KR102637471B1 - Data processing apparatus for detecting respiratory event according to polysomnography and operating method of the same - Google Patents

Abstract

The present invention relates to a data processing device and operating method for detecting breathing events according to polysomnography, and more specifically, to the technical idea of signal data processing for automatic determination of sleep state according to analysis of polysomnography results. According to one embodiment of the present invention, the data processing device collects signal data detected by polysomnography, pre-processes the collected signal data to extract a breathing signal, and breathes from the extracted breathing signal. The event (respiratory event) section is determined, the signal excursion for the determined respiratory event section is calculated and processed as first candidate signal data, and the calculated signal excursion is normalized to create a second candidate. Using a signal data processing unit that processes signal data, the processed first candidate signal data, and the processed second candidate signal data, the determined respiratory event section is divided into a sleep apnea section, sleep hypopnea ( It may include a sleep state detection processing unit that detects any one of a hypopnea section, a RERA (Respiratory Effort Related Arousal) section, and a normal breathing section.

Description

Data processing device and operating method for detecting breathing events according to polysomnography {DATA PROCESSING APPARATUS FOR DETECTING RESPIRATORY EVENT ACCORDING TO POLYSOMNOGRAPHY AND OPERATING METHOD OF THE SAME}

The present invention relates to a data processing device and operating method for detecting breathing events according to polysomnography, and more specifically, to the technical idea of signal data processing for automatic determination of sleep state according to analysis of polysomnography results. will be.

Polysomnography is a test to diagnose sleep disorders. It comprehensively measures brain waves, eye movements, muscle movements, respiration, and electrocardiogram during sleep, and simultaneously records the sleeping state on video. By analyzing records, it is used to diagnose sleep-related diseases and determine treatment policies.

The above-described polysomnography can diagnose symptoms such as sleep apnea, sleep disorders, and sleep gait, and as indices for determining these diseases, sleep stage and apnea index (AHI) are used. , Respiratory Effort Related Arousal (RERA) index, etc. are being used.

Polysomnography is a test performed to diagnose sleep disorders while the patient sleeps overnight at a well-equipped sleep center. At least 24 sensors of 10 types are attached and the test is performed.

The test time varies depending on each individual's sleep time, but is approximately 6 to 8 hours, and it generally takes about 4 hours to read the test results, even if a qualified and skilled technician reads the test results.

In Korea, polysomnography has been included in health insurance since July 2018, and the number of patients who have undergone polysomnography has increased rapidly. In the United States, polysomnography has been available since 2014 when public insurance for polysomnography was implemented. The number of patients receiving treatment is rapidly increasing.

In Korea, looking at the reimbursement standards for polysomnography, it is recognized as once upon diagnosis and once after diagnosis, such as positive airway pressure treatment or surgery, only for sleep apnea and narcolepsy that have proven therapeutic usefulness.

Therefore, market demand for diagnostic automation and accuracy in polysomnography, especially for sleep apnea-related sleep conditions, is increasing.

Sleep apnea-related sleep states can be divided into sleep apnea state, sleep hypopnea state, RERA state, and normal breathing state.

Sleep apnea is defined as a decrease in the maximum amplitude of the oronasal thermal flow sensor by more than 90% of the baseline, and a decrease in apnea amplitude of more than 10 seconds for more than 90% of the event duration. In some cases, it can be classified as a sleep apnea section.

Here, the baseline can be defined as the average amplitude of stable respiration and oxygen supply for 2 minutes before the event period or the average amplitude of the three largest respirations over 2 minutes.

Sleep hypopnea occurs when 1) the output of the nasal pressure sensor decreases by more than 30% relative to the pre-event baseline of the respiratory curve, and 2) the duration of the decrease by more than 30% is 10 seconds. 3) This may be the case when the oxygen saturation falls by more than 3% below the respiratory curve standard value or is accompanied by arousal.

In addition, sleep hypopnea occurs when 1) the output of the nasal pressure sensor decreases by more than 30% with respect to the previous respiratory curve baseline (pre-event baseline), and 2) the duration of the decrease by more than 30% occurs. This may apply if it is more than 10 seconds and 3) the oxygen saturation falls by more than 4% compared to the breathing curve standard value.

Looking at products released on the existing market, there is a problem in that they cannot detect the exact signal generation section like a sleep technician in relation to the detection of sleep apnea and sleep hypopnea that occur during sleep.

Instead, the AHI (Apenea-Hypopnea index), which is the standard for diagnosing sleep disorders, is derived by only determining whether or not a breathing event is detected in 30-second epoch units.

The AHI index is calculated by multiplying the sum of the number of sleep apneas and the number of sleep hypopneas by 60 and dividing by the total sleep time. In this case, there are two or more breathing event sections within one epoch or a breathing event exists between epochs. In this case, errors occur in sleep apnea diagnosis.

Therefore, detection of an accurate breathing event period may be a very important factor in diagnosing sleep apnea, one of the important sleep disease indices.

RERA refers to continuous breathing of more than 10 seconds in which the flattening waveform of the output of the nasal pressure sensor causes arousal during sleep without meeting the criteria for apnea or hypopnea.

Looking at the products released in the existing market, there is a problem that it is difficult to detect the exact signal generation section like a sleep specialist in detecting the RERA event section that occurs during sleep.

Instead, it is possible to derive the RDI (Respiratory Disturbance Index), which is used to diagnose sleep disorders, by only determining whether or not the RERA event section is detected in 30-second epoch units.

The RDI index can be calculated as the sum of the number of intersleep RERA sections, the number of intersleep apnea sections, and the number of intersleep hypopnea sections multiplied by 60 and divided by the total sleep time.

In this case, if there are two or more RERA event sections in one epoch or if a RERA event section exists between epochs, an error is bound to occur in RDI diagnosis.

Therefore, detection of an accurate RERA event section may be a very important factor in confirming RDI, one of the important sleep disease indices.

Korean Patent No. 10-1868888, “Sleep/wake classification device and method for patients with sleep-disordered breathing using nasal pressure signals” Korean Patent No. 10-2068484, “Method for creating a sleep apnea prediction model and method for predicting sleep apnea using this model” Korean Patent No. 10-2258726, “Data processing device and method of operation for automatic determination of sleep disorders using deep learning”

The present invention provides data for automatically determining sleep status according to polysomnography, which can minimize the influence of noise signals depending on individual differences and measurement environments after polysomnography through signal data processing to automate the analysis of polysomnography results. The purpose is to provide a processing device and method.

The purpose of the present invention is to accurately detect sleep apnea and hypopnea signals and to provide results of detecting breathing signals that can cause RERA (Respiratory Effort Related arousal).

The purpose of the present invention is to detect the amplitude and duration of a breathing signal in order to detect a sleep apnea state section, a sleep hypopnea state section, and a RERA section.

The purpose of the present invention is to more accurately detect and distinguish sleep apnea state sections, sleep hypopnea state sections, and RERA sections by using signal excursion calculation and signal normalization.

The present invention differentiates the output signal of a nasal pressure sensor, takes the absolute value of the differentiated output signal, and accurately detects a flattening signal through moving average filtering on the differential value. The purpose is to detect and distinguish RERA sections more accurately.

According to one embodiment of the present invention, the data processing device collects signal data detected by polysomnography, pre-processes the collected signal data to extract a breathing signal, and respiratory events in the extracted breathing signal ( respiratory event) section is determined, signal excursion for the determined respiratory event section is calculated and processed as first candidate signal data, and normalization processing is performed on the calculated signal excursion to produce second candidate signal data. Using a signal data processing unit and the processed first candidate signal data and the processed second candidate signal data, the determined respiratory event section is divided into a sleep apnea section and a sleep hypopnea section. It may include a sleep state detection processing unit that detects any one of the sleep state section, RERA (Respiratory Effort Related Arousal) section, and normal breathing section.

The signal data processing unit may include a signal data collection unit, a signal data extraction unit, and a signal data processing unit.

The signal data collection unit collects at least one signal data from among signal data of a nasal pressure sensor, signal data of a temperature sensor, signal data of oxygen saturation, and arousal signal data through the polysomnography. It can be collected as signal data detected by .

The signal data extraction unit may extract a respiratory signal from the collected signal data after signal preprocessing of low pass filtering and DC component removal.

The signal data processing unit extracts a positive peak signal and a negative peak signal from the extracted respiration signal, determines a respiration event section in the extracted respiration signal, and determines a respiration event section for the determined respiration event section. Signal excursion may be calculated and processed as the first candidate signal data for the number of positive signal excursions corresponding to the positive peak signal and the number of negative signal excursions corresponding to the negative peak signal.

The signal data processing unit may perform normalization processing on the calculated signal fluctuations, extract positive peak signals and negative peak signals for the normalized signal fluctuations, and process them as the second candidate signal data. .

The sleep state detection processing unit compares the first candidate signal data and the second candidate signal data to detect the amplitude and duration of the signal in the determined breathing event section and detects the number of the detected amplitudes. is compared with a threshold value of signal fluctuation, and the detected number of durations is compared with a threshold value of durations to detect any one of the sleep state sections.

The sleep state detection processing unit may output one of an apnea section detection signal, a hypopnea section detection signal, a RERA section detection signal, and a normal breathing section detection signal in relation to any one of the detected sleep state sections. .

The sleep state detection processor compares sleep apnea and sleep hypopnea in the determined respiratory event section using the first candidate signal data, and uses the second candidate signal data to determine the determined respiratory event ( The ranges related to the sleep apnea section and the sleep hypopnea section in the respiratory event section can be compared.

According to one embodiment of the present invention, a method of operating a data processing device includes the steps of collecting signal data detected by polysomnography, preprocessing the collected signal data to extract a respiration signal, and the extracted respiration signal. determining a respiratory event section, calculating signal excursion for the determined respiratory event section and processing it as first candidate signal data, and normalizing the calculated signal excursion. Processing the second candidate signal data and converting the determined respiratory event section into a sleep apnea section and sleep using the processed first candidate signal data and the processed second candidate signal data. It may include a step of detecting any one of a hypopnea section, a Respiratory Effort Related Arousal (RERA) section, and a normal breathing section as a sleep state section.

The step of collecting signal data detected by the polysomnography includes signal data from a nasal pressure sensor, signal data from a temperature sensor, signal data from oxygen saturation, and arousal signal data. It may include collecting at least one signal data as signal data detected by the polysomnography.

The step of calculating signal excursion for the determined breathing event section and processing it as first candidate signal data includes extracting a positive peak signal and a negative peak signal from the extracted breathing signal to obtain the extracted breathing signal. Determine the breathing event section in and calculate the signal excursion for the determined breathing event section to calculate the number of positive signal excursions corresponding to the positive peak signal and the negative signal corresponding to the negative peak signal. The first candidate signal data for the variation number can be processed.

The step of processing the calculated signal fluctuations into second candidate signal data by performing normalization processing on the calculated signal fluctuations includes performing normalization processing on the calculated signal fluctuations, a positive peak signal for the normalized signal fluctuations, and It may include extracting a negative peak signal and processing it as the second candidate signal data.

Using the processed first candidate signal data and the processed second candidate signal data, the determined respiratory event section is divided into a sleep apnea section, a sleep hypopnea section, and a RERA (Respiratory Effort Related) section. The step of detecting any one of the arousal) section and the normal breathing section as a sleep state section includes comparing the first candidate signal data and the second candidate signal data to determine the amplitude of the signal in the determined breathing event section and Detecting the duration, comparing the detected number of magnitudes with a threshold value of signal fluctuation, and comparing the detected number of durations with a threshold value of the duration to detect the sleep state section. May include steps.

The present invention provides data for automatically determining sleep status according to polysomnography, which can minimize the influence of noise signals depending on individual differences and measurement environments after polysomnography through signal data processing to automate the analysis of polysomnography results. Processing devices and methods can be provided.

The present invention can accurately detect sleep apnea and hypopnea signals and provide results of detecting breathing signals that can cause RERA (Respiratory Effort Related arousal).

The present invention can detect the amplitude and duration of a breathing signal to detect a sleep apnea state section, a sleep hypopnea state section, and a RERA section.

The present invention can more accurately detect and distinguish sleep apnea state sections, sleep hypopnea state sections, and RERA sections by using signal excursion calculation and signal normalization.

The present invention differentiates the output signal of a nasal pressure sensor, takes the absolute value of the differentiated output signal, and accurately detects a flattening signal through moving average filtering on the differential value. RERA sections can be detected and distinguished more accurately.

1 is a diagram illustrating a data processing device for automatically determining sleep state according to an embodiment of the present invention.
Figure 2 is a diagram explaining a signal data processing unit according to an embodiment of the present invention.
Figure 3 is a diagram explaining a sleep state detection processor according to an embodiment of the present invention.
4A and 4B are diagrams illustrating a method of operating a signal data processor that calculates signal excursion in an embodiment of the present invention.
FIG. 4C is a diagram illustrating the operation method of the signal data processing unit that performs normalization processing of signal fluctuations according to an embodiment of the present invention.
Figure 5A illustrates an example of sleep hypopnea detected according to an embodiment of the present invention.
5B to 5D illustrate examples of noise in sleep apnea and sleep hypopnea sections according to an embodiment of the present invention.
Figures 6a and 6b are diagrams illustrating respiratory signals related to RERA (Respiratory Effort Related Arousal) according to an embodiment of the present invention.
Figure 7 is a diagram illustrating respiration signal processing for detecting a sleep state section from signal data according to an embodiment of the present invention.
8 to 10 are diagrams illustrating a method of operating a data processing device for automatically determining sleep state according to an embodiment of the present invention.
Figure 11 is a diagram illustrating an example of a screen displaying a sleep state section detected by processing signal data in a data processing device for automatically determining sleep state according to an embodiment of the present invention.

Hereinafter, various embodiments of this document are described with reference to the attached drawings.

The embodiments and terms used herein are not intended to limit the technology described in this document to a specific embodiment, and should be understood to include various changes, equivalents, and/or substitutes for the embodiments.

In the following description of various embodiments, if a detailed description of a related known function or configuration is judged to unnecessarily obscure the gist of the invention, the detailed description will be omitted.

The terms described below are terms defined in consideration of functions in various embodiments, and may vary depending on the intention or custom of the user or operator. Therefore, the definition should be made based on the contents throughout this specification.

In connection with the description of the drawings, similar reference numbers may be used for similar components.

Singular expressions may include plural expressions, unless the context clearly indicates otherwise.

In this document, expressions such as “A or B” or “at least one of A and/or B” may include all possible combinations of the items listed together.

Expressions such as “first,” “second,” “first,” or “second,” can modify the corresponding components regardless of order or importance and are used to distinguish one component from another. It is only used and does not limit the corresponding components.

When a component (e.g. a first) component is said to be "connected (functionally or communicatively)" or "connected" to another (e.g. a second) component, it means that the component is connected to the other component. It may be connected directly to a component or may be connected through another component (e.g., a third component).

In this specification, “configured to” means “suitable for,” “having the ability to,” or “changed to,” depending on the situation, for example, in terms of hardware or software. ," can be used interchangeably with "made to," "capable of," or "designed to."

In some contexts, the expression “a device configured to” may mean that the device is “capable of” working with other devices or components.

For example, the phrase "processor configured (or set) to perform A, B, and C" refers to a processor dedicated to performing the operations (e.g., an embedded processor), or by executing one or more software programs stored on a memory device. , may refer to a general-purpose processor (e.g., CPU or application processor) capable of performing the corresponding operations.

Additionally, the term 'or' means 'inclusive or' rather than 'exclusive or'.

That is, unless otherwise stated or clear from the context, the expression 'x uses a or b' means any of the natural inclusive permutations.

Terms such as '..unit' and '..unit' used hereinafter refer to a unit that processes at least one function or operation, and may be implemented as hardware, software, or a combination of hardware and software.

1 is a diagram illustrating a data processing device for automatically determining sleep state according to an embodiment of the present invention.

Figure 1 illustrates the components of a data processing device for automatically determining sleep state according to an embodiment of the present invention.

Referring to FIG. 1, the data processing device 100 according to an embodiment of the present invention may include a signal data processing unit 110 and a sleep state detection processing unit 120.

For example, a data processing device for automatically determining a sleep state may include a data processing device for detecting a breathing event according to polysomnography.

According to one embodiment of the present invention, the signal data processing unit 110 can collect signal data detected by polysomnography, and extract a respiratory signal by pre-processing the collected signal data.

In addition, the signal data processing unit 110 determines a respiratory event section in the extracted breathing signal, calculates the signal excursion for the determined respiratory event section, and processes it as first candidate signal data, The calculated signal fluctuations can be processed as second candidate signal data by performing normalization processing.

In other words, the signal data processing unit 110 continues the signal excursion calculation result for the respiratory event section determined from the breathing signal according to the normalization process for the first candidate signal data and the signal excursion calculation result. The duration calculation result can be processed as second candidate signal data.

According to one embodiment of the present invention, the signal data processing unit 110 processes signal data after polysomnography to minimize the influence of noise signals according to individual differences and measurement environments as candidate signal data through signal variation calculation and normalization processing. As a result, the diagnosis of sleep apnea can be automated.

For example, the sleep state detection processing unit 120 uses the first candidate signal data and the second candidate signal data processed by the signal data processing unit 110 to divide the respiratory event section into a sleep apnea section, It can be detected as one of the following sleep state sections: hypopnea section, RERA (Respiratory Effort Related Arousal) section, and normal breathing section.

According to one embodiment of the present invention, the sleep state detection processor 120 determines the amplitude and duration of the signal in the breathing event section determined by comparing the first candidate signal data and the second candidate signal data. It can be detected.

In addition, the sleep state detection processing unit 120 compares the number of previously detected sizes with a threshold value of signal fluctuation and compares the number of previously detected durations with a threshold value of the duration to detect any one sleep state section. can do. For example, a threshold value of duration may be referred to as a threshold time.

According to one embodiment of the present invention, the sleep state detection processor 120 determines the value (Mag) of the number of signal fluctuations in the first candidate signal data by subtracting the odd number from the even number and the number of signal fluctuations in the second candidate signal data. You can calculate the value (Delta) of the number of durations by subtracting the odd number from the even number.

Additionally, the sleep state detection processor 120 may calculate the number of sleep apnea durations and the number of sleep hypopnea durations.

Additionally, the sleep state detection processor 120 may set the threshold value (M _th ) and threshold time corresponding to the average value of up to three signal fluctuations. For example, the threshold time may be 10 seconds.

For example, the sleep state detection processing unit 120 includes a value (Mag), a value (Delta), the number of sleep apnea (apena) durations, the number of sleep hypopnea (hypopnea) durations, and a threshold (M _th ) and threshold. Using time, a predetermined respiratory event section is detected as a sleep state section among sleep apnea section, hypopnea section, RERA (Respiratory Effort Related Arousal) section, and normal breathing section. can do.

According to one embodiment of the present invention, the sleep state detection processor 120 detects the amplitude and duration of the signal in the breathing event section determined by comparing the first candidate signal data and the second candidate signal data. can do.

In addition, the sleep state detection processing unit 120 determines the number of detected magnitudes by setting a threshold value (M _th ) corresponding to the average value of up to three signal fluctuations with reference to the American Academy of Sleep Medicine (AASM) standard. It can be set to the threshold value (M _th ).

Specifically, the average value for the three maximums is a threshold (M _th ) determined by extracting the three average values of the maximum signal fluctuations in the 2 minutes preceding the point at which the negative peak signal at the start of the respiratory event interval is extracted. It can be.

According to one embodiment of the present invention, the sleep state detection processing unit 120 detects any one of an apnea section detection signal, a hypopnea section detection signal, a RERA section detection signal, and a normal breathing section detection signal in relation to any one detected sleep state section. One detection signal can be output as an output signal.

For example, the output signal is related to the detection of apnea zone, hypopnea zone, RERA zone and normal breathing zone, oxygen saturation-related signal can be utilized for detection of hypopnea zone, and arousal-related signal can be used for detection of hypopnea zone. It can be used to detect breathing sections and RERA sections.

According to one embodiment of the present invention, the sleep state detection processor 120 compares sleep apnea and sleep hypopnea in the respiratory event section using the first candidate signal data, and uses the second candidate signal data. Thus, the ranges related to the sleep apnea section and the sleep hypopnea section in the determined respiratory event section can be compared.

The present invention provides data for automatically determining sleep status according to polysomnography, which can minimize the influence of noise signals depending on individual differences and measurement environments after polysomnography through signal data processing to automate the analysis of polysomnography results. Processing devices and methods can be provided.

In addition, the present invention can detect the amplitude and duration of the breathing signal to detect the sleep apnea state section, sleep hypopnea state section, and RERA section.

Figure 2 is a diagram explaining a signal data processing unit according to an embodiment of the present invention.

Figure 2 illustrates the components of a signal data processing unit according to an embodiment of the present invention.

Referring to FIG. 2, the signal data processing unit 200 according to an embodiment of the present invention includes a signal data collection unit 210, a signal data extraction unit 220, and a signal data processing unit 230.

For example, the signal data collection unit 210 collects at least one of signal data from a nasal pressure sensor, signal data from a temperature sensor, signal data from oxygen saturation, and arousal signal data. Data can be collected as signal data detected by polysomnography.

For example, signal data detected by polysomnography includes EEG (Electroencephalogram) sensor, EOG (Electrooculography) sensor, EMG (Electromyogram) sensor, EKG (Electrokardiogramme) sensor, PPG (Photoplethysmography) sensor, and chest belt. , It can be interpreted as biometric data measured from the subject through at least one sensing means of an Abdomen belt, thermistor, flow sensor, and microphone.

Additionally, for example, signal data detected by polysomnography may be interpreted as data collected in real time, or may be interpreted as data recorded in a polysomnography database by a polysomnography performed in advance.

According to one embodiment of the present invention, the signal data extraction unit 220 may extract a respiratory signal from the collected signal data after signal preprocessing of low pass filtering and DC component removal.

In other words, the signal data extraction unit 220 may use a 0.01 to 0.35 Hz band pass filter to extract only the respiration signal by removing the DC signal from the collected signal data corresponding to the input signal.

According to one embodiment of the present invention, the signal data processing unit 230 extracts a positive peak signal and a negative peak signal from the extracted breathing signal to determine a breathing event section in the extracted breathing signal. You can.

In addition, the signal data processing unit 230 calculates the signal excursion for the determined breathing event section and calculates the number of positive signal excursions corresponding to the positive peak signal and the number of negative signal excursions corresponding to the negative peak signal. It can be processed as the first candidate signal data for.

That is, the first candidate signal data may include the number of positive signal fluctuations corresponding to the calculation result of the signal fluctuations and the number of negative signal fluctuations corresponding to the negative peak signal.

For example, the positive peak signal and the negative peak signal may be extracted by the signal data extraction unit 220 and then transmitted to the signal data processing unit 230.

According to one embodiment of the present invention, the signal data processing unit 230 performs normalization processing on the previously calculated signal fluctuations, then extracts the positive peak signal and the negative peak signal for the normalized signal fluctuations to produce 2 Can be processed as candidate signal data.

That is, the signal data processing unit 230 processes the signal variation calculation result into first candidate signal data, and processes the duration calculation result according to the normalization process for the signal variation calculation result into second candidate signal data. It can be processed.

For example, the signal data processing unit 230 may differentiate a breathing signal based on a specific pressure sensor among the breathing signals and process the absolute value of the differentiated signal into RERA section judgment data.

That is, in order to detect the RERA section, the signal data processing unit 230 calculates the differential value for the breathing signal based on the nasal pressure sensor among the signal data of the polysomnography, and then calculates the absolute value to apply moving average filtering. By taking , it is possible to prevent the occurrence of a section where part of the flattening signal is larger than the differential value.

Accordingly, the present invention differentiates the output signal of a nasal pressure sensor, takes the absolute value of the differentiated output signal, and produces a flattening signal through moving average filtering on the differential value. By detecting accurately, RERA sections can be detected and distinguished more accurately.

Figure 3 is a diagram explaining a sleep state detection processor according to an embodiment of the present invention.

Figure 3 illustrates the components of a sleep state detection processor according to an embodiment of the present invention.

Referring to Figure 3, the sleep state detection processing unit 300 according to an embodiment of the present invention includes a sleep apnea section detection unit 310, a sleep hypopnea section detection unit 320, and a RERA section detection unit 330. .

For example, the sleep state detection processing unit 300 includes a sleep apnea section detection unit 310, a sleep hypopnea section detection unit 320, and a RERA section detection unit 330, thereby detecting breathing signal data from polysomnography data. For this reason, the breathing event section can be detected, and the detected breathing event section can be classified and detected as whether it is an apnea section, a hypopnea section, a RERA section, or a normal breathing section.

The sleep apnea section detection unit determines the value (Mag) minus the odd number from the even number of the number of signal fluctuations in the first candidate signal data for up to three signal fluctuations during the 2 minutes immediately before the signal fluctuation starts. Determine whether it is less than (equal to or exceeds) the threshold value (M _th ) corresponding to the average value divided by 10.

In addition, the sleep apnea section detection unit 210 determines that the sum of the apnea duration (A_Delta) and the even number minus the odd number for the number of durations in the second candidate signal data is the apnea duration. Determine if it is not the same as (A_Delta).

In addition, the sleep apnea section detection unit 210 detects sleep apnea as a judgment target when the above-mentioned determination result is met, and detects the breathing event section when the apnea duration (A_Delta) exceeds the critical time of 10 seconds. It can be detected as a sleep apnea section.

According to one embodiment of the present invention, the sleep hypopnea section detection unit 220 detects the value (Delta) and the hypopnea duration (H_Delta) while the value (Mag) is not less than the threshold value (M _th ) divided by 70. ) Determine whether the sum of the hypoventilation duration (H_Delta) is not equal to the above.

In addition, the sleep hypopnea section detection unit 220 detects the determined breathing event section as a judgment target of sleep hypopnea if the condition is met, and the hypopnea duration (H_Delta) exceeds the threshold time of 10 seconds. It can be detected as a sleep hypopnea section.

According to one embodiment of the present invention, the RERA section detection unit 330 flattens the RERA section determination data as long as the value (Mag) is not less than the threshold value (M _th ) divided by 70 and the threshold value (M _th ). If it is not less than the threshold and the sum of the value (Delta) and the hypoventilation duration (H_Delta) is not equal to the hypoventilation duration (H_Delta), it is detected as the subject of RERA judgment.

Additionally, the RERA section detection unit 330 may detect the determined breathing event section as the RERA section when the hypoventilation duration (H_Delta) exceeds the threshold time of 10 seconds.

Accordingly, the sleep state detection processor 300 according to an embodiment of the present invention calculates the number of signal fluctuations in the first candidate signal data by subtracting the odd number from the even number (Mag), the second candidate signal data. The value of the odd number minus the even number for the number of durations (Delta), the number of sleep apnea durations, the number of sleep hypopnea durations, and up to three of the signal fluctuations. The respiratory event section determined using the threshold value (M _th ) and threshold time corresponding to the average value, RERA section judgment data, and the flattening threshold value is divided into the sleep apnea section, the sleep hypopnea section, the RERA section, and It can be detected as any sleep state section among the normal breathing sections. For example, the flattening threshold may correspond to a threshold voltage value (V _th ).

4A and 4B are diagrams illustrating a method of operating a signal data processor that calculates signal excursion in an embodiment of the present invention.

Referring to FIG. 4A, the output waveform of the graph 400 is the signal data detected by polysomnography corresponding to the input signal, other than the DC change and respiration signal due to noise due to the external environment or patient movement. Illustrates the respiratory signal with the signal removed.

The operating method of the signal data processing unit extracts a respiration signal from which signals other than the respiration signal and DC changes caused by noise caused by the external environment or patient movement are removed. The extracted output waveform may be as shown in the graph 400. there is.

In the graph 400, the positive peak signal 401 and the negative peak signal 402 can be extracted by the signal data processor.

The operating method of the signal data processing unit is a section in which the size of the waveforms of the positive peak signal 401 and the negative peak signal 402 decreases along the extraction point of the positive peak signal 401 and the negative peak signal 402. Can be detected as a breathing event section 403.

Referring to FIG. 4B, the arrow in the graph 410 represents the result of calculating the signal variation of the respiration signal.

The operation method of the signal data processing unit can calculate the signal variation calculation result for the breathing signal as shown in the graph 410.

In the graph 410, the signal variation calculation results for the positive peak signal 411 and the negative peak signal 412 may be calculated by the signal data processor.

The operation method of the signal data processing unit is that the size of the signal variation calculation result of the positive peak signal 411 and the negative peak signal 412 is determined according to the signal variation of the positive peak signal 411 and the negative peak signal 412. The section that becomes smaller can be detected as the breathing event section 413.

FIG. 4C is a diagram illustrating the operation method of the signal data processing unit that performs normalization processing of signal fluctuations according to an embodiment of the present invention.

Referring to FIG. 4C, the arrow in the graph 420 represents the result of normalization processing for the signal variation calculation result of the respiration signal.

The operation method of the signal data processing unit can calculate the normalization processing result of the signal variation calculation result for the breathing signal as shown in the graph 420.

In the graph 420, normalization of the signal variation calculation results for the positive peak signal 421 and the negative peak signal 422 may be performed by the signal data processor.

The operation method of the signal data processing unit is based on the signal variation calculation results of the positive peak signal 421 and the negative peak signal 422 according to the signal variation of the positive peak signal 421 and the negative peak signal 422. Calculation results for the duration of the signal can be derived.

Figure 5A illustrates an example of sleep hypopnea detected according to an embodiment of the present invention.

Figure 5a illustrates sleep hypopnea detected according to an embodiment of the present invention.

Referring to FIG. 5A, the output screen 500 displays the breathing signal as a waveform, and the sleep hypopnea section can be confirmed by referring to the area 501 in the displayed waveform.

In area 501, it can be confirmed that the waveform decreases by more than 30% of the breathing curve standard value, has a duration of more than 10 seconds, and falls by more than 3% than the breathing curve standard value, accompanied by arousal, or falls by more than 4%. there is.

5B to 5D illustrate examples of noise in sleep apnea and sleep hypopnea sections according to an embodiment of the present invention.

Figure 5b illustrates a noise section in sleep hypopnea detected according to an embodiment of the present invention.

Referring to FIG. 5B, the output screen 510 displays the breathing signal as a waveform, and by referring to the area 511 in the displayed waveform, it can be confirmed that noise exists in the sleep hypopnea section.

Here, noise may act as a factor that causes problems in detection of the sleep hypopnea section.

Figure 5c illustrates a noise section in sleep apnea detected according to an embodiment of the present invention.

Referring to FIG. 5C, the output screen 520 displays the breathing signal as a waveform, and by referring to the area 521 in the displayed waveform, it can be confirmed that noise exists in the sleep apnea section.

Here, noise may act as a factor that causes problems in detection of the sleep apnea section.

Figure 5D illustrates a noise section in sleep hypopnea detected according to an embodiment of the present invention.

Referring to FIG. 5D, the output screen 530 displays the breathing signal as a waveform, and referring to the area 531 in the displayed waveform, noise exists in the sleep hypopnea section, and the peak signal 532 is relatively high due to the noise. Indicates what is output.

Here, noise may act as a factor that causes problems in detection of the sleep hypopnea section.

In other words, in order to detect the actual breathing event section, it is generally necessary to detect the peak value and set the section.

However, when detecting only the peak signal, it is difficult to accurately detect the section due to noise.

Accordingly, as the value of the peak signal is output relatively high due to noise, it may be difficult to detect the breathing event section.

Figures 6a and 6b are diagrams illustrating respiratory signals related to RERA (Respiratory Effort Related Arousal) according to an embodiment of the present invention.

Figure 6a illustrates the RERA section detected according to an embodiment of the present invention compared to the normal breathing section.

Referring to FIG. 6A, the output screen 600 illustrates the RERA section, and the output screen 610 illustrates the normal breathing section.

Comparing the output screen 600 and the output screen 610, the output screen 600 shows a RERA section in the area 601, but the output screen 600 confirms that the output waveform repeats with the same size and duration. You can.

For example, area 601 is the output data of a nasal flow sensor, and it can be confirmed that it is a flattening signal that is a rising curve in the output waveform.

Figure 6b illustrates the RERA section detected according to an embodiment of the present invention compared to the normal breathing section.

Referring to FIG. 6B, the output screen 620 illustrates the RERA section and confirms that there is a point 621 where a peak occurs in the flattening signal in the RERA section.

In order to detect the RERA section, it is important to detect the section where the signal is flattened according to the definition of RERA. In order to detect the flattened signal, the signal from the input specific pressure sensor must be differentiated.

If the differential value is below a set threshold, it can be regarded as a flattening signal and the RERA section can be set. However, as there is a part such as point 621 where the differential value is greater than the differential value, the differential value is obtained and the absolute value is taken to calculate the differential value. Moving average filtering is required.

Figure 7 is a diagram illustrating respiration signal processing for detecting a sleep state section from signal data according to an embodiment of the present invention.

Figure 7 illustrates breathing signal processing for detecting a sleep state section in signal data according to an embodiment of the present invention using a graph.

Referring to FIG. 7, a graph 700 shows a respiratory signal extracted by collecting signal data detected by polysomnography and preprocessing the collected signal data.

The graph 710 shows an enlarged respiration event section among the respiration signals of the graph 700.

The graph 720 shows the result of calculating the signal change for the graph 710 corresponding to the breathing event section among the breathing signals of the graph 700.

More specifically, the graph 720 extracts positive peak signals and negative peak signals from the graph 710 and displays signal variation calculation results for the extracted positive peak signals and negative peak signals.

For example, the graph 720 may correspond to first candidate signal data.

The graph 720 represents odd and even numbers of signal fluctuations, with odd numbers corresponding to the number of negative peak signals and even numbers corresponding to the number of positive peak signals.

The graph 730 can be derived by normalizing the signal variation calculated in the graph 720.

The graph 740 represents the signal variation calculation results for the graph 730.

More specifically, the graph 740 extracts positive peak signals and negative peak signals from the graph 730 and displays signal variation calculation results for the extracted positive peak signals and negative peak signals.

For example, the graph 740 may correspond to second candidate signal data.

Graph 740 represents odd and even numbers of signal fluctuations, with odd numbers corresponding to the number of normalization results for negative peak signals, and even numbers corresponding to the number of normalization results for positive peak signals. .

According to one embodiment of the present invention, the operating method of the data processing device can calculate the size and duration of the signal in the breathing event section using the graph 720 and the graph 740.

Here, since the starting standard of the breathing event section is based on the negative peak signal, an algorithm for detecting the sleep apnea section and the sleep hypopnea section and the algorithm for detecting the RERA section may be performed based on the negative peak signal.

The algorithm for detecting the sleep apnea section and the sleep hypopnea section is explained in Figure 9, and the algorithm for detecting the RERA section is further explained in Figure 10.

8 to 10 are diagrams illustrating a method of operating a data processing device for automatically determining sleep state according to an embodiment of the present invention.

Figure 8 illustrates an embodiment in which a method of operating a data processing device for automatically determining a sleep state according to an embodiment of the present invention detects a sleep state section using signal fluctuation calculation and signal normalization.

Referring to FIG. 8, in step 801, the method of operating a data processing device according to an embodiment of the present invention collects signal data from polysomnography.

In other words, the operating method of the data processing device collects signal data detected by polysomnography.

Here, the collected signal data may include at least one of signal data from a nasal pressure sensor, signal data from a temperature sensor, signal data from oxygen saturation, and arousal signal data. You can.

In step 802, the method of operating a data processing device according to an embodiment of the present invention extracts a respiratory signal by pre-processing signal data.

That is, the method of operating the data processing device may extract a respiratory signal after pre-processing the signal data collected in step 801 by performing low pass filtering and DC component removal.

In step 803, the method of operating the data processing device according to an embodiment of the present invention extracts the first peak signal and the second peak signal of the respiration signal to determine the respiration event section. Here, the first peak signal may represent a positive peak signal, and the second peak signal may represent a negative peak signal.

That is, the operating method of the data processing device may extract a positive peak signal and a negative peak signal from the respiration signal and determine a respiration event section corresponding to a section in which the size of the waveform in the extracted respiration signal rapidly decreases.

In step 804, the method of operating a data processing device according to an embodiment of the present invention calculates signal fluctuations and processes them as first candidate signal data.

In other words, the operating method of the data processing device calculates the signal excursion for the breathing event section and calculates the number of positive signal excursions corresponding to the positive peak signal and the number of negative signal excursions corresponding to the negative peak signal. It can be processed as first candidate signal data.

In step 805, the method of operating a data processing device according to an embodiment of the present invention may process the signal fluctuation calculated in step 804 as second candidate signal data through normalization processing.

In other words, the operating method of the data processing device is to perform normalization processing on the calculated signal fluctuations, then extract the positive peak signal and negative peak signal for the normalized signal fluctuations and process them as second candidate signal data. there is.

In step 806, the method of operating a data processing device according to an embodiment of the present invention detects a sleep state section using first candidate signal data and second candidate signal data.

In other words, the operating method of the data processing device is to use the first candidate signal data and the second candidate signal data to transform the respiratory event section into a sleep apnea section, a sleep hypopnea section, and a RERA (Respiratory Effort) section. It can be detected as either a sleep state section (Related Arousal) section or a normal breathing section.

Therefore, the present invention can more accurately detect and distinguish the sleep apnea state section, the sleep hypopnea state section, and the RERA section by using signal excursion calculation and signal normalization.

Figure 9 shows a method of operating a data processing device for automatically determining sleep state according to an embodiment of the present invention, detecting a sleep apnea state section and a sleep hypopnea state section among sleep state sections by calculating signal fluctuations and signal normalization. An example using the algorithm is illustrated.

The method of operating a data processing device for automatically determining a sleep state according to an embodiment of the present invention may perform the embodiment described in FIG. 9 after the signal processing step described in FIG. 7 has already been performed.

Referring to FIG. 9, in step 901, the method of operating a data processing device according to an embodiment of the present invention sets a threshold value (M _th ) corresponding to the average value of a plurality of signal fluctuations before the breathing event section. .

That is, the method of operating the data processing device according to an embodiment of the present invention detects the amplitude and duration of the signal in the breathing event section by comparing the first candidate signal data and the second candidate signal data. And, referring to the American Academy of Sleep Medicine (AASM) standard, the threshold value (M _th ) corresponding to the average value of up to three signal fluctuations can be set as the threshold value of signal fluctuation (M _th ).

In step 902, the method of operating the data processing device according to an embodiment of the present invention calculates a signal variation-related variable (Mag) and a duration-related variable (Delta) of the breathing event section.

That is, the operating method of the data processing device calculates the value (Mag) of the number of signal fluctuations in the first candidate signal data by subtracting the odd number from the even number, and calculates the number of durations in the second candidate signal data. You can calculate the value (Delta) by subtracting the odd numbers from the even numbers.

In step 903, the method of operating the data processing device according to an embodiment of the present invention is the sum of the variable (Delta) and the apnea duration (A_Delta) while the variable (Mag) is less than the threshold value (Mth) divided by 10. Determine whether this is the same as the apnea duration (A_Delta).

The operation method of the data processing device returns to step 902 if the case is satisfied, and if the case is not satisfied, detects sleep apnea as a determination target and proceeds to step 904 or determines sleep hypopnea. Proceed to step 907.

In step 904, the operating method of the data processing device according to an embodiment of the present invention determines whether the apnea duration (A_Delta) exceeds the threshold time of 10 seconds.

If the operation method of the data processing device is satisfied, the procedure is terminated in step 905 by detecting the breathing event section as a sleep apnea section.

Meanwhile, if the operation method of the data processing device is not satisfied with the corresponding case, the apnea duration time (A_Delta) is determined to be 0 and returns to step 902.

In step 907, the method of operating the data processing device according to an embodiment of the present invention is that the variable (Mag) is less than the threshold value (Mth) divided by 70 and the variable (Delta) and the hypoventilation duration (H_Delta) are Determine whether the sum is equal to the hypoventilation duration (H_Delta).

The operation method of the data processing device returns to step 902 if the case is satisfied, and if the case is not satisfied, detects sleep hypopnea as a judgment target and proceeds to step 908 or proceeds to step 911. Proceed.

In step 908, the operating method of the data processing device according to an embodiment of the present invention determines whether the hypoventilation duration (H_Delta) exceeds the threshold time of 10 seconds.

If the operation method of the data processing device is satisfied, the procedure is terminated in step 909 by detecting the breathing event section as a sleep hypopnea section.

Meanwhile, if the operating method of the data processing device is not satisfied with the corresponding case, the hypoventilation duration time (H_Delta) is determined to be 0 and returns to step 902.

In step 911, the operation method of the data processing device detects the breathing event section as a normal breathing section or a RERA section and ends the procedure.

Figure 10 shows an embodiment in which the operating method of the data processing device for automatically determining the sleep state according to an embodiment of the present invention uses an algorithm to detect the RERA state section among the sleep state sections using signal fluctuation calculation and signal normalization. Illustrate.

The method of operating a data processing device for automatically determining a sleep state according to an embodiment of the present invention may perform the embodiment described in FIG. 10 after the signal processing step described in FIG. 7 has already been performed.

Referring to FIG. 10, in step 1001, the method of operating the data processing device collects signal data from polysomnography.

In other words, the operating method of the data processing device collects signal data detected by polysomnography.

Here, the collected signal data may include at least one of signal data from a nasal pressure sensor, signal data from a temperature sensor, signal data from oxygen saturation, and arousal signal data. You can.

In step 1002, the method of operating the data processing device according to an embodiment of the present invention differentiates a respiration signal based on a specific pressure sensor among signal data, and processes the absolute value of the differentiated respiration signal as RERA section judgment data (flattened). do.

In step 1003, the method of operating a data processing device according to an embodiment of the present invention sets a threshold value (M _th ) corresponding to the average value of a plurality of signal fluctuations before the breathing event section.

In other words, the method of operating the data processing device detects the amplitude and duration of the signal in the breathing event section by comparing the first candidate signal data and the second candidate signal data, and detects the amplitude and duration of the signal in the breathing event section and AASM (American Academy of Sleep Referring to the Medicine) standard, the threshold value (M _th ) corresponding to the average value of up to three signal fluctuations can be set as the threshold value of signal fluctuation (M _th ).

In step 1004, the method of operating the data processing device according to an embodiment of the present invention calculates a signal variation-related variable (Mag) and a duration-related variable (Delta) of the breathing event section.

That is, the operating method of the data processing device calculates the value (Mag) of the number of signal fluctuations in the first candidate signal data by subtracting the odd number from the even number, and calculates the number of durations in the second candidate signal data. You can calculate the value (Delta) by subtracting the odd numbers from the even numbers.

In step 1005, the method of operating a data processing device according to an embodiment of the present invention detects an apnea section and a hypopnea section.

That is, the operating method of the data processing device detects the apnea section and the hypopnea section by performing steps 903 to 909 in FIG. 9.

In step 1006, the method of operating the data processing device according to an embodiment of the present invention is to detect the normal breathing section and the RERA section by dividing the variable (Mag) by 70 and dividing the threshold value (M _th ) by 70. (M _th ), the RERA section judgment data is less than the threshold voltage (V _th ) corresponding to the flattening threshold, and the sum of the variable (Delta) and the hypopnea duration (H_Delta) is the hypopnea duration (H_Delta). Determine whether it is the same as .

The operation method of the data processing device returns to step 1005 when the above-mentioned conditions are satisfied and determines again whether it is one of the apnea section and the hypopnea section.

In step 1007, the operating method of the data processing device according to an embodiment of the present invention can detect a RERA section or a normal breathing section when the hypoventilation variable (H_Delta) is less than 10 seconds. Here, the hypopnea variable (H_Delta) is equal to the hypopnea duration (H_Delta).

In addition, the operation method of the data processing device determines that the hypopnea variable (H_Delta) is 0 when the hypopnea variable (H_Delta) is greater than 10 seconds, and returns to step 1005 to check once again whether it is one of the apnea section and the hypopnea section. You can also judge.

For example, a method of operating a data processing device for automatically determining a sleep state may include a data processing method for detecting a breathing event according to polysomnography.

FIG. 11 is a diagram illustrating an example of a screen displaying a sleep state section detected by processing signal data in a data processing device for automatically determining sleep state according to an embodiment of the present invention.

Referring to FIG. 11, the output screen 1100 can output waveforms for polysomnography data and provide the results as text content.

The data processing device for automatically determining the sleep state displays the detected hypoventilation section through area 1101 on the output screen 1100, and displays the breathing event type, start time, end time, and duration through area 1102. can be provided together as text content.

Therefore, the present invention can accurately detect sleep apnea and hypopnea signals and provide results of detecting breathing signals that can cause Respiratory Effort Related arousal (RERA).

The device described above may be implemented with hardware components, software components, and/or a combination of hardware components and software components. For example, devices and components described in embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), It may be implemented using one or more general-purpose or special-purpose computers, such as a programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions. A processing device may execute an operating system (OS) and one or more software applications that run on the operating system. Additionally, a processing device may access, store, manipulate, process, and generate data in response to the execution of software. For ease of understanding, a single processing device may be described as being used; however, those skilled in the art will understand that a processing device includes multiple processing elements and/or multiple types of processing elements. It can be seen that it may include. For example, a processing device may include a plurality of processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are possible.

Software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing unit to operate as desired, or may be processed independently or collectively. You can command the device. Software and/or data may be used on any type of machine, component, physical device, virtual equipment, computer storage medium or device to be interpreted by or to provide instructions or data to a processing device. , or may be permanently or temporarily embodied in a transmitted signal wave. Software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

Although the embodiments have been described with limited drawings as described above, various modifications and variations can be made by those skilled in the art from the above description. For example, the described techniques are performed in a different order than the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form than the described method, or other components are used. Alternatively, appropriate results may be achieved even if substituted or substituted by an equivalent.

Therefore, other implementations, other embodiments, and equivalents of the claims also fall within the scope of the claims described below.

100: data processing device
110: signal data processing unit 120: sleeping state detection processing unit

Claims (14)

Collecting signal data detected by polysomnography, preprocessing the collected signal data to extract a respiratory signal, determining a respiratory event section in the extracted respiratory signal, and determining the determined respiratory event A signal data processing unit that calculates signal excursion for a section and processes it as first candidate signal data, and normalizes the calculated signal excursion to process it as second candidate signal data; and
Calculate a value (Mag) minus the odd number from the even number for the number of signal fluctuations in the processed first candidate signal data, and calculate the even number for the number of durations in the processed second candidate signal data. Calculate the value (Delta) excluding odd numbers, calculate the number of sleep apnea durations, the number of sleep hypopnea durations, and the threshold corresponding to the average value for signal fluctuations ( Mth) and threshold time, and the value (Mag), the value (Delta), the number of sleep apnea (apena) durations, the number of sleep hypopnea (hypopnea) durations, and the threshold (Mth) And using the threshold time, the determined respiratory event section is classified into any one of sleep apnea section, sleep hypopnea section, RERA (Respiratory Effort Related Arousal) section, and normal breathing section. Characterized by comprising a sleep state detection processing unit that detects by section
Data processing device. According to paragraph 1,
The signal data processing unit includes a signal data collection unit, a signal data extraction unit, and a signal data processing unit.
Data processing device. According to paragraph 2,
The signal data collection unit collects at least one signal data from among signal data of a nasal pressure sensor, signal data of a temperature sensor, signal data of oxygen saturation, and arousal signal data through the polysomnography. Characterized by collecting signal data detected by
Data processing device. According to paragraph 3,
The signal data extraction unit extracts a respiratory signal after signal preprocessing of the collected signal data by low-pass filtering and DC component removal.
Data processing device. According to clause 4,
The signal data processing unit extracts a positive peak signal and a negative peak signal from the extracted respiration signal, determines a respiration event section in the extracted respiration signal, and determines a respiration event section for the determined respiration event section. Calculating signal excursion and processing the number of positive signal excursions corresponding to the positive peak signal and the number of negative signal excursions corresponding to the negative peak signal as the first candidate signal data. doing
Data processing device. According to clause 5,
The signal data processing unit performs normalization processing on the calculated signal fluctuations, extracts positive peak signals and negative peak signals for the normalized signal fluctuations, and processes them as the second candidate signal data. to do
Data processing device. According to clause 6,
The sleep state detection processing unit compares the first candidate signal data and the second candidate signal data to detect the amplitude and duration of the signal in the determined breathing event section and detects the number of the detected amplitudes. Comparing with a threshold value of signal fluctuation, and comparing the number of detected durations with a threshold value of duration to detect any one of the sleep state sections.
Data processing device. In clause 7,
The sleep state detection processing unit outputs one of an apnea section detection signal, a hypopnea section detection signal, a RERA section detection signal, and a normal breathing section detection signal in relation to any one of the detected sleep state sections. to do
Data processing device. According to clause 6,
The sleep state detection processor compares sleep apnea and sleep hypopnea in the determined respiratory event section using the first candidate signal data, and uses the second candidate signal data to determine the determined respiratory event ( Characterized by comparing the ranges related to the sleep apnea section and the sleep hypopnea section in the respiratory event section.
Data processing device. In a signal data processing unit, collecting signal data detected by polysomnography;
In the signal data processing unit, pre-processing the collected signal data to extract a respiratory signal;
In the signal data processing unit, determining a respiratory event section in the extracted respiratory signal;
In the signal data processing unit, calculating a signal excursion for the determined breathing event section and processing it as first candidate signal data;
In the signal data processing unit, normalizing the calculated signal fluctuations and processing them into second candidate signal data; and
In the sleep state detection processing unit, a value (Mag) is calculated by subtracting the odd number from the even number for the number of signal fluctuations in the processed first candidate signal data, and the duration of the processed second candidate signal data is calculated. Calculate the value (Delta) by subtracting the odd number from the even number, calculate the number of sleep apnea durations, the number of sleep hypopnea durations, and the average value for signal fluctuations. Set to the threshold value (Mth) and threshold time corresponding to the value (Mag), the value (Delta), the number of sleep apnea durations, the number of sleep hypopnea durations, and Using the threshold value (Mth) and the threshold time, the determined respiratory event sections are divided into sleep apnea section, sleep hypopnea section, RERA (Respiratory Effort Related Arousal) section, and normal breathing section. Characterized in that it includes the step of detecting any one of the sleep state sections.
Method of operation of a data processing device. According to clause 10,
The step of collecting signal data detected by the polysomnography is
A signal detected by the polysomnography at least one of signal data of a nasal pressure sensor, signal data of a temperature sensor, signal data of oxygen saturation, and arousal signal data. Characterized by comprising the step of collecting data
Method of operation of a data processing device. According to clause 11,
The step of calculating signal excursion for the determined breathing event section and processing it as first candidate signal data,
A positive peak signal and a negative peak signal are extracted from the extracted respiration signal to determine a respiration event section in the extracted respiration signal, and a signal excursion for the determined respiration event section is calculated to determine the positive peak signal and the negative peak signal. Characterized in that processing the first candidate signal data for the number of positive signal fluctuations corresponding to the peak signal and the number of negative signal fluctuations corresponding to the negative peak signal.
Method of operation of a data processing device. According to clause 12,
The step of normalizing the calculated signal fluctuations and processing them into second candidate signal data is:
After normalizing the calculated signal fluctuations, extracting positive peak signals and negative peak signals for the normalized signal fluctuations and processing them as the second candidate signal data. doing
Method of operation of a data processing device. According to clause 13,
Calculate a value (Mag) minus the odd number from the even number for the number of signal fluctuations in the processed first candidate signal data, and calculate the even number for the number of durations in the processed second candidate signal data. Calculate the value (Delta) excluding odd numbers, calculate the number of sleep apnea durations, the number of sleep hypopnea durations, and the threshold corresponding to the average value for signal fluctuations ( Mth) and threshold time, and the value (Mag), the value (Delta), the number of sleep apnea (apena) durations, the number of sleep hypopnea (hypopnea) durations, and the threshold (Mth) And using the threshold time, the determined respiratory event section is converted to any one of sleep apnea section, sleep hypopnea section, RERA (Respiratory Effort Related Arousal) section, and normal breathing section. The stage of detecting a section is,
By comparing the first candidate signal data and the second candidate signal data, the amplitude and duration of the signal in the determined breathing event section are detected, and the number of detected magnitudes is set as a threshold value for signal fluctuation. , and comparing the detected number of durations with a threshold value of duration to detect the one sleep state section.
Method of operation of a data processing device.

Images