CN104545818A - Sleep apnea syndrome detection method based on pulse and blood oxygen signals - Google Patents
Sleep apnea syndrome detection method based on pulse and blood oxygen signals Download PDFInfo
- Publication number
- CN104545818A CN104545818A CN201510046672.0A CN201510046672A CN104545818A CN 104545818 A CN104545818 A CN 104545818A CN 201510046672 A CN201510046672 A CN 201510046672A CN 104545818 A CN104545818 A CN 104545818A
- Authority
- CN
- China
- Prior art keywords
- pulse
- blood oxygen
- sleep apnea
- apnea syndrome
- oxygen signal
- 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
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
-
- 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
Abstract
The invention relates to a sleep apnea syndrome detection method based on pulse and blood oxygen signals. The method includes: digging characteristic parameters related to SAS (statistical analysis system) from the pulse and blood oxygen signals by means of probability principal component analysis, setting up an SAS detection characteristic matrix, adopting a pouch decision tree for mapping the relation between the characteristic matrix and SAS disease degrees, and setting up an SAS detection model. By means of probability principal component analysis, input data dimensionality is decreased while quality of the characteristic matrix is improved; owing to the pouch decision tree, SAS detection accuracy is improved; by utilization of the sleep apnea syndrome detection method based on the pulse and blood oxygen signals, the SAS disease degrees of testees can be outputted only by inputting the pulse and blood oxygen signals of the testees.
Description
Technical field:
The present invention relates to a kind of sleep apnea syndrome detection method based on pulse and blood oxygen signal, particularly relate to Data Dimensionality Reduction technology and the application of supervised learning technology in sleep apnea syndrome classification.
Background technology:
Sleep apnea syndrome (SAS) has become the stealthy killer jeopardizing human life, is only patient populations's 20% in developed country through the SAS patient of diagnosis and treatment at present according to statistics, and China is then only 2 ‰.Trace it to its cause and be mainly standard detecting method sleep and lead figure PSG flow process complexity, somewhat expensive more, be difficult to popularization and application.Therefore, the detection method that the convenience of research substituting PS G, accuracy rate are high provides strong technical support by for SAS patient's early detection and diagnosis, has higher practical value.
Inventive point or the innovative point of current detection sleep apnea syndrome SAS mainly concentrate on: 1) reduce the physiological parameter that PSG measures, measure the major parameter relevant with sleep apnea.CN103202687A discloses a kind of Obstructive sleep apnea hypopnea syndrome just screen system, by gathering mouth and nose air-flow, thorax abdomen motor message, blood oxygen heart rate signal, use video module shooting sleeping position and mouth and nose to go out close-up shot, complete the detection to sleep apnea situation; CN202776304U discloses a kind of device analyzing sleep apnea event, it is characterized in that by mouth and nose pneumatic sensor, breast abdominal respiration motion sensor, body position sensor, lower limb dynamic sensor and sound of snoring sensor detect multiple physiological signal, differentiate the generation of sleep-respiratory event; CN203468609U discloses a kind of monitoring and therapy system of sleep apnea syndrome, is detected, judge the generation of sleep apnea with this by sound of snoring signal, respiratory air flow, blood oxygen saturation, thorax abdomen motor message.Although above method is fewer than the monitoring parameter of PSG, but still need testee to wear multiple sensors, affect the sleep state of testee.
2) use less monitor signal or adopt non-contact measurement mode, simple to operate, little on user sleep impact.CN102138795A discloses a kind of method determining the order of severity of obstructive sleep apnea syndrome by analyzing sound of snoring acoustic features; CN102429648A discloses a kind of sleep-respiratory, heartbeat monitoring pillow, the breathing of non-contact measurement patient and heart beating; CN103687540A discloses a kind of mark of respiratory disorder, discriminating and diagnostic method, device and system, has set forth and has used the sound of snoring to detect the deficiency of SAS, proposing the amplitude by analyzing respiratory murmur, periodically differentiating SAS type with aperiodicity; CN102973244A discloses one and utilizes nostril and oral cavity thermal infrared images in thermal imaging system Real-time Collection sleep quality process, realizes the apneic method of contactless monitoring sleep by analysis chart picture; CN103070683A discloses a kind of method by gather and analysis breast, abdominal respiration electrical impedance Signal analysis sleep quality breathing pattern; CN1803089A discloses a kind of method of being placed the non-contact measurement sleep disorder of the mike collection mouth and nose sound of snoring by medicated pillow both sides; CN102641125A discloses a kind of sleep apnea decision maker, sends microwave, judge based on its echo to breathing from suspending state the breast of examinee, abdominal part in sleep; CN101536904A discloses a kind of based on cardiac electrical sleep apnea checkout gear, determines whether to there occurs sleep apnea by heart rate variability analysis; CN103006182A discloses a kind of preliminary detection system of domestic sleep apnea syndrome, mainly through analyzing the sound of snoring, designing different threshold values and carrying out Preliminary detection to SAS; CN203029237U discloses a kind of hypnotic instrument of contactless monitoring sleep breathing state, by radar sensor monitoring respiration information and heartbeat message, and then carries out sleep apnea analysis; CN102579049A discloses a kind of sleep breathing state monitoring device based on piezoelectric cable sensor, it is characterized in that detecting thorax abdomen cyclic deformation by piezoelectric cable, obtains respiratory frequency, asphyxia number of times and asphyxia time; CN101577974A is analyzed pulmonary sound and cardiac murmur by mike, in order to differentiate asphyxia and hypopnea.Although above method is simple and convenient, little on user sleep impact, the physiological parameter used is few, and Detection accuracy awaits improving.
3) using artificial intellectual technology, improves Detection accuracy.CN103083770A discloses a kind of system and method for Diagnosis and Treat patient's breathing patterns, by the respiratory air flow of oral nose mask monitoring patient, and the intermediate value of recording respiration parameter, meansigma methods, scope and standard deviation, gathered and be input to neutral net, detects patient's states by neutral net; CN1468077A discloses a kind of for detecting and treating the method for disordered breathing between sleep period, and end user's artificial neural networks is analyzed respiratory air flow data; CN103690168A discloses a kind of detection method of obstructive sleep apnea syndrome, by detecting blood oxygen saturation, the sound of snoring, breath signal three kinds of parameters, according to artificial intelligence's decision-making theory, build obscuring element set membership function, then determine whether to suffer from the comprehensive syndrome of sleep apnea by maximum subjection principle.
The present invention more adequately detects SAS based on how using less physiological signal.Based on the pulse of monitoring between testee's sleep period and blood oxygen signal, utilize Probabilistic Principal Component Analysis dimensionality reduction and extract principal character, then by the relation between packed decision tree height mapping eigenmatrix and SAS extent.
Summary of the invention:
The object of the invention is probability of use principal component analysis and reduce sleep apnea syndrome (SAS) data dimension, improve eigenmatrix and extract quality, use packed decision tree to promote the Detection accuracy of SAS.Use the method only need input the pulse and blood oxygen signal of monitoring between testee's sleep period, i.e. the extent of exportable tested SAS.
Accompanying drawing illustrates:
Fig. 1 is the sleep apnea syndrome detection method flow chart that the present invention is based on pulse and blood oxygen signal.
Detailed description of the invention:
For making technical scheme of the present invention more clear, below in conjunction with accompanying drawing embodiments of the present invention being done and describing in detail further.
Consulting shown in Fig. 1, is the sleep apnea syndrome detection method flow chart that the present invention is based on pulse and blood oxygen signal.
Step S101, obtains pulse and blood oxygen signal.
Specifically, continuously pulse the whole night and blood oxygen signal under record testee natural sleep state.
Step S102: pretreatment is carried out to pulse and blood oxygen signal.Concrete steps are as follows:
By minute in units of collected pulse and blood oxygen signal are split, form pulse and the blood oxygen signal of unit interval, and adopt the pulse crest of dyadic wavelet Study of modulus maximum algorithm positioning units time, calculate peak-peak interval, peak-peak interval series heterogeneous is carried out cubic spline interpolation, then use 2.5Hz resampling, form uniform pulse peak-peak interval series.
Step S103: extract SAS and detect eigenmatrix.Concrete steps are as follows:
(1) respectively principal component analysis is carried out to uniform pulse peak-peak interval series and the blood oxygen signal of unit interval, calculate main constituent contribution rate of accumulative total, determine the main constituent number that contribution rate of accumulative total is greater than 99%, as the output dimension of Probabilistic Principal Component Analysis.
(2) carry out Probabilistic Principal Component Analysis to the blood oxygen signal of unit interval, concrete steps are as follows:
1. utilize the non-main constituent in Gaussian noise ε Expressive Features space, set up the mapping relations between oximetry data s and its eigenmatrix x by latent variable model:
s＝Wx+μ+ε
Wherein W is Factor load-matrix, and μ is Non-zero Mean.
2. under this latent variable model, set up the probability distribution of oximetry data s under eigenmatrix x condition, according to Bayes formula, derive the posterior probability Density Distribution of eigenmatrix about oximetry data.
3. solved the maximal possibility estimation parameter of oximetry data log-likelihood function by EM algorithm: W, μ, ε, can derive hemorrhage oxygen eigenmatrix x.
(3) in like manner above analysis is carried out to the even pulse peak-peak interval series formed after resampling, form pulse peak-peak interval eigenmatrix.
(4) combine blood oxygen eigenmatrix and pulse peak-peak interval eigenmatrix, and then form sleep apnea syndrome detection eigenmatrix.
Step S104: set up detection model, output detections result.Concrete steps are as follows:
(1) eigenmatrix that the pulse of unit interval and blood oxygen signal are formed is defined as a sample, then all compositions of sample training set H.
(2) T time calling is carried out to C4.5 algorithm, each only call certain subset in training set H as current training set, after T time call, the individual different decision tree of T can be formed, when classifying to unknown test sample book, call this T decision tree and obtain T classification results, give unknown test sample book classification maximum for occurrence number in this T classification results.
(3) can determine whether each sample of testee there occurs sleep apnea event according to described method, the number of times of statistics testee generation sleep apnea event per hour, according to standard shown in table 1, export the extent of testee's sleep apnea syndrome.
Table 1
Claims (4)
1., based on a sleep apnea syndrome detection method for pulse and blood oxygen signal, it is characterized in that, the method comprises the steps:
(1) pulse and blood oxygen signal is obtained;
(2) pretreatment is carried out to pulse and blood oxygen signal;
(3) extract sleep apnea syndrome and detect eigenmatrix;
(4) detection model is set up, output detections result.
2. according to the sleep apnea syndrome detection method based on pulse and blood oxygen signal according to claim 1, it is characterized in that described step (2) by minute in units of respectively pulse and blood oxygen signal are split, form pulse and the blood oxygen signal of unit interval, and the pulse signal of unit interval is processed, by 2.5Hz resampling, form uniform pulse peak-peak interval series.
3. according to the sleep apnea syndrome detection method based on pulse and blood oxygen signal according to claim 1, it is characterized in that described step (3) carries out principal component analysis respectively to uniform pulse peak-peak interval series and the blood oxygen signal of unit interval, calculate main constituent contribution rate of accumulative total, determine the main constituent number that contribution rate of accumulative total is greater than 99%, use identical main constituent number, Probabilistic Principal Component Analysis is carried out to uniform pulse peak-peak interval series and blood oxygen signal, forms sleep apnea syndrome and detect eigenmatrix.
4. according to the sleep apnea syndrome detection method based on pulse and blood oxygen signal according to claim 1, it is characterized in that described step (4) adopts packed decision tree to set up detection model, the input matrix of model is eigenmatrix, output matrix is the diagnostic comments of clinical expert, and diagnostic comments can be sleep apnea syndrome severe, moderate, slight, critical and normal type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510046672.0A CN104545818A (en) | 2015-01-29 | 2015-01-29 | Sleep apnea syndrome detection method based on pulse and blood oxygen signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510046672.0A CN104545818A (en) | 2015-01-29 | 2015-01-29 | Sleep apnea syndrome detection method based on pulse and blood oxygen signals |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104545818A true CN104545818A (en) | 2015-04-29 |
Family
ID=53063686
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510046672.0A Pending CN104545818A (en) | 2015-01-29 | 2015-01-29 | Sleep apnea syndrome detection method based on pulse and blood oxygen signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104545818A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105725993A (en) * | 2016-04-13 | 2016-07-06 | 思澜科技（成都）有限公司 | Portable sleep monitoring equipment and monitoring method thereof |
| CN106037759A (en) * | 2016-06-30 | 2016-10-26 | 西安交通大学 | Cerebral self-adjusting index detection method for sleep apnea |
| CN106709251A (en) * | 2016-12-23 | 2017-05-24 | 李进让 | Evaluation method and device |
| CN107595243A (en) * | 2017-07-28 | 2018-01-19 | 深圳和而泰智能控制股份有限公司 | A kind of illness appraisal procedure and terminal device |
| CN108520781A (en) * | 2018-03-28 | 2018-09-11 | 北京大学人民医院 | A method of calculating test-tube baby's success final result probability |
| CN108778102A (en) * | 2016-02-01 | 2018-11-09 | 威里利生命科学有限责任公司 | The machine learning model of rapid-eye-movement sleep period is detected using the spectrum analysis of heart rate and movement |
| 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 |
| CN109303564A (en) * | 2018-10-24 | 2019-02-05 | 中山大学 | A kind of SAHS screening, Breathing mode and dormant cascade recognition methods |
| CN109431465A (en) * | 2018-10-26 | 2019-03-08 | 南京邮电大学 | A kind of old quality of human sleeping monitoring method based on radio frequency tag technology |
| CN110742621A (en) * | 2019-10-31 | 2020-02-04 | 京东方科技集团股份有限公司 | Signal processing method and computer equipment |
| CN111227793A (en) * | 2020-01-10 | 2020-06-05 | 京东方科技集团股份有限公司 | Apnea recognition method and system, electronic equipment and storage medium |
| CN111904393A (en) * | 2020-08-06 | 2020-11-10 | 苏州国科医工科技发展(集团)有限公司 | Intelligent child sleep monitoring system with adenoid hypertrophy probability judging function |
| CN112420195A (en) * | 2020-11-06 | 2021-02-26 | 清华大学 | Hypertension risk prediction method and device |
| CN113143263A (en) * | 2021-03-12 | 2021-07-23 | 杭州电子科技大学 | System for constructing optimal sleep apnea discrimination model |
| CN115952450A (en) * | 2023-03-14 | 2023-04-11 | 首都医科大学附属北京同仁医院 | Sleep apnea syndrome recognition method, device, computer and storage medium |
-
2015
- 2015-01-29 CN CN201510046672.0A patent/CN104545818A/en active Pending
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108778102A (en) * | 2016-02-01 | 2018-11-09 | 威里利生命科学有限责任公司 | The machine learning model of rapid-eye-movement sleep period is detected using the spectrum analysis of heart rate and movement |
| CN105725993A (en) * | 2016-04-13 | 2016-07-06 | 思澜科技（成都）有限公司 | Portable sleep monitoring equipment and monitoring method thereof |
| CN106037759B (en) * | 2016-06-30 | 2018-11-09 | 西安交通大学 | A kind of brain self-regulation index detection method towards sleep apnea |
| CN106037759A (en) * | 2016-06-30 | 2016-10-26 | 西安交通大学 | Cerebral self-adjusting index detection method for sleep apnea |
| CN106709251B (en) * | 2016-12-23 | 2019-06-28 | 李进让 | A kind of appraisal procedure and device |
| CN106709251A (en) * | 2016-12-23 | 2017-05-24 | 李进让 | Evaluation method and device |
| CN107595243A (en) * | 2017-07-28 | 2018-01-19 | 深圳和而泰智能控制股份有限公司 | A kind of illness appraisal procedure and terminal device |
| CN108520781A (en) * | 2018-03-28 | 2018-09-11 | 北京大学人民医院 | A method of calculating test-tube baby's success final result probability |
| CN109303564B (en) * | 2018-10-24 | 2021-02-12 | 中山大学 | SAHS screening and cascade identification method for respiratory mode and sleep state |
| CN109303564A (en) * | 2018-10-24 | 2019-02-05 | 中山大学 | A kind of SAHS screening, Breathing mode and dormant cascade recognition methods |
| 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 |
| CN109431465A (en) * | 2018-10-26 | 2019-03-08 | 南京邮电大学 | A kind of old quality of human sleeping monitoring method based on radio frequency tag technology |
| CN109431465B (en) * | 2018-10-26 | 2021-06-22 | 南京邮电大学 | Old people sleep quality monitoring method based on radio frequency tag technology |
| CN110742621A (en) * | 2019-10-31 | 2020-02-04 | 京东方科技集团股份有限公司 | Signal processing method and computer equipment |
| CN111227793A (en) * | 2020-01-10 | 2020-06-05 | 京东方科技集团股份有限公司 | Apnea recognition method and system, electronic equipment and storage medium |
| CN111904393A (en) * | 2020-08-06 | 2020-11-10 | 苏州国科医工科技发展(集团)有限公司 | Intelligent child sleep monitoring system with adenoid hypertrophy probability judging function |
| CN111904393B (en) * | 2020-08-06 | 2021-08-06 | 苏州国科医工科技发展(集团)有限公司 | Intelligent child sleep monitoring system with adenoid hypertrophy probability judging function |
| CN112420195A (en) * | 2020-11-06 | 2021-02-26 | 清华大学 | Hypertension risk prediction method and device |
| CN113143263A (en) * | 2021-03-12 | 2021-07-23 | 杭州电子科技大学 | System for constructing optimal sleep apnea discrimination model |
| CN115952450A (en) * | 2023-03-14 | 2023-04-11 | 首都医科大学附属北京同仁医院 | Sleep apnea syndrome recognition method, device, computer and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104545818A (en) | Sleep apnea syndrome detection method based on pulse and blood oxygen signals | |
| CN108388912A (en) | Sleep stage method based on multisensor feature optimization algorithm | |
| CN104545888B (en) | A kind of sleep apnea acquisition analysis system based on dynamic electrocardiogram and respiratory wave acquisition | |
| US20170258366A1 (en) | Apparatus and method for continuous noninvasive measurement of respiratory function and events | |
| TW201918222A (en) | Method and apparatus for high accuracy photoplethysmogram based atrial fibrillation detection using wearable device | |
| CN104257353A (en) | Sleep apnea syndrome detecting system | |
| Haidar et al. | Convolutional neural networks on multiple respiratory channels to detect hypopnea and obstructive apnea events | |
| CN103153183A (en) | Apparatus and method for diagnosing obstructive sleep apnea | |
| WO2007143535A2 (en) | Apparatus, system, and method for monitoring physiological signs | |
| CN106175770B (en) | The judgement system of apnea during a kind of sleep | |
| CN107361770A (en) | Sleep apnea event discriminating gear | |
| Liu et al. | Identifying obstructive sleep apnea by exploiting fine-grained BCG features based on event phase segmentation | |
| CN110477887A (en) | A kind of monitoring device of non-invasive long-range apnea syndrome | |
| WO2023179757A1 (en) | Lung function detection method, system and apparatus, and computer device and storage medium | |
| Shouldice et al. | Real time breathing rate estimation from a non contact biosensor | |
| CN113633260A (en) | Multi-sleep monitoring method, monitor, computer device and readable storage medium | |
| CN104274173A (en) | Bio-impedance-based wearable sleep respiration state monitoring system | |
| US11291406B2 (en) | System for determining a set of at least one cardio-respiratory descriptor of an individual during sleep | |
| CN204765621U (en) | Sleep apnea acquisition and analysis system based on developments electrocardio and respiratory wave collection | |
| Khandoker et al. | Modeling respiratory movement signals during central and obstructive sleep apnea events using electrocardiogram | |
| CN103027684A (en) | Device and method for removing noise caused by body motion in respiratory movement monitoring | |
| CN111657870A (en) | Apnea judgment system based on D-S evidence theory | |
| Alshaer | New technologies for the diagnosis of sleep apnea | |
| Heise et al. | Unobtrusively detecting apnea and hypopnea events via a hydraulic bed sensor | |
| WO2022269936A1 (en) | Sleeping state estimation system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150429 |
|
| WD01 | Invention patent application deemed withdrawn after publication |