CN112089413A - Blocking type sleep apnea syndrome screening system - Google Patents
Blocking type sleep apnea syndrome screening system Download PDFInfo
- Publication number
- CN112089413A CN112089413A CN202010936602.3A CN202010936602A CN112089413A CN 112089413 A CN112089413 A CN 112089413A CN 202010936602 A CN202010936602 A CN 202010936602A CN 112089413 A CN112089413 A CN 112089413A
- Authority
- CN
- China
- Prior art keywords
- unit
- apnea
- signal
- screening system
- sleep apnea
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
- 201000002859 sleep apnea Diseases 0.000 title claims abstract description 57
- 238000012216 screening Methods 0.000 title claims abstract description 26
- 230000000903 blocking effect Effects 0.000 title description 4
- 208000008784 apnea Diseases 0.000 claims abstract description 53
- 230000029058 respiratory gaseous exchange Effects 0.000 claims abstract description 32
- 238000004458 analytical method Methods 0.000 claims abstract description 31
- 230000000241 respiratory effect Effects 0.000 claims abstract description 22
- 208000001797 obstructive sleep apnea Diseases 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 11
- 238000000034 method Methods 0.000 claims description 47
- 238000012549 training Methods 0.000 claims description 30
- 238000001514 detection method Methods 0.000 claims description 29
- 238000012360 testing method Methods 0.000 claims description 16
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000012937 correction Methods 0.000 claims description 11
- 230000016507 interphase Effects 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 9
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 238000013145 classification model Methods 0.000 claims description 6
- 238000002790 cross-validation Methods 0.000 claims description 6
- 238000012706 support-vector machine Methods 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000012952 Resampling Methods 0.000 claims description 3
- 230000005856 abnormality Effects 0.000 claims description 3
- 208000011338 SATB2 associated disease Diseases 0.000 claims 3
- 208000013959 SATB2-associated syndrome Diseases 0.000 claims 3
- 208000011580 syndromic disease Diseases 0.000 description 3
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 2
- 230000035565 breathing frequency Effects 0.000 description 2
- 238000007635 classification algorithm Methods 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 229910052760 oxygen Inorganic materials 0.000 description 2
- 239000001301 oxygen Substances 0.000 description 2
- 208000019116 sleep disease Diseases 0.000 description 2
- 208000024891 symptom Diseases 0.000 description 2
- 206010020772 Hypertension Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 208000029078 coronary artery disease Diseases 0.000 description 1
- 206010012601 diabetes mellitus Diseases 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 238000000537 electroencephalography Methods 0.000 description 1
- 238000002567 electromyography Methods 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000718 qrs complex Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003860 sleep quality Effects 0.000 description 1
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
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7225—Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation
-
- 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
- A61B5/7246—Details of waveform analysis using correlation, e.g. template matching or determination of similarity
-
- 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
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Physics & Mathematics (AREA)
- Animal Behavior & Ethology (AREA)
- Pathology (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Signal Processing (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Power Engineering (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention relates to the field of sleep, and particularly discloses an obstructive sleep apnea syndrome screening system which comprises an electrocardiosignal analysis module, an apnea recognition module and an apnea degree discrimination module, wherein the electrocardiosignal analysis module is used for analyzing and processing electrocardiosignals, the apnea recognition module is used for identifying and diagnosing the apnea condition of people, and the apnea degree discrimination module is used for analyzing and processing respiration signals and analyzing the morbidity degree of sleep apnea syndrome. The invention is scientific and reasonable, is safe and convenient to use, can search the wave crests and the wave troughs in the respiratory signal by utilizing the apnea degree judging module, and judges whether the tested person has the sleep apnea syndrome according to the time interval between two adjacent wave crests.
Description
Technical Field
The invention relates to the technical field of sleep diseases, in particular to an obstructive sleep apnea syndrome screening system.
Background
The Sleep Apnea Syndrome (SAS) is a common sleep disease, easily induces various diseases such as hypertension, diabetes, coronary heart disease and the like, seriously threatens the life safety of a patient, and needs to be treated in time, and is characterized in that the SAS is characterized in that the apnea or the respiratory volume is reduced in sleep, when the sleep apnea occurs, the respiratory airflow stops or the airflow intensity is reduced by 30 percent compared with the basic level, the duration time is more than ten seconds, the breathing is not smooth and the death is easily caused, so a polysomnogram appears in the market;
polysomnography (PSG) is the current gold standard for diagnosing OSAS, and not only can diagnose the degree of the SAS, but also can quantitatively analyze the sleep structure of a patient, and in addition, PSG can detect various data in the sleep process, such as: electroencephalography (EEG), Electrocardiogram (ECG), Electrooculogram (EOG), Electromyography (EMG), oxygen saturation during airflow and sleep, blood oxygen saturation signals, etc., the measurement accuracy is high, but PSG also has the following disadvantages:
1. the detection time is too long, and the patient needs to sleep for at least one night in a sleep laboratory;
2. the sleep laboratory is expensive in cost and needs related technicians to operate;
3. the patient needs to wear various sensors for monitoring for a long time, the comfort is poor, and the sleep quality of the patient can be affected;
research shows that sleep apnea has certain influence on HRV (small difference of instantaneous heart rate between continuous heartbeats), and the sleep apnea causes slow heartbeat and then aggravation, so that an apnea patient can be screened by monitoring the HRV in sleep, and therefore, an obstructive sleep apnea syndrome screening system is needed to solve the problem.
Disclosure of Invention
The present invention is directed to an obstructive sleep apnea syndrome screening system, which solves the above-mentioned problems of the background art.
In order to solve the technical problems, the invention provides the following technical scheme: a blocking sleep apnea syndrome screening system comprises an electrocardiosignal analysis module, an apnea identification module and an apnea degree discrimination module;
the electrocardiosignal analysis module is used for analyzing and processing electrocardiosignals, so that no noise is interfered in the electrocardiosignals, the apnea recognition module is used for recognizing and diagnosing the breath of people, so as to judge the difference between normal people and the breath minutes with SAS, the apnea degree discrimination module is used for analyzing and processing the breath signals, so that the breath signals can be stably output, the respiration rate is calculated, the morbidity degree of sleep apnea syndrome is analyzed, the electrocardiosignal analysis module is connected with the apnea recognition module, and the apnea recognition module is connected with the apnea degree discrimination module.
The electrocardiosignal analysis module comprises a signal denoising unit, an electrocardiosignal analysis unit and a characteristic selection unit, wherein the signal denoising unit is used for removing noise in electrocardiosignals, a median filter is used for removing baseline drift in the electrocardiosignals, the phenomenon of the current truncation error is calculated by data in the process of proceeding the median, the approximate baseline profile of the electrocardiosignals is obtained through the phenomenon, the baseline profile is subtracted, namely the baseline is removed, meanwhile, the wavelet threshold denoising method is used for removing power frequency interference and myoelectricity interference in the electrocardiosignals, the electrocardiosignal analysis unit is used for analyzing the related indexes of HRV so as to be convenient for knowing the related parameters of the HRV, the characteristic selection unit is used for extracting the characteristics in RR interphase signals, and the heart rate is calculated through the related indexes of the HRV so as to know the difference between normal breathing and abnormal breathing frequencies, the output ends of the electrocardiosignal analysis unit and the characteristic selection unit are connected with the input end of the signal denoising unit;
when the median filtering is performed, because the data in the database is 100HZ electrocardiosignals, the signals are interpolated into 200HZ signals, the original signals are processed by two median filters on the interpolated signals, the first median filter can filter QRS complexes and P waves in the electrocardiosignals, the window width is 200ms, the second median filter can filter T waves in the electrocardiosignals, and the window width is 600 ms;
when the wavelet threshold denoising method is used for removing power frequency interference, a bior3.5 wavelet basis is selected for five-layer decomposition, denoising is carried out by using high-frequency or low-frequency scale parameters, and a threshold is calculated, so that an interference signal can be removed, and a low-frequency effective part in the signal can be prevented from being removed;
the apnea recognition module comprises a signal distinguishing unit, a model training unit and a model updating unit, wherein the signal distinguishing unit is used for distinguishing and judging the minutes of people with SAS and normal people, whether the people have SAS symptoms can be judged according to the breathing condition of the people, and the model training unit is used for training and testing the model by data, so that the optimal model can be selected;
when the kernel function is used for classifying the vector machine, the radial basis kernel function, the linear kernel function and the polynomial kernel function are usually used for classifying the model, through experimental comparison, the classification performance of the radial basis kernel function is found to be optimal, the accuracy of a training set can reach 97.42%, the F value reaches 95.25%, the accuracy of a test set reaches 88.42%, and the F value reaches 86.81%, when a classification algorithm of the radial basis kernel function support vector machine is used for distinguishing normal and abnormal minutes in electrocardiosignals, secondary classification can be carried out in a high-dimensional space, wherein the discriminant function of the radial basis kernel function is as follows:
wherein: se is the correct identification rate of the SAS, SP is the detectable rate of the normal sample, and F is the comprehensive judgment value of the classification performance of the model.
Preferably, the apnea degree judging module comprises a respiration signal preprocessing unit and a morbidity degree judging unit, wherein the respiration signal preprocessing unit is used for removing interference in respiration signal waveforms, so that various actions in sleep can not influence the respiration signals, the morbidity degree judging unit is used for judging the morbidity degree of an SAS, the morbidity degree of sleep apnea syndrome can be judged according to the number of times of sleep apnea, and the output end of the respiration signal preprocessing unit is connected with the input end of the morbidity degree judging unit;
the degree of the sleep apnea syndrome is as follows:
degree of | AHI (second/hour) |
Mild degree of | 5-14 |
Of moderate degree | 15-29 |
Severe degree | ≥30 |
An obstructive sleep apnea syndrome screening system, the screening system comprising the steps of:
s1: processing interference signals in the electrocardiosignals by using an electrocardiosignal analysis module, and detecting HRV signals to judge whether heart rate abnormality exists;
s2: distinguishing the respiratory minute segment of a normal person from the minute segment with SAS by using an apnea identification module and a method of a support vector machine;
s3: training a model of the vector machine by using a cross validation method by using a model training unit;
s4: carrying out classification test on the data of the whole test set by using a model updating unit to obtain an optimal classification model;
s5: and (3) utilizing an apnea degree judging module to remove clutter from the respiratory signal by using a cubic spline interpolation fitting method and a waveform method.
In step S1, removing baseline wander generated in the electrocardiosignal by using a median filtering method, detecting the RR interval signal by using a time domain analysis method, and determining whether the heart rate is abnormal;
according to the formula:
wherein, RRmeanMean values of RR intervals, representing the level of heart rate population, SDANN the standard deviation of mean values of RR intervals every 5min, representing the slow variation characterizing HRV, HR representing heart rate, RR interval signals being the time interval between beat-to-beat.
Preferably, in step S2, before extracting the features in the RR interval signal, all R waves are detected by using a QRS wave detection algorithm, and in the preliminary detection process, noise with a higher amplitude is regarded as a QRS wave, which causes an error detection phenomenon and an undetected phenomenon, so that the signal is detected and corrected by using an RR interval correction algorithm, and the detection method is as follows:
a1: judging whether the phenomenon of missing detection occurs by utilizing the five-point average calculation of the RR interphase;
a2: judging whether the phenomenon of false detection occurs by utilizing the four-point average calculation of the RR interphase;
in step A1, the average value of the five points of the RR interval is calculated, namely, the RR is calculatediAnd RRiCalculating the mean value of the two points before and after the step (2), according to a formula:
when RR isi>1.6×RR5meanIn time, detection missing occurs, and the RR interval correction algorithm can correct RRiSplitting into K equal parts untilWhen the value of (A) is minimum, K are usedTo replace RRiThereby correcting the missed detection;
in step a2, the RR interval correction algorithm performs a four-point average calculation on any two adjacent RR interval data according to the formula:
RR4mean>RRi;
RR4mean>RRi+1;
|RRi+RRi+1-2×RR4mean|>|RRi+RRi+1-RR4mean|;
when the above conditions are met, the error detection can be corrected;
resampling the RR' interval signal, and obtaining 3 frequency domain characteristics according to decomposition: LF, HF and VLF and several time domain features.
In the steps S3-S4, a K-fold cross-validation method is used to train the vector machine model and determine the classification model of the vector machine, and the steps are as follows:
q1: dividing the whole training set S into K disjoint subsets, wherein the number of training samples in S is set to be m, each subset has m/K training samples, and the corresponding subset is { S1, S2, … sk };
q2: selecting one from the subsets as a test set, and taking other ones as training sets;
q3: putting the model on a test set to obtain a classification rate;
q4: and calculating the average value of the classification rate obtained by K times to serve as the real classification rate of the model.
In the step 5, fitting the respiratory signal by utilizing a cubic spline interpolation fitting method, fitting the maximum value and the minimum value of the envelope curve, and calculating the average value of the envelope curve so as to obtain the peak and the trough of the respiratory signal;
according to the formula:
S(xj)=yj(j=0,1,…,n);
the second derivative is performed on the above formula to obtain:
S″(x0)=M0=f″(x0);
in the function s (x), each interval is a cubic polynomial with four unknowns, and since the conditions of interpolation and spline are 4n-2, the boundary provides the following two conditions:
S″(xn)=Mn=0;
S″(x0)=M0=0;
wherein: s (x) is a cubic spline interpolation function;
in MATLAB, calculating the average value of the envelope through a ppval function, finding out the maximum value M (t) of the envelope function by using a findpeak function, and finding out the time Z corresponding to the peak2(t) is Y (t);
wherein: z1The sum (t) is the envelope signal.
In step S5, the number of apneas is calculated by a waveform method;
according to the formula:
Y(f)=Y(i+1)-Y(i);
when Y (i +1) -Y (i) <2.4, this wave is an interference wave;
when Y (i +1) -Y (i) >10, the occurrence of one sleep apnea condition is indicated;
normal breathing is indicated when 2.4< Y (i +1) -Y (i) < 10;
wherein: and Y (f) is the apnea time.
Compared with the prior art, the invention has the following beneficial effects:
1. by using the system, the effective detection time is not overlong, the use cost is low, the signal stability is high, and the system is convenient to carry;
2. the electrocardiosignal analysis module is utilized to remove noise in the electrocardiosignals, extract time-frequency domain characteristics from RR interphase signals, detect and calculate heart rate through the time-frequency domain characteristics, and judge whether the person breathes normally through the electrocardiosignals;
3. by utilizing the apnea identification module, the classification algorithm of a radial basis kernel function support vector machine can be utilized to distinguish the minute segment with SAS and the minute segment of a normal person in the electrocardiosignal;
4. by utilizing the apnea degree judging module, the times of sleep apnea can be calculated by a cubic spline fitting value method and a waveform method, and the attack degree of the sleep apnea syndrome is judged.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a schematic block diagram of an obstructive sleep apnea syndrome screening system according to the present invention;
FIG. 2 is a schematic flow chart of support vector machine classification of an obstructive sleep apnea syndrome screening system of the present invention;
FIG. 3 is a schematic flow chart of SAS disease degree judgment of the obstructive sleep apnea syndrome screening system according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
A blocking sleep apnea syndrome screening system comprises an electrocardiosignal analysis module, an apnea identification module and an apnea degree discrimination module;
the electrocardiosignal analysis module is used for analyzing and processing electrocardiosignals, the apnea identification module is used for identifying and diagnosing human breath, the apnea degree discrimination module is used for analyzing and processing respiration signals and analyzing the morbidity degree of sleep apnea syndrome, the electrocardiosignal analysis module is connected with the apnea identification module, and the apnea identification module is connected with the apnea degree discrimination module.
The electrocardiosignal analysis module comprises a signal denoising unit, a signal analysis unit and a characteristic selection unit, wherein the signal denoising unit is used for removing noise in electrocardiosignals, the electrocardiosignal analysis unit is used for analyzing related indexes of HRV, the characteristic selection unit is used for extracting characteristics in RR interval signals, and output ends of the electrocardiosignal analysis unit and the characteristic selection unit are connected with an input end of the signal denoising unit.
The apnea recognition module comprises a signal distinguishing unit and a model training unit, the signal distinguishing unit is used for distinguishing and judging minutes of people suffering from SAS and normal people, the model training unit is used for training and testing a model by data, and the output end of the model training unit is connected with the input end of the signal distinguishing unit.
The apnea degree judging module comprises a respiration signal preprocessing unit and a morbidity degree judging unit, the respiration signal preprocessing unit is used for removing interference in respiration signal waveforms, the morbidity degree judging unit is used for judging the morbidity degree of the SAS, and the output end of the respiration signal preprocessing unit is connected with the input end of the morbidity degree judging unit.
The screening system comprises the following steps:
s1: processing interference signals in the electrocardiosignals by using an electrocardiosignal analysis module, and detecting HRV signals to judge whether heart rate abnormality exists;
in step S1, removing baseline wander generated in the electrocardiosignal by using a median filtering method, detecting the RR interval signal by using a time domain analysis method, and determining whether the heart rate is abnormal;
according to the formula:
wherein, RRmeanMean values of RR intervals, indicating the level of heart rate population, SDANN the standard deviation of mean values per 5minRR intervals, indicating slow changes characterizing HRV, HR indicating heart rate.
S2: correcting the signal of the RR interphase by using a characteristic selection unit and extracting time domain characteristics;
in step S2, before extracting features in the RR interval signal, all R waves are detected by using a QRS wave detection algorithm, and in the preliminary detection process, noise with a higher amplitude is regarded as a QRS wave, which causes an error detection phenomenon and an undetected phenomenon, so that the RR interval correction algorithm is used to perform undetected and overdetected correction on the signal, and the detection method is as follows:
a1: judging whether the phenomenon of missing detection occurs by utilizing the five-point average calculation of the RR interphase;
a2: judging whether the phenomenon of false detection occurs by utilizing the four-point average calculation of the RR interphase;
in step A1, the average value of the five points of the RR interval is calculated, namely, the RR is calculatediAnd RRiCalculating the mean value of the two points before and after the step (2), according to a formula:
when RR isi>1.6×RR5meanIn time, detection missing occurs, and the RR interval correction algorithm can correct RRiSplitting into K equal parts untilWhen the value of (A) is minimum, K are usedTo replace RRiThereby correcting the missed detection;
in step a2, the RR interval correction algorithm performs a four-point average calculation on any two adjacent RR interval data according to the formula:
RR4mean>RRi;
RR4mean>RRi+1;
|RRi+RRi+1-2×RR4mean|>|RRi+RRi+1-RR4mean|;
when the above conditions are met, the error detection can be corrected;
resampling the RR' interval signal, and obtaining 3 frequency domain characteristics according to decomposition: LF, HF and VLF and several time domain features.
S3: distinguishing the respiratory minute segment of a normal person from the minute segment with SAS by using an apnea identification module and a method of a support vector machine;
s4: training a model of the vector machine by using a cross validation method by using a model training unit;
in the steps S3-S4, a K-fold cross-validation method is used to train the vector machine model and determine the classification model of the vector machine, and the steps are as follows:
q1: dividing the whole training set S into K disjoint subsets, wherein the number of training samples in S is set to be m, each subset has m/K training samples, and the corresponding subset is { S1, S2, … sk };
q2: selecting one from the subsets as a test set, and taking other ones as training sets;
q3: putting the model on a test set to obtain a classification rate;
q4: and calculating the average value of the classification rate obtained by K times to serve as the real classification rate of the model.
In the step 5, fitting the respiratory signal by utilizing a cubic spline interpolation fitting method, fitting the maximum value and the minimum value of the envelope curve, and calculating the average value of the envelope curve so as to obtain the peak and the trough of the respiratory signal;
according to the formula:
S(xj)=yj(j=0,1,…,n);
the second derivative is performed on the above formula to obtain:
S″(x0)=M0=f″(x0);
in the function s (x), each interval is a cubic polynomial with four unknowns, and since the conditions of interpolation and spline are 4n-2, the boundary provides the following two conditions:
S″(xn)=Mn=0;
S″(x0)=M0=0;
wherein: s (x) is a cubic spline interpolation function;
in MATLAB, calculating the average value of the envelope through a ppval function, finding out the maximum value M (t) of the envelope function by using a findpeak function, and finding out the time Z corresponding to the peak2(t) is Y (t);
wherein: z1The sum (t) is the envelope signal.
S5: carrying out classification test on the data of the whole test set by using a model updating unit to obtain an optimal classification model;
in step S5, the number of apneas is calculated by a waveform method;
according to the formula:
Y(f)=Y(i+1)-Y(i);
when Y (i +1) -Y (i) <2.4, this wave is an interference wave;
when Y (i +1) -Y (i) >10, the occurrence of one sleep apnea condition is indicated;
normal breathing is indicated when 2.4< Y (i +1) -Y (i) < 10;
wherein: and Y (f) is the apnea time.
S6: and (3) utilizing an apnea degree judging module to remove clutter from the respiratory signal by using a cubic spline interpolation fitting method and a waveform method.
Example 1: the method comprises the steps of detecting a central electric signal and a respiratory signal of a testee in a sleeping process, judging whether the heart rate beat accords with a heart rate value in the sleeping process or not through the heart rate beat of the testee in the sleeping process, distinguishing the heart rate value of normal breathing from the heart rate value of an abnormal signal in the heart rate signal by utilizing a radial basis kernel function, calculating an average value of an envelope line by utilizing a cubic spline fitting value method, enabling peaks and troughs in the respiratory signal to be more obvious, and judging the morbidity degree in the sleeping breathing process by utilizing a waveform method for envelope peak points and trough points in the respiratory signal.
Detecting time interval in the respiratory signal by using findpeak function, and setting the first respiratory signal of a certain adjacent peak point in the sleep respiratory signal of the detected person as Yi、Yi+1The time interval between two adjacent peak points is Yi+1-YiThe breathing time interval of a normal person is set to YvWhen a time interval Y of two peak points in the signal is detectedi+1-Yi<At 2.4s, the two peak points are considered as interference signals to be removed, and when the time interval Y of the two peak points in the signals is detectedi+1-YiWhen the time interval of the peak point is more than 10s, accumulating the apnea condition by using the AHI;
when the AHI is 5 to 14, the degree of the sleep breathing syndrome is mild, when the AHI is 15 to 29, the degree of the sleep breathing syndrome is moderate, and when the AHI is more than 30 times, the degree of the sleep breathing syndrome is severe;
the present embodiment determines whether the subject is normal or not with respect to the peak interval in the respiratory signal of the subject.
The working principle of the invention is as follows: in the electrocardiosignal analysis module, a median filter is used for removing baseline drift in electrocardiosignals, the phenomenon of the existing truncation error is calculated by data on the process of the median, the approximate baseline profile of the electrocardiosignals is obtained by the phenomenon, the baseline profile is subtracted, namely, the baseline is removed, simultaneously, the power frequency interference and the electromyographic interference in the electrocardiosignals are removed by a wavelet threshold denoising method, after the signal interference is removed, the characteristics in RR interval signals are extracted, the heart rate is calculated by the related index of HRV, so that the difference between normal breathing and abnormal breathing frequency is known, in the apnea identification module, the radial basis kernel function is used for distinguishing and judging the minute segments of persons suffering from SAS and normal persons, whether the persons suffer from SAS symptoms or not is judged according to the breathing condition of the persons, and the apnea degree judgment module is used for removing the interference in the respiration signal waveform, and judging the sleep apnea times by utilizing a cubic spline interpolation function and a waveform method, and judging the morbidity degree of the sleep apnea syndrome.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An obstructive sleep apnea syndrome screening system, characterized by: the screening system comprises an electrocardiosignal analysis module, an apnea recognition module and an apnea degree discrimination module;
the electrocardiosignal analysis module is used for analyzing and processing electrocardiosignals, the apnea identification module is used for identifying and diagnosing the apnea condition of people, the apnea degree discrimination module is used for analyzing and processing the respiration signals and analyzing the morbidity degree of sleep apnea syndrome, the electrocardiosignal analysis module is connected with the apnea identification module, and the apnea identification module is connected with the apnea degree discrimination module.
2. The obstructive sleep apnea syndrome screening system of claim 1, wherein: the electrocardiosignal analysis module comprises a signal denoising unit, an electrocardiosignal analysis unit and a characteristic selection unit, wherein the signal denoising unit is used for removing noise in electrocardiosignals, the electrocardiosignal analysis unit is used for analyzing related indexes of HRV, the characteristic selection unit is used for extracting characteristics in RR interval signals, and output ends of the electrocardiosignal analysis unit and the characteristic selection unit are connected with an input end of the signal denoising unit.
3. The obstructive sleep apnea syndrome screening system of claim 1, wherein: the apnea recognition module comprises a signal distinguishing unit and a model training unit, the signal distinguishing unit is used for distinguishing and judging minutes of people suffering from SAS and normal people, the model training unit is used for training and testing a model by data, and the output end of the model training unit is connected with the input end of the signal distinguishing unit.
4. The obstructive sleep apnea syndrome screening system of claim 1, wherein: the apnea degree judging module comprises a respiration signal preprocessing unit and a morbidity degree judging unit, the respiration signal preprocessing unit is used for removing interference in respiration signal waveforms, the morbidity degree judging unit is used for judging the morbidity degree of the SAS, and the output end of the respiration signal preprocessing unit is connected with the input end of the morbidity degree judging unit.
5. The obstructive sleep apnea syndrome screening system of claim 1, wherein: the screening system comprises the following steps:
s1: processing interference signals in the electrocardiosignals by using an electrocardiosignal analysis module, and detecting HRV signals to judge whether heart rate abnormality exists;
s2: correcting the signal of the RR interphase by using a characteristic selection unit and extracting time domain characteristics;
s3: distinguishing the respiratory minute segment of a normal person from the minute segment with SAS by using an apnea identification module and a method of a support vector machine;
s4: training a model of the vector machine by using a cross validation method by using a model training unit;
s5: carrying out classification test on the data of the whole test set by using a model updating unit to obtain an optimal classification model;
s6: and (3) utilizing an apnea degree judging module to remove clutter from the respiratory signal by using a cubic spline interpolation fitting method and a waveform method.
6. The obstructive sleep apnea syndrome screening system of claim 5, wherein: in step S1, removing baseline wander generated in the electrocardiosignal by using a median filtering method, detecting the RR interval signal by using a time domain analysis method, and determining whether the heart rate is abnormal;
according to the formula:
wherein, RRmeanMean values of RR intervals, indicating the level of heart rate population, SDANN is the standard deviation of mean values of RR intervals every 5 minutes, indicating slow changes in HRV, HR indicating heart rate.
7. The obstructive sleep apnea syndrome screening system of claim 5, wherein: in step S2, before extracting features in the RR interval signal, all R waves are detected by using a QRS wave detection algorithm, and in the preliminary detection process, noise with a higher amplitude is regarded as a QRS wave, which causes an error detection phenomenon and an undetected phenomenon, so that the RR interval correction algorithm is used to perform undetected and overdetected correction on the signal, and the detection method is as follows:
a1: judging whether the phenomenon of missing detection occurs by utilizing the five-point average calculation of the RR interphase;
a2: judging whether the phenomenon of false detection occurs by utilizing the four-point average calculation of the RR interphase;
in step A1, the average value of the five points of the RR interval is calculated, namely, the RR is calculatediAnd RRiCalculating the mean value of the two points before and after the step (2), according to a formula:
when RR isi>1.6×RR5meanIn time, detection missing occurs, and the RR interval correction algorithm can correct RRiSplitting into K equal parts untilWhen the value of (A) is minimum, K are usedTo replace RRiThereby correcting the missed detection;
in step a2, the RR interval correction algorithm performs a four-point average calculation on any two adjacent RR interval data according to the formula:
RR4mean>RRi;
RR4mean>RRi+1;
|RRi+RRi+1-2×RR4mean|>|RRi+RRi+1-RR4mean|;
when the above conditions are met, the error detection can be corrected;
resampling the RR' interval signal, and obtaining 3 frequency domain characteristics according to decomposition: LF, HF and VLF and several time domain features.
8. The obstructive sleep apnea syndrome screening system of claim 5, wherein: in the steps S3-S4, a K-fold cross-validation method is used to train the vector machine model and determine the classification model of the vector machine, and the steps are as follows:
q1: dividing the whole training set S into K disjoint subsets, wherein the number of training samples in S is set to be m, each subset has m/K training samples, and the corresponding subset is { S1, S2, … sk };
q2: selecting one from the subsets as a test set, and taking other ones as training sets;
q3: putting the model on a test set to obtain a classification rate;
q4: and calculating the average value of the classification rate obtained by K times to serve as the real classification rate of the model.
9. The obstructive sleep apnea syndrome screening system of claim 5, wherein: in the step 5, fitting the respiratory signal by utilizing a cubic spline interpolation fitting method, fitting the maximum value and the minimum value of the envelope curve, and calculating the average value of the envelope curve so as to obtain the peak and the trough of the respiratory signal;
according to the formula:
S(xj)=yj(j=0,1,…,n);
the second derivative is performed on the above formula to obtain:
S″(x0)=M0=f″(x0);
in the function s (x), each interval is a cubic polynomial with four unknowns, and since the conditions of interpolation and spline are 4n-2, the boundary provides the following two conditions:
S″(xn)=Mn=0;
S″(x0)=M0=0;
wherein: s (x) is a cubic spline interpolation function;
in MATLAB, calculating the average value of the envelope through a ppval function, finding out the maximum value M (t) of the envelope function by using a findpeak function, and finding out the time Z corresponding to the peak2(t) is Y (t);
wherein: z1The sum (t) is the envelope signal.
10. The obstructive sleep apnea syndrome screening system of claim 5, wherein: in step S5, the number of apneas is calculated by a waveform method;
according to the formula:
Y(f)=Y(i+1)-Y(i);
when Y (i +1) -Y (i) <2.4, this wave is an interference wave;
when Y (i +1) -Y (i) >10, the occurrence of one sleep apnea condition is indicated;
normal breathing is indicated when 2.4< Y (i +1) -Y (i) < 10;
wherein: and Y (f) is the apnea time.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010936602.3A CN112089413A (en) | 2020-09-08 | 2020-09-08 | Blocking type sleep apnea syndrome screening system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010936602.3A CN112089413A (en) | 2020-09-08 | 2020-09-08 | Blocking type sleep apnea syndrome screening system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112089413A true CN112089413A (en) | 2020-12-18 |
Family
ID=73751693
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010936602.3A Pending CN112089413A (en) | 2020-09-08 | 2020-09-08 | Blocking type sleep apnea syndrome screening system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112089413A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113205022A (en) * | 2021-04-23 | 2021-08-03 | 湖南万脉医疗科技有限公司 | Respiratory anomaly monitoring method and system based on wavelet analysis |
CN113243890A (en) * | 2021-05-10 | 2021-08-13 | 清华大学深圳国际研究生院 | Sleep apnea syndrome recognition device |
WO2021253125A1 (en) * | 2020-06-18 | 2021-12-23 | University Of Manitoba | Systems and methods for screening obstructive sleep apnea during wakefulness using anthropometric information and tracheal breathing sounds |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104545888A (en) * | 2014-12-27 | 2015-04-29 | 迪姆软件(北京)有限公司 | Sleep apnea collecting and analyzing system based on dynamic electrocardiogram and respiratory wave collection |
CN109674474A (en) * | 2018-11-30 | 2019-04-26 | 深圳和而泰智能控制股份有限公司 | Sleep apnea recognition methods, equipment and computer-readable medium |
-
2020
- 2020-09-08 CN CN202010936602.3A patent/CN112089413A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104545888A (en) * | 2014-12-27 | 2015-04-29 | 迪姆软件(北京)有限公司 | Sleep apnea collecting and analyzing system based on dynamic electrocardiogram and respiratory wave collection |
CN109674474A (en) * | 2018-11-30 | 2019-04-26 | 深圳和而泰智能控制股份有限公司 | Sleep apnea recognition methods, equipment and computer-readable medium |
Non-Patent Citations (2)
Title |
---|
余晓敏: "基于心电信号的睡眠呼吸暂停综合征检测算法", 《中国优秀硕士学位论文全文数据库医药卫生科技辑》 * |
李忠海: "基于 PVDF 的睡眠监测系统设计与实现", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021253125A1 (en) * | 2020-06-18 | 2021-12-23 | University Of Manitoba | Systems and methods for screening obstructive sleep apnea during wakefulness using anthropometric information and tracheal breathing sounds |
CN113205022A (en) * | 2021-04-23 | 2021-08-03 | 湖南万脉医疗科技有限公司 | Respiratory anomaly monitoring method and system based on wavelet analysis |
CN113243890A (en) * | 2021-05-10 | 2021-08-13 | 清华大学深圳国际研究生院 | Sleep apnea syndrome recognition device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Adnane et al. | Development of QRS detection algorithm designed for wearable cardiorespiratory system | |
Singh et al. | ECG to individual identification | |
US7578793B2 (en) | Sleep staging based on cardio-respiratory signals | |
JP5303802B2 (en) | Apnea / hypopnea index derived from ECG | |
Mendez et al. | Automatic screening of obstructive sleep apnea from the ECG based on empirical mode decomposition and wavelet analysis | |
US7343198B2 (en) | System, software, and method for detection of sleep-disordered breathing using an electrocardiogram | |
US7623912B2 (en) | Method, apparatus and system for characterizing sleep | |
KR101868888B1 (en) | Classification of Sleep/Wakefulness using Nasal Pressure for Patients with Sleep-disordered Breathing | |
CN112089413A (en) | Blocking type sleep apnea syndrome screening system | |
US20020002327A1 (en) | Method for detecting cheyne-stokes respiration in patients with congestive heart failure | |
Dong et al. | Frequency network analysis of heart rate variability for obstructive apnea patient detection | |
WO2008132736A2 (en) | Method and device for characterizing sleep | |
CN110755049A (en) | Anesthesia state monitoring method based on sign signal analysis | |
Cho et al. | Detection of arousals in patients with respiratory sleep disorders using a single channel EEG | |
Shinar et al. | Obstructive sleep apnea detection based on electrocardiogram analysis | |
Hugeng et al. | Development of the ‘Healthcor’system as a cardiac disorders symptoms detector using an expert system based on arduino uno | |
Mendez et al. | Automatic detection of sleep macrostructure based on bed sensors | |
CN113827202A (en) | Sleep quality detection method and device based on machine learning | |
Chatlapalli et al. | Accurate derivation of heart rate variability signal for detection of sleep disordered breathing in children | |
Rohan et al. | Classification of cardiac arrhythmia diseases from obstructive sleep apnea signals using decision tree classifier | |
Sadr et al. | Non-invasive diagnosis of sleep apnoea using ecg and respiratory bands | |
Hossen | A novel method for identification of obstructive sleep apnea | |
Herrero et al. | Electrocardiogram-based detection of central sleep apnea: a full-record signal processing approach | |
Tyagi et al. | Identification of QRS Segments of Electrocardiogram signals using Feature Extraction | |
Garma et al. | Diagnosis of sleep apnea using heart rate features |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201218 |
|
RJ01 | Rejection of invention patent application after publication |