -0.5, the adjustment threshold is set to -0.8, If the specific value is set at 50%, the threshold value is adjusted to -0.8 when the number of peak and valley pairs having a difference between the peak value and the valley value less than -0.5 is less than 50%, that is, 50 or more.

If the number of peak and valley pairs having a difference between the peak value and the valley value of less than -0.8 is detected and the number of peak and valley pairs is less than 50%, that is, less than 50, the peak and valley pair Lt; / RTI >

On the other hand, when the first filter unit 130 determines that the number of pairs of peaks and valleys whose difference between the peak value and the valley value is less than the threshold value is greater than or equal to a predetermined percentage, the first filter unit 130 determines that the error is a measurement error. That is, it is judged that the peak and valley pair samples for the respiration rate measurement are not sufficient.

The second filter unit 140 calculates peak values and valleys located in the non-dense region in consideration of the waveform size and spacing for the remaining peak and valley pairs except for the peak and valley pairs removed by the first filter unit 130. [ Remove pairs.

For example, if the measured waveform is a measure of the inspiratory and temporal changes in temperature at the time of the person to be measured, the waveform size of the peak and valley pairs corresponds to the intensity of the respiration and the adjacent peak and valley pair interval corresponds to the respiratory interval The peak and valley pairs having abnormally large or small waveform magnitudes, that is, the magnitudes of the respiratory waves, and the peak and valley pairs, that is, the peak and valley pairs having abnormally large or small respiratory intervals, Remove.

The respiration rate calculation unit 150 calculates the respiration rate using the remaining pairs of the peak and valley except for the peak and valley pairs removed by the second filter unit 140. Since the peak and valley pairs correspond to one breath, the number of remaining peak and valley pairs can be counted to determine the number of breaths. In this case, the respiratory rate calculator 150 may calculate the number of breaths per minute by converting the number of remaining peak and valley pairs into a 60-second interval.

By implementing such a method, the present invention can calculate the accurate breathing number even when the waveform of the measurement waveform for breath detection is irregular. The present invention is also applicable to various types of breathing waveforms that may appear in actual clinical practice.

According to an additional aspect of the present invention, the second filter unit 140 may include a size quadrature range arithmetic operation unit 141. The magnitude quadrature range arithmetic operation unit 141 computes a quadrature range (IQR) of the waveform sizes of the remaining peak and valley pairs excluding the peak and valley pairs removed by the first filter unit 130, To remove peak and valley pairs with an outlier that deviates from the normal distribution. In this case, the weight can be further applied in the calculation of the quadrature range (IQR).

For example, the waveform size of the peak and valley pairs is sorted from the smallest waveform size through the IQR (Interquartile Range) calculation for the waveform sizes of the peak and valley pairs, so that the waveform size ranges from 25% -a to 75% + a Peak and valley pairs that are likely to be unlikely to belong to the range of waveform sizes below 25% -a and above 75% + a, leaving only the peak and valley pairs to which they belong, are removed to yield peak and valley pairs with reliable waveform size Can be obtained. The weight a may be 1.5 * IQR. Here, IQR is in the range of 25% to 75%.

At this time, the size quadrangle range arithmetic operation unit 141 repeats the quadrature range range (IQR) operation for the waveform sizes of the peak and valley pairs until an outlier whose waveform size is out of the normal distribution is not detected it may be implemented to iterate.

That is, for the peak and valley pairs left over from the quadrature range (IQR) operation, the quadrature range (IQR) operation is repeated until no outliers whose waveform size deviates from the normal distribution are detected, Peak and valley pairs with dimensions can be obtained.

According to a further aspect of the present invention, the second filter unit 140 may further include an interval quadrature range arithmetic unit 142. The interval quadrature range arithmetic operation unit 142 calculates the waveform quadrature intervals of the peak and valley pairs from which peak and valley pairs having an outlier whose waveform size is out of the normal distribution are removed by the size quadrature range arithmetic operation unit 141 Calculate the IQR (Interquartile Range) to remove peak and valley pairs with outliers whose waveform intervals are outside the normal distribution. In this case, the weight can be further applied in the calculation of the quadrature range (IQR).

For example, a waveform interval between peak and valley pairs is sorted from a waveform interval of 25% -a to 75% + a through a quartile range (IQR) calculation for waveform intervals of peak and valley pairs Peak and valley pairs that are likely to be unlikely to fall within the range of waveform intervals less than 25% -a and greater than 75% + a, leaving only the peak and valley pairs to which they belong, are removed to yield peak and valley pairs with reliable waveform spacing Can be obtained. The weight a may be 1.5 * IQR. Here, IQR is in the range of 25% to 75%.

At this time, the interval quadrature range arithmetic operation unit 142 repeatedly performs a quadrature range (IQR) operation on the waveform intervals of the peak and valley pairs until an outlier whose waveform interval is out of the normal distribution is not detected it may be implemented to iterate.

That is, for the peak and valley pairs left over from the quadrature range (IQR) operation, the quadrature range (IQR) operation is repeated until no outliers whose waveform interval deviates from the normal distribution are detected, Peak and valley pairs with spacing can be obtained.

On the other hand, by performing a quadrature range (IQR) operation on the waveform interval after the operation of the quadrature range (IQR) for the relatively large waveform size in the breath count calculation, a peak and valley pair having a more reliable waveform interval Can be obtained.

According to a further aspect of the invention, the respiratory rate detection apparatus 100 may further include an error warning unit 160. [ The error warning unit 160 performs a warning when a measurement error occurs. For example, when the error warning unit 160 generates a measurement error, it can be implemented to perform audible warning through a visual warning or an alarm output through an LED or a screen display. Thereby, in accordance with the measurement error warning by the error warning section 160, the measurer can recognize the measurement error.

According to an additional aspect of the present invention, the breathing rate detection apparatus 100 may further include a breathing rate display unit 170. The breathing rate display unit 170 displays the breathing rate calculated by the breathing rate calculation unit 150. For example, the respiratory rate display unit 170 may be configured to display the respiration rate per minute through an LCD or an LED screen. Accordingly, the measurer can check the respiratory rate of the subject displayed by the respiratory rate display unit 170 and know the respiratory rate of the subject.

According to an additional aspect of the present invention, the respiratory rate detection apparatus 100 may further include a measurement waveform acquisition unit 180. [ The measurement waveform acquisition unit 180 acquires a measurement waveform for breath detection.

At this time, the measurement waveform acquisition unit 180 may include various sensors for detecting changes in the biometric information, such as a temperature measurement sensor, a flow measurement sensor, a pulse measurement sensor, a blood pressure measurement sensor, a blood volume measurement sensor, Lt; RTI ID = 0.0 > a < / RTI > breathing detection.

For example, when the change in the biometric information detected by the measurement waveform obtaining unit 180 is a change in temperature during respiration or a change in body volume due to air flow or pulse or blood pressure, blood volume, or respiration caused by breathing or body movement or snoring or irregularity Lt; / RTI > and / or < / RTI > Accordingly, various types of respiratory waveforms that may appear in actual clinical practice can be acquired as measurement waveforms for respiration detection.

According to a further aspect of the present invention, the respiratory rate detection apparatus 100 may further include a resampler 190. The resampling unit 190 resamples the measurement waveform acquired by the measurement waveform acquisition unit 180 by a specific time unit. At this time, the resampling unit 190 may be configured to perform a moving average operation on the resampled measured waveform data.

The peak and valley pair detecting section 110 detects peak and valley pairs from the measured waveform by the measured waveform data resampled by the resampling section 190 and outputs the peak and valley pair to the peak / valley difference calculating section 120 The difference between the peak value and the valley value included in the peak and valley pairs detected by the first filter unit 130 and the second filter unit 140 is calculated and the remaining peak after the invalid peak and valley pairs are removed through the first filter unit 130 and the second filter unit 140 And respiratory rate calculation unit 150 calculates the breath count using the pair of valleys.

The process of detecting the breathing number of the breathing apparatus according to the present invention will now be described with reference to FIG. FIG. 4 is a flowchart illustrating an embodiment of a breathing detection method according to the present invention.

First, in the measurement waveform acquisition step 410, the breathing detection apparatus acquires a measurement waveform for breath detection. At this time, measurement waveforms for breath detection are detected through various sensors for detecting changes in biometric information, such as a temperature measurement sensor, a flow measurement sensor, a pulse measurement sensor, a blood pressure measurement sensor, a blood volume measurement sensor, a volumetric measurement sensor, Can be obtained.

For example, the biometric information change may be at least one of a temperature change at the time of breathing, an air flow caused by breathing, a pulse or blood pressure, a body volume change due to blood volume or respiration, or an acoustic change due to bodily or snoring or irritation.

Then, in resampling step 420, the respiratory rate detection device samples the measured waveform acquired by the measured waveform acquisition step 410 in a specific time unit. In this case, the moving average may be calculated for the resampled measured waveform data in the resampling step 420.

Next, in the peak and valley pair detection step 430, the respiratory rate detection device detects pairs of peaks and valleys from the measurement waveform for respiration detection. For example, the peak and valley pair detection step 430 may be implemented to differentiate the measured waveform to detect peak and valley pairs.

Then, in the peak / valley difference calculation step 440, the breath detection apparatus calculates the difference between the peak value and the valley value included in the peak and valley pairs detected by the peak and valley pair detection step 430. [ The reason for calculating the difference between the peak value and the valley value included in the peak and valley pairs is that the peak point and the valley point included in the peak and valley pair correspond to one breath.

Then, in a first filtering step 450, the respiratory rate detection device removes the peak and valley pairs for which the difference between the peak value and the valley value calculated by the peak / valley difference calculation step 440 is less than the threshold. The peak and valley pair detection step 430 removes the peak and valley pairs that have a difference between the peak value and the valley value below the threshold to remove noise signals that are not correlated with normal breathing when detecting peak and valley pairs through the first derivative.

In this case, in the first filtering step 450, the difference between the peak value and the valley value may be normalized, and the peak and valley pairs whose difference between the normalized peak value and the valley value are less than the threshold value may be removed. Normalizing the difference between the peak value and the valley value means that the difference between the peak value and the valley value is converted into a value within a predetermined normalization interval. For example, the difference between the peak value and the valley value can be normalized to a range from -1 to 1.

Meanwhile, if the number of peak and valley pairs having a difference between the peak value and the valley value in the first filtering step 450 is equal to or greater than a predetermined percentage, the threshold may be adjusted and filtered.

For example, if the detected peak and valley pairs are 100, the difference between the peak value and the valley value is normalized to a section from -1 to 1, the initial threshold is set to -0.5, the adjustment threshold is set to -0.8, If the specific value is set to 50%, the threshold value is adjusted to -0.8 when the number of peak and valley pairs having a difference between the peak value and the valley value of less than -0.5 is 50% or more, that is, 50 or more.

If the number of peak and valley pairs having a difference between the peak value and the valley value of less than -0.8 is detected and the number of peak and valley pairs is less than 50%, that is, less than 50, the peak and valley pair Lt; / RTI >

On the other hand, if the number of peak and valley pairs less than the threshold value in which the difference between the peak value and the valley value is adjusted in the first filtering step 450 is greater than or equal to a predetermined percentage, it is determined to be a measurement error. That is, it is judged that the peak and valley pair samples for the respiration rate measurement are not sufficient.

Then, in a second filtering step 460, the respiratory rate detecting device determines the remaining peak and valley pairs excluding the peak and valley pairs removed by the first filtering step 450, ≪ / RTI > remove the peak and valley pairs located in the region.

For example, if the measured waveform is a measure of the inspiratory and temporal changes in temperature at the time of the person to be measured, the waveform size of the peak and valley pairs corresponds to the intensity of the respiration and the adjacent peak and valley pair interval corresponds to the respiratory interval The second filtering step 460 is performed to determine peak and valley pairs that are abnormally large or small in waveform size, that is, respiration intensity, and peak and valley pairs, that is, pairs of peaks and valleys with abnormally large or small respiratory intervals Remove.

Meanwhile, the second filtering step 460 may include a magnitude quadrature range calculation step 461 for calculating a quadrature range (IQR) of the waveform sizes of the peak and valley pairs, And an interval quartile range calculation step 462 for performing a quintile range (IQR) operation.

In the magnitude quadrature range calculation step 461, the respiratory quantity detection apparatus sets the waveform sizes of the remaining peak and valley pairs except for the peak and valley pairs removed by the first filtering step 450 to a quadrature range (IQR) ) To remove peak and valley pairs with outliers whose waveform size is outside the normal distribution. In this case, the weight can be further applied in the calculation of the quadrature range (IQR).

For example, the waveform size of the peak and valley pairs is sorted from the smallest waveform size through the IQR (Interquartile Range) calculation for the waveform sizes of the peak and valley pairs, so that the waveform size ranges from 25% -a to 75% + a Peak and valley pairs that are likely to be unlikely to belong to the range of waveform sizes below 25% -a and above 75% + a, leaving only the peak and valley pairs to which they belong, are removed to yield peak and valley pairs with reliable waveform size Can be obtained. The weight a may be 1.5 * IQR. Here, IQR is in the range of 25% to 75%.

At this time, the quadrature range (IQR) calculation for the waveform sizes of the peak and valley pairs is performed until the outlier of the waveform size is out of the normal distribution in the magnitude quadrature range calculation step 461 It can be implemented to iterate.

That is, for the peak and valley pairs left over from the quadrature range (IQR) operation, the quadrature range (IQR) operation is repeated until no outliers whose waveform size deviates from the normal distribution are detected, Peak and valley pairs with dimensions can be obtained.

In the interval quartile range calculating step 462, the respiratory rate detecting apparatus calculates the peak quadrangle having the peak out of the normal distribution by the amplitude quadrangle range calculating step 461, And IQR (Interquartile Range) of the waveform intervals of the valley pairs to remove peak and valley pairs with outliers whose waveform intervals are outside the normal distribution. In this case, the weight can be further applied in the calculation of the quadrature range (IQR).

For example, a waveform interval between peak and valley pairs is sorted from a waveform interval of 25% -a to 75% + a through a quartile range (IQR) calculation for waveform intervals of peak and valley pairs Peak and valley pairs that are likely to be unlikely to fall within the range of waveform intervals less than 25% -a and greater than 75% + a, leaving only the peak and valley pairs to which they belong, are removed to yield peak and valley pairs with reliable waveform spacing Can be obtained. The weight a may be 1.5 * IQR. Here, IQR is in the range of 25% to 75%.

At this time, the quadrature range (IQR) operation for the waveform intervals of the peak and valley pairs is performed until the outlier of the waveform interval is out of the normal distribution in the interval quartile range calculation step 462 It can be implemented to iterate.

That is, for the peak and valley pairs left over from the quadrature range (IQR) operation, the quadrature range (IQR) operation is repeated until no outliers whose waveform interval deviates from the normal distribution are detected, Peak and valley pairs with spacing can be obtained.

On the other hand, by performing a quadrature range (IQR) operation on the waveform interval after the operation of the quadrature range (IQR) for the relatively large waveform size in the breath count calculation, a peak and valley pair having a more reliable waveform interval Can be obtained.

The respiratory rate detection device then calculates the respiratory rate using the remaining peak and valley pairs except for the peak and valley pairs removed by the second filtering step 460 in the respiratory rate calculation step 470. Since the peak and valley pairs correspond to one breath, the number of remaining peak and valley pairs can be counted to determine the number of breaths. In this case, the number of remaining peak and valley pairs in the breath count calculation step 470 may be converted into a 60 second interval to calculate the breath per minute.

Then, in the breathing rate display step 480, the breathing rate detection device displays the breathing rate calculated by the breathing rate calculation step 470. For example, the respiratory rate may be displayed in the number of breaths per minute through the LCD or LED screen in the respiratory rate display step 480. Accordingly, the measurer can recognize the respiration rate of the subject by checking the respiration rate of the subject to be displayed by the respiration rate display step 480. [

As described above, the present invention can calculate the accurate number of breaths even when the waveform of the measurement waveform for breath detection is irregular, and can be applied to various types of breathing waveforms that can appear in actual clinical situations. Therefore, The object of the present invention can be achieved.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. .

The present invention is industrially applicable in the field of respiration detection technology and its application fields.

100: Breathing water detection device
110: Peak and valley pair detector
120: peak / valley difference calculation section
130:
140:
141: Size Quartile Range Operation Unit
142: interval quartile range computing unit
150: respiratory rate calculation unit
160: error warning section
170: Respiration number display section
180: Measurement waveform acquisition unit
190: resampling unit

Claims (20)

A first filter section for normalizing the difference between the peak value and the valley value of the peak and valley pairs detected from the measurement waveform for breath detection and removing the peak and valley pairs whose difference between the normalized peak value and the valley value is less than the threshold value, Wow;
A second filter portion for removing peak and valley pairs located in the non-dense region in consideration of the waveform size and spacing for the remaining pairs of peaks and valleys excluding the pairs of peaks and valleys removed by the first filter portion;
A respiration rate calculation unit for calculating a respiration rate using the remaining pairs of peaks and valleys excluding the pairs of peaks and valleys removed by the second filter unit;
And a control unit for controlling the operation of the respiration detection unit. The method according to claim 1,
The second filter portion:
(IQR: Interquartile Range) of the remaining peak and valley pairs except for the peak and valley pairs removed by the first filter section to determine a peak having an outlier whose waveform size deviates from the normal distribution and A size quadrature range arithmetic unit for removing the valley pairs;
Wherein the respiration detection unit comprises: 3. The method of claim 2,
The second filter portion:
(IQR: Interquartile Range) of peak and valley pairs in which peak and valley pairs having an outlier whose waveform size is out of the normal distribution is removed by the size quadrature range arithmetic operation unit, An interval interval quadrature range arithmetic unit for removing peak and valley pairs having outliers out of this normal distribution;
Further comprising: means for detecting a respiratory rate of the subject. 3. The method of claim 2,
The size quadrature range arithmetic unit:
And repeats a quadrature range (IQR) operation on waveform sizes of the peak and valley pairs until an outlier whose waveform size deviates from the normal distribution is not detected. The method of claim 3,
The interval quartile range calculating unit:
And repeats a quadrature range (IQR) operation on waveform intervals of the peak and valley pairs until an outlier whose waveform interval is out of the normal distribution is not detected. delete The method according to claim 1,
The first filter portion:
And when the number of peak and valley pairs in which the difference between the peak value and the valley value is less than the threshold is equal to or more than a certain percentage, the threshold value is adjusted and re-filtered. 8. The method of claim 7,
The first filter portion:
When the number of pairs of peaks and valleys whose difference between the peak value and the valley value is less than the adjusted threshold value is equal to or more than a certain percentage, is determined as a measurement error. 9. The method of claim 8,
Wherein the respiratory rate detection device comprises:
An error warning unit for performing a warning when a measurement error occurs;
Further comprising: means for detecting a respiratory rate of the subject. The method according to claim 1,
Wherein the respiratory rate detection device comprises:
A breath count display unit for displaying breath counts calculated by the breath count calculation unit;
Further comprising: means for detecting a respiratory rate of the subject. The method according to claim 1,
Wherein the respiratory rate detection device comprises:
A peak and valley pair detecting section for differentiating a measured waveform for breath detection to detect peak and valley pairs;
A peak / valley difference calculation unit for calculating a difference between a peak value and a valley value included in pairs of peaks and valleys detected by the peak and valley pair detecting unit;
Further comprising: means for detecting a respiratory rate of the subject. The method according to claim 1,
Wherein the respiratory rate detection device comprises:
A measurement waveform acquisition unit for acquiring a measurement waveform for breath detection;
Further comprising: means for detecting a respiratory rate of the subject. 13. The method of claim 12,
Wherein the respiratory rate detection device comprises:
A resampler for sampling the measured waveform acquired by the measured waveform acquisition unit at a specific time unit;
Further comprising: means for detecting a respiratory rate of the subject. 14. The method of claim 13,
The resampling unit:
And performs a moving average calculation on the resampled measured waveform data. 13. The method of claim 12,
Wherein the measured waveform for breathing detection comprises:
And detecting the change in the biometric information. 16. The method of claim 15,
Wherein the biometric information change is:
Wherein the respiratory rate is at least one of a temperature change at the time of breathing, an air flow caused by breathing, a change in body volume caused by a pulse or blood pressure, a blood volume or breathing, or a change in sound caused by bodily injury or snoring or irregularity. A peak and valley pair detection step in which the breathing detection apparatus detects pairs of peaks and valleys from a measurement waveform for breath detection;
A peak / valley difference calculation step of calculating a difference between a peak value and a valley value included in the pairs of peaks and valleys detected by the respiratory rate detection device by the peak and valley pair detection step;
Wherein the respiratory rate detection apparatus normalizes the difference between the peak value and the valley value calculated by the peak / valley difference calculation step, and the first and second pairs of peak and valley pairs, wherein the difference between the normalized peak value and the valley value is less than the threshold, Filtering;
A second filtering for removing peak and valley pairs located in the non-dense region in consideration of the waveform size and spacing for the remaining peak and valley pairs except for the peak and valley pairs removed by the first filtering step, ;
Calculating a breathing number using the remaining pairs of peaks and valleys except the pairs of peaks and valleys removed by the second filtering step;
Wherein the respiration detection unit detects the respiration rate of the subject. 18. The method of claim 17,
Wherein the second filtering step comprises:
Calculating the waveform sizes of the remaining peak and valley pairs excluding the peak and valley pairs removed by the first filtering step to obtain a peak having an outlier whose waveform size is out of the normal distribution, A size quartile range calculation step of removing valley pairs;
Calculating a quadrature range (IQR: Interquartile Range) of the waveform intervals of the peak and valley pairs from which peak and valley pairs having an outlier whose waveform size deviates from the normal distribution by the size quadrature range calculation step, An interval interval quadrature range calculation step of removing peak and valley pairs having an outlier whose interval is out of normal distribution;
Wherein the respiratory rate detection method comprises the steps of: 19. The method of claim 18,
In the step of calculating the size quadrangle range,
(IQR) operation for the waveform sizes of the peak and valley pairs until an outlier whose waveform size deviates from the normal distribution is not detected. 19. The method of claim 18,
In the interval quartile range calculating step:
The method comprising: iterating a quadrature range (IQR) operation on waveform intervals of peak and valley pairs until an outlier whose waveform interval is out of normal distribution is not detected.

Images