CN110742621A - Signal processing method and computer equipment - Google Patents
Signal processing method and computer equipment Download PDFInfo
- Publication number
- CN110742621A CN110742621A CN201911055292.8A CN201911055292A CN110742621A CN 110742621 A CN110742621 A CN 110742621A CN 201911055292 A CN201911055292 A CN 201911055292A CN 110742621 A CN110742621 A CN 110742621A
- Authority
- CN
- China
- Prior art keywords
- blood oxygen
- signal
- signals
- oxygen saturation
- time domain
- 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.)
- Granted
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
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/14542—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue for measuring blood gases
-
- 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
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
- A61B5/0826—Detecting or evaluating apnoea events
-
- 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
- A61B5/7235—Details of waveform analysis
-
- 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
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
Abstract
The invention provides a signal processing method and computer equipment, and relates to the technical field of signals. Wherein the method comprises the following steps: acquiring an original blood oxygen signal; preprocessing the original blood oxygen signal to obtain a first blood oxygen signal; determining a first time domain characteristic of the first blood oxygen signal; the first time domain feature is used for characterizing whether the blood oxygen saturation of the first blood oxygen signal is reduced or not; and inputting the first time domain characteristic into a preset respiratory classification model to obtain the target respiratory category to which the original blood oxygen signal belongs. In the embodiment of the present invention, the signal processing device may classify the first blood oxygen signal according to a time domain feature that can represent whether the blood oxygen saturation is decreased, and classify the first blood oxygen signal through a preset respiration classification model, so as to determine a respiration type to which the original blood oxygen signal corresponding to the first blood oxygen signal belongs.
Description
Technical Field
The present invention relates to the field of signal technology, and in particular, to a signal processing method and a computer device.
Background
In recent years, with the development of signal technology, the application field of signal technology is becoming wider and wider, for example, in the medical field, medical signal data such as blood oxygen signals and electrocardiosignals can be processed, and the obtained processing result can assist clinical analysis.
At present, people pay more and more attention to sleep quality, so the problem in sleep gradually begins to be emphasized by people. Apnea (apnea) refers to complete cessation of oronasal airflow for more than 10s during sleep, hypopnea (hypopnea) refers to a decrease in the amplitude of the respiratory airflow by more than 50% from the baseline level, while the blood oxygen saturation decreases by more than 4% from the baseline level, and the apnea-hypoventilation index (AHI) refers to the sum of the number of apneas and hypopneas per hour of sleep time.
In the related art, whether the sleep apnea is a normal sleep apnea type or not can be determined only by a simple threshold discrimination method according to the sleep apnea hypopnea index, but the accuracy of the threshold discrimination method is low.
Disclosure of Invention
The invention provides a signal processing method and computer equipment, which aim to solve the problem of low accuracy rate due to the fact that a simple threshold value distinguishing mode is adopted in the existing sleep breathing type distinguishing.
In order to solve the above problem, the present invention discloses a signal processing method, comprising:
acquiring an original blood oxygen signal;
preprocessing the original blood oxygen signal to obtain a first blood oxygen signal;
determining a first time domain characteristic of the first blood oxygen signal; the first time domain feature is used for characterizing whether the blood oxygen saturation of the first blood oxygen signal is reduced or not;
and inputting the first time domain characteristic into a preset respiratory classification model to obtain the target respiratory category to which the original blood oxygen signal belongs.
Optionally, the determining the first time domain characteristic of the first blood oxygen signal comprises:
performing windowing processing on the first blood oxygen signal to obtain n windowed blood oxygen signals;
respectively carrying out segmentation processing on each windowing blood oxygen signal to obtain n multiplied by m segmented blood oxygen signals in total;
determining the blood oxygen saturation corresponding to each segmented blood oxygen signal;
and determining a first time domain characteristic of the first blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals in each of the windowed blood oxygen signals.
Optionally, the first time domain feature comprises at least one of a mean blood oxygen saturation, a standard deviation blood oxygen saturation, a local maximum blood oxygen saturation, a local minimum blood oxygen saturation, a local range of blood oxygen saturation, an average primary fall percentage, an average primary sustained fall time, a secondary fall percentage, and a secondary sustained fall time.
Optionally, the determining a first time domain characteristic of the first blood oxygen signal according to the blood oxygen saturation levels corresponding to the m segmented blood oxygen signals in each of the windowed blood oxygen signals includes:
for m segmented blood oxygen signals in each of the windowed blood oxygen signals, determining a mean value of blood oxygen saturation of the segmented blood oxygen signals according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals;
determining the blood oxygen saturation standard deviation of the windowing blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals and the blood oxygen saturation mean value;
determining the local maximum value of the blood oxygen saturation of the windowing blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals;
determining the local minimum value of the blood oxygen saturation of the windowing blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals;
determining local range of blood oxygen saturation of each of the windowed blood oxygen signals according to each of the local maxima of blood oxygen saturation and each of the local minima of blood oxygen saturation;
determining the average first-level reduction percentage of the windowing blood oxygen signals according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals;
determining the average first-stage continuous falling time of the windowing blood oxygen signals according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals and a first preset constant;
determining a second-level reduction percentage of the windowed blood oxygen signal according to the mean value of the blood oxygen saturation;
and determining the secondary continuous falling time of the windowing blood oxygen signal according to the blood oxygen saturation mean value and a second preset constant.
Optionally, the determining an average primary reduction percentage of the windowed blood oxygen signal according to the blood oxygen saturation levels corresponding to the m segmented blood oxygen signals includes:
determining the average first-level reduction percentage of the windowing blood oxygen signals according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals through the following formula;
wherein the content of the first and second substances,
MD1_ wi represents the average first-order drop percentage of the ith said windowed oximetry signal of the n said windowed oximetry signals; d1_ di represents the percentage of the first order drop of the ith said segment oximetry signal of the n × m said segment oximetry signals; sd represents the blood oxygen saturation corresponding to the segmented blood oxygen signal;
determining the average first-level continuous falling time of the windowing blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals and a first preset constant, wherein the determining comprises the following steps:
determining the average first-stage continuous falling time of the windowing blood oxygen signals according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals and a first preset constant through the following formula;
wherein the content of the first and second substances,
MT1_ wi represents the average first-order duration falling time of the ith one of the n windowed oximetry signals; t1_ di represents the one-level continuous falling time of the ith one of the n × m segmented oximetry signals; a represents the first preset constant;
determining a second level reduction percentage of the windowed blood oxygen signal according to the mean value of blood oxygen saturation, comprising:
determining the second-level reduction percentage of the windowed blood oxygen signal according to the blood oxygen saturation mean value through the following formula;
wherein D2_ wi represents the two-level reduction percentage of the ith windowed oximetry signal of the n windowed oximetry signals; mean _ wi represents the Mean value of the blood oxygen saturation of the ith windowed blood oxygen signal in the n windowed blood oxygen signals;
determining the second-level continuous falling time of the windowed blood oxygen signal according to the mean value of the blood oxygen saturation and a second preset constant, wherein the determining comprises the following steps:
determining the secondary continuous falling time of the windowing blood oxygen signal according to the mean value of the blood oxygen saturation and a second preset constant by the following formula;
wherein T2_ wi represents the average first-level continuous falling time of the ith windowed oximetry signal of the n windowed oximetry signals, and b represents the second predetermined constant.
Optionally, before acquiring the raw blood oxygen signal, the method further includes:
acquiring a plurality of sample blood oxygen signals and a breathing category corresponding to each sample blood oxygen signal;
preprocessing each sample blood oxygen signal respectively to obtain a plurality of second blood oxygen signals;
determining a second time domain characteristic of each of the second blood oxygen signals; the second time domain feature is used for characterizing whether the blood oxygen saturation of the second blood oxygen signal is reduced or not;
constructing an initial respiration classification model;
and training the initial breath classification model by taking the second time domain features and the breath classes corresponding to each group as training parameters to obtain the preset breath classification model.
Optionally, the preset breathing classification model includes a stepwise linear discriminant analysis model, a linear discriminant analysis model, or a support vector machine model.
Optionally, the preset breathing classification model is a stepwise linear discriminant analysis model, and the initial breathing classification model is the stepwise linear discriminant analysis model to be trained; each of the second time-domain features comprises at least two sub-time-domain features;
the training of the initial breath classification model is performed by using the second time domain feature and the breath category corresponding to each group as training parameters to obtain the preset breath classification model, which includes:
inputting the step-by-step linear discriminant analysis model to be trained by taking the second time domain feature and the breathing category corresponding to each group as training parameters;
according to the second time domain features and the breathing categories corresponding to each group, performing significance test on each sub-time domain feature in the second time domain features to obtain sub-time domain features with significance weights exceeding a preset threshold;
and training to obtain the step-by-step linear discriminant analysis model according to the sub-time domain characteristics of which the significance weights exceed a preset threshold.
Optionally, the attribute of the sub-time domain feature of the first time domain feature and the attribute of the sub-time domain feature of the second time domain feature whose significance weight exceeds the preset threshold are the same.
In order to solve the above problem, the present invention also discloses a computer device comprising a processor, a memory and a computer program stored on the memory and operable on the processor, wherein the computer program, when executed by the processor, implements the steps of the signal processing method as described above.
Compared with the prior art, the invention has the following advantages:
in the embodiment of the present invention, the signal processing device may first obtain an original blood oxygen signal, may then perform preprocessing on the original blood oxygen signal to obtain a first blood oxygen signal, and may then determine a first time domain feature of the first blood oxygen signal, where the first time domain feature may characterize whether the blood oxygen saturation of the first blood oxygen signal is decreased, and then the signal processing device may input a preset respiratory classification model by using the first time domain feature as an input parameter, so as to obtain a target respiratory category to which the original blood oxygen signal belongs. In the embodiment of the present invention, the signal processing device may classify the first blood oxygen signal according to a time domain feature that can represent whether the blood oxygen saturation is decreased, and classify the first blood oxygen signal through a preset respiration classification model, so as to determine a respiration type to which the original blood oxygen signal corresponding to the first blood oxygen signal belongs.
Drawings
Fig. 1 is a flow chart of a signal processing method according to a first embodiment of the present invention;
fig. 2 shows a flow chart of a signal processing method according to a second embodiment of the invention;
FIG. 3 is a flow chart illustrating a first-level continuous falling time calculation according to a second embodiment of the present invention;
fig. 4 shows a flowchart of calculating the two-step continuous falling time according to the second embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Example one
Referring to fig. 1, a flowchart illustrating steps of a signal processing method according to a first embodiment of the present invention is shown, where the method includes the following steps:
step 101: a raw blood oxygen signal is acquired.
In the embodiment of the present invention, after the blood oxygen signal detecting device is connected to the human body, the blood oxygen signal detecting device may collect the original blood oxygen signal of the human body within a period of time, and may introduce the collected original blood oxygen signal data into the signal processing device, so that the signal processing device may obtain the original blood oxygen signal. Optionally, the blood oxygen signal may be specifically a pulse blood oxygen signal, which is not specifically limited in this embodiment of the present invention, as long as the blood oxygen content can be measured.
Step 102: the method comprises the steps of preprocessing an original blood oxygen signal to obtain a first blood oxygen signal.
In the embodiment of the present invention, after the signal processing device obtains the original blood oxygen signal, the signal processing device may perform preprocessing for noise reduction, such as filtering, on the original blood oxygen signal, so as to reduce noise interference in the original blood oxygen signal, thereby obtaining the first blood oxygen signal after noise reduction.
Step 103: determining a first time domain characteristic of a first blood oxygen signal; the first time domain characteristic is used for representing whether the blood oxygen saturation of the first blood oxygen signal is reduced or not.
In this step, the signal processing device may extract a first time domain feature of the first blood oxygen saturation signal, where the first time domain feature is some time domain features capable of characterizing whether the blood oxygen saturation of the first blood oxygen saturation signal is decreased. For example, the first time domain feature may include a mean value of blood oxygen saturation, a standard deviation of blood oxygen saturation, a local maximum value of blood oxygen saturation, a local minimum value of blood oxygen saturation, a local range of blood oxygen saturation, and the like, and these time domain features may directly or indirectly indicate whether the blood oxygen saturation of the first blood oxygen signal is decreasing, and the embodiment of the present invention is not limited to the first time domain feature.
Step 104: and inputting the first time domain characteristics into a preset breath classification model to obtain the target breath category to which the original blood oxygen signal belongs.
In this step, a preset breath classification model may be constructed in advance in the signal processing device, and is used to determine the breath category to which the blood oxygen signal belongs according to the time domain feature of the input blood oxygen signal, and output the breath category. The breathing classes may include a normal breathing class and a sleep apnea class, as the case may be. The signal processing device may input the preset respiration classification model with the first time-domain feature as an input parameter, so that the preset respiration classification model may output a target respiration category corresponding to the first time-domain feature, that is, a target respiration category to which the output original blood oxygen signal belongs.
In the embodiment of the present invention, the signal processing device may first obtain an original blood oxygen signal, may then perform preprocessing on the original blood oxygen signal to obtain a first blood oxygen signal, and may then determine a first time domain feature of the first blood oxygen signal, where the first time domain feature may characterize whether the blood oxygen saturation of the first blood oxygen signal is decreased, and then the signal processing device may input a preset respiratory classification model by using the first time domain feature as an input parameter, so as to obtain a target respiratory category to which the original blood oxygen signal belongs. In the embodiment of the present invention, the signal processing device may classify the first blood oxygen signal according to a time domain feature that can represent whether the blood oxygen saturation is decreased, and classify the first blood oxygen signal through a preset respiration classification model, so as to determine a respiration type to which the original blood oxygen signal corresponding to the first blood oxygen signal belongs.
Example two
Referring to fig. 2, a flowchart illustrating steps of a signal processing method according to a second embodiment of the present invention is shown, where the method includes the following steps:
step 201: and obtaining a preset breath classification model through training.
In the embodiment of the present invention, before classifying the respiratory category based on the original blood oxygen signal, the signal processing device may first obtain a preset respiratory classification model through training, and specifically includes the following steps: acquiring a plurality of sample blood oxygen signals and the corresponding breath category of each sample blood oxygen signal; preprocessing each sample blood oxygen signal respectively to obtain a plurality of second blood oxygen signals; determining a second time domain characteristic of each second blood oxygen signal; the second time domain characteristic is used for representing whether the blood oxygen saturation of the second blood oxygen signal is reduced or not; constructing an initial respiration classification model; and training the initial breath classification model by taking the second time domain characteristics and the breath classes corresponding to each group as training parameters to obtain a preset breath classification model.
First, a plurality of sample oximetry signals may be input to the signal processing device, and the breathing category labeled for each sample oximetry signal may be corresponding to each sample oximetry signal. Then, the signal processing device may perform pre-processing for noise reduction, such as filtering, on each sample oximetry signal, respectively, so as to reduce noise interference in each sample oximetry signal, thereby obtaining a plurality of noise-reduced second oximetry signals. Then, the signal processing device may extract a second time domain feature of each second blood oxygen signal, where the second time domain feature is some time domain features capable of characterizing whether the blood oxygen saturation of the second blood oxygen signal is decreased, and has the same attribute as the first time domain feature. The signal processing device may then construct an initial breathing classification model in which the model parameters are to be trained. Then, the signal processing device may construct the second time domain features and the labeled breathing classes corresponding to each group as a training feature matrix, and input the training feature matrix as a training parameter into the initial breathing classification model to perform parameter adjustment, thereby implementing training of the initial breathing classification model. After the training is finished, determining each model parameter in the initial breathing classification model, thereby obtaining the trained preset breathing classification model.
Alternatively, in practical applications, the preset respiration classification model may include a stepwise Linear Discriminant Analysis (SWLDA) model, a Linear Discriminant Analysis (LDA) model, or a Support Vector Machine (SVM) model. The step-by-step linear discriminant analysis model can be combined with two methods, namely LDA (linear discriminant analysis) and bidirectional step-by-step analysis, significance inspection of each dimension is carried out on input features, and finally only the feature combination which contributes most to classification results is reserved to establish a classification model, so that the number of the features can be greatly reduced, and an overfitting phenomenon is avoided. In specific application, the step-by-step linear discriminant analysis model can optimize the model by adjusting the p value of the significant features, the p value of the eliminated significant features and the total number of the significant features, and the three important parameters, so that the classification result of the model is more accurate.
Specifically, under the condition that the preset breath classification model is a step-by-step linear discriminant analysis model, the initial breath classification model is a step-by-step linear discriminant analysis model to be trained, and each second time domain feature comprises at least two sub-time domain features, the initial breath classification model is trained by using the second time domain features and the breath classes corresponding to each group as training parameters, and the step of obtaining the preset breath classification model can be specifically realized by the following method, including:
inputting a step-by-step linear discriminant analysis model to be trained by taking each group of corresponding second time domain features and breathing categories as training parameters; according to the second time domain features and the breathing categories corresponding to each group, performing significance test on each sub-time domain feature in the second time domain features to obtain the sub-time domain features of which the significance weight exceeds a preset threshold; and training to obtain a step-by-step linear discriminant analysis model according to the sub-time domain characteristics with the significance weight exceeding a preset threshold value.
The signal processing device may input the step-by-step linear discriminant analysis model to be trained by using the second time domain feature and the breathing category corresponding to each group as training parameters. Wherein each second time-domain feature may include at least two sub-time-domain features, and the sub-time-domain features may be, for example, a mean value of blood oxygen saturation, a standard deviation of blood oxygen saturation, a local maximum value of blood oxygen saturation, a local minimum value of blood oxygen saturation, a local range of blood oxygen saturation, an average primary falling percentage, an average primary sustained falling time, a secondary falling percentage, a secondary sustained falling time, and the like. The step-by-step linear discriminant analysis model to be trained can combine two modes of LDA and bidirectional step-by-step analysis, and based on the parameter adjusting result of each group of corresponding second time domain features and breathing categories on model parameters, the influence ratio of each sub-time domain feature on the classification result is determined, so that the significance test can be performed on each sub-time domain feature. The higher the significance weight is, the more significant the influence on the accuracy of the classification result is. And then the signal processing equipment can train to obtain a stepwise linear discriminant analysis model according to the sub-time domain features of which the significance weights exceed the preset threshold, namely one or more sub-time domain features before the maximum influence on the accuracy of the classification result.
In the embodiment of the invention, the attribute of the sub-time domain feature with the significance weight exceeding the preset threshold in the first time domain feature and the second time domain feature is the same. Because the step-by-step linear discriminant analysis model can be obtained by training the sub-time domain features which have obvious influence on the accuracy of the classification result in the second time domain feature, when the original blood oxygen signal is classified by the trained preset respiratory classification model, only the time domain features which have the same attributes as the sub-time domain features which have obvious influence can be extracted from the original blood oxygen signal, and the time domain features which have the same attributes as all the sub-time domain features in the second time domain feature do not need to be extracted from the original blood oxygen signal, so that the number of the time domain features extracted from the original blood oxygen signal can be reduced, and the time domain feature calculation amount of the original blood oxygen signal is further reduced.
For example, the second time-domain feature may include the following 9 sub-time-domain features: mean blood oxygen saturation, standard deviation blood oxygen saturation, local maximum blood oxygen saturation, local minimum blood oxygen saturation, local range of blood oxygen saturation, average primary falling percentage, average primary sustained falling time, secondary falling percentage, and secondary sustained falling time. During model training, it can be determined that the number of the sub-time domain features having significant influence on the accuracy of the classification result in the second time domain feature is 6, and the sub-time domain features are the blood oxygen saturation mean value, the local range of the blood oxygen saturation, the average primary descending percentage, the average primary continuous descending time, the secondary descending percentage and the secondary continuous descending time. The signal processing device can train to obtain a step-by-step linear discriminant analysis model according to the 6 sub-time domain features with obvious influence. Therefore, when the first time domain feature needing to be input is extracted from the original blood oxygen signal subsequently, only 6 time domain features of the blood oxygen saturation mean value, the local range of the blood oxygen saturation, the average primary falling percentage, the average primary continuous falling time, the secondary falling percentage and the secondary continuous falling time of the original blood oxygen signal can be extracted as input parameters of the step-by-step linear discriminant analysis model, wherein the attributes of the 6 time domain features are the same as those of the 6 sub-time domain features which have significant influence on the accuracy of the classification result in the second time domain feature, so that the number of the time domain features extracted from the original blood oxygen signal is reduced, the blood oxygen saturation standard deviation, the blood oxygen saturation local maximum and the blood oxygen saturation local minimum of the original blood oxygen signal do not need to be calculated, and the time domain feature calculation amount of the original blood oxygen signal is reduced.
After the trained preset breath classification model is obtained, the signal processing device can acquire a new original blood oxygen signal, and then the classification of the breath category is carried out through the trained preset breath classification model according to the time domain characteristics corresponding to the new original blood oxygen signal.
Step 202: a raw blood oxygen signal is acquired.
The specific implementation manner of this step may refer to the implementation process of step 101 described above, and this embodiment is not described in detail here.
Step 203: the method comprises the steps of preprocessing an original blood oxygen signal to obtain a first blood oxygen signal.
The specific implementation manner of this step may refer to the implementation process of step 102 described above, and this embodiment is not described in detail here.
Step 204: and performing windowing processing on the first blood oxygen signal to obtain n windowed blood oxygen signals.
In the embodiment of the present invention, the signal processing device may perform windowing on the preprocessed first blood oxygen signal according to a preset window length, so as to obtain n pieces of windowed blood oxygen signals, where a time duration corresponding to each windowed blood oxygen signal is the preset window length. For example, the preset window length may be 60s, and then the first blood oxygen signal may be windowed every 60s to obtain n 60s windowed blood oxygen signals.
Step 205: and respectively carrying out segmentation processing on each windowing blood oxygen signal to obtain n multiplied by m segmented blood oxygen signals in total.
In the embodiment of the present invention, the signal processing device may perform segmentation processing on each of the segmented window blood oxygen signals according to a preset segment length, where each of the segmented window blood oxygen signals may be segmented into m segmented blood oxygen signals, and n × m segmented blood oxygen signals may be obtained in total, where a time duration corresponding to each of the segmented blood oxygen signals is the preset segment length. For example, the preset segment length may be 10s, and then a 60s windowed oximetry signal may be segmented every 10s, and each windowed oximetry signal may be segmented into 6 10s segmented oximetry signals, for a total of 6n segmented oximetry signals.
Step 206: and determining the blood oxygen saturation corresponding to each segmented blood oxygen signal.
In this step, the signal processing device may determine the blood oxygen saturation corresponding to each segmented blood oxygen signal. Optionally, the partial pressure of blood oxygen may be measured based on an electrochemical principle, and then the oxygen saturation level may be calculated according to the partial pressure of blood oxygen, which may refer to related technologies.
Step 207: and determining a first time domain characteristic of the first blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals in each windowing blood oxygen signal.
Optionally, the first time domain feature may include at least one of a mean blood oxygen saturation, a standard deviation blood oxygen saturation, a local maximum blood oxygen saturation, a local minimum blood oxygen saturation, a local range of blood oxygen saturation, an average primary fall percentage, an average primary sustained fall time, a secondary fall percentage, and a secondary sustained fall time.
Correspondingly, the step can be specifically realized by the following modes, including:
substep 2071: and for m segmented blood oxygen signals in each windowing blood oxygen signal, determining the mean value of the blood oxygen saturation of the windowing blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals.
Taking a 60s windowed blood oxygen saturation signal, a 10s segmented blood oxygen saturation signal and m ═ 6 as examples, under the condition of considering the windowing limit, the blood oxygen saturation levels corresponding to the 6n segmented blood oxygen saturation signals can be represented as Sw1d1, Sw1d2, Sw1d3, Sw1d4, Sw1d5, Sw1d6, Sw2d1, Sw2d2, Sw2d3, Sw2d4, Sw2d5, Sw2d6, …, Swnd5 and Swnd6, and can be recorded as a secondary blood oxygen saturation level sequence. Without considering the window boundary, the blood oxygen saturation levels corresponding to the 6n segmented blood oxygen signals can be represented as Sd1, Sd2, Sd3, …, Sd6n, which can be recorded as a primary blood oxygen saturation sequence. The values of the primary blood oxygen saturation sequence and the secondary blood oxygen saturation sequence are the same, but the representation modes are different, the representation mode of the secondary blood oxygen saturation sequence can embody a windowed blood oxygen signal to which the segmented blood oxygen signal belongs, the representation mode of the primary blood oxygen saturation sequence does not show a windowed boundary, and the primary blood oxygen saturation sequence can be named according to requirements in specific applications, which is not limited in the embodiment of the invention.
Optionally, the m segmented blood oxygen signals in a certain windowed blood oxygen signal are used for illustration, and the signal processing device may determine the mean value of the blood oxygen saturation of the windowed blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals in the windowed blood oxygen signal through the following formula (1). In the following formula (1), Mean _ wi represents the Mean value of the blood oxygen saturation of the ith windowed blood oxygen signal in the n windowed blood oxygen signals, and Swidj represents the blood oxygen saturation corresponding to the jth segmented blood oxygen signal in the ith windowed blood oxygen signal.
For example, when m is 6, the formula (1) can be embodied as the following formula (1-1).
Substep 2072: and determining the blood oxygen saturation standard deviation of the windowing blood oxygen signal according to the blood oxygen saturation and the blood oxygen saturation mean value corresponding to the m segmented blood oxygen signals.
Optionally, the m segmented blood oxygen signals in a certain windowed blood oxygen signal are used for illustration, and the signal processing device may determine the blood oxygen saturation standard deviation of the windowed blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals in the windowed blood oxygen signal and the blood oxygen saturation mean value of the windowed blood oxygen signal through the following formula (2). Wherein, in the following formula (2), Sd _ wi represents the standard deviation of blood oxygen saturation of the ith windowed blood oxygen signal.
For example, when m is 6, the formula (2) can be embodied as the following formula (2-1).
Substep 2073: and determining the local maximum value of the blood oxygen saturation of the windowing blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals.
Optionally, the m segmented blood oxygen signals in a certain windowed blood oxygen signal are used for illustration, and the signal processing device may determine the local maximum value of the blood oxygen saturation of the windowed blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals in the windowed blood oxygen signal by the following formula (3). In the following formula (3), Max _ wi represents the local maximum value of blood oxygen saturation of the ith windowed blood oxygen signal.
Max_wi＝max{Swidj},(j＝1,2,3,…,m,i＝1,2,3,…,n) (3)
For example, when m is 6, the formula (3) can be embodied as the following formula (3-1).
Max_wi＝max{Swid1,Swid2,Swid3,Swid4,Swid5,Swid6},(j＝1,2,3,…,m,i＝1,2,3,…,n)
(3-1)
Substep 2074: and determining the local minimum value of the blood oxygen saturation of the windowing blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals.
Optionally, the m segmented blood oxygen signals in a certain windowed blood oxygen signal are used for illustration, and the signal processing device may determine the local minimum value of the blood oxygen saturation of the windowed blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals in the windowed blood oxygen signal through the following formula (4). In the following formula (4), Min _ wi represents the local minimum of blood oxygen saturation of the ith windowed blood oxygen signal.
Min_wi＝min{Swidj},(j＝1,2,3,…,m,i＝1,2,3,…,n) (4)
For example, when m is 6, the formula (4) can be embodied as the following formula (4-1).
Min_wi＝min{Swid1,Swid2,Swid3,Swid4,Swid5,Swid6},(j＝1,2,3,…,m,i＝1,2,3,…,n)
(4-1)
Substep 2075: and determining local range of blood oxygen saturation of each windowed blood oxygen saturation signal according to each local maximum of blood oxygen saturation and each local minimum of blood oxygen saturation.
Alternatively, describing a certain windowed oximetry signal, the signal processing device may determine the local range of oximetry of the windowed oximetry signal according to the local maximum value and the local minimum value of oximetry of the windowed oximetry signal by the following formula (5). Wherein, in the following formula (5), R _ wi represents the local range of blood oxygen saturation of the ith windowed blood oxygen signal.
R_wi＝Max_wi-Min_wi,(i＝1,2,3,…,n) (5)
Substep 2076: and determining the average first-level reduction percentage of the windowed blood oxygen signals according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals.
Alternatively, the signal processing device may determine the percentage of one-step decrease of each segment oximetry signal according to the corresponding blood oxygen saturation of the n × m segment oximetry signals by the following formula (6). Wherein, in the following formula (6), D1_ di represents the first order drop percentage of the ith segmented blood oxygen signal in the n × m segmented blood oxygen signals.
For example, when m is 6, the formula (6) can be embodied as the following formula (6-1).
Then, taking m segmented blood oxygen signals in a certain windowed blood oxygen signal for illustration, the signal processing device may determine the average primary reduction percentage of the windowed blood oxygen signal according to the primary reduction percentage of the m segmented blood oxygen signals in the windowed blood oxygen signal by the following formula (7). In the following formula (7), MD1_ wi represents the average first-order drop percentage of the ith windowed oximetry signal in the n windowed oximetry signals.
For example, when m is 6, the formula (7) can be embodied as the following formula (7-1).
Substep 2077: and determining the average first-stage continuous falling time of the windowing blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals and a first preset constant.
Alternatively, the signal processing device may determine the percentage of the first-order drop of each of the segmented blood oxygen signals according to the corresponding blood oxygen saturation levels of the n × m segmented blood oxygen signals and a first preset constant by the following formula (8). In the following formula (8), T1_ di represents the one-step continuous falling time of the ith segmented oximetry signal in the n × m segmented oximetry signals, and a represents a first predetermined constant equal to the predetermined segment length. The calculation process represented by the formula (8) can refer to fig. 3.
For example, when m is 6 and the preset segment length is 10s, the formula (8) may be embodied as the following formula (8-1).
Then, taking m segment oximetry signals in a certain windowed oximetry signal for illustration, the signal processing device may determine the average primary continuous falling time of the windowed oximetry signal according to the primary falling percentage of the m segment oximetry signals in the windowed oximetry signal by the following formula (9). Wherein, in the following formula (9), MT1_ wi represents the average first-order continuous falling time of the ith windowed oximetry signal of the n windowed oximetry signals.
For example, when m is 6, the formula (9) can be embodied as the following formula (9-1).
Substep 2078: and determining the second-level reduction percentage of the windowed blood oxygen signal according to the mean value of the blood oxygen saturation.
Alternatively, the signal processing device may determine the second-order reduction percentage of each windowed blood oxygen signal according to the following formula (10) according to the blood oxygen saturation mean value corresponding to each windowed blood oxygen signal. In the following formula (10), D2_ wi represents the two-step reduction percentage of the ith windowed oximetry signal in the n windowed oximetry signals.
Substep 2079: and determining the secondary continuous falling time of the windowed blood oxygen signal according to the mean value of the blood oxygen saturation and a second preset constant.
Alternatively, the signal processing device may determine the secondary continuous falling time of each of the windowed blood oxygen signals according to the average value of the blood oxygen saturation of each of the windowed blood oxygen signals and a second preset constant by the following formula (11). In the following formula (11), T2_ wi represents the average first-level continuous falling time of the ith windowed oximetry signal in the n windowed oximetry signals, and b represents a second predetermined constant equal to the predetermined window length. The calculation process represented by equation (11) may refer to fig. 4.
For example, when the preset window length is 60s, the equation (8) may be embodied as the following equation (8-1).
It should be noted that, in practical application, the preset breath classification model may determine, from the time domain features of each dimension, a time domain feature dimension that significantly affects the classification result during training, and therefore, in practical application, the first time domain feature of the same dimension may be extracted according to the time domain feature dimension that significantly affects the classification result and is determined by the preset breath classification model. For example, the time domain feature dimensions determined by the preset respiratory classification model and having a significant influence on the classification result are the average primary reduction percentage, the average primary continuous reduction time, the secondary reduction percentage and the secondary continuous reduction time, and then the signal processing device extracts the time domain features of the average primary reduction percentage, the average primary continuous reduction time, the secondary reduction percentage and the secondary continuous reduction time from the first blood oxygen signal.
In addition, in the embodiment of the present invention, only the features of the first blood oxygen signal in the time domain may be extracted, and the features in the frequency domain, the spectrum, the nonlinearity, and the like do not need to be extracted, so that the complexity of the feature extraction and the operation of the classification may be low.
Step 208: and inputting the first time domain characteristics into a preset breath classification model to obtain the target breath category to which the original blood oxygen signal belongs.
In the embodiment of the present invention, the signal processing device may construct the first time domain feature as a first feature matrix, and input the first feature matrix as an input parameter into the trained preset respiration classification model, so that the preset respiration classification model may output a target respiration class corresponding to the first time domain feature, that is, a target respiration class to which the original blood oxygen signal belongs, thereby determining whether the original blood oxygen signal belongs to a normal respiration class or a sleep apnea class.
In addition, in practical application, optionally, different preset breathing classification models may be trained for different persons to whom blood oxygen signals belong, so that when sleep breathing types are classified, the preset breathing classification models corresponding to the current persons to whom blood oxygen signals belong are adopted for classification, and thus classification accuracy can be improved. Of course, optionally, the same preset breathing classification model may also be used for the blood oxygen signals of different blood oxygen signals, that is, the preset breathing classification model may be universal, so that the training time of the model may be saved.
In the embodiment of the invention, the signal processing device can firstly obtain the preset breath classification model through training, then, the original blood oxygen signal can be obtained and preprocessed to obtain the first blood oxygen signal, and then, the signal processing device may perform windowing and segmentation processing on the first blood oxygen signal to obtain n windowed blood oxygen signals, each windowed blood oxygen signal including m segmented blood oxygen signals, and determines the first time domain characteristic of the first blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals in each windowing blood oxygen signal, the first time domain feature can represent whether the blood oxygen saturation of the first blood oxygen signal is reduced or not, and then the signal processing equipment inputs the first time domain feature into a preset respiration classification model by taking the first time domain feature as an input parameter, so that the target respiration category of the original blood oxygen signal can be obtained. In the embodiment of the invention, the signal processing device can extract the time domain characteristics of the first blood oxygen signal in a windowing and segmenting manner so as to obtain the multi-dimensional and multi-quantity time domain characteristics, and classify the first blood oxygen signal through the preset respiration classification model according to the time domain characteristics which can represent whether the blood oxygen saturation degree is reduced or not in the first blood oxygen signal so as to determine the respiration type of the original blood oxygen signal corresponding to the first blood oxygen signal.
An embodiment of the present invention further provides a computer device, which includes a processor, a memory, and a computer program stored in the memory and executable on the processor, and when the computer program is executed by the processor, the steps of the signal processing method described above are implemented.
While, for purposes of simplicity of explanation, the foregoing method embodiments have been described as a series of acts or combination of acts, it will be appreciated by those skilled in the art that the present invention is not limited by the illustrated ordering of acts, as some steps may occur in other orders or concurrently with other steps in accordance with the invention. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required by the invention.
The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
Finally, it should also be 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. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The foregoing detailed description of the signal processing method and computer device provided by the present invention has been presented, and the principles and embodiments of the present invention have been explained by applying specific examples, and the descriptions of the above embodiments are only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A method of signal processing, the method comprising:
acquiring an original blood oxygen signal;
preprocessing the original blood oxygen signal to obtain a first blood oxygen signal;
determining a first time domain characteristic of the first blood oxygen signal; the first time domain feature is used for characterizing whether the blood oxygen saturation of the first blood oxygen signal is reduced or not;
and inputting the first time domain characteristic into a preset respiratory classification model to obtain the target respiratory category to which the original blood oxygen signal belongs.
2. The method of claim 1, wherein said determining a first time domain characteristic of said first blood oxygen signal comprises:
performing windowing processing on the first blood oxygen signal to obtain n windowed blood oxygen signals;
respectively carrying out segmentation processing on each windowing blood oxygen signal to obtain n multiplied by m segmented blood oxygen signals in total;
determining the blood oxygen saturation corresponding to each segmented blood oxygen signal;
and determining a first time domain characteristic of the first blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals in each of the windowed blood oxygen signals.
3. The method of claim 2, wherein the first time domain characteristic comprises at least one of a mean blood oxygen saturation, a standard deviation blood oxygen saturation, a local maximum blood oxygen saturation, a local minimum blood oxygen saturation, a local range of blood oxygen saturation, an average primary drop percentage, an average primary sustained drop time, a secondary drop percentage, and a secondary sustained drop time.
4. The method of claim 3, wherein said determining a first time domain characteristic of said first blood oxygen signal according to the blood oxygen saturation levels corresponding to m segment blood oxygen signals in each of said windowed blood oxygen signals comprises:
for m segmented blood oxygen signals in each of the windowed blood oxygen signals, determining a mean value of blood oxygen saturation of the segmented blood oxygen signals according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals;
determining the blood oxygen saturation standard deviation of the windowing blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals and the blood oxygen saturation mean value;
determining the local maximum value of the blood oxygen saturation of the windowing blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals;
determining the local minimum value of the blood oxygen saturation of the windowing blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals;
determining local range of blood oxygen saturation of each of the windowed blood oxygen signals according to each of the local maxima of blood oxygen saturation and each of the local minima of blood oxygen saturation;
determining the average first-level reduction percentage of the windowing blood oxygen signals according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals;
determining the average first-stage continuous falling time of the windowing blood oxygen signals according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals and a first preset constant;
determining a second-level reduction percentage of the windowed blood oxygen signal according to the mean value of the blood oxygen saturation;
and determining the secondary continuous falling time of the windowing blood oxygen signal according to the blood oxygen saturation mean value and a second preset constant.
5. The method of claim 4, wherein said determining an average first-order reduction percentage of said windowed blood oxygen signal based on the blood oxygen saturation levels corresponding to said m segmented blood oxygen signals comprises:
determining the average first-level reduction percentage of the windowing blood oxygen signals according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals through the following formula;
wherein the content of the first and second substances,
MD1_ wi represents the average first-order drop percentage of the ith said windowed oximetry signal of the n said windowed oximetry signals; d1_ di represents the percentage of the first order drop of the ith said segment oximetry signal of the n × m said segment oximetry signals; sd represents the blood oxygen saturation corresponding to the segmented blood oxygen signal;
determining the average first-level continuous falling time of the windowing blood oxygen signal according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals and a first preset constant, wherein the determining comprises the following steps:
determining the average first-stage continuous falling time of the windowing blood oxygen signals according to the blood oxygen saturation corresponding to the m segmented blood oxygen signals and a first preset constant through the following formula;
wherein the content of the first and second substances,
MT1_ wi represents the average first-order duration falling time of the ith one of the n windowed oximetry signals; t1_ di represents the one-level continuous falling time of the ith one of the n × m segmented oximetry signals; a represents the first preset constant;
determining a second level reduction percentage of the windowed blood oxygen signal according to the mean value of blood oxygen saturation, comprising:
determining the second-level reduction percentage of the windowed blood oxygen signal according to the blood oxygen saturation mean value through the following formula;
wherein D2_ wi represents the two-level reduction percentage of the ith windowed oximetry signal of the n windowed oximetry signals; mean _ wi represents the Mean value of the blood oxygen saturation of the ith windowed blood oxygen signal in the n windowed blood oxygen signals;
determining the second-level continuous falling time of the windowed blood oxygen signal according to the mean value of the blood oxygen saturation and a second preset constant, wherein the determining comprises the following steps:
determining the secondary continuous falling time of the windowing blood oxygen signal according to the mean value of the blood oxygen saturation and a second preset constant by the following formula;
wherein T2_ wi represents the average first-level continuous falling time of the ith windowed oximetry signal of the n windowed oximetry signals, and b represents the second predetermined constant.
6. The method of claim 1, wherein said obtaining a raw blood oxygen signal further comprises:
acquiring a plurality of sample blood oxygen signals and a breathing category corresponding to each sample blood oxygen signal;
preprocessing each sample blood oxygen signal respectively to obtain a plurality of second blood oxygen signals;
determining a second time domain characteristic of each of the second blood oxygen signals; the second time domain feature is used for characterizing whether the blood oxygen saturation of the second blood oxygen signal is reduced or not;
constructing an initial respiration classification model;
and training the initial breath classification model by taking the second time domain features and the breath classes corresponding to each group as training parameters to obtain the preset breath classification model.
7. The method of claim 1, wherein the preset breathing classification model comprises a stepwise linear discriminant analysis model, a linear discriminant analysis model, or a support vector machine model.
8. The method of claim 7, wherein the preset breathing classification model is a stepwise linear discriminant analysis model, and the initial breathing classification model is the stepwise linear discriminant analysis model to be trained; each of the second time-domain features comprises at least two sub-time-domain features;
the training of the initial breath classification model is performed by using the second time domain feature and the breath category corresponding to each group as training parameters to obtain the preset breath classification model, which includes:
inputting the step-by-step linear discriminant analysis model to be trained by taking the second time domain feature and the breathing category corresponding to each group as training parameters;
according to the second time domain features and the breathing categories corresponding to each group, performing significance test on each sub-time domain feature in the second time domain features to obtain sub-time domain features with significance weights exceeding a preset threshold;
and training to obtain the step-by-step linear discriminant analysis model according to the sub-time domain characteristics of which the significance weights exceed a preset threshold.
9. The method of claim 8, wherein the attribute of the sub-temporal feature of the first temporal feature and the second temporal feature whose significance weight exceeds the preset threshold is the same.
10. A computer arrangement comprising a processor, a memory and a computer program stored on the memory and executable on the processor, the computer program, when executed by the processor, implementing the steps of the signal processing method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911055292.8A CN110742621B (en) | 2019-10-31 | 2019-10-31 | Signal processing method and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911055292.8A CN110742621B (en) | 2019-10-31 | 2019-10-31 | Signal processing method and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110742621A true CN110742621A (en) | 2020-02-04 |
| CN110742621B CN110742621B (en) | 2022-09-13 |
Family
ID=69281651
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911055292.8A Active CN110742621B (en) | 2019-10-31 | 2019-10-31 | Signal processing method and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110742621B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111297327A (en) * | 2020-02-20 | 2020-06-19 | 京东方科技集团股份有限公司 | Sleep analysis method, system, electronic equipment and storage medium |
| CN114795133A (en) * | 2022-06-29 | 2022-07-29 | 华南师范大学 | Sleep apnea detection method, device, equipment and storage medium |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6450957B1 (en) * | 1999-08-02 | 2002-09-17 | Denso Corporation | Respiratory disease monitoring system |
| US20110006901A1 (en) * | 2009-07-09 | 2011-01-13 | Cassidy Harry J | Breathing disorder treatment system and method |
| US20110208539A1 (en) * | 2001-05-17 | 2011-08-25 | Lynn Lawrence A | System and method for automatic detection of a plurality of sp02 time series pattern types |
| CN102458245A (en) * | 2009-04-20 | 2012-05-16 | 雷斯梅德有限公司 | Discrimination of cheyne -stokes breathing patterns by use of oximetry signals |
| US20120220846A1 (en) * | 2008-03-27 | 2012-08-30 | Nellcor Puritan Bennett Llc | System and method for diagnosing sleep apnea |
| WO2014100084A1 (en) * | 2012-12-18 | 2014-06-26 | Covidien Lp | Systems and methods for distinguishing between central apnea and obstructive apnea |
| US20150031969A1 (en) * | 2012-09-14 | 2015-01-29 | Mohammad Khair | System and method for monitoring arterial and venous blood oxygen, blood glucose, and blood constituent concentration |
| CN104545818A (en) * | 2015-01-29 | 2015-04-29 | 吉林大学 | Sleep apnea syndrome detection method based on pulse and blood oxygen signals |
| US20160120464A1 (en) * | 2004-12-23 | 2016-05-05 | Resmed Limited | Discrimination of cheyne-stokes breathing patterns by use of oximetry signals |
| CN106137219A (en) * | 2016-08-04 | 2016-11-23 | 天津大学 | The absolute difference of dual wavelength adds and calculates arterial oxygen saturation method and device thereof |
| WO2017168029A1 (en) * | 2016-04-01 | 2017-10-05 | Consejo Superior De Investigaciones Científicas (Csic) | Non-invasive sensor, device, system and method for determining training parametres during the performance of physical exercises |
| CN108283489A (en) * | 2017-12-16 | 2018-07-17 | 湖南明康中锦医疗科技发展有限公司 | Sleep-respiratory system and method |
| CN109044302A (en) * | 2018-07-03 | 2018-12-21 | 京东方科技集团股份有限公司 | Measure device, electronic equipment and the computer readable storage medium of blood pressure |
| CN109259733A (en) * | 2018-10-25 | 2019-01-25 | 深圳和而泰智能控制股份有限公司 | Apnea detection method, apparatus and detection device in a kind of sleep |
| CN109528216A (en) * | 2019-01-18 | 2019-03-29 | 京东方科技集团股份有限公司 | The detection method and device of fetal hemoglobin saturation |
| CN109965848A (en) * | 2019-04-18 | 2019-07-05 | 天津大学 | A kind of detection method of the sleep apnea syndrome based on blood oxygen signal classification |
| CN110302460A (en) * | 2019-08-09 | 2019-10-08 | 丹阳慧创医疗设备有限公司 | A kind of attention training method, device, equipment and system |
-
2019
- 2019-10-31 CN CN201911055292.8A patent/CN110742621B/en active Active
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6450957B1 (en) * | 1999-08-02 | 2002-09-17 | Denso Corporation | Respiratory disease monitoring system |
| US20110208539A1 (en) * | 2001-05-17 | 2011-08-25 | Lynn Lawrence A | System and method for automatic detection of a plurality of sp02 time series pattern types |
| US20160120464A1 (en) * | 2004-12-23 | 2016-05-05 | Resmed Limited | Discrimination of cheyne-stokes breathing patterns by use of oximetry signals |
| US20120220846A1 (en) * | 2008-03-27 | 2012-08-30 | Nellcor Puritan Bennett Llc | System and method for diagnosing sleep apnea |
| CN104757975A (en) * | 2009-04-20 | 2015-07-08 | 瑞思迈有限公司 | Discrimination of cheyne -stokes breathing patterns by use of oximetry signals |
| NZ717439A (en) * | 2009-04-20 | 2017-09-29 | Resmed Ltd | Discrimination of cheyne-stokes breathing patterns by use of oximetry signals |
| CN102458245A (en) * | 2009-04-20 | 2012-05-16 | 雷斯梅德有限公司 | Discrimination of cheyne -stokes breathing patterns by use of oximetry signals |
| US20110006901A1 (en) * | 2009-07-09 | 2011-01-13 | Cassidy Harry J | Breathing disorder treatment system and method |
| US20150031969A1 (en) * | 2012-09-14 | 2015-01-29 | Mohammad Khair | System and method for monitoring arterial and venous blood oxygen, blood glucose, and blood constituent concentration |
| WO2014100084A1 (en) * | 2012-12-18 | 2014-06-26 | Covidien Lp | Systems and methods for distinguishing between central apnea and obstructive apnea |
| CN104545818A (en) * | 2015-01-29 | 2015-04-29 | 吉林大学 | Sleep apnea syndrome detection method based on pulse and blood oxygen signals |
| WO2017168029A1 (en) * | 2016-04-01 | 2017-10-05 | Consejo Superior De Investigaciones Científicas (Csic) | Non-invasive sensor, device, system and method for determining training parametres during the performance of physical exercises |
| CN106137219A (en) * | 2016-08-04 | 2016-11-23 | 天津大学 | The absolute difference of dual wavelength adds and calculates arterial oxygen saturation method and device thereof |
| CN108283489A (en) * | 2017-12-16 | 2018-07-17 | 湖南明康中锦医疗科技发展有限公司 | Sleep-respiratory system and method |
| CN109044302A (en) * | 2018-07-03 | 2018-12-21 | 京东方科技集团股份有限公司 | Measure device, electronic equipment and the computer readable storage medium of blood pressure |
| CN109259733A (en) * | 2018-10-25 | 2019-01-25 | 深圳和而泰智能控制股份有限公司 | Apnea detection method, apparatus and detection device in a kind of sleep |
| CN109528216A (en) * | 2019-01-18 | 2019-03-29 | 京东方科技集团股份有限公司 | The detection method and device of fetal hemoglobin saturation |
| CN109965848A (en) * | 2019-04-18 | 2019-07-05 | 天津大学 | A kind of detection method of the sleep apnea syndrome based on blood oxygen signal classification |
| CN110302460A (en) * | 2019-08-09 | 2019-10-08 | 丹阳慧创医疗设备有限公司 | A kind of attention training method, device, equipment and system |
Non-Patent Citations (2)
| Title |
|---|
| 彭嘉鹏: ""一种多参数便携式睡眠监测仪的研制"", 《中国优秀硕士学位论文全文数据库工程科技辑》 * |
| 闫佳运: ""可穿戴式睡眠监护设备中若干关键技术研究"", 《中国优秀硕士学位论文全文数据库医药卫生科技辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111297327A (en) * | 2020-02-20 | 2020-06-19 | 京东方科技集团股份有限公司 | Sleep analysis method, system, electronic equipment and storage medium |
| CN111297327B (en) * | 2020-02-20 | 2023-12-01 | 京东方科技集团股份有限公司 | Sleep analysis method, system, electronic equipment and storage medium |
| CN114795133A (en) * | 2022-06-29 | 2022-07-29 | 华南师范大学 | Sleep apnea detection method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110742621B (en) | 2022-09-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Mendonca et al. | A review of obstructive sleep apnea detection approaches | |
| Karar et al. | Automated diagnosis of heart sounds using rule-based classification tree | |
| CN103730130B (en) | A kind of detection system of pathological voice | |
| Dafna et al. | Sleep-wake evaluation from whole-night non-contact audio recordings of breathing sounds | |
| CN105448291A (en) | Parkinsonism detection method and detection system based on voice | |
| CN110742621B (en) | Signal processing method and computer equipment | |
| CN107945817A (en) | Heart and lung sounds signal sorting technique, detection method, device, medium and computer equipment | |
| CN111920420B (en) | Patient behavior multi-modal analysis and prediction system based on statistical learning | |
| CN110570880A (en) | Snore signal identification method | |
| US20200365271A1 (en) | Method for predicting sleep apnea from neural networks | |
| Orlandi et al. | Effective pre-processing of long term noisy audio recordings: An aid to clinical monitoring | |
| WO2020077413A1 (en) | A method and apparatus for diagnosis of maladies from patient sounds | |
| DE102015218948A1 (en) | Apparatus and method for determining a medical health parameter of a subject by means of voice analysis | |
| Dafna et al. | OSA severity assessment based on sleep breathing analysis using ambient microphone | |
| Chitkara et al. | Voice based detection of type 2 diabetes mellitus | |
| Razi et al. | Sleep apnea classification using random forest via ECG | |
| EP3581098B1 (en) | Method of generating respiratory status classifier and method of determining respiratory status | |
| Usman et al. | Heart rate detection and classification from speech spectral features using machine learning | |
| CN111938650A (en) | Method and device for monitoring sleep apnea | |
| CN105962897A (en) | Self-adaptive snoring sound signal detection method | |
| KR102076759B1 (en) | Multi-sensor based noncontact sleep monitoring method and apparatus using ensemble of deep neural network and random forest | |
| CN111759277B (en) | Detection device, method, equipment and storage medium for sleep apnea hypopnea | |
| CN111227793B (en) | Apnea recognition method and system, electronic equipment and storage medium | |
| Ghaemmaghami et al. | Normal probability testing of snore signals for diagnosis of obstructive sleep apnea | |
| CN111685731A (en) | Sleep data processing method, device, equipment and storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |