WO2020042897A1 - 呼吸暂停事件类型的判断方法、装置和存储介质 - Google Patents

呼吸暂停事件类型的判断方法、装置和存储介质 Download PDF

Info

Publication number
WO2020042897A1
WO2020042897A1 PCT/CN2019/100238 CN2019100238W WO2020042897A1 WO 2020042897 A1 WO2020042897 A1 WO 2020042897A1 CN 2019100238 W CN2019100238 W CN 2019100238W WO 2020042897 A1 WO2020042897 A1 WO 2020042897A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
original
flow
res
derivative
Prior art date
Application number
PCT/CN2019/100238
Other languages
English (en)
French (fr)
Inventor
周文丽
Original Assignee
深圳融昕医疗科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳融昕医疗科技有限公司 filed Critical 深圳融昕医疗科技有限公司
Publication of WO2020042897A1 publication Critical patent/WO2020042897A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/48—Other medical applications
    • A61B5/4806—Sleep evaluation
    • A61B5/4818—Sleep apnoea
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/113—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb occurring during breathing
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/113—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb occurring during breathing
    • A61B5/1135—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb occurring during breathing by monitoring thoracic expansion
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems

Definitions

  • the present application relates to the field of medical devices, and in particular, to a method, device, and storage medium for determining the type of apnea event.
  • Sleep apnea is a type of breathing disorder, which refers to more than 30 apneas during continuous 7h sleep, each time the airflow is stopped for more than 10s (including 10s), or the average hourly low ventilation (breathing Disturbance index) more than 5 times, causing clinical syndromes of chronic hypoxemia and hypercapnia. It can be divided into central type, obstructive type and mixed type.
  • Obstructive Sleep Apnea is the collapse of the upper airway causing the upper airway obstruction or narrowing of the airway leading to sleep apnea during sleep, which is manifested by the nasal and airflow cessation and thoracoabdominal breathing.
  • Central nervous sleep apnea CSA
  • CSA Central nervous sleep apnea
  • MSA Mixed Sleep Apnea
  • Sleep apnea affects patients' lives and health to varying degrees, and accurately distinguishing the type of sleep apnea is a prerequisite for choosing the appropriate treatment.
  • the existing methods for determining the type of apnea events are mainly as follows. One is to determine the occurrence of apnea events by monitoring the statistical characteristics of the respiratory flow signal. One of the specific methods is to use the peak flow velocity of the patient during normal breathing or The percentage of the average value is used as the threshold value, and then the amplitude of each respiratory flow signal is compared with this threshold value. The period below this threshold is marked as the apnea event period; the second method is to use a short period Compare with the variance of the long-term respiratory flow signal (considered as the normal breathing period).
  • the apnea event period When the variance of the short-term respiratory flow signal is lower than a certain percentage of the long-term respiratory flow signal variance, mark this period as the apnea event period; determine After the period of apnea event, the type of apnea event is determined by monitoring the frequency characteristics of the apnea signal.
  • the specific method is to convert the apnea signal to a frequency signal, and to distinguish different apnea events by monitoring the amplitude of the frequency signal Types of.
  • Another method is to judge the apnea event by monitoring the amplitude and frequency characteristics of the thoraco-abdominal band signal. Specifically, the time period when the amplitude of the thoraco-abdominal band signal is significantly reduced is marked as the apnea event period, and The frequency of the vibration is used to distinguish between different types of apnea events.
  • the existing methods for determining the type of apnea events have the following problems.
  • the information contained in a single apnea signal is insufficient, and the accuracy of apnea events is not high.
  • CSA and OSA type events are not obvious in the characteristics of respiratory flow signals, and it is difficult to determine the type of event based on the frequency characteristics of respiratory flow signals.
  • the technique of judging apnea events based on the variance of short-term and long-term respiratory flow signals At each moment, the variance information of the respiratory flow signal needs to be stored for a long period of time. The complexity of the algorithm is high and it is not easy to implement.
  • a method, a device, and a storage medium for determining the type of apnea event are provided.
  • a method for determining the type of apnea event including:
  • the type of the apnea event is determined according to the magnitude of the derivative of the respiratory flow signal and the magnitude of the thoracic and abdominal motion signal.
  • the determining the type of the apnea event according to the amplitude of the respiratory flow signal derivative and the chest and abdominal motion signal derivative includes:
  • the pre-processing the original breathing flow signal and the original thoracoabdominal motion signal to obtain the breathing flow signal and thoracoabdominal motion signal include:
  • the original signal of the respiratory flow (Flow original) of the original and thoracoabdominal motion signal (Res original) truncation processing are done according to the formula, to obtain respiratory flow signal processing (Flow at) and thoracoabdominal motion signal processing (Res office):
  • Flow former (i) represents the i-th sampling point of the original respiratory flow signal
  • Flow at (i) represents the respiratory flow signal after processing the i-th sampling point
  • Res original (i) represents the original i-th sampling point Thoracoabdominal motion signal
  • at Res represents the processed thoracoabdominal motion signal at the ith sampling point
  • A is the original truncation threshold of Flow
  • A>0 B is the original truncation threshold of Res, And B>0;
  • the calculating the derivative of the respiratory flow signal and the thoracic and abdominal motion signal to obtain the amplitude of the respiratory flow signal derivative and the thoracic and abdominal motion signal derivative includes:
  • step is the derivative step size
  • N is the total length of the respiratory flow signal sequence and the thoracoabdominal band signal sequence
  • Flow (i) represents the respiratory flow signal at the i-th sampling point
  • Flow (i-step) represents the i- The respiratory flow signal at step sampling points
  • Flow (i-2step) represents the breathing flow signal at the i-2step sampling point
  • Flow "(i) represents the magnitude of the second derivative of the breathing flow signal at the i sampling point
  • Res (i) represents the thoracoabdominal motion signal at the i-th sampling point
  • Res (i-step) represents the thoracoabdominal motion signal at the i-step sampling point
  • Res (i-2step) represents the i-2step sampling point
  • Res" (i) represents the second derivative amplitude of the thoracoabdominal motion signal at the i-th sampling point.
  • the value of Flow ′′ (i) is the first preset value
  • the value of Res ′′ (i) is the second preset value
  • determining the apnea event period according to the amplitude of the derivative of the respiratory flow signal and a first preset threshold Thres_flow includes:
  • the duration of the first period corresponding to the i-th sampling point to the M-th sampling point is greater than 10s, the first period is marked as the first period of the apnea event.
  • determining the type of the apnea event according to the amplitude of the thoracoabdominal exercise signal corresponding to the apnea event period and a second preset threshold Thres_Res includes:
  • Res ′′ (i) corresponding to the first period of the apnea event.
  • Res ′′ (i) ⁇ Thres_res, 0 ⁇ i ⁇ M
  • the type of the apnea event in the first period of the apnea event is Obstructive Sleep Apnea (Obstructive Sleep Apnea, OSA);
  • the type of the apnea event in the first period of the apnea event is central nervous sleep apnea (CSA);
  • a method for determining the type of apnea event including:
  • the original respiratory flow signal (Flow original ) of the patient within a preset time, the original respiratory flow signal representing the respiratory airflow of the patient, wherein the number of the original respiratory flow signals is greater than 1;
  • the preprocessing the original respiratory flow signal to obtain a respiratory flow signal (Flow) includes:
  • Flow original (i) represents the original breathing flow signal at the i-th sampling point, (i) at Flow represents the processed breathing flow signal at the i-th sampling point, A is the amplitude truncation threshold of the Flow original , and A>0;
  • Respiratory flow signal after the treatment (Flow at) band-pass filtering process, to obtain the respiratory flow signal (Flow).
  • the preprocessing the raw abdominal motion signal (original Res), to give thoracoabdominal signal (Res) comprises:
  • Res original (i) represents the original thoracoabdominal motion signal at the ith sampling point
  • Res at (i) represents the processed thoracoabdominal motion signal at the ith sampling point
  • B is the amplitude truncation threshold of the original Res, and B>0;
  • An apnea event type judgment device includes:
  • An obtaining unit configured to obtain a patient's original respiratory flow signal ( primary Flow) and original thoracoabdominal motion signal (Res original ) within a preset time, the original respiratory flow signal representing the patient's respiratory airflow, the original chest Abdominal motion signals characterize thoracoabdominal motion of the patient, wherein the number of the original respiratory flow signals is greater than 1, and the number of the original thoracoabdominal motion signals is greater than 1;
  • a preprocessing unit configured to preprocess the original breathing flow signal and the original thoracoabdominal motion signal to obtain a breathing flow signal (Flow) and thoracoabdominal motion signal (Res);
  • a calculation unit configured to calculate a derivative of the respiratory flow signal and a derivative of the chest-abdominal motion signal to obtain a magnitude of the derivative of the respiratory flow signal and a derivative of the chest-abdominal motion signal;
  • the judging unit is configured to judge the type of the apnea event according to the amplitude of the derivative of the respiratory flow signal and the amplitude of the thoracic and abdominal motion signal.
  • the determining unit includes:
  • the apnea event period determination module is configured to determine an apnea event period according to the magnitude of a derivative of the respiratory flow signal and a first preset threshold Thres_flow.
  • An apnea event type determination module is configured to determine the apnea event type according to the thoracoabdominal motion signal derivative amplitude corresponding to the apnea event period and a second preset threshold Thres_Res.
  • the pre-processing unit includes:
  • Truncation processing module for the original respiratory flow signal (Flow original) of the original and thoracoabdominal motion signal (Res original) truncation processing are done according to the formula, to obtain respiratory flow signal processing (Flow at) and treated after thoracoabdominal signal (Res): the
  • Flow former (i) represents the i-th sampling point of the original respiratory flow signal
  • Flow at (i) represents the respiratory flow signal after processing the i-th sampling point
  • Res original (i) represents the original i-th sampling point Thoracoabdominal motion signal
  • at Res represents the processed thoracoabdominal motion signal at the ith sampling point
  • A is the original truncation threshold of Flow
  • B is the original truncation threshold of Res
  • Band-pass filter processing module for the respiratory flow signal processing (Flow at) and the abdominal motion signal processing (at Res) band-pass filtering process, to obtain the respiratory flow signal (Flow) And the thoracoabdominal exercise signal (Res).
  • the calculation unit is specifically configured to calculate a second derivative of the breathing flow signal (Flow) and the thoracoabdominal motion signal (Res) according to the following formulas respectively to obtain the amplitude of the second derivative of the respiratory flow signal and the chest
  • the magnitude of the second derivative of the abdominal motion signal is specifically configured to calculate a second derivative of the breathing flow signal (Flow) and the thoracoabdominal motion signal (Res) according to the following formulas respectively to obtain the amplitude of the second derivative of the respiratory flow signal and the chest.
  • step is the derivative step size
  • N is the total length of the respiratory flow signal sequence and the thoracoabdominal band signal sequence
  • Flow (i) represents the respiratory flow signal at the i-th sampling point
  • Flow (i-step) represents the i- The respiratory flow signal at step sampling points
  • Flow (i-2step) represents the respiratory flow signal at the i-2step sampling point
  • Flow "(i) represents the magnitude of the second derivative of the respiratory flow signal at the i sampling point
  • Res (i) represents the thoracoabdominal motion signal at the i-th sampling point
  • Res (i-step) represents the thoracoabdominal motion signal at the i-step sampling point
  • Res (i-2step) represents the i-2step sampling point
  • Res" (i) represents the second derivative amplitude of the thoracoabdominal motion signal at the i-th sampling point.
  • An apnea event type judgment device includes:
  • a first acquiring unit is configured to acquire a patient's original respiratory flow signal (Flow original ) within a preset time, the original respiratory flow signal representing the respiratory airflow of the patient, wherein the number of the original respiratory flow signals is greater than 1 ;
  • a first preprocessing unit configured to preprocess the original breathing flow signal to obtain a breathing flow signal (Flow);
  • a first calculation unit configured to calculate a derivative of the respiratory flow signal to obtain a magnitude of a derivative of the respiratory flow signal
  • a first determining unit configured to determine an apnea event period according to a magnitude of a derivative of the respiratory flow signal and a first preset threshold Thres_flow;
  • a second acquisition unit for acquiring an original signal thoracoabdominal movements (formerly Res) the period apnea event
  • a second pre-processing unit configured to pre-process the original thoracoabdominal motion signal (Res original ) to obtain a thoracoabdominal motion signal (Res);
  • a second calculation unit configured to calculate a derivative of the thoracoabdominal motion signal to obtain a magnitude of the thoracoabdominal motion signal derivative
  • a second determining unit is configured to determine the apnea event type according to the magnitude of the thoracic and abdominal motion signal derivative and a second preset threshold Thres_Res.
  • the first pre-processing unit includes:
  • a first truncation processing module for the original respiratory flow signal (Flow ogen) do truncation processing according to the following formula, to obtain respiratory flow signal processing (Flow):
  • Flow original (i) represents the original breathing flow signal at the i-th sampling point
  • (i) at Flow represents the processed breathing flow signal at the i-th sampling point
  • A is the amplitude truncation threshold of the Flow original , and A> 0.
  • the second pre-processing unit includes:
  • the second truncation processing module for the original signal thoracoabdominal movements (formerly Res) according to the formula do truncation processing, to obtain processed signals thoracoabdominal movements (Res):
  • Res original (i) represents the original thoracoabdominal motion signal at the ith sampling point
  • Res at (i) represents the processed thoracoabdominal motion signal at the ith sampling point
  • B is the amplitude truncation threshold of the original Res
  • Second band-pass filter processing means for band-pass filtering process on the thoracoabdominal signal processing (at Res), to give the thoracoabdominal signal (Res).
  • a computer-readable storage medium having computer-executable instructions stored on the computer-readable storage medium.
  • the processor causes the processor to perform the steps of any one of the foregoing methods. .
  • Embodiment of the present application by obtaining the original patient respiratory flow signal within a preset time (Flow ogen) and thoracoabdominal original signal (original Res), the original respiratory flow signal (Flow former) and the original signal thoracoabdominal movements ( Res original ) to preprocess to eliminate the effects of invalid signals and interference signals, and perform derivative operations on the respiratory flow signal (Flow) and thoracoabdominal motion signal (Res) obtained after preprocessing, to obtain the respiratory flow signal derivative amplitude and The magnitude of the thoracic and abdominal motion signal derivative, the apnea event period is determined according to the magnitude of the respiratory flow signal derivative and the first preset threshold Thres_flow, and then the CSA is distinguished according to the different trends of the thoracoabdominal motion signal corresponding to the apnea event period / OSA / MSA, the three types of apnea events, the solution has low complexity and high accuracy of judgment results.
  • Flow ogen a preset
  • FIG. 1 is a flowchart of a method for determining an apnea event type according to an embodiment
  • FIG. 2 is a flowchart of a method for determining an apnea event type according to another embodiment
  • FIG. 3 is a structural block diagram of a device for determining an apnea event type according to an embodiment
  • FIG. 4 is a structural block diagram of a device for determining an apnea event type according to another embodiment
  • FIG. 5 is a schematic diagram of an internal structure of a computer device according to an embodiment.
  • the patient's original breathing flow signal (Flow original ) and original thoracoabdominal motion signal (Res original ) can be monitored and collected by the nasal airflow and thoracoabdominal exercise box, and can also be monitored and collected by other sleep monitors.
  • the nasal airflow and thoracoabdominal exercise box or other sleep monitor sends the collected original respiratory flow signals and the original thoracoabdominal exercise signals to the apnea event type determination device in real time or at preset intervals.
  • the apnea event type determination device executes the process steps included in the apnea event type determination method described in this application.
  • a method for determining an apnea event type includes the following steps: Step S101: Obtain an original breathing flow signal (Flow original ) and an original thoracoabdominal motion signal of a patient within a preset time. (Res original ), the original respiratory flow signal characterizes the patient's respiratory airflow, the original thoracoabdominal motion signal characterizes the patient's thoracoabdominal motion, wherein the number of the original respiratory flow signal is greater than 1, the The number of original thoracoabdominal motion signals is greater than one.
  • the original breathing flow signal (the original Flow) and the original thoracoabdominal exercise signal (the original Res) of the patient are obtained within a preset time, and the preset time may be the time when the patient wears a sleep monitor at night, such as 7 hours or 8 hours.
  • step S102 the original respiratory flow signal and the original thoracoabdominal motion signal are preprocessed to obtain a respiratory flow signal (Flow) and a thoracoabdominal motion signal (Res).
  • Flow respiratory flow signal
  • Res thoracoabdominal motion signal
  • the sleep monitor collects information while the patient is asleep, and during this period, the patient may experience a situation such as turning over and sudden large-scale movement of the limb. Therefore, abnormal spike signals are usually generated, which affects the accuracy of judgment. Therefore, these spike signals need to be removed.
  • amplitude truncation technology can be used to perform truncation processing.
  • Step S103 Calculate a derivative of the respiratory flow signal and a derivative of the thoracoabdominal motion signal to obtain a magnitude of the respiratory flow signal derivative and a thoracoabdominal motion signal derivative.
  • the first derivative of the respiratory flow signal and the first derivative of the thoracoabdominal motion signal can be calculated, and the second derivative of the respiratory flow signal and the second derivative of the thoracoabdominal motion signal can also be calculated to obtain the amplitude of the respiratory flow signal derivative and the chest Abdominal motion signal derivative magnitude.
  • step S104 the type of the apnea event is determined according to the magnitude of the derivative of the respiratory flow signal and the magnitude of the thoracic and abdominal motion signal.
  • step S104 further includes:
  • Step S1041 determining an apnea event period according to the magnitude of the respiratory flow signal derivative and a first preset threshold Thres_flow, wherein the first preset threshold Thres_flow is a derivative of a respiratory flow signal corresponding to a stable normal breathing period of the patient in the past Value, or the mean value of the derivative amplitude of the respiratory flow signal corresponding to the past several stable normal breaths, and the first preset threshold Thres_flow may be dynamically changed according to the change of the patient's normal breath;
  • Step S1042 determining the type of the apnea event according to the magnitude of the thoracic and abdominal motion signal derivative corresponding to the apnea event period and a second preset threshold Thres_Res, where the second preset threshold Thres_Res is stable according to the past period
  • the magnitude of the derivative of the thoracoabdominal motion signal corresponding to the normal breathing of the patient, or the average of the magnitude of the derivative of the thoracoabdominal motion signal corresponding to the past several stable normal breathing, and the second preset threshold Thres_Res can be based on the patient's Normal breathing changes dynamically.
  • the respiratory flow signal to obtain the original patient within a preset time (Flow ogen) and thoracoabdominal original signal (original Res), the original respiratory flow signal (Flow ogen) thoracoabdominal movements and the original signal (original Res ) Perform preprocessing to exclude the effects of invalid signals and interference signals, and perform derivative operations on the respiratory flow signal (Flow) and thoracoabdominal motion signal (Res) obtained after preprocessing, to obtain the respiratory flow signal derivative amplitude and thoracoabdominal The magnitude of the derivative of the motion signal.
  • the apnea event period is determined according to the magnitude of the respiratory flow signal derivative and the first preset threshold Thres_flow, and then the CSA / OSA is distinguished according to the different trends of the chest and abdominal motion signals corresponding to the apnea event period.
  • the three apnea events of / MSA have low complexity and high accuracy.
  • step S102 preprocessing the original breathing flow signal and the original thoracoabdominal motion signal to obtain the breathing flow signal (Flow) and thoracoabdominal motion signal (Res) include:
  • Step S1021 the original respiratory flow signal (Flow original) of the original and thoracoabdominal motion signal (Res original) truncation processing are done according to the formula, to obtain respiratory flow signal processing (Flow at) and post-treatment chest abdominal motion signal (Res): the
  • Flow former (i) represents the i-th sampling point of the original respiratory flow signal
  • Flow at (i) represents the respiratory flow signal after processing the i-th sampling point
  • Res original (i) represents the original i-th sampling point Thoracoabdominal motion signal
  • at Res represents the processed thoracoabdominal motion signal at the ith sampling point
  • A is the original truncation threshold of Flow
  • B is the original truncation threshold of Res
  • amplitude truncation technology can be used to perform truncation processing to reduce the effect of invalid signals, so that the processed respiratory flow signal and the processed thoracoabdominal motion signal quickly converge to a reasonable range.
  • A can be set according to the peak value of the original respiratory flow signal corresponding to the stable normal breathing in the past, or the average value of the peak value of the original respiratory flow signal corresponding to the stable normal breathing in the past.
  • the peak value of the original thoracoabdominal motion signal corresponding to the normal breathing of the patient, or the average of the peak value of the original thoracoabdominal motion signal corresponding to the past several stable normal breathing; A and B can be based on the patient's original breathing during different normal breathing periods.
  • the peak value of the flow signal and the peak value of the original thoracoabdominal motion signal are adaptively adjusted. For example, for patients with large breathing amplitude under normal breathing conditions, a larger Flow original amplitude truncation threshold and a larger value can be set in each period. Res original amplitude truncation threshold, and for patients with weak breathing amplitude under normal breathing conditions, the Flow original amplitude truncation threshold and Res original amplitude truncation threshold can be appropriately reduced.
  • Step S1022 the respiratory flow signal after the treatment (Flow at) band-pass filtering process and the abdominal motion signal processing (at Res), to obtain the respiratory flow signal (Flow) and the chest Motion Signal (Res).
  • a respiratory flow signal processing (Flow at) and thoracoabdominal motion signal processing (at Res) respectively band-pass filter performs a filtering process at a high frequency and low frequency signals outside the band pass filtered, to obtain a useful signal of the respiratory flow (flow) within a band-limited frequency band, and Thoracoabdominal exercise signal (Res) signal.
  • the upper and lower pass band of the pass band of the upper and lower respiratory flow signal processing (Flow at) filtering process can be set on the abdominal motion signal processing (at Res) filtering process may be performed the upper limit of the pass band is set, for example, of respiratory flow signal processing (flow at) filtering process may be provided 3Hz, the lower limit of the pass band can be set to 0.05Hz.
  • the influence of abnormal signals and interference signals can be eliminated, and the accuracy of the apnea event type judgment can be improved.
  • step S103 calculating the derivative of the respiratory flow signal and the thoracic and abdominal motion signal to obtain the amplitude of the respiratory flow signal derivative and the thoracic and abdominal motion signal amplitude include:
  • step is the derivative step size
  • N is the total length of the respiratory flow signal sequence and the thoracoabdominal band signal sequence
  • Flow (i) represents the respiratory flow signal at the i-th sampling point
  • Flow (i-step) represents the i- The respiratory flow signal at step sampling points
  • Flow (i-2step) represents the respiratory flow signal at the i-2step sampling point
  • Flow "(i) represents the magnitude of the second derivative of the respiratory flow signal at the i sampling point
  • Res (i) represents the thoracoabdominal motion signal at the i-th sampling point
  • Res (i-step) represents the thoracoabdominal motion signal at the i-step sampling point
  • Res (i-2step) represents the i-2step sampling point
  • Res" (i) represents the second derivative amplitude of the thoracoabdominal motion signal at the i-th sampling point.
  • the value of Flow ′′ (i) is the first preset value
  • the value of Res ′′ (i) is the second preset value
  • the probability of apnea occurring at the beginning of falling asleep is relatively small, so when 1 ⁇ i ⁇ 2step, the value of Flow ′′ (i) can be set to the first preset value, and Res ′′ (i) can be set to the value It is a second preset value to facilitate subsequent calculations.
  • the first preset value may be a value greater than the first preset threshold Thres_flow
  • the second preset value may be a value greater than the second preset threshold Thres_Res.
  • first-order derivatives can also be calculated for the respiratory flow signal (Flow) and the thoracoabdominal motion signal (Res) respectively, to obtain the first-order derivative amplitude of the respiratory flow signal and the first-order derivative amplitude of the thoracoabdominal motion signal.
  • Flow respiratory flow signal
  • Res thoracoabdominal motion signal
  • step is the derivative step size
  • N is the total length of the respiratory flow signal sequence and the thoracoabdominal band signal sequence
  • Flow (i) represents the respiratory flow signal at the i-th sampling point
  • Flow (i-step) represents the i- The respiratory flow signal at step sampling points
  • Flow ′ (i) represents the first derivative magnitude of the respiratory flow signal at the i-th sampling point
  • Res (i) represents the thoracoabdominal motion signal at the i-th sampling point
  • Res (i -step) represents the thoracoabdominal motion signal at the i-step sampling point
  • Res ′ (i) represents the magnitude of the first derivative of the thoracoabdominal motion signal at the ith sampling point.
  • step S1041 determining the apnea event period according to the magnitude of the respiratory flow signal derivative and the first preset threshold Thres_flow is specifically:
  • the duration of the first period corresponding to the i-th sampling point to the M-th sampling point is greater than 10s, the first period is marked as the first period of the apnea event.
  • the first preset threshold Thres_flow is based on the magnitude of the second derivative of the respiratory flow signal corresponding to the patient's stable normal breathing in the past, or the magnitude of the second derivative of the respiratory flow signal corresponding to the stable normal breathing in the past.
  • the average value is set, and the first preset threshold Thres_flow may be dynamically changed according to a change in the normal breathing of the patient.
  • the duration of the first period corresponding to the i-th sampling point to the M-th sampling point is greater than 10s, then the first period is marked as the first period of the apnea event.
  • a time threshold Time_Threshold may be set, as long as the spike original between the two apnea event periods is original If the duration of the respiratory flow signal is less than Time_Threshold, it can be considered that the patient only experienced sudden wheezing or other movements during a period of apnea event. In order to make the judgment result closer to the actual situation of the patient, this embodiment uses the two periods of apnea event. Combined into a period of apnea events.
  • Step S1042 determining the type of the apnea event according to the magnitude of the thoracic and abdominal motion signal derivative corresponding to the apnea event period and a second preset threshold Thres_Res is:
  • Res ′′ (i) corresponding to the first period of the apnea event.
  • Res ′′ (i) ⁇ Thres_res, 0 ⁇ i ⁇ M
  • the type of the apnea event in the first period of the apnea event is Obstructive Sleep Apnea (Obstructive Sleep Apnea, OSA);
  • the type of the apnea event in the first period of the apnea event is central nervous sleep apnea (CSA);
  • the second preset threshold Thres_Res is based on the magnitude of the second derivative of the thoracoabdominal motion signal corresponding to the patient's stable normal breathing in the past, or the magnitude of the second derivative of the thoracoabdominal motion signal corresponding to the stable normal breathing in the past.
  • the average value of the values is set, and the second preset threshold Thres_Res may be dynamically changed according to a change in the normal breathing of the patient.
  • the OSA event is caused by the obstruction or narrowing of the airway, it is mainly manifested in the significant reduction of the respiratory flow signal, the amplitude of the thoracoabdominal motion signal is weakened but there are still certain fluctuations, which makes the second order of thoracoabdominal motion signals during the OSA event period.
  • the magnitude of the derivative is above the second preset threshold Thres_res; and in the CSA event, due to the respiratory central nervous disorder, the chest and abdomen movement trend changes little, and the movement is approximately stopped, so the magnitude of the second derivative of the CSA event period is below Thres_res, In this way, the type of apnea event can be distinguished by monitoring the magnitude of the second derivative of the thoracoabdominal motion signal corresponding to the period of apnea event.
  • Res ′′ (i) corresponding to the first period of the apnea event.
  • ⁇ Thres_res, 0 ⁇ i ⁇ M the first period of the apnea event is Obstructive Sleep Apnea , OSA
  • ⁇ Thres_res, 0 ⁇ i ⁇ M the central sleep Apnea (CSA) in the first period of the apnea event; when from the i-th sampling point to the first M 1 sampling point corresponding to a second time period satisfies
  • the magnitude of the derivative of the respiratory flow signal in step S1041 may be the magnitude of the first derivative of the respiratory flow signal
  • the first preset threshold Thres_flow may be the first derivative of the respiratory flow signal corresponding to the patient's stable normal breathing in the past.
  • the amplitude or the average value of the first derivative amplitude of the respiratory flow signal corresponding to the past several stable normal breaths is set, and the first preset threshold Thres_flow may be dynamically changed according to changes in the normal breathing of the patient.
  • the specific method for determining the apnea event period according to the amplitude of the first derivative of the respiratory flow signal and the first preset threshold Thres_flow may refer to the above description, and will not be repeated here.
  • the magnitude of the thoracic and abdominal motion signal derivative in step S1042 may be the magnitude of the first derivative of the thoracic and abdominal motion signal
  • the second preset threshold Thres_Res is the magnitude of the first derivative of the thoracic and abdominal motion signal corresponding to the patient's stable normal breathing in the past.
  • the average value of the first derivative amplitude of the thoracoabdominal motion signal corresponding to the past several stable normal breaths is set, and the second preset threshold Thres_Res can be dynamically changed according to changes in the normal breathing of the patient.
  • the specific method for determining the type of the apnea event according to the amplitude of the first-order derivative of the thoracoabdominal motion signal corresponding to the apnea event period and the second preset threshold Thres_Res can refer to the above description, and will not be repeated here.
  • the amplitude of the derivative of the respiratory flow signal (the first derivative of the respiratory flow signal or the second derivative of the respiratory flow signal) and the amplitude of the derivative of the thoracoabdominal motion signal (the first derivative of the thoracoabdominal motion signal or the thorax
  • the magnitude of the second derivative of the abdominal motion signal reflects the change trend of the respiratory flow signal and the thoracoabdominal motion signal corresponding to different apnea events.
  • the type of the apnea event is determined by the magnitude of the derivative of the respiratory flow signal and the magnitude of the thoracoabdominal motion signal.
  • a method for determining an apnea event type includes the following steps:
  • Step S201 Obtain the original respiratory flow signal (Flow original ) of the patient within a preset time, and the original respiratory flow signal represents the respiratory airflow of the patient, wherein the number of the original respiratory flow signals is greater than one.
  • the original respiratory flow signal (Flow original ) of the patient is acquired, and the preset time may be the time when the patient wears the sleep monitor at night, such as 7 hours or 8 hours.
  • step S202 the original breathing flow signal is pre-processed to obtain a breathing flow signal (Flow).
  • step S202 includes:
  • Step S2021 the original respiratory flow signal (Flow ogen) do truncation processing according to the following formula, to obtain respiratory flow signal processing (Flow):
  • Flow original (i) represents the original breathing flow signal at the i-th sampling point
  • (i) at Flow represents the processed breathing flow signal at the i-th sampling point
  • A is the amplitude truncation threshold of the Flow original , and A> 0.
  • amplitude truncation technology can be used to perform truncation processing to reduce the effect of invalid signals, and to make the processed respiratory flow signal quickly converge to a reasonable range.
  • A can be set according to the peak value of the original respiratory flow signal corresponding to the stable normal breathing in the past, or the average value of the peak value of the original respiratory flow signal corresponding to the stable stable breathing in the past. The peak value of the original respiratory flow signal during the breathing period is adaptively adjusted.
  • a larger Flow original amplitude truncation threshold can be set in each period, and for normal breathing conditions For patients with weak lower breathing amplitude, the original truncation threshold of Flow can be appropriately reduced.
  • Step S2022 the respiratory flow signal after the treatment (Flow at) band-pass filtering process, to obtain the respiratory flow signal (Flow).
  • a band pass filter need respiratory flow signal processing (Flow at) filtering Processing to filter out high-frequency signals and low-frequency signals outside the passband to obtain useful breathing flow signals (Flow) in the band-limited band.
  • Flow at respiratory flow signal processing
  • the upper limit of the passband may be set to 3Hz
  • the lower limit of the passband may be set to 0.05Hz.
  • Step S203 Calculate the derivative of the respiratory flow signal to obtain the amplitude of the derivative of the respiratory flow signal.
  • the first derivative of the respiratory flow signal is calculated to obtain the amplitude of the first derivative of the respiratory flow signal
  • the second derivative of the respiratory flow signal may also be calculated to obtain the amplitude of the second derivative of the respiratory flow signal.
  • the first derivative of the respiratory flow signal can be calculated according to the following formula:
  • step is the derivative step size
  • N is the total length of the respiratory flow signal sequence and the thoracoabdominal band signal sequence
  • Flow (i) represents the respiratory flow signal at the i-th sampling point
  • Flow (i-step) represents the i- The respiratory flow signal at step sampling points
  • Flow ′ (i) represents the amplitude of the first derivative of the respiratory flow signal at the ith sampling point.
  • the second derivative of the respiratory flow signal can be calculated according to the following formula:
  • step is the derivative step size
  • N is the total length of the respiratory flow signal sequence and the thoracoabdominal band signal sequence
  • Flow (i) represents the respiratory flow signal at the i-th sampling point
  • Flow (i-step) represents the i- The breath flow signal at step sampling points
  • Flow (i-2step) represents the breathing flow signal at the i-2step sampling point
  • Flow "(i) represents the magnitude of the second derivative of the breathing flow signal at the i sampling point.
  • the value of Flow ′′ (i) is the first preset value.
  • Step S204 Determine an apnea event period according to the magnitude of the derivative of the respiratory flow signal and a first preset threshold Thres_flow.
  • the flow "(i) corresponding to each sampling point i is compared with the first preset threshold Thres_flow, and when
  • the duration of the first period corresponding to the i-th sampling point to the M-th sampling point is greater than 10s, the first period is marked as the first period of the apnea event.
  • the first preset threshold Thres_flow is set according to the magnitude of the second derivative of the respiratory flow signal corresponding to a stable normal breathing in the past, or the mean value of the second derivative of the respiratory flow signal corresponding to the stable normal breathing in the past. And the first preset threshold Thres_flow may be dynamically changed according to changes in the normal breathing of the patient.
  • a time threshold Time_Threshold may be set, as long as the spike original between the two apnea event periods is original If the duration of the respiratory flow signal is less than Time_Threshold, it can be considered that the patient only experienced sudden wheezing or other movements during a period of apnea event. In order to make the judgment result closer to the actual situation of the patient, this embodiment uses the two periods of apnea event. Combined into a period of apnea events.
  • the amplitude of the derivative of the respiratory flow signal in the above step may be the amplitude of the first derivative of the respiratory flow signal
  • the first preset threshold Thres_flow may be the first derivative of the respiratory flow signal corresponding to the patient's stable normal breathing in the past.
  • the amplitude or the average value of the first derivative amplitude of the respiratory flow signal corresponding to the past several stable normal breaths is set, and the first preset threshold Thres_flow may be dynamically changed according to changes in the normal breathing of the patient.
  • the specific method for determining the apnea event period according to the amplitude of the first derivative of the respiratory flow signal and the first preset threshold Thres_flow may refer to the above description, and will not be repeated here.
  • Step S205 acquires the original signal thoracoabdominal (formerly Res) said apnea event period.
  • the original thoracoabdominal motion signal (Res original ) corresponding to the first period of the apnea event is acquired.
  • Step S206 the original abdominal motion signal (Res original) pretreatment, to obtain thoracoabdominal signal (Res).
  • step S206 includes:
  • Step S2061 the original abdominal motion signal (formerly Res) according to the formula do truncation processing, to obtain processed signals thoracoabdominal movements (Res):
  • Res original (i) represents the original thoracoabdominal motion signal at the ith sampling point
  • Res at (i) represents the processed thoracoabdominal motion signal at the ith sampling point
  • B is the amplitude truncation threshold of the original Res
  • the sleep monitor collects information while the patient is asleep, and during this period, the patient may experience a situation such as turning over and sudden large-scale movement of the limb. Therefore, abnormal spike signals are usually generated, which affects the accuracy of judgment. Therefore, these spike signals need to be removed.
  • the amplitude truncation technique can be used to perform truncation processing to reduce the effect of invalid signals, so that the processed thoracoabdominal motion signals quickly converge to a reasonable range.
  • B can be set according to the peak value of the original thoracoabdominal motion signal corresponding to the patient's stable normal breathing in the past, or the average value of the peak value of the original thoracoabdominal motion signal corresponding to the stable stable breathing in the past several times; B can be set according to the patient's
  • the peaks of the original thoracoabdominal motion signals in different normal breathing periods are adjusted adaptively. For example, for patients with large breathing amplitude under normal breathing conditions, a larger Res original amplitude truncation threshold can be set in each period. patients with normal breathing breathing magnitude weaker, may be appropriate to reduce the amplitude of the original Res cutoff threshold.
  • Step S2062 the motion of the chest and abdomen of the signal processing (at Res) band-pass filtering process, to obtain the thoracoabdominal signal (Res).
  • a band-pass filter need thoracoabdominal motion signal processing (at Res) Filtering is performed to filter out high-frequency signals and low-frequency signals outside the pass band, and obtain useful thoracoabdominal motion signals (Res) in the band-limited band.
  • Step S207 Calculate the derivative of the thoracoabdominal motion signal to obtain the amplitude of the thoracoabdominal motion signal derivative.
  • the first derivative of the thoracoabdominal motion signal is calculated to obtain the amplitude of the first derivative of the thoracoabdominal motion signal
  • the second derivative of the thoracoabdominal motion signal can also be calculated to obtain the amplitude of the second derivative of the thoracoabdominal motion signal.
  • the first derivative of the thoracoabdominal motion signal can be calculated according to the following formula:
  • step is the derivative step
  • N is the total length of the thoracoabdominal band signal sequence
  • Res (i) is the thoracoabdominal motion signal at the ith sampling point
  • Res (i-step) is the i-step sampling point
  • Res ′ (i) represents the first derivative amplitude of the thoracoabdominal motion signal at the ith sampling point.
  • the second derivative of the thoracoabdominal motion signal can be calculated according to the following formula:
  • step is the derivative step size
  • N is the total length of the respiratory flow signal sequence
  • Res (i) represents the thoracoabdominal motion signal at the i-th sampling point
  • Res (i-step) represents the i-step sampling point.
  • Res (i-2step) represents the thoracoabdominal motion signal at the i-2step sampling point
  • Res "(i) represents the second derivative amplitude of the thoracoabdominal motion signal at the ith sampling point.
  • the value of Res ′′ (i) is the second preset value.
  • Step S208 Determine the type of the apnea event according to the amplitude of the thoracic and abdominal motion signal derivative and a second preset threshold Thres_Res.
  • Res ′′ (i) corresponding to the first period of the apnea event, when
  • OSA Obstructive Sleep Apnea
  • the type of the apnea event in the first period of the apnea event is central nervous sleep apnea (CSA);
  • the second preset threshold Thres_Res is the mean value of the second derivative of the thoracoabdominal motion signal corresponding to the patient's past stable stable breath in the past, or the mean value of the second derivative of the thoracoabdominal motion signal corresponding to the past several stable normal breathing. It is set, and the second preset threshold Thres_Res can be dynamically changed according to the change of the normal breathing of the patient.
  • the derivative of the respiratory flow signal data is first obtained to obtain the amplitude of the derivative of the respiratory flow signal, the apnea event period is determined according to the amplitude of the respiratory flow signal derivative, and then the thoracoabdominal exercise signal data corresponding to the apnea event period is determined.
  • Derivative operation is performed to obtain the magnitude of the thoracic and abdominal motion signal derivative, and then the type of apnea event in the apnea event period is determined based on the magnitude of the thoracic and abdominal motion signal derivative.
  • the complexity of the solution is low, and the accuracy of the judgment result is high, To a certain extent, computing resources are saved.
  • steps in the embodiments of the present application are not necessarily performed sequentially in the order indicated by the step numbers. Unless explicitly stated in this document, the execution of these steps is not strictly limited, and these steps can be performed in other orders. Moreover, at least a part of the steps in each embodiment may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily performed at the same time, but may be performed at different times. The execution of these sub-steps or stages The sequence is not necessarily performed sequentially, but may be performed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.
  • a device for determining an apnea event type includes:
  • An obtaining unit configured to obtain a patient's original respiratory flow signal ( primary Flow) and original thoracoabdominal motion signal (Res original ) within a preset time, the original respiratory flow signal representing the patient's respiratory airflow, the original chest Abdominal motion signals characterize thoracoabdominal motion of the patient, wherein the number of the original respiratory flow signals is greater than 1, and the number of the original thoracoabdominal motion signals is greater than 1;
  • a preprocessing unit configured to preprocess the original breathing flow signal and the original thoracoabdominal motion signal to obtain a breathing flow signal (Flow) and thoracoabdominal motion signal (Res);
  • a calculation unit configured to calculate a derivative of the respiratory flow signal and a derivative of the chest-abdominal motion signal to obtain a magnitude of the derivative of the respiratory flow signal and a derivative of the chest-abdominal motion signal;
  • the judging unit is configured to judge the type of the apnea event according to the amplitude of the derivative of the respiratory flow signal and the amplitude of the thoracic and abdominal motion signal.
  • the determining unit includes:
  • the apnea event period determination module is configured to determine an apnea event period according to the magnitude of a derivative of the respiratory flow signal and a first preset threshold Thres_flow.
  • An apnea event type determination module is configured to determine the apnea event type according to the thoracoabdominal motion signal derivative amplitude corresponding to the apnea event period and a second preset threshold Thres_Res.
  • the pre-processing unit includes:
  • Truncation processing module for the original respiratory flow signal (Flow original) of the original and thoracoabdominal motion signal (Res original) truncation processing are done according to the formula, to obtain respiratory flow signal processing (Flow at) and treated after thoracoabdominal signal (Res): the
  • Flow former (i) represents the i-th sampling point of the original respiratory flow signal
  • Flow at (i) represents the respiratory flow signal after processing the i-th sampling point
  • Res original (i) represents the original i-th sampling point Thoracoabdominal motion signal
  • at Res represents the processed thoracoabdominal motion signal at the ith sampling point
  • A is the original truncation threshold of Flow
  • B is the original truncation threshold of Res
  • Band-pass filter processing module for the respiratory flow signal processing (Flow at) and the abdominal motion signal processing (at Res) band-pass filtering process, to obtain the respiratory flow signal (Flow) And the thoracoabdominal exercise signal (Res).
  • the calculation unit is specifically configured to calculate a second derivative of the breathing flow signal (Flow) and the thoracoabdominal motion signal (Res) according to the following formulas respectively to obtain the amplitude of the second derivative of the respiratory flow signal and the chest
  • the magnitude of the second derivative of the abdominal motion signal is specifically configured to calculate a second derivative of the breathing flow signal (Flow) and the thoracoabdominal motion signal (Res) according to the following formulas respectively to obtain the amplitude of the second derivative of the respiratory flow signal and the chest.
  • step is the derivative step size
  • N is the total length of the respiratory flow signal sequence and the thoracoabdominal band signal sequence
  • Flow (i) represents the respiratory flow signal at the i-th sampling point
  • Flow (i-step) represents the i- The respiratory flow signal at step sampling points
  • Flow (i-2step) represents the respiratory flow signal at the i-2step sampling point
  • Flow "(i) represents the magnitude of the second derivative of the respiratory flow signal at the i sampling point
  • Res (i) represents the thoracoabdominal motion signal at the i-th sampling point
  • Res (i-step) represents the thoracoabdominal motion signal at the i-step sampling point
  • Res (i-2step) represents the i-2step sampling point
  • Res" (i) represents the second derivative amplitude of the thoracoabdominal motion signal at the i-th sampling point.
  • the calculation unit may be further configured to calculate first-order derivatives of the respiratory flow signal (Flow) and the chest-abdominal motion signal (Res), respectively, to obtain the magnitude of the first-order derivative of the respiratory flow signal and the chest-abdomen.
  • the magnitude of the first derivative of the motion signal can be calculated according to the following formula:
  • step is the derivative step size
  • N is the total length of the respiratory flow signal sequence and the thoracoabdominal band signal sequence
  • Flow (i) represents the respiratory flow signal at the i-th sampling point
  • Flow (i-step) represents the i- The respiratory flow signal at step sampling points
  • Flow ′ (i) represents the first derivative magnitude of the respiratory flow signal at the i-th sampling point
  • Res (i) represents the thoracoabdominal motion signal at the i-th sampling point
  • Res (i -step) represents the thoracoabdominal motion signal at the i-step sampling point
  • Res ′ (i) represents the magnitude of the first derivative of the thoracoabdominal motion signal at the ith sampling point.
  • the apnea event period determination module is specifically configured to compare the flow "(i) corresponding to each sampling point i with the first preset threshold Thres_flow, when
  • ⁇ Thres_flow, then mark the time corresponding to the sampling point i as the starting point of the possible apnea event, and continue to traverse subsequent sampling points Flow ′′ (j), j i + 1, ... N, From i sampling point to M sampling point:
  • the duration of the first period corresponding to the i-th sampling point to the M-th sampling point is greater than 10s, then the first period is marked as the first period of the apnea event.
  • the apnea event type determination module is specifically configured for Res ′′ (i) corresponding to the first period of the apnea event.
  • Res ′′ (i)
  • a device for determining an apnea event type includes:
  • a first acquiring unit is configured to acquire a patient's original respiratory flow signal (Flow original ) within a preset time, the original respiratory flow signal representing the respiratory airflow of the patient, wherein the number of the original respiratory flow signals is greater than 1 .
  • the first preprocessing unit is configured to preprocess the original breathing flow signal to obtain a breathing flow signal (Flow).
  • the first pre-processing unit includes:
  • a first truncation processing module for the original respiratory flow signal (Flow ogen) do truncation processing according to the following formula, to obtain respiratory flow signal processing (Flow):
  • Flow original (i) represents the original breathing flow signal at the i-th sampling point
  • (i) at Flow represents the processed breathing flow signal at the i-th sampling point
  • A is the amplitude truncation threshold of the Flow original , and A> 0.
  • a first calculation unit is configured to calculate a derivative of the respiratory flow signal to obtain a magnitude of a derivative of the respiratory flow signal. Specifically, it is used to calculate the first derivative of the respiratory flow signal to obtain the amplitude of the first derivative of the respiratory flow signal, or to calculate the second derivative of the respiratory flow signal to obtain the amplitude of the second derivative of the respiratory flow signal.
  • the first derivative of the respiratory flow signal can be calculated according to the following formula:
  • step is the derivative step size
  • N is the total length of the respiratory flow signal sequence and the thoracoabdominal band signal sequence
  • Flow (i) represents the respiratory flow signal at the i-th sampling point
  • Flow (i-step) represents the i- The respiratory flow signal at step sampling points
  • Flow ′ (i) represents the amplitude of the first derivative of the respiratory flow signal at the ith sampling point.
  • the second derivative of the respiratory flow signal can be calculated according to the following formula:
  • step is the derivative step size
  • N is the total length of the respiratory flow signal sequence and the thoracoabdominal band signal sequence
  • Flow (i) represents the respiratory flow signal at the i-th sampling point
  • Flow (i-step) represents the i- The respiratory flow signal at step sampling points
  • Flow (i-2step) represents the respiratory flow signal at the i-2step sampling point
  • Flow "(i) represents the magnitude of the second derivative of the respiratory flow signal at the i-th sampling point.
  • a first determining unit is configured to determine an apnea event period according to a magnitude of a derivative of the breathing flow signal and a first preset threshold Thres_flow.
  • the flow "(i) corresponding to each sampling point i is compared with the first preset threshold Thres_flow, and when
  • the duration of the first period corresponding to the i-th sampling point to the M-th sampling point is greater than 10s, the first period is marked as the first period of the apnea event.
  • a second acquisition unit for acquiring an original signal thoracoabdominal movements (formerly Res) said apnea event period.
  • the second pre-processing unit is configured to pre-process the original thoracoabdominal motion signal (Res original ) to obtain the thoracoabdominal motion signal (Res).
  • the second pre-processing unit includes:
  • the second truncation processing module for the original signal thoracoabdominal movements (formerly Res) according to the formula do truncation processing, to obtain processed signals thoracoabdominal movements (Res):
  • Res original (i) represents the original thoracoabdominal motion signal at the ith sampling point
  • Res at (i) represents the processed thoracoabdominal motion signal at the ith sampling point
  • B is the amplitude truncation threshold of the original Res
  • Second band-pass filter processing means for band-pass filtering process on the thoracoabdominal signal processing (at Res), to give the thoracoabdominal signal (Res).
  • the second calculation unit is configured to calculate a derivative of the thoracoabdominal motion signal to obtain a magnitude of the thoracoabdominal motion signal derivative. Specifically, the first derivative of the thoracoabdominal motion signal is calculated to obtain the amplitude of the first derivative of the thoracoabdominal motion signal, and the second derivative of the thoracoabdominal motion signal can also be calculated to obtain the amplitude of the second derivative of the thoracoabdominal motion signal.
  • the first derivative of the thoracoabdominal motion signal can be calculated according to the following formula:
  • step is the derivative step
  • N is the total length of the thoracoabdominal band signal sequence
  • Res (i) is the thoracoabdominal motion signal at the ith sampling point
  • Res (i-step) is the i-step sampling point
  • Res ′ (i) represents the first derivative amplitude of the thoracoabdominal motion signal at the ith sampling point.
  • the second derivative of the thoracoabdominal motion signal can be calculated according to the following formula:
  • step is the derivative step size
  • N is the total length of the respiratory flow signal sequence
  • Res (i) represents the thoracoabdominal motion signal at the i-th sampling point
  • Res (i-step) represents the i-step sampling point.
  • Res (i-2step) represents the thoracoabdominal motion signal at the i-2step sampling point
  • Res "(i) represents the second derivative amplitude of the thoracoabdominal motion signal at the ith sampling point.
  • a second determining unit is configured to determine the apnea event type according to the magnitude of the thoracic and abdominal motion signal derivative and a second preset threshold Thres_Res.
  • Res ′′ (i) corresponding to the first period of the apnea event, when
  • OSA Obstructive Sleep Apnea
  • FIG. 5 it is a schematic diagram of an internal structure of a computer device in an embodiment.
  • the computer device connects a processor, a nonvolatile storage medium, an internal memory, and a network interface through a system connection bus.
  • the non-volatile storage medium of the computer device can store an operating system and computer-readable instructions.
  • the processor can execute a method for determining an apnea event type.
  • the processor of the computer equipment is used to provide computing and control capabilities to support the operation of the entire computer equipment.
  • the internal memory may store computer-readable instructions.
  • the processor may cause the processor to execute a method for determining an apnea event type.
  • the network interface of the computer device is used for network communication.
  • the computer device may be a server, and the server may be implemented by an independent server or a server cluster composed of multiple servers.
  • the computer equipment can also be a terminal.
  • the display screen of the terminal can be a liquid crystal display or an electronic ink display.
  • the input device of the computer equipment can be the touch layer covered on the display screen, or the keys and trackballs provided on the computer equipment casing. Or a touchpad, or an external keyboard, touchpad, or mouse.
  • the touch layer and the display screen form a touch screen.
  • a non-transitory computer-readable storage medium including instructions such as a storage device including a computer program (instruction), the program (instruction) may be executed by a processor of the computer device to complete the present application.
  • the non-transitory computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
  • This application adopts the above technical solution, and the technical effect brought by the method is: by acquiring the original respiratory flow signal (the original Flow) and the original chest and abdominal motion signal (the original Res) of the patient within a preset time, the original respiratory flow signal (Flow The original) and the original thoracoabdominal motion signal (Res original) are preprocessed to eliminate the effects of invalid signals and interference signals.
  • the respiratory flow signal (Flow) and thoracoabdominal motion signal (Res) obtained after preprocessing are separately calculated.

Abstract

一种呼吸暂停事件类型的判断方法、装置和存储介质。呼吸暂停事件类型的判断方法包括：获取预设时间内的患者的原始呼吸流信号(Flow 原)和原始胸腹运动信号(Res 原)(S101)，原始呼吸流信号表征患者的呼吸气流，原始胸腹运动信号表征患者的胸腹运动，其中原始呼吸流信号的个数大于1，原始胸腹运动信号的个数大于1；对原始呼吸流信号和原始胸腹运动信号进行预处理，得到呼吸流信号(Flow)和胸腹运动信号(Res)(S102)；计算呼吸流信号的导数和胸腹运动信号的导数，得到呼吸流信号导数幅值和胸腹运动信号导数幅值(S103)；根据呼吸流信号导数幅值和胸腹运动信号导数幅值判断呼吸暂停事件类型(S104)。通过呼吸暂停事件类型的判断方法可以降低呼吸暂停事件类型判断过程的复杂程度，同时提高呼吸暂停事件类型判断结果的准确性。

Description

呼吸暂停事件类型的判断方法、装置和存储介质
本申请要求于2018年8月29日提交中国专利局，申请号为201810993543.6，申请名称为“呼吸暂停事件类型的判断方法、装置和存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及医疗器械领域，尤其涉及一种呼吸暂停事件类型的判断方法、装置和存储介质。
背景技术
睡眠呼吸暂停(Sleep Apnea，SA)是一种呼吸障碍，是指在连续7h睡眠中发生30次以上的呼吸暂停，每次气流中止10s以上(含10s)，或平均每小时低通气次数(呼吸紊乱指数)超过5次，而引起慢性低氧血症及高碳酸血症的临床综合征.可分为中枢型、阻塞型及混合型。
阻塞性睡眠呼吸暂停(Obstructive Sleep Apnea，OSA)是上气道塌陷而造成上呼吸道阻塞或者呼吸道收窄引致睡眠时呼吸暂停，表现为口鼻腔气流停止而胸腹呼吸动作尚存在。中枢神经性睡眠呼吸暂停(Central Sleep Apnea，CSA)是由于中枢神经系统的呼吸中枢神经功能障碍或支配呼吸肌的神经或呼吸肌病变，导致气道无阻塞但发生呼吸暂停，表现为口鼻腔气流和胸腹呼吸动作同时停止。混合性睡眠呼吸暂停(Mixed Sleep Apnea，MSA)：混合阻塞性睡眠呼吸暂停和中枢神经性睡眠呼吸暂停。
睡眠呼吸暂停对患者的生活和健康造成不同程度的影响，准确区分睡眠呼吸暂停的类型是选择合适治疗方法的前提。
现有的判断呼吸暂停事件类型的方法主要有如下几种，其中一种是通过监测呼吸流信号的统计特性来判断呼吸暂停事件的发生，具体方法之一是以病人正常呼吸时的流速峰值或者平均值的百分比作为门限值，再将每段呼吸流信号的幅值与此门限值做比较，低于此门限值的时段标记为呼吸暂停事件时段；具体方法之二是将短时段和长时段呼吸流信号(认为是正常呼吸时段)的方差进行比较，当短时段呼吸流信号的方差低于长时段呼吸流信号方差的一定百分比时，将此时段标定为呼吸暂停事件时段；确定了呼吸暂停事件时段之后，再通过监测呼吸流信号的频率特性来判断呼吸暂停事件的类型，具体方法是将呼吸流信号转换成频率信号， 通过监测频率信号幅值大小来区分不同的呼吸暂停事件类型。
另一种方法是通过监测胸腹带信号的幅度和频率特性来判断呼吸暂停事件，具体是将胸腹带信号幅度明显降低的时段标记为呼吸暂停事件时段，并通过不同时段胸腹带信号的震动频率快慢来区分不同的呼吸暂停事件类型。
现有判断呼吸暂停事件类型的方法存在以下问题，在采用呼吸流信号进行呼吸暂停事件判断的技术中，单一的呼吸流信号包含的信息不够充分，呼吸暂停事件判断的准确性不高；同时由于CSA和OSA类型事件的呼吸流信号特性不明显，很难根据呼吸流信号的频率特性来判断事件类型；而对于将短时段和长时段呼吸流信号的方差进行呼吸暂停事件判断的技术中，在每个时刻均需要存储很长时段的呼吸流信号方差信息，算法的复杂性很高，不易于实现。
在利用胸腹带信号进行呼吸暂停事件判断的技术中，虽然在呼吸暂停时段能检测到信号幅值的降低，但是很难区分是因为呼吸阻塞还是其它原因导致的胸腹运动减弱，而且在呼吸暂停阶段很难监测胸腹运动的频率，对监测仪器的要求较高，很难保证测量的精准性。
发明内容
根据本申请提供了一种呼吸暂停事件类型的判断方法、装置和存储介质。
一种呼吸暂停事件类型的判断方法，包括：
获取预设时间内的患者的原始呼吸流信号(Flow 原)和原始胸腹运动信号(Res 原)，所述原始呼吸流信号表征所述患者的呼吸气流，所述原始胸腹运动信号表征所述患者的胸腹运动，其中所述原始呼吸流信号的个数大于1，所述原始胸腹运动信号的个数大于1；
对所述原始呼吸流信号和所述原始胸腹运动信号进行预处理，得到呼吸流信号(Flow)和胸腹运动信号(Res)；
计算所述呼吸流信号的导数和所述胸腹运动信号的导数，得到呼吸流信号导数幅值和胸腹运动信号导数幅值；
根据所述呼吸流信号导数幅值和所述胸腹运动信号导数幅值判断呼吸暂停事件类型。
在一个实施例中，所述根据所述呼吸流信号导数幅值和所述胸腹运动信号导数幅值判断呼吸暂停事件类型包括：
根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时 段；
根据所述呼吸暂停事件时段对应的所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型。
在一个实施例中，所述对所述原始呼吸流信号和所述原始胸腹运动信号进行预处理，得到呼吸流信号和胸腹运动信号包括：
对所述原始呼吸流信号(Flow 原)和所述原始胸腹运动信号(Res 原)分别按照以下公式做截断处理，得到处理后的呼吸流信号(Flow 处)和处理后的胸腹运动信号(Res 处)：
Figure PCTCN2019100238-appb-000001
Figure PCTCN2019100238-appb-000002
其中Flow 原(i)表示第i个采样点的原始呼吸流信号，Flow 处(i)表示第i个采样点的处理后的呼吸流信号，Res 原(i)表示第i个采样点的原始胸腹运动信号，Res 处(i)表示第i个采样点的处理后的胸腹运动信号，A为Flow 原的幅值截断门限，且A>0，B为Res 原的幅值截断门限，且B>0；
对所述处理后的呼吸流信号(Flow 处)和所述处理后的胸腹运动信号(Res 处)进行带通滤波处理，得到所述呼吸流信号(Flow)和所述胸腹运动信号(Res)。在一个实施例中，所述计算所述呼吸流信号的导数和所述胸腹运动信号的导数，得到呼吸流信号导数幅值和胸腹运动信号导数幅值包括：
对所述呼吸流信号(Flow)和所述胸腹运动信号(Res)分别按照以下公式计算二阶导数，得到呼吸流信号二阶导数幅值和胸腹运动信号二阶导数幅值：
Figure PCTCN2019100238-appb-000003
2step＜i≤N
Figure PCTCN2019100238-appb-000004
2step＜i≤N
其中step表示求导的步长，N为呼吸流信号序列和胸腹带信号序列的总长度，Flow(i)表示第i个采样点的呼吸流信号，Flow(i-step)表示第i-step个采样点的呼吸流信号，Flow(i-2step)表示第i-2step个采样点的呼吸流信号，Flow″(i) 表示第i个采样点的呼吸流信号的二阶导数幅值，Res(i)表示第i个采样点的胸腹运动信号，Res(i-step)表示第i-step个采样点的胸腹运动信号，Res(i-2step)表示第i-2step个采样点的胸腹运动信号，Res″(i)表示第i个采样点的胸腹运动信号的二阶导数幅值。
在一个实施例中，当1≤i≤2step时，Flow″(i)取值为第一预设值，Res″(i)取值为第二预设值。
在一个实施例中，所述根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段，包括：
将每个采样点i对应的Flow″(i)与所述第一预设阈值Thres_flow进行比较，当在一采样点i存在|Flow″(i)|＜Thres_flow，则将该采样点i对应的时刻标记为可能发生呼吸暂停事件的起点，继续遍历后续采样点Flow″(j),j＝i+1,…N，当存在从第i采样点到第M采样点均有：
Figure PCTCN2019100238-appb-000005
且从第i采样点到第M采样点对应的第一时段的持续时间大于10s，则将所述第一时段标定为呼吸暂停事件第一时段。
在一个实施例中，所述根据所述呼吸暂停事件时段对应的所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型包括：
所述呼吸暂停事件第一时段对应的Res″(i)，当|Res″(i)|≥Thres_res,0＜i≤M时，所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为阻塞性睡眠呼吸暂停(Obstructive Sleep Apnea，OSA)；
当|Res″(i)|＜Thres_res,0＜i≤M时，所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为中枢神经性睡眠呼吸暂停(Central Sleep Apnea，CSA)；
当从第i采样点到第M 1采样点对应的第二时段满足|Res″(j)|＜Thres_res,i≤j≤M 1，从第M 1采样点到第M采样点对应的第三时段满足|Res″(j)|≥Thres_res,M 1＜j≤M，且所述第二时段对应的时长占所述第一时 段对应的时长的比例在预设比例范围内，则所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为混合性睡眠呼吸暂停(Mixed Sleep Apnea，MSA)。
一种呼吸暂停事件类型的判断方法，包括：
获取预设时间内的患者的原始呼吸流信号(Flow 原)，所述原始呼吸流信号表征所述患者的呼吸气流，其中所述原始呼吸流信号的个数大于1；
对所述原始呼吸流信号进行预处理，得到呼吸流信号(Flow)；
计算所述呼吸流信号的导数，得到呼吸流信号导数幅值；
根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段；
获取所述呼吸暂停事件时段的原始胸腹运动信号(Res 原)；
对所述原始胸腹运动信号(Res 原)进行预处理，得到胸腹运动信号(Res)；
计算所述胸腹运动信号的导数，得到胸腹运动信号导数幅值；
根据所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型。
在一个实施例中，所述对所述原始呼吸流信号进行预处理，得到呼吸流信号(Flow)包括：
对所述原始呼吸流信号(Flow 原)按照以下公式做截断处理，得到处理后的呼吸流信号(Flow 处)：
Figure PCTCN2019100238-appb-000006
其中Flow 原(i)表示第i个采样点的原始呼吸流信号，Flow 处(i)表示第i个采样点的处理后的呼吸流信号，A为Flow 原的幅值截断门限，且A>0；
对所述处理后的呼吸流信号(Flow 处)进行带通滤波处理，得到所述呼吸流信号(Flow)。
在一个实施例中，所述对所述原始胸腹运动信号(Res 原)进行预处理，得到胸腹运动信号(Res)包括：
对所述原始胸腹运动信号(Res 原)按照以下公式做截断处理，得到处理后的胸腹运动信号(Res 处)：
Figure PCTCN2019100238-appb-000007
其中Res 原(i)表示第i个采样点的原始胸腹运动信号，Res 处(i)表示第i个采样点的处理后的胸腹运动信号，B为Res 原的幅值截断门限，且B>0；
对所述处理后的胸腹运动信号(Res 处)进行带通滤波处理，得到所述胸腹运动信号(Res)。
一种呼吸暂停事件类型的判断装置，包括：
获取单元，用于获取预设时间内的患者的原始呼吸流信号(Flow 原)和原始胸腹运动信号(Res 原)，所述原始呼吸流信号表征所述患者的呼吸气流，所述原始胸腹运动信号表征所述患者的胸腹运动，其中所述原始呼吸流信号的个数大于1，所述原始胸腹运动信号的个数大于1；
预处理单元，用于对所述原始呼吸流信号和所述原始胸腹运动信号进行预处理，得到呼吸流信号(Flow)和胸腹运动信号(Res)；
计算单元，用于计算所述呼吸流信号的导数和所述胸腹运动信号的导数，得到呼吸流信号导数幅值和胸腹运动信号导数幅值；
判断单元，用于根据所述呼吸流信号导数幅值和所述胸腹运动信号导数幅值判断呼吸暂停事件类型。
在一个实施例中，判断单元包括：
呼吸暂停事件时段确定模块，用于根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段。
呼吸暂停事件类型确定模块，用于根据所述呼吸暂停事件时段对应的所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型。
在一个实施例中，预处理单元包括：
截断处理模块，用于对所述原始呼吸流信号(Flow 原)和所述原始胸腹运动信号(Res 原)分别按照以下公式做截断处理，得到处理后的呼吸流信号(Flow 处)和处理后的胸腹运动信号(Res 处)：
Figure PCTCN2019100238-appb-000008
Figure PCTCN2019100238-appb-000009
其中Flow 原(i)表示第i个采样点的原始呼吸流信号，Flow 处(i)表示第i个采样点的处理后的呼吸流信号，Res 原(i)表示第i个采样点的原始胸腹运动信号，Res 处(i)表示第i个采样点的处理后的胸腹运动信号，A为Flow 原的幅值截断门限，且A>0，B为Res 原的幅值截断门限，且B>0。
带通滤波处理模块，用于对所述处理后的呼吸流信号(Flow 处)和所述处理后的胸腹运动信号(Res 处)进行带通滤波处理，得到所述呼吸流信号(Flow)和所述胸腹运动信号(Res)。
在一个实施例中，计算单元具体用于对所述呼吸流信号(Flow)和所述胸腹运动信号(Res)分别按照以下公式计算二阶导数，得到呼吸流信号二阶导数幅值和胸腹运动信号二阶导数幅值：
Figure PCTCN2019100238-appb-000010
2step＜i≤N
Figure PCTCN2019100238-appb-000011
2step＜i≤N
其中step表示求导的步长，N为呼吸流信号序列和胸腹带信号序列的总长度，Flow(i)表示第i个采样点的呼吸流信号，Flow(i-step)表示第i-step个采样点的呼吸流信号，Flow(i-2step)表示第i-2step个采样点的呼吸流信号，Flow″(i)表示第i个采样点的呼吸流信号的二阶导数幅值，Res(i)表示第i个采样点的胸腹运动信号，Res(i-step)表示第i-step个采样点的胸腹运动信号，Res(i-2step)表示第i-2step个采样点的胸腹运动信号，Res″(i)表示第i个采样点的胸腹运动信号的二阶导数幅值。
一种呼吸暂停事件类型的判断装置，包括：
第一获取单元，用于获取预设时间内的患者的原始呼吸流信号(Flow 原)，所述原始呼吸流信号表征所述患者的呼吸气流，其中所述原始呼吸流信号的个数大于1；
第一预处理单元，用于对所述原始呼吸流信号进行预处理，得到呼吸流信号(Flow)；
第一计算单元，用于计算所述呼吸流信号的导数，得到呼吸流信号导数幅值；
第一确定单元，用于根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段；
第二获取单元，用于获取所述呼吸暂停事件时段的原始胸腹运动信号(Res 原)；
第二预处理单元，用于对所述原始胸腹运动信号(Res 原)进行预处理，得到胸腹运动信号(Res)；
第二计算单元，用于计算所述胸腹运动信号的导数，得到胸腹运动信号导数幅值；
第二确定单元，用于根据所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型。
在一个实施例中，第一预处理单元包括：
第一截断处理模块，用于对所述原始呼吸流信号(Flow 原)按照以下公式做截断处理，得到处理后的呼吸流信号(Flow 处)：
Figure PCTCN2019100238-appb-000012
其中Flow 原(i)表示第i个采样点的原始呼吸流信号，Flow 处(i)表示第i个采样点的处理后的呼吸流信号，A为Flow 原的幅值截断门限，且A>0。
第一带通滤波处理模块，用于对所述处理后的呼吸流信号(Flow 处)进行带通滤波处理，得到所述呼吸流信号(Flow)。
在一个实施例中，第二预处理单元包括：
第二截断处理模块，用于对所述原始胸腹运动信号(Res 原)按照以下公式做截断处理，得到处理后的胸腹运动信号(Res 处)：
Figure PCTCN2019100238-appb-000013
其中Res 原(i)表示第i个采样点的原始胸腹运动信号，Res 处(i)表示第i个采样点的处理后的胸腹运动信号，B为Res 原的幅值截断门限，且B>0。
第二带通滤波处理模块，用于对所述处理后的胸腹运动信号(Res 处)进行带通滤波处理，得到所述胸腹运动信号(Res)。
一种计算机可读存储介质，所述计算机可读存储介质上存储有计算机可执行指令，所述计算机可执行指令被处理器执行时，使得所述处理器执行上述任意一项所述方法的步骤。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征、目的和优点将从说明书、附图以及权利要求书变得明显。
本申请的实施例通过获取预设时间内的患者的原始呼吸流信号(Flow 原)和原始胸腹运动信号(Res 原)，对该原始呼吸流信号(Flow 原)和原始胸腹运动信号(Res 原)进行预处理，排除无效信号和干扰信号的影响，对预处理后得到的呼吸流信号(Flow)和胸腹运动信号(Res)分别进行求导运算，得到呼吸流信号导数幅值和胸腹运动信号导数幅值，根据该呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段，再根据该呼吸暂停事件时段对应的胸腹运动信号地变化趋势的不同来区分CSA/OSA/MSA这三种呼吸暂停事件，该方案复杂度低，判断结果准确性高。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为一个实施例的呼吸暂停事件类型的判断方法的流程图；
图2为另一个实施例的呼吸暂停事件类型的判断方法的流程图；
图3为一个实施例的呼吸暂停事件类型的判断装置的结构方框图；
图4为另一个实施例的呼吸暂停事件类型的判断装置的结构方框图；
图5为一个实施例的计算机设备的内部结构示意图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
患者的原始呼吸流信号(Flow 原)和原始胸腹运动信号(Res 原)可以由鼻气流和胸腹运动盒子进行监测和采集，也可以由其他的睡眠监测仪进行监测和采集。鼻气流和胸腹运动盒子或其他睡眠监测仪将采集的原始呼吸流信号和原始胸腹运 动信号实时或者按照预设时间间隔发送给呼吸暂停事件类型判断装置。该呼吸暂停事件类型判断装置执行本申请所述的呼吸暂停事件类型判断方法包括的流程步骤。
如图1所示，在一实施例中一种呼吸暂停事件类型的判断方法，包括以下步骤：步骤S101，获取预设时间内的患者的原始呼吸流信号(Flow 原)和原始胸腹运动信号(Res 原)，所述原始呼吸流信号表征所述患者的呼吸气流，所述原始胸腹运动信号表征所述患者的胸腹运动，其中所述原始呼吸流信号的个数大于1，所述原始胸腹运动信号的个数大于1。
具体地，获取预设时间内的患者的原始呼吸流信号(Flow 原)和原始胸腹运动信号(Res 原)，该预设时间可以是患者夜间佩戴睡眠监测仪的时间，比如7个小时或者8个小时。
步骤S102，对所述原始呼吸流信号和所述原始胸腹运动信号进行预处理，得到呼吸流信号(Flow)和胸腹运动信号(Res)。
具体地，由于睡眠监测仪是在患者睡着时采集信息，而在此时段内，病人可能会发生翻身、肢体突然大幅度运动等情况，因此通常会产生赋值异常的尖峰信号，影响判断精度，因此需要去除这些尖峰信号，比如可以采用幅值截断技术进行截断处理；除了尖峰信号外，还会有各种干扰信号，比如噪音等，为了防止干扰信号影响事件判断结果，需要对截断处理后的信号数据进行滤波处理。
步骤S103，计算所述呼吸流信号的导数和所述胸腹运动信号的导数，得到呼吸流信号导数幅值和胸腹运动信号导数幅值。
具体地，计算呼吸流信号的一阶导数和胸腹运动信号的一阶导数，也可以计算呼吸流信号的二阶导数和胸腹运动信号的二阶导数，得到呼吸流信号导数幅值和胸腹运动信号导数幅值。
步骤S104，根据所述呼吸流信号导数幅值和所述胸腹运动信号导数幅值判断呼吸暂停事件类型。
具体地，步骤S104，进一步包括：
步骤S1041，根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段，其中，第一预设阈值Thres_flow是根据患者过去一段稳定的正常呼吸对应的呼吸流信号的导数幅值，或者过去几段稳定的正常呼吸对应的呼吸流信号的导数幅值的均值来设定，且该第一预设阈值Thres_flow可以根据患 者的正常呼吸的变化而动态改变；
步骤S1042，根据所述呼吸暂停事件时段对应的所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型，其中该第二预设阈值Thres_Res是根据患者过去一段稳定的正常呼吸对应的胸腹运动信号的导数幅值，或者过去几段稳定的正常呼吸对应的胸腹运动信号的导数幅值的均值来设定，且该第二预设阈值Thres_Res可以根据患者的正常呼吸的变化而动态改变。
本实施例通过获取预设时间内的患者的原始呼吸流信号(Flow 原)和原始胸腹运动信号(Res 原)，对该原始呼吸流信号(Flow 原)和原始胸腹运动信号(Res 原)进行预处理，排除无效信号和干扰信号的影响，对预处理后得到的呼吸流信号(Flow)和胸腹运动信号(Res)分别进行求导运算，得到呼吸流信号导数幅值和胸腹运动信号导数幅值，根据该呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段，再根据该呼吸暂停事件时段对应的胸腹运动信号地变化趋势的不同来区分CSA/OSA/MSA这三种呼吸暂停事件，该方案复杂度低，判断结果准确性高。
在一个实施例中，步骤S102，对所述原始呼吸流信号和所述原始胸腹运动信号进行预处理，得到呼吸流信号(Flow)和胸腹运动信号(Res)包括：
步骤S1021，对所述原始呼吸流信号(Flow 原)和所述原始胸腹运动信号(Res 原)分别按照以下公式做截断处理，得到处理后的呼吸流信号(Flow 处)和处理后的胸腹运动信号(Res 处)：
Figure PCTCN2019100238-appb-000014
Figure PCTCN2019100238-appb-000015
其中Flow 原(i)表示第i个采样点的原始呼吸流信号，Flow 处(i)表示第i个采样点的处理后的呼吸流信号，Res 原(i)表示第i个采样点的原始胸腹运动信号，Res 处(i)表示第i个采样点的处理后的胸腹运动信号，A为Flow 原的幅值截断门限，且A>0，B为Res 原的幅值截断门限，且B>0。
具体地，由于睡眠监测仪是在患者睡着时采集信息，而在此时段内，病人可能会 发生翻身、肢体突然大幅度运动等情况，因此通常会产生赋值异常的尖峰信号，影响判断精度，因此需要去除这些尖峰信号，比如可以采用幅值截断技术进行截断处理，以降低无效信号的影响，使得处理后的呼吸流信号和处理后的胸腹运动信号快速收敛到合理范围。其中，A可以根据患者过去一段稳定的正常呼吸对应的原始呼吸流信号的峰值，或者过去几段稳定的正常呼吸对应的原始呼吸流信号的峰值的均值来设定；B可以根据患者过去一段稳定的正常呼吸对应的原始胸腹运动信号的峰值，或者过去几段稳定的正常呼吸对应的原始胸腹运动信号的峰值的均值来设定；A和B可以根据患者在不同正常呼吸时段的原始呼吸流信号的峰值和原始胸腹运动信号的峰值做自适应调整，例如：对于正常呼吸情况下呼吸幅度较大的患者，可在各个时段都设置较大的Flow 原的幅值截断门限和较大的Res 原的幅值截断门限，而对于正常呼吸情况下呼吸幅度较弱的患者，可适当降低Flow 原的幅值截断门限和Res 原的幅值截断门限。
步骤S1022，对所述处理后的呼吸流信号(Flow 处)和所述处理后的胸腹运动信号(Res 处)进行带通滤波处理，得到所述呼吸流信号(Flow)和所述胸腹运动信号(Res)。
具体地，由于原始呼吸流信号和原始胸腹运动信号中还参杂着各种干扰信号，比如噪声等，为了防止干扰信号影响事件判断结果，需要对处理后的呼吸流信号(Flow 处)和处理后的胸腹运动信号(Res 处)分别采用带通滤波器进行滤波处理，以滤除通频带以外的高频信号和低频信号，获得带限频带内的有用的呼吸流信号(Flow)和胸腹运动信号(Res)信号。对处理后的呼吸流信号(Flow 处)进行滤波处理的通频带的上限和下限可以进行设定，对处理后的胸腹运动信号(Res 处)进行滤波处理的通频带的上限和下限也可以进行设定，比如，对处理后的呼吸流信号(Flow 处)进行滤波处理的通频带的上限可以设置为3Hz，通频带的下限可以设置为0.05Hz。
本实施例中，通过对原始呼吸流信号和所述原始胸腹运动信号进行预处理，可以排除异常信号和干扰信号的影响，提高呼吸暂停事件类型判断的准确性。
在一个实施例中，步骤S103，计算所述呼吸流信号的导数和所述胸腹运动信号的导数，得到呼吸流信号导数幅值和胸腹运动信号导数幅值包括：
对所述呼吸流信号(Flow)和所述胸腹运动信号(Res)分别按照以下公式计算 二阶导数，得到呼吸流信号二阶导数幅值和胸腹运动信号二阶导数幅值：
Figure PCTCN2019100238-appb-000016
2step＜i≤N
Figure PCTCN2019100238-appb-000017
2step＜i≤N
其中step表示求导的步长，N为呼吸流信号序列和胸腹带信号序列的总长度，Flow(i)表示第i个采样点的呼吸流信号，Flow(i-step)表示第i-step个采样点的呼吸流信号，Flow(i-2step)表示第i-2step个采样点的呼吸流信号，Flow″(i)表示第i个采样点的呼吸流信号的二阶导数幅值，Res(i)表示第i个采样点的胸腹运动信号，Res(i-step)表示第i-step个采样点的胸腹运动信号，Res(i-2step)表示第i-2step个采样点的胸腹运动信号，Res″(i)表示第i个采样点的胸腹运动信号的二阶导数幅值。
具体地，在二阶导数的计算中，step值的取值直接影响计算结果，如果step太小，则二阶导数的值对信号的变化很敏感，获得的二阶导数曲线Flow″(i),i＝1,…N；Res″(i),i＝1,…N波动很大，不易于进行事件类型的判断；同时如果step值太大，则在每一次求解过程中只用到少量的呼吸流信号数据和胸腹运动信号数据，丢弃了大量游泳数据，结果不能精确反应呼吸流信号数据的趋势和胸腹运动信号数据的趋势。因此需要动态调整step的大小，根据过去一段时间内呼吸流信号的平均变化趋势及胸腹运动信号的平均变化趋势来确定，使得求解出的二阶导数可准确用于进行呼吸暂停事件的判断。
可选地，当1≤i≤2step时，Flow″(i)取值为第一预设值，Res″(i)取值为第二预设值。
具体地，当刚开始入睡的时候发生呼吸暂停的概率比较小，因此当1≤i≤2step时，可以设定Flow″(i)取值为第一预设值，Res″(i)取值为第二预设值，以方便后面的计算，第一预设值可以是大于第一预设阈值Thres_flow的一个值，第二预设值 可以是大于第二预设阈值Thres_Res的一个值。
可选地，也可以对所述呼吸流信号(Flow)和所述胸腹运动信号(Res)分别计算一阶导数，得到呼吸流信号一阶导数幅值和胸腹运动信号一阶导数幅值，具体可以按照如下公式进行计算：
Figure PCTCN2019100238-appb-000018
step＜i≤N
Figure PCTCN2019100238-appb-000019
step＜i≤N
其中step表示求导的步长，N为呼吸流信号序列和胸腹带信号序列的总长度，Flow(i)表示第i个采样点的呼吸流信号，Flow(i-step)表示第i-step个采样点的呼吸流信号，Flow′(i)表示第i个采样点的呼吸流信号的一阶导数幅值，Res(i)表示第i个采样点的胸腹运动信号，Res(i-step)表示第i-step个采样点的胸腹运动信号，Res′(i)表示第i个采样点的胸腹运动信号的一阶导数幅值。
在一个实施例中，步骤S1041，根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段具体为：
将每个采样点i对应的Flow″(i)与所述第一预设阈值Thres_flow进行比较，当在一采样点i存在|Flow″(i)|＜Thres_flow，则将该采样点i对应的时刻标记为可能发生呼吸暂停事件的起点，继续遍历后续采样点Flow″(j),j＝i+1,…N，当存在从第i采样点到第M采样点均有：
Figure PCTCN2019100238-appb-000020
且从第i采样点到第M采样点对应的第一时段的持续时间大于10s，则将所述第一时段标定为呼吸暂停事件第一时段。
具体地，第一预设阈值Thres_flow是根据患者过去一段稳定的正常呼吸对应的呼吸流信号的二阶导数幅值，或者过去几段稳定的正常呼吸对应的呼吸流信号的二阶导数幅值的均值来设定，且该第一预设阈值Thres_flow可以根据患者的正 常呼吸的变化而动态改变。当在一采样点i存在|Flow″(i)|＜Thres_flow，则将该采样点i对应的时刻标记为可能发生呼吸暂停事件的起点，继续遍历后续采样点Flow″(j),j＝i+1,…N，当存在从第i采样点到第M采样点均有：
Figure PCTCN2019100238-appb-000021
且从第i采样点到第M采样点对应的第一时段的持续时间大于10s，则将该第一时段标定为呼吸暂停事件第一时段。
可选地，当标定的两段呼吸暂停事件时段中间出现很短时间的A值范围内的尖峰原始呼吸流信号时，可设定时间门限Time_Threshold，只要两段呼吸暂停事件时段之间的尖峰原始呼吸流信号持续时间小于Time_Threshold，则可认为患者只是在一段呼吸暂停事件时段中出现突然的喘气或其他动作，为了使得判断结果更加接近患者的实际情况，本实施例将该两段呼吸暂停事件时段合并为一段呼吸暂停事件时段。
步骤S1042，根据所述呼吸暂停事件时段对应的所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型具体为：
所述呼吸暂停事件第一时段对应的Res″(i)，当|Res″(i)|≥Thres_res,0＜i≤M时，所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为阻塞性睡眠呼吸暂停(Obstructive Sleep Apnea，OSA)；
当|Res″(i)|＜Thres_res,0＜i≤M时，所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为中枢神经性睡眠呼吸暂停(Central Sleep Apnea，CSA)；
当从第i采样点到第M 1采样点对应的第二时段满足|Res″(j)|＜Thres_res,i≤j≤M 1，从第M 1采样点到第M采样点对应的第三时段满足|Res″(j)|≥Thres_res,M 1＜j≤M，且所述第二时段对应的时长占所述第一时段对应的时长的比例在预设比例范围内，则所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为混合性睡眠呼吸暂停(Mixed Sleep Apnea，MSA)。
具体地，第二预设阈值Thres_Res是根据患者过去一段稳定的正常呼吸对应的胸腹运动信号的二阶导数幅值，或者过去几段稳定的正常呼吸对应的胸腹运动信号 的二阶导数幅值的均值来设定，且该第二预设阈值Thres_Res可以根据患者的正常呼吸的变化而动态改变。由于OSA事件是由呼吸道阻塞或者变窄导致的，因此主要表现在呼吸流信号的大幅度减弱，胸腹运动信号幅度减弱但是依然存在一定的波动，使得OSA事件时段的胸腹运动信号的二阶导数幅值在第二预设阈值Thres_res以上；而在CSA事件中，由于呼吸中枢神经障碍使得胸腹部运动趋势变化很小，近似停止运动，因此CSA事件时段的二阶导数幅值在Thres_res以下，这样便可以通过监测呼吸暂停事件时段对应的胸腹运动信号的二阶导数幅值来区别呼吸暂停事件类型。
该呼吸暂停事件第一时段对应的Res″(i)，当|Res″(i)|≥Thres_res,0＜i≤M时，该呼吸暂停事件第一时段为阻塞性睡眠呼吸暂停(Obstructive Sleep Apnea，OSA)；当|Res″(i)|＜Thres_res,0＜i≤M时，该呼吸暂停事件第一时段中枢神经性睡眠呼吸暂停(Central Sleep Apnea，CSA)；当从第i采样点到第M 1采样点对应的第二时段满足|Res″(j)|＜Thres_res,i≤j≤M 1，从第M 1采样点到第M采样点对应的第三时段满足|Res″(j)|≥Thres_res,M 1＜j≤M，且该第二时段对应的时长占该第一时段对应的时长的比例在预设比例范围内，则该呼吸暂停事件第一时段为混合性睡眠呼吸暂停(Mixed Sleep Apnea，MSA)。该预设比例范围可以设置为20-70％，也可以设置为其他比例范围。
可选地，步骤S1041中的呼吸流信号导数幅值可以是呼吸流信号一阶导数幅值，第一预设阈值Thres_flow可以是根据患者过去一段稳定的正常呼吸对应的呼吸流信号的一阶导数幅值，或者过去几段稳定的正常呼吸对应的呼吸流信号的一阶导数幅值的均值来设定，且该第一预设阈值Thres_flow可以根据患者的正常呼吸的变化而动态改变。具体地根据改呼吸流信号一阶导数幅值和该第一预设阈值Thres_flow确定呼吸暂停事件时段的具体方法可以参照上述描述，在此不再赘述。
步骤S1042中的胸腹运动信号导数幅值可以是胸腹运动信号一阶导数幅值，第二预设阈值Thres_Res是根据患者过去一段稳定的正常呼吸对应的胸腹运动信号的一阶导数幅值，或者过去几段稳定的正常呼吸对应的胸腹运动信号的一阶导数 幅值的均值来设定，且该第二预设阈值Thres_Res可以根据患者的正常呼吸的变化而动态改变。具体地根据该呼吸暂停事件时段对应的该胸腹运动信号一阶导数幅值和该第二预设阈值Thres_Res确定呼吸暂停事件类型的具体方法可以参照上述描述，在此不再赘述。
上述实施例中，呼吸流信号导数幅值(呼吸流信号一阶导数幅值或呼吸流信号二阶导数幅值)和胸腹运动信号导数幅值(胸腹运动信号一阶导数幅值或胸腹运动信号二阶导数幅值)体现了不同呼吸暂停事件对应的呼吸流信号和胸腹运动信号的变化趋势，通过呼吸流信号导数幅值和胸腹运动信号导数幅值判断呼吸暂停事件类型准确性高；并且在计算呼吸流信号导数幅值和胸腹运动信号导数幅值过程中加入了步长step这一变量，因此导数的求解具有一定的记忆功能，能更好地反应一段时间内的整体呼吸状况，同时将异常信号对呼吸暂停事件判断的影响降低。如图2所示，在一个实施例中，一种呼吸暂停事件类型的判断方法，包括以下步骤：
步骤S201，获取预设时间内的患者的原始呼吸流信号(Flow 原)，所述原始呼吸流信号表征所述患者的呼吸气流，其中所述原始呼吸流信号的个数大于1。
具体地，获取预设时间内的患者的原始呼吸流信号(Flow 原)，该预设时间可以是患者夜间佩戴睡眠监测仪的时间，比如7个小时或者8个小时。
步骤S202，对所述原始呼吸流信号进行预处理，得到呼吸流信号(Flow)。
具体地，步骤S202包括：
步骤S2021，对所述原始呼吸流信号(Flow 原)按照以下公式做截断处理，得到处理后的呼吸流信号(Flow 处)：
Figure PCTCN2019100238-appb-000022
其中Flow 原(i)表示第i个采样点的原始呼吸流信号，Flow 处(i)表示第i个采样点的处理后的呼吸流信号，A为Flow 原的幅值截断门限，且A>0。
具体地，由于睡眠监测仪是在患者睡着时采集信息，而在此时段内，病人可能会发生翻身、肢体突然大幅度运动等情况，因此通常会产生赋值异常的尖峰信号，影响判断精度，因此需要去除这些尖峰信号，比如可以采用幅值截断技术进行截断处理，以降低无效信号的影响，使得处理后的呼吸流信号快速收敛到合理范围。 其中，A可以根据患者过去一段稳定的正常呼吸对应的原始呼吸流信号的峰值，或者过去几段稳定的正常呼吸对应的原始呼吸流信号的峰值的均值来设定；A可以根据患者在不同正常呼吸时段的原始呼吸流信号的峰值做自适应调整，例如：对于正常呼吸情况下呼吸幅度较大的患者，可在各个时段都设置较大的Flow 原的幅值截断门限，而对于正常呼吸情况下呼吸幅度较弱的患者，可适当降低Flow 原的幅值截断门限。
步骤S2022，对所述处理后的呼吸流信号(Flow 处)进行带通滤波处理，得到所述呼吸流信号(Flow)。
具体地，由于原始呼吸流信号中还参杂着各种干扰信号，比如噪声等，为了防止干扰信号影响事件判断结果，需要对处理后的呼吸流信号(Flow 处)采用带通滤波器进行滤波处理，以滤除通频带以外的高频信号和低频信号，获得带限频带内的有用的呼吸流信号(Flow)。可选的，通频带的上限可以设置为3Hz，通频带的下限可以设置为0.05Hz。
步骤S203，计算所述呼吸流信号的导数，得到呼吸流信号导数幅值。
具体地，计算呼吸流信号的一阶导数，得到呼吸流信号一阶导数幅值，也可以计算呼吸流信号的二阶导数，得到呼吸流信号二阶导数幅值。
具体可以按照如下公式计算呼吸流信号的一阶导数：
Figure PCTCN2019100238-appb-000023
step＜i≤N
其中step表示求导的步长，N为呼吸流信号序列和胸腹带信号序列的总长度，Flow(i)表示第i个采样点的呼吸流信号，Flow(i-step)表示第i-step个采样点的呼吸流信号，Flow′(i)表示第i个采样点的呼吸流信号的一阶导数幅值。
具体可以按照如下公式计算呼吸流信号的二阶导数：
Figure PCTCN2019100238-appb-000024
2step＜i≤N
其中step表示求导的步长，N为呼吸流信号序列和胸腹带信号序列的总长度，Flow(i)表示第i个采样点的呼吸流信号，Flow(i-step)表示第i-step个采样点的呼吸流信号，Flow(i-2step)表示第i-2step个采样点的呼吸流信号，Flow″(i) 表示第i个采样点的呼吸流信号的二阶导数幅值。
可选地，当1≤i≤2step时，Flow″(i)取值为第一预设值。
步骤S204，根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段。
具体地，将每个采样点i对应的Flow″(i)与所述第一预设阈值Thres_flow进行比较，当在一采样点i存在|Flow″(i)|＜Thres_flow，则将该采样点i对应的时刻标记为可能发生呼吸暂停事件的起点，继续遍历后续采样点Flow″(j),j＝i+1,…N，当存在从第i采样点到第M采样点均有：
Figure PCTCN2019100238-appb-000025
且从第i采样点到第M采样点对应的第一时段的持续时间大于10s，则将所述第一时段标定为呼吸暂停事件第一时段。
第一预设阈值Thres_flow是根据患者过去一段稳定的正常呼吸对应的呼吸流信号的二阶导数幅值，或者过去几段稳定的正常呼吸对应的呼吸流信号的二阶导数幅值的均值来设定，且该第一预设阈值Thres_flow可以根据患者的正常呼吸的变化而动态改变。
可选地，当标定的两段呼吸暂停事件时段中间出现很短时间的A值范围内的尖峰原始呼吸流信号时，可设定时间门限Time_Threshold，只要两段呼吸暂停事件时段之间的尖峰原始呼吸流信号持续时间小于Time_Threshold，则可认为患者只是在一段呼吸暂停事件时段中出现突然的喘气或其他动作，为了使得判断结果更加接近患者的实际情况，本实施例将该两段呼吸暂停事件时段合并为一段呼吸暂停事件时段。
可选地，以上步骤中的呼吸流信号导数幅值可以是呼吸流信号一阶导数幅值，第一预设阈值Thres_flow可以是根据患者过去一段稳定的正常呼吸对应的呼吸流信号的一阶导数幅值，或者过去几段稳定的正常呼吸对应的呼吸流信号的一阶导数幅值的均值来设定，且该第一预设阈值Thres_flow可以根据患者的正常呼吸的变化而动态改变。具体地根据改呼吸流信号一阶导数幅值和该第一预设阈值Thres_flow确定呼吸暂停事件时段的具体方法可以参照上述描述，在此不再赘 述。
步骤S205，获取所述呼吸暂停事件时段的原始胸腹运动信号(Res 原)。
具体地，获取该呼吸暂停事件第一时段对应的原始胸腹运动信号(Res 原)。
步骤S206，对所述原始胸腹运动信号(Res 原)进行预处理，得到胸腹运动信号(Res)。
具体地，步骤S206包括：
步骤S2061，对所述原始胸腹运动信号(Res 原)按照以下公式做截断处理，得到处理后的胸腹运动信号(Res 处)：
Figure PCTCN2019100238-appb-000026
其中Res 原(i)表示第i个采样点的原始胸腹运动信号，Res 处(i)表示第i个采样点的处理后的胸腹运动信号，B为Res 原的幅值截断门限，且B>0。
具体地，由于睡眠监测仪是在患者睡着时采集信息，而在此时段内，病人可能会发生翻身、肢体突然大幅度运动等情况，因此通常会产生赋值异常的尖峰信号，影响判断精度，因此需要去除这些尖峰信号，比如可以采用幅值截断技术进行截断处理，以降低无效信号的影响，使得处理后的胸腹运动信号快速收敛到合理范围。其中，B可以根据患者过去一段稳定的正常呼吸对应的原始胸腹运动信号的峰值，或者过去几段稳定的正常呼吸对应的原始胸腹运动信号的峰值的均值来设定；B可以根据患者在不同正常呼吸时段的原始胸腹运动信号的峰值做自适应调整，例如：对于正常呼吸情况下呼吸幅度较大的患者，可在各个时段都设置较大的Res 原的幅值截断门限，而对于正常呼吸情况下呼吸幅度较弱的患者，可适当降低Res 原的幅值截断门限。
步骤S2062，对所述处理后的胸腹运动信号(Res 处)进行带通滤波处理，得到所述胸腹运动信号(Res)。
具体地，由于原始胸腹运动信号中还参杂着各种干扰信号，比如噪声等，为了防止干扰信号影响事件判断结果，需要对处理后的胸腹运动信号(Res 处)采用带通滤波器进行滤波处理，以滤除通频带以外的高频信号和低频信号，获得带限频带内的有用的胸腹运动信号(Res)。
步骤S207，计算所述胸腹运动信号的导数，得到胸腹运动信号导数幅值。
具体地，计算胸腹运动信号的一阶导数，得到胸腹运动信号一阶导数幅值，也可以计算胸腹运动信号的二阶导数，得到胸腹运动信号二阶导数幅值。
具体可以按照如下公式计算胸腹运动信号的一阶导数：
Figure PCTCN2019100238-appb-000027
step＜i≤N
其中step表示求导的步长，N为胸腹带信号序列的总长度，Res(i)表示第i个采样点的胸腹运动信号，Res(i-step)表示第i-step个采样点的胸腹运动信号，Res′(i)表示第i个采样点的胸腹运动信号的一阶导数幅值。
具体可以按照如下公式计算胸腹运动信号的二阶导数：
Figure PCTCN2019100238-appb-000028
2step＜i≤N
其中step表示求导的步长，N为呼吸流信号序列的总长度，Res(i)表示第i个采样点的胸腹运动信号，Res(i-step)表示第i-step个采样点的胸腹运动信号，Res(i-2step)表示第i-2step个采样点的胸腹运动信号，Res″(i)表示第i个采样点的胸腹运动信号的二阶导数幅值。
可选地，当1≤i≤2step时，Res″(i)取值为第二预设值。
步骤S208，根据所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型。
具体地，所述呼吸暂停事件第一时段对应的Res″(i)，当|Res″(i)|≥Thres_res,0＜i≤M时，所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为阻塞性睡眠呼吸暂停(Obstructive Sleep Apnea，OSA)；
当|Res″(i)|＜Thres_res,0＜i≤M时，所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为中枢神经性睡眠呼吸暂停(Central Sleep Apnea，CSA)；
当从第i采样点到第M 1采样点对应的第二时段满足|Res″(j)|＜Thres_res,i≤j≤M 1，从第M 1采样点到第M采样点对应的第三时段满足|Res″(j)|≥Thres_res,M 1＜j≤M，且所述第二时段对应的时长占所述第一时 段对应的时长的比例在预设比例范围内，则所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为混合性睡眠呼吸暂停(Mixed Sleep Apnea，MSA)。
第二预设阈值Thres_Res是根据患者过去一段稳定的正常呼吸对应的胸腹运动信号的二阶导数幅值，或者过去几段稳定的正常呼吸对应的胸腹运动信号的二阶导数幅值的均值来设定，且该第二预设阈值Thres_Res可以根据患者的正常呼吸的变化而动态改变。
上述实施例中，首先对呼吸流信号数据进行求导运算得到呼吸流信号导数幅值，根据呼吸流信号导数幅值确定呼吸暂停事件时段，再对该呼吸暂停事件时段对应的胸腹运动信号数据进行求导运算得到胸腹运动信号导数幅值，再根据该胸腹运动信号导数幅值判断该呼吸暂停事件时段的呼吸暂停事件类型，该方案复杂度低，判断结果准确性高，并且在一定程度上节约了运算资源。
应该理解的是，本申请各实施例中的各个步骤并不是必然按照步骤标号指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，各实施例中至少一部分步骤可以包括多个子步骤或者多个阶段，这些子步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些子步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤的子步骤或者阶段的至少一部分轮流或者交替地执行。
如图3所示，在一个实施例中，一种呼吸暂停事件类型的判断装置，包括：
获取单元，用于获取预设时间内的患者的原始呼吸流信号(Flow 原)和原始胸腹运动信号(Res 原)，所述原始呼吸流信号表征所述患者的呼吸气流，所述原始胸腹运动信号表征所述患者的胸腹运动，其中所述原始呼吸流信号的个数大于1，所述原始胸腹运动信号的个数大于1；
预处理单元，用于对所述原始呼吸流信号和所述原始胸腹运动信号进行预处理，得到呼吸流信号(Flow)和胸腹运动信号(Res)；
计算单元，用于计算所述呼吸流信号的导数和所述胸腹运动信号的导数，得到呼吸流信号导数幅值和胸腹运动信号导数幅值；
判断单元，用于根据所述呼吸流信号导数幅值和所述胸腹运动信号导数幅值判断呼吸暂停事件类型。
在一个实施例中，判断单元包括：
呼吸暂停事件时段确定模块，用于根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段。
呼吸暂停事件类型确定模块，用于根据所述呼吸暂停事件时段对应的所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型。
在一个实施例中，预处理单元包括：
截断处理模块，用于对所述原始呼吸流信号(Flow 原)和所述原始胸腹运动信号(Res 原)分别按照以下公式做截断处理，得到处理后的呼吸流信号(Flow 处)和处理后的胸腹运动信号(Res 处)：
Figure PCTCN2019100238-appb-000029
Figure PCTCN2019100238-appb-000030
其中Flow 原(i)表示第i个采样点的原始呼吸流信号，Flow 处(i)表示第i个采样点的处理后的呼吸流信号，Res 原(i)表示第i个采样点的原始胸腹运动信号，Res 处(i)表示第i个采样点的处理后的胸腹运动信号，A为Flow 原的幅值截断门限，且A>0，B为Res 原的幅值截断门限，且B>0。
带通滤波处理模块，用于对所述处理后的呼吸流信号(Flow 处)和所述处理后的胸腹运动信号(Res 处)进行带通滤波处理，得到所述呼吸流信号(Flow)和所述胸腹运动信号(Res)。
在一个实施例中，计算单元具体用于对所述呼吸流信号(Flow)和所述胸腹运动信号(Res)分别按照以下公式计算二阶导数，得到呼吸流信号二阶导数幅值和胸腹运动信号二阶导数幅值：
Figure PCTCN2019100238-appb-000031
2step＜i≤N
Figure PCTCN2019100238-appb-000032
2step＜i≤N
其中step表示求导的步长，N为呼吸流信号序列和胸腹带信号序列的总长度，Flow(i)表示第i个采样点的呼吸流信号，Flow(i-step)表示第i-step个采样点 的呼吸流信号，Flow(i-2step)表示第i-2step个采样点的呼吸流信号，Flow″(i)表示第i个采样点的呼吸流信号的二阶导数幅值，Res(i)表示第i个采样点的胸腹运动信号，Res(i-step)表示第i-step个采样点的胸腹运动信号，Res(i-2step)表示第i-2step个采样点的胸腹运动信号，Res″(i)表示第i个采样点的胸腹运动信号的二阶导数幅值。
在一个实施例中，计算单元具体还可以用于对所述呼吸流信号(Flow)和所述胸腹运动信号(Res)分别计算一阶导数，得到呼吸流信号一阶导数幅值和胸腹运动信号一阶导数幅值，具体可以按照如下公式进行计算：
Figure PCTCN2019100238-appb-000033
step＜i≤N
Figure PCTCN2019100238-appb-000034
step＜i≤N
其中step表示求导的步长，N为呼吸流信号序列和胸腹带信号序列的总长度，Flow(i)表示第i个采样点的呼吸流信号，Flow(i-step)表示第i-step个采样点的呼吸流信号，Flow′(i)表示第i个采样点的呼吸流信号的一阶导数幅值，Res(i)表示第i个采样点的胸腹运动信号，Res(i-step)表示第i-step个采样点的胸腹运动信号，Res′(i)表示第i个采样点的胸腹运动信号的一阶导数幅值。
在一个实施例中，呼吸暂停事件时段确定模块具体用于将每个采样点i对应的Flow″(i)与所述第一预设阈值Thres_flow进行比较，当在一采样点i存在|Flow″(i)|＜Thres_flow，则将该采样点i对应的时刻标记为可能发生呼吸暂停事件的起点，继续遍历后续采样点Flow″(j),j＝i+1,…N，当存在从第i采样点到第M采样点均有：
Figure PCTCN2019100238-appb-000035
且从第i采样点到第M采样点对应的第一时段的持续时间大于10s，则将所述第 一时段标定为呼吸暂停事件第一时段。
呼吸暂停事件类型确定模块具体用于所述呼吸暂停事件第一时段对应的Res″(i)，当|Res″(i)|≥Thres_res,0＜i≤M时，确定所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为阻塞性睡眠呼吸暂停(Obstructive Sleep Apnea，OSA)；
当|Res″(i)|＜Thres_res,0＜i≤M时，确定所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为中枢神经性睡眠呼吸暂停(Central Sleep Apnea，CSA)；
当从第i采样点到第M 1采样点对应的第二时段满足|Res″(j)|＜Thres_res,i≤j≤M 1，从第M 1采样点到第M采样点对应的第三时段满足|Res″(j)|≥Thres_res,M 1＜j≤M，且所述第二时段对应的时长占所述第一时段对应的时长的比例在预设比例范围内，则确定所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为混合性睡眠呼吸暂停(Mixed Sleep Apnea，MSA)。
如图4所示在一个实施例中，一种呼吸暂停事件类型的判断装置，包括：
第一获取单元，用于获取预设时间内的患者的原始呼吸流信号(Flow 原)，所述原始呼吸流信号表征所述患者的呼吸气流，其中所述原始呼吸流信号的个数大于1。
第一预处理单元，用于对所述原始呼吸流信号进行预处理，得到呼吸流信号(Flow)。
第一预处理单元包括：
第一截断处理模块，用于对所述原始呼吸流信号(Flow 原)按照以下公式做截断处理，得到处理后的呼吸流信号(Flow 处)：
Figure PCTCN2019100238-appb-000036
其中Flow 原(i)表示第i个采样点的原始呼吸流信号，Flow 处(i)表示第i个采样点的处理后的呼吸流信号，A为Flow 原的幅值截断门限，且A>0。
第一带通滤波处理模块，用于对所述处理后的呼吸流信号(Flow 处)进行带通滤波处理，得到所述呼吸流信号(Flow)。
第一计算单元，用于计算所述呼吸流信号的导数，得到呼吸流信号导数幅值。具体用于计算呼吸流信号的一阶导数，得到呼吸流信号一阶导数幅值，也可以计 算呼吸流信号的二阶导数，得到呼吸流信号二阶导数幅值。
具体可以按照如下公式计算呼吸流信号的一阶导数：
Figure PCTCN2019100238-appb-000037
step＜i≤N
其中step表示求导的步长，N为呼吸流信号序列和胸腹带信号序列的总长度，Flow(i)表示第i个采样点的呼吸流信号，Flow(i-step)表示第i-step个采样点的呼吸流信号，Flow′(i)表示第i个采样点的呼吸流信号的一阶导数幅值。
具体可以按照如下公式计算呼吸流信号的二阶导数：
Figure PCTCN2019100238-appb-000038
2step＜i≤N
其中step表示求导的步长，N为呼吸流信号序列和胸腹带信号序列的总长度，Flow(i)表示第i个采样点的呼吸流信号，Flow(i-step)表示第i-step个采样点的呼吸流信号，Flow(i-2step)表示第i-2step个采样点的呼吸流信号，Flow″(i)表示第i个采样点的呼吸流信号的二阶导数幅值。
第一确定单元，用于根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段。
具体地，将每个采样点i对应的Flow″(i)与所述第一预设阈值Thres_flow进行比较，当在一采样点i存在|Flow″(i)|＜Thres_flow，则将该采样点i对应的时刻标记为可能发生呼吸暂停事件的起点，继续遍历后续采样点Flow″(j),j＝i+1,…N，当存在从第i采样点到第M采样点均有：
Figure PCTCN2019100238-appb-000039
且从第i采样点到第M采样点对应的第一时段的持续时间大于10s，则将所述第一时段标定为呼吸暂停事件第一时段。
第二获取单元，用于获取所述呼吸暂停事件时段的原始胸腹运动信号(Res 原)。
第二预处理单元，用于对所述原始胸腹运动信号(Res 原)进行预处理，得到胸腹运动信号(Res)。
第二预处理单元包括：
第二截断处理模块，用于对所述原始胸腹运动信号(Res 原)按照以下公式做截断处理，得到处理后的胸腹运动信号(Res 处)：
Figure PCTCN2019100238-appb-000040
其中Res 原(i)表示第i个采样点的原始胸腹运动信号，Res 处(i)表示第i个采样点的处理后的胸腹运动信号，B为Res 原的幅值截断门限，且B>0。
第二带通滤波处理模块，用于对所述处理后的胸腹运动信号(Res 处)进行带通滤波处理，得到所述胸腹运动信号(Res)。
第二计算单元，用于计算所述胸腹运动信号的导数，得到胸腹运动信号导数幅值。具体地，计算胸腹运动信号的一阶导数，得到胸腹运动信号一阶导数幅值，也可以计算胸腹运动信号的二阶导数，得到胸腹运动信号二阶导数幅值。
具体可以按照如下公式计算胸腹运动信号的一阶导数：
Figure PCTCN2019100238-appb-000041
step＜i≤N
其中step表示求导的步长，N为胸腹带信号序列的总长度，Res(i)表示第i个采样点的胸腹运动信号，Res(i-step)表示第i-step个采样点的胸腹运动信号，Res′(i)表示第i个采样点的胸腹运动信号的一阶导数幅值。
具体可以按照如下公式计算胸腹运动信号的二阶导数：
Figure PCTCN2019100238-appb-000042
2step＜i≤N
其中step表示求导的步长，N为呼吸流信号序列的总长度，Res(i)表示第i个采样点的胸腹运动信号，Res(i-step)表示第i-step个采样点的胸腹运动信号，Res(i-2step)表示第i-2step个采样点的胸腹运动信号，Res″(i)表示第i个采样点的胸腹运动信号的二阶导数幅值。
第二确定单元，用于根据所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型。
具体地，所述呼吸暂停事件第一时段对应的Res″(i)，当|Res″(i)|≥Thres_res,0＜i≤M时，确定所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为阻塞性睡眠呼吸暂停(Obstructive Sleep Apnea，OSA)；
当|Res″(i)|＜Thres_res,0＜i≤M时，确定所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为中枢神经性睡眠呼吸暂停(Central Sleep Apnea，CSA)；
当从第i采样点到第M 1采样点对应的第二时段满足|Res″(j)|＜Thres_res,i≤j≤M 1，从第M 1采样点到第M采样点对应的第三时段满足|Res″(j)|≥Thres_res,M 1＜j≤M，且所述第二时段对应的时长占所述第一时段对应的时长的比例在预设比例范围内，则确定所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为混合性睡眠呼吸暂停(Mixed Sleep Apnea，MSA)。
如图5所示，为一个实施例中计算机设备的内部结构示意图，该计算机设备通过系统连接总线连接处理器、非易失性存储介质、内存储器和网络接口。其中，该计算机设备的非易失性存储介质可存储操作系统和计算机可读指令，该计算机可读指令被执行时，可使得处理器执行一种呼吸暂停事件类型的判断方法。该计算机设备的处理器用于提供计算和控制能力，支撑整个计算机设备的运行。该内存储器中可储存有计算机可读指令，该计算机可读指令被处理器执行时，可使得处理器执行一种呼吸暂停事件类型的判断方法。计算机设备的网络接口用于进行网络通信。该计算机设备可以是服务器，服务器可以用独立的服务器或者是多个服务器组成的服务器集群来实现。计算机设备也可以是终端，终端的显示屏可以是液晶显示屏或者电子墨水显示屏，计算机设备的输入装置可以是显示屏上覆盖的触摸层，也可以是计算机设备外壳上设置的按键、轨迹球或触控板，还可以是外接的键盘、触控板或鼠标等。触摸层和显示屏构成触控屏。
在一个实施例中，还提供了一种包括指令的非临时性计算机可读存储介质，例如包括计算机程序(指令)的存储设备，上述程序(指令)可由计算机设备的处理器执行以完成本申请各个实施例所示的呼吸暂停事件类型的判断方法。例如，所述非临时性计算机可读存储介质可以是ROM、随机存取存储器(RAM)、CD-ROM、磁带、软盘和光数据存储设备等。
本申请采用上述技术方案，带来的技术效果为：通过获取预设时间内的患者的原 始呼吸流信号(Flow原)和原始胸腹运动信号(Res原)，对该原始呼吸流信号(Flow原)和原始胸腹运动信号(Res原)进行预处理，排除无效信号和干扰信号的影响，对预处理后得到的呼吸流信号(Flow)和胸腹运动信号(Res)分别进行求导运算，得到呼吸流信号导数幅值和胸腹运动信号导数幅值，根据该呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段，再根据该呼吸暂停事件时段对应的胸腹运动信号地变化趋势的不同来区分CSA/OSA/MSA这三种呼吸暂停事件，该方案复杂度低，判断结果准确性高。
本领域技术人员在考虑说明书及实践这里公开的发明后，将容易想到本申请的其它实施方案。本申请旨在涵盖本申请的任何变型、用途或者适应性变化，这些变型、用途或者适应性变化遵循本申请的一般性原理并包括本申请未公开的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的，本申请的真正范围和精神由本申请的权利要求指出。
应当理解的是，本申请并不局限于上面已经描述并在附图中示出的精确结构，并且可以在不脱离其范围进行各种修改和改变。本申请的范围仅由所附的权利要求来限制。

Claims (11)

  1. 一种呼吸暂停事件类型的判断方法，其特征在于，包括：
    获取预设时间内的患者的原始呼吸流信号(Flow 原)和原始胸腹运动信号(Res 原)，所述原始呼吸流信号表征所述患者的呼吸气流，所述原始胸腹运动信号表征所述患者的胸腹运动，其中所述原始呼吸流信号的个数大于1，所述原始胸腹运动信号的个数大于1；
    对所述原始呼吸流信号和所述原始胸腹运动信号进行预处理，得到呼吸流信号(Flow)和胸腹运动信号(Res)；
    计算所述呼吸流信号的导数和所述胸腹运动信号的导数，得到呼吸流信号导数幅值和胸腹运动信号导数幅值；
    根据所述呼吸流信号导数幅值和所述胸腹运动信号导数幅值判断呼吸暂停事件类型。
  2. 根据权利要求1所述的呼吸暂停事件类型判断方法，其特征在于，所述根据所述呼吸流信号导数幅值和所述胸腹运动信号导数幅值判断呼吸暂停事件类型包括：
    根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段；
    根据所述呼吸暂停事件时段对应的所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型。
  3. 根据权利要求2所述的呼吸暂停事件类型判断方法，其特征在于，所述对所述原始呼吸流信号和所述原始胸腹运动信号进行预处理，得到呼吸流信号和胸腹运动信号包括：
    对所述原始呼吸流信号(Flow 原)和所述原始胸腹运动信号(Res 原)分别按照以下公式做截断处理，得到处理后的呼吸流信号(Flow 处)和处理后的胸腹运动信号(Res 处)：
    Figure PCTCN2019100238-appb-100001
    Figure PCTCN2019100238-appb-100002
    其中Flow 原(i)表示第i个采样点的原始呼吸流信号，Flow 处(i)表示第i个采样点的处理后的呼吸流信号，Res 原(i)表示第i个采样点的原始胸腹运动信号，Res 处(i)表示第i个采样点的处理后的胸腹运动信号，A为Flow 原的幅值截断门限，且A>0，B为Res 原的幅值截断门限，且B>0；
    对所述处理后的呼吸流信号(Flow 处)和所述处理后的胸腹运动信号(Res 处)进行带通滤波处理，得到所述呼吸流信号(Flow)和所述胸腹运动信号(Res)。
  4. 根据权利要求3所述的呼吸暂停事件类型判断方法，其特征在于，所述计算所述呼吸流信号的导数和所述胸腹运动信号的导数，得到呼吸流信号导数幅值和胸腹运动信号导数幅值包括：
    对所述呼吸流信号(Flow)和所述胸腹运动信号(Res)分别按照以下公式计算二阶导数，得到呼吸流信号二阶导数幅值和胸腹运动信号二阶导数幅值：
    Figure PCTCN2019100238-appb-100003
    Figure PCTCN2019100238-appb-100004
    其中step表示求导的步长，N为呼吸流信号序列和胸腹带信号序列的总长度，Flow(i)表示第i个采样点的呼吸流信号，Flow(i-step)表示第i-step个采样点的呼吸流信号，Flow(i-2step)表示第i-2step个采样点的呼吸流信号，Flow″(i)表示第i个采样点的呼吸流信号的二阶导数幅值，Res(i)表示第i个采样点的胸腹运动信号，Res(i-step)表示第i-step个采样点的胸腹运动信号，Res(i-2step)表示第i-2step个采样点的胸腹运动信号，Res″(i)表示第i个采样点的胸腹运动信号的二阶导数幅值。
  5. 根据权利要求4所述的呼吸暂停事件类型判断方法，其特征在于，当1≤i≤2step时，Flow″(i)取值为第一预设值，Res″(i)取值为第二预设值。
  6. 根据权利要求4所述的呼吸暂停事件类型判断方法，其特征在于，所述根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段， 包括：
    将每个采样点i对应的Flow″(i)与所述第一预设阈值Thres_flow进行比较，当在一采样点i存在|Flow″(i)|＜Thres_flow，则将该采样点i对应的时刻标记为可能发生呼吸暂停事件的起点，继续遍历后续采样点Flow″(j),j＝i+1,…N，当存在从第i采样点到第M采样点均有：
    Figure PCTCN2019100238-appb-100005
    且从第i采样点到第M采样点对应的第一时段的持续时间大于10s，则将所述第一时段标定为呼吸暂停事件第一时段。
  7. 根据权利要求6所述的呼吸暂停事件类型判断方法，其特征在于，所述根据所述呼吸暂停事件时段对应的所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型包括：
    所述呼吸暂停事件第一时段对应的Res″(i)，当|Res″(i)|≥Thres_res,0＜i≤M时，所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为阻塞性睡眠呼吸暂停(Obstructive Sleep Apnea，OSA)；
    当|Res″(i)|＜Thres_res,0＜i≤M时，所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为中枢神经性睡眠呼吸暂停(Central Sleep Apnea，CSA)；
    当从第i采样点到第M 1采样点对应的第二时段满足|Res″(j)|＜Thres_res,i≤j≤M 1，从第M 1采样点到第M采样点对应的第三时段满足|Res″(j)|≥Thres_res,M 1＜j≤M，且所述第二时段对应的时长占所述第一时段对应的时长的比例在预设比例范围内，则所述呼吸暂停事件第一时段的所述呼吸暂停事件类型为混合性睡眠呼吸暂停(Mixed Sleep Apnea，MSA)。
  8. 一种呼吸暂停事件类型的判断方法，其特征在于，包括：
    获取预设时间内的患者的原始呼吸流信号(Flow 原)，所述原始呼吸流信号表征所述患者的呼吸气流，其中所述原始呼吸流信号的个数大于1；
    对所述原始呼吸流信号进行预处理，得到呼吸流信号(Flow)；
    计算所述呼吸流信号的导数，得到呼吸流信号导数幅值；
    根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段；
    获取所述呼吸暂停事件时段的原始胸腹运动信号(Res 原)；
    对所述原始胸腹运动信号(Res 原)进行预处理，得到胸腹运动信号(Res)；
    计算所述胸腹运动信号的导数，得到胸腹运动信号导数幅值；
    根据所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型。
  9. 一种呼吸暂停事件类型的判断装置，其特征在于，包括：
    获取单元，用于获取预设时间内的患者的原始呼吸流信号(Flow 原)和原始胸腹运动信号(Res 原)，所述原始呼吸流信号表征所述患者的呼吸气流，所述原始胸腹运动信号表征所述患者的胸腹运动，其中所述原始呼吸流信号的个数大于1，所述原始胸腹运动信号的个数大于1；
    预处理单元，用于对所述原始呼吸流信号和所述原始胸腹运动信号进行预处理，得到呼吸流信号(Flow)和胸腹运动信号(Res)；
    计算单元，用于计算所述呼吸流信号的导数和所述胸腹运动信号的导数，得到呼吸流信号导数幅值和胸腹运动信号导数幅值；
    判断单元，用于根据所述呼吸流信号导数幅值和所述胸腹运动信号导数幅值判断呼吸暂停事件类型。
  10. 一种呼吸暂停事件类型的判断装置，其特征在于，包括：
    第一获取单元，用于获取预设时间内的患者的原始呼吸流信号(Flow 原)，所述原始呼吸流信号表征所述患者的呼吸气流，其中所述原始呼吸流信号的个数大于1；
    第一预处理单元，用于对所述原始呼吸流信号进行预处理，得到呼吸流信号(Flow)；
    第一计算单元，用于计算所述呼吸流信号的导数，得到呼吸流信号导数幅值；
    第一确定单元，用于根据所述呼吸流信号导数幅值和第一预设阈值Thres_flow确定呼吸暂停事件时段；
    第二获取单元，用于获取所述呼吸暂停事件时段的原始胸腹运动信号(Res 原)；
    第二预处理单元，用于对所述原始胸腹运动信号(Res 原)进行预处理，得到胸 腹运动信号(Res)；
    第二计算单元，用于计算所述胸腹运动信号的导数，得到胸腹运动信号导数幅值；
    第二确定单元，用于根据所述胸腹运动信号导数幅值和第二预设阈值Thres_Res确定所述呼吸暂停事件类型。
  11. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质上存储有计算机可执行指令，所述计算机可执行指令被处理器执行时，使得所述处理器执行权利要求1至8中任一项所述方法的步骤。
PCT/CN2019/100238 2018-08-29 2019-08-12 呼吸暂停事件类型的判断方法、装置和存储介质 WO2020042897A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810993543.6 2018-08-29
CN201810993543.6A CN109091112B (zh) 2018-08-29 2018-08-29 呼吸暂停事件类型的判断装置

Publications (1)

Publication Number Publication Date
WO2020042897A1 true WO2020042897A1 (zh) 2020-03-05

Family

ID=64863989

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/100238 WO2020042897A1 (zh) 2018-08-29 2019-08-12 呼吸暂停事件类型的判断方法、装置和存储介质

Country Status (2)

Country Link
CN (1) CN109091112B (zh)
WO (1) WO2020042897A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109091112B (zh) * 2018-08-29 2020-09-29 深圳融昕医疗科技有限公司 呼吸暂停事件类型的判断装置
CN114269238A (zh) * 2019-09-18 2022-04-01 深圳迈瑞生物医疗电子股份有限公司 一种呼吸识别方法及装置、通气设备、存储介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080275349A1 (en) * 2007-05-02 2008-11-06 Earlysense Ltd. Monitoring, predicting and treating clinical episodes
CN102641125A (zh) * 2011-02-18 2012-08-22 西铁城控股株式会社 睡眠呼吸暂停判定装置
CN103202687A (zh) * 2012-01-12 2013-07-17 谢汝石 一种阻塞性睡眠呼吸暂停低通气综合征患者初筛系统
CN104706355A (zh) * 2015-03-26 2015-06-17 北京怡和嘉业医疗科技有限公司 呼吸暂停事件的种类的确定方法及系统
CN106175770A (zh) * 2016-08-01 2016-12-07 华南师范大学 一种睡眠期间呼吸暂停的判断方法及系统
CN107361770A (zh) * 2017-08-15 2017-11-21 杭州医电园科技有限公司 睡眠呼吸暂停事件判别装置
CN109091112A (zh) * 2018-08-29 2018-12-28 深圳融昕医疗科技有限公司 呼吸暂停事件类型的判断方法、装置和存储介质

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE69433699T2 (de) * 1993-11-05 2005-03-31 Resmed Ltd., North Ryde Bestimmung der Durchgängigkeit der Luftwege
CN1241394A (zh) * 1998-06-18 2000-01-19 塔埃玛 测定使用者睡眠呼吸期的装置
DE102007006689B4 (de) * 2006-03-31 2021-07-29 Löwenstein Medical Technology S.A. Vorrichtung und Verfahren zur Obstruktionserkennung während Apnoephasen durch eine zusätzliche Druckstufe
JP5694139B2 (ja) * 2011-12-28 2015-04-01 日本光電工業株式会社 睡眠中における無呼吸低呼吸状態の検出装置
CN107280674A (zh) * 2017-06-02 2017-10-24 南京理工大学 基于音频录取设备的呼吸模式判决方法
CN108392186B (zh) * 2018-04-19 2021-04-13 广西万云科技有限公司 一种非接触式睡眠呼吸暂停症检测方法及系统

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080275349A1 (en) * 2007-05-02 2008-11-06 Earlysense Ltd. Monitoring, predicting and treating clinical episodes
CN102641125A (zh) * 2011-02-18 2012-08-22 西铁城控股株式会社 睡眠呼吸暂停判定装置
CN103202687A (zh) * 2012-01-12 2013-07-17 谢汝石 一种阻塞性睡眠呼吸暂停低通气综合征患者初筛系统
CN104706355A (zh) * 2015-03-26 2015-06-17 北京怡和嘉业医疗科技有限公司 呼吸暂停事件的种类的确定方法及系统
CN106175770A (zh) * 2016-08-01 2016-12-07 华南师范大学 一种睡眠期间呼吸暂停的判断方法及系统
CN107361770A (zh) * 2017-08-15 2017-11-21 杭州医电园科技有限公司 睡眠呼吸暂停事件判别装置
CN109091112A (zh) * 2018-08-29 2018-12-28 深圳融昕医疗科技有限公司 呼吸暂停事件类型的判断方法、装置和存储介质

Also Published As

Publication number Publication date
CN109091112A (zh) 2018-12-28
CN109091112B (zh) 2020-09-29

Similar Documents

Publication Publication Date Title
JP6815362B2 (ja) 睡眠覚醒状態を分類する方法、装置及びシステム
WO2020042897A1 (zh) 呼吸暂停事件类型的判断方法、装置和存储介质
CN108388912A (zh) 基于多传感器特征优化算法的睡眠分期方法
US10219740B2 (en) Systems, methods and apparatuses for the alleviation and outcome monitoring of sleep disordered breathing
EP3639748A1 (en) System and method for monitoring pathological breathing patterns
KR20200024855A (ko) 잡음을 필터링하고 정맥 파형 신호들을 분석하기 위한 시스템들 및 방법들
WO2016150227A1 (zh) 一种睡眠呼吸暂停的检测方法及设备
WO2010080405A1 (en) Detection and prediction of physiological events in people with sleep disordered breathing using a lamstar neural network
US20200281523A1 (en) Apparatus for monitoring a subject
WO2020143161A1 (zh) 一种心率检测方法和装置
DE102018204868A1 (de) Verfahren, Systeme und Vorrichtung zur Detektion von Atemphasen
US20210000385A1 (en) Systems, methods and apparatuses for monitoring hypoxic events
CN107898432A (zh) 鼾声检测方法、装置、设备及介质
US20160128629A1 (en) Apnea detection system and method
CN106419846B (zh) 一种呼吸暂停检测方法
CN111281349A (zh) 睡眠呼吸障碍监测方法及系统
WO2020227856A1 (zh) 一种呼吸中止检测系统、方法、计算设备和储存介质
Daly et al. SleepCare: obstructive sleep apnoea screening using mobile health technology
TWI669104B (zh) 睡眠呼吸中止生理訊號檢測系統
CN110870763B (zh) 鼻口共鸣比测量和视听反馈技术的鼻共鸣康复仪器及方法
RU2749068C2 (ru) Система и способ исследования сна
Guul et al. Portable prescreening system for sleep apnea
WO2024024200A1 (ja) 情報処理装置
CN109567752A (zh) 睡眠觉醒状态的判断方法、装置、睡眠监测仪和存储介质
CN114488841B (zh) 一种智能穿戴设备的数据搜集处理方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19855909

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19855909

Country of ref document: EP

Kind code of ref document: A1