CN113576401A - Sleep apnea syndrome rapid diagnosis device based on convolutional neural network - Google Patents
Sleep apnea syndrome rapid diagnosis device based on convolutional neural network Download PDFInfo
- Publication number
- CN113576401A CN113576401A CN202110655249.6A CN202110655249A CN113576401A CN 113576401 A CN113576401 A CN 113576401A CN 202110655249 A CN202110655249 A CN 202110655249A CN 113576401 A CN113576401 A CN 113576401A
- Authority
- CN
- China
- Prior art keywords
- data
- neural network
- apnea syndrome
- sleep apnea
- sleep
- 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
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/113—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb occurring during breathing
- A61B5/1135—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb occurring during breathing by monitoring thoracic expansion
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/113—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb occurring during breathing
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—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
- A61B5/48—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
- 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
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Veterinary Medicine (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Physiology (AREA)
- Artificial Intelligence (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Dentistry (AREA)
- Data Mining & Analysis (AREA)
- Optics & Photonics (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Pulmonology (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a sleep apnea syndrome rapid diagnosis device based on a convolutional neural network, which comprises: the data acquisition and acquisition module is used for monitoring night sleep monitoring data of the testee of the messenger by using the polysomnography or the portable sleep monitoring equipment and acquiring a corresponding medical diagnosis report; the data preprocessing module comprises a data extracting unit, a signal down-sampling unit and a data acquiring unit, and the neural network module is used for inputting the sleep monitoring data of the person to be detected into the trained sleep apnea syndrome recognition model to predict the sleep apnea syndrome of the person to be detected and outputting the severity of the sleep apnea syndrome of the person to be detected. The invention can more efficiently and directly realize the diagnosis of the severity of the sleep apnea syndrome, assist doctors in diagnosis and improve the diagnosis efficiency.
Description
Technical Field
The application relates to the technical field of neural networks and medical treatment, in particular to a sleep apnea syndrome rapid diagnosis device based on a convolutional neural network.
Background
Sleep Apnea Syndrome (SAS) refers to a Sleep disorder disease in which Apnea (Sleep Apnea) and hypoventilation (Hypopnea) repeatedly occur during Sleep in a patient. Apnea and hypopnea repeatedly occurring at night cause chronic intermittent hypoxia, carbon dioxide retention, increase of sympathetic excitability, systemic inflammatory reaction, and oxidative stress reaction, and insufficient antioxidant capacity, thereby causing or aggravating cardiovascular and cerebrovascular diseases and metabolic disorders. SAS can seriously affect human health.
The prevalence rate of SAS in China is about 4%, and the SAS is a common disease. However, diagnosis of SAS requires a special device, Polysomnography (PSG). At present, only three-level hospitals or partial two-level hospitals in large and medium cities can carry out standard diagnosis and treatment on the disease, so that a large number of patients cannot be diagnosed and treated in time, and great harm is caused to the health of people. In addition, in the prior art, PSG is mainly used for detecting SAS, but PSG detects a lot of signals, and professional technicians are required for diagnosis, which causes a lot of labor cost and equipment cost.
The AHI index (Apnea hypnea index Apnea hypopnea index) is an important index for measuring the tested sleep Apnea syndrome. In the prior art, physiological signal data of sleep monitoring is obtained firstly, noise reduction and filtering are carried out on the signals firstly, then the signals are segmented, signal characteristics are extracted manually, a neural network is trained to predict whether the signal segment belongs to a normal segment or a sleep respiratory event occurs, an AHI value is calculated according to the sleep respiratory event, and finally the tested SAS disease degree is judged according to the AHI value. These methods require noise reduction and filtering of the signal, and are weak in signal interference resistance. In addition, since the signal is segmented, event annotation data is also required. However, in practice, event annotation data is not readily available, and this segmented approach does not allow a direct diagnosis of the severity of sleep apnea syndrome, and an AHI value is also calculated. Most importantly, these methods require manual extraction of signal features, which is time consuming and laborious. Expert domain knowledge is also required, and important information hidden in the signal cannot be represented, so that the robustness and the universality are poor.
Disclosure of Invention
In order to enable potential patients with SAS to be diagnosed and treated in time, the invention provides a sleep apnea syndrome rapid diagnosis device based on a convolutional neural network, and aims to solve the problems of low efficiency, poor universality and the like in the existing method.
In order to realize the task, the invention adopts the following technical scheme:
a sleep apnea syndrome rapid diagnosis device based on a convolutional neural network comprises:
the data acquisition and acquisition module is used for monitoring night sleep monitoring data of the testee of the messenger by using the polysomnography or the portable sleep monitoring equipment and acquiring a corresponding medical diagnosis report;
the data preprocessing module comprises a data extraction unit, a signal down-sampling unit and a data acquisition unit, wherein:
the data extraction unit is used for extracting 4 physiological signals of a blood oxygen saturation degree signal, an airflow signal, a chest respiration signal and an abdomen respiration signal from the sleep monitoring data of each testee, extracting label data from a medical diagnosis report, and forming a group of original data by each group of the physiological signals and the label data; the signal downsampling unit is used for downsampling 4 physiological signals of the blood oxygen saturation signal, the airflow signal, the chest respiration signal and the abdomen respiration signal in each set of original data; the data acquisition unit is used for extracting data positioned in the middle from the original data of each testee after the downsampling processing as training data of the testee; constructing training data of different testees into a data set;
and the neural network module is used for inputting the sleep monitoring data of the person to be tested into the trained sleep apnea syndrome recognition model to predict the sleep apnea syndrome of the person to be tested and outputting the severity of the sleep apnea syndrome of the person to be tested.
Further, the construction method of the sleep apnea syndrome recognition model comprises the following steps:
constructing a convolutional neural network, wherein the convolutional neural network comprises 4 one-dimensional convolutional layers, 1 batch normalization layer, 1 maximum pooling layer and 1 full-connection layer; the input of the convolutional neural network is training data, firstly, the input normalization processing is completed through a batch normalization layer, then, the training data is activated through a one-dimensional convolutional layer provided with 4 convolutional kernels with the size of 1 × 8 and a Relu function, then, the maximum pooling down-sampling is performed through a maximum pooling layer with the size of 1 × 2, then, the training data is respectively passed through a one-dimensional convolutional layer provided with 128 convolutional kernels with the size of 1 × 8, a one-dimensional convolutional layer provided with 64 convolutional kernels with the size of 1 × 8 and a one-dimensional convolutional layer provided with 16 convolutional kernels with the size of 1 × 8, and finally, the output is converted into the severity of the tested sleep apnea syndrome through a full connection layer and an output function: and 4 normal, mild, moderate and severe classification probabilities, and taking the class with the highest probability as the prediction result of the neural network.
Further, the construction method further comprises: and training the neural network by using a training set by adopting a random gradient descent algorithm, adjusting parameters of the neural network by combining a test set and a verification set, and storing the trained model as a trained sleep apnea syndrome recognition model.
Further, the night sleep monitoring data of the testee is collected for at least 7 hours; correspondingly, the data acquisition unit extracts data located in the middle 7 hours from the raw data of each subject after the down-sampling process.
Further, the 4 physiological signals are down-sampled to 1 Hz.
Further, the data preprocessing module further comprises a data set dividing unit for dividing the data set into a training set, a verification set and a test set according to a ratio of 6:2: 2.
Further, the sleep monitoring data of the person to be detected are monitoring data of 4 signals of blood oxygen saturation, nasal airflow, chest and abdomen movement in the sleep state of the person to be detected.
Compared with the prior art, the invention has the following technical characteristics:
the invention trains the convolutional neural network by using the blood oxygen saturation signal, the airflow signal, the chest breathing signal and the abdomen breathing signal at 7 hours at night, automatically extracts the signal characteristics by using 4 convolutional layers of the convolutional neural network, is basically different from the prior method, does not need to segment the signals, and does not need event marking data. Moreover, because the training signature of the present invention is the severity of sleep apnea syndrome: the convolutional neural network directly predicts the severity of the tested sleep apnea syndrome without judging whether a sleep apnea event occurs in a signal, and the method is more direct. In addition, because the noise in the physiological signal is also used as the input of the convolutional neural network, the invention has stronger anti-interference capability on the signal. In conclusion, the invention can more efficiently and directly realize the diagnosis of the severity of the sleep apnea syndrome, assist the diagnosis of doctors and improve the diagnosis efficiency.
Drawings
FIG. 1 is a schematic flow chart of the operation of the apparatus of the present invention;
FIG. 2 is a block diagram of a convolutional neural network model;
FIG. 3 shows the judgment of the SAS disease level and AHI and/or hypoxemia level.
Detailed Description
The invention provides a sleep apnea syndrome severity diagnostic device based on a convolutional neural network, which uses the existing portable sleep monitoring equipment or a polysomnograph to obtain 4 kinds of physiological signal data of a tested blood oxygen saturation signal, an airflow signal, a chest respiration signal and an abdomen respiration signal, trains a convolutional neural network according to the 4 kinds of original physiological signal data, and finally predicts the severity of the tested sleep apnea syndrome through the convolutional neural network so as to realize the rapid diagnosis of the sleep apnea syndrome; the core of the realization of the invention is the convolutional neural network structure which is reasonably designed. The flow chart is as follows
Referring to fig. 1, the fast sleep apnea syndrome diagnosing apparatus based on a convolutional neural network of the present invention includes:
1. data acquisition and acquisition module
Monitoring sleep monitoring data of a testee of the testee for at least 7 hours at night by using a polysomnography or portable sleep monitoring equipment, and acquiring a corresponding medical diagnosis report; wherein the report is self-administered by the device or assisted by a physician.
In this embodiment, the data is actual clinical data provided by a hospital in Guangzhou, which is collected from a PSG portable sleep monitoring device. The data is composed of 1755 records, and each record comprises a plurality of physiological signal data such as a blood oxygen saturation signal, an airflow signal, a chest respiration signal, an abdomen respiration signal and the like and a medical diagnosis report (containing the diagnosis result of the severity of the tested sleep apnea syndrome made by a doctor).
2. Data preprocessing module
The device comprises a data extraction unit, a signal down-sampling unit, a data acquisition unit and a data set division unit, wherein:
the data extraction unit is used for extracting 4 physiological signals of a blood oxygen saturation degree signal, an airflow signal, a chest respiration signal and an abdomen respiration signal from the sleep monitoring data of each testee, extracting label data from a medical diagnosis report, and forming a group of original data by each group of the physiological signals and the label data.
The signal down-sampling unit is used for down-sampling 4 physiological signals of the blood oxygen saturation signal, the airflow signal, the chest respiration signal and the abdomen respiration signal in each set of original data to 1 Hz; this is because the sampling frequency of the monitored data to be tested acquired by the sleep monitoring device is not necessarily the same, and the downsampling process is performed to reduce the data dimension and unify the sampling frequency.
A data acquisition unit, which is used for extracting data (including 4 kinds of physiological signals and corresponding label data) positioned in the middle 7 hours from the original data of each testee after the down-sampling processing as training data of the testee; the training data of different subjects constitutes a data set. The design purpose of the unit is to extract the sleep monitoring data of the middle 7 hours of the testee, because the monitoring time of the tested monitoring data collected by using the sleep monitoring equipment is not always the same, and the traditional medicine diagnoses the sleep apnea syndrome according to the sleep monitoring data of at least 7 hours tested.
The data set dividing unit is used for dividing the data set into a training set, a verification set and a test set according to the proportion of 6:2:2 so as to scientifically train and test the model; in this embodiment, the training set has a total of 1051 records. The verification set has a total of 352 records and the test set has a total of 352 records.
3. Neural network module
The method is used for inputting the sleep monitoring data of the testee into a trained sleep apnea syndrome recognition model so as to predict the sleep apnea syndrome of the testee.
The sleep monitoring data of the person to be detected are monitoring data of 4 signals of blood oxygen saturation, nasal airflow, chest and abdomen movement of the person to be detected in a sleep state. Then, the monitoring data of the 4 signals are directly input into a trained sleep apnea syndrome recognition model, whether the tested person suffers from the sleep apnea syndrome can be directly predicted through the trained model, and if the tested person suffers from the sleep apnea syndrome, the model can recognize the severity of the tested sleep apnea syndrome.
The construction method of the sleep apnea syndrome recognition model comprises the following steps:
constructing a convolutional neural network, wherein the convolutional neural network comprises 4 one-dimensional convolutional layers, 1 batch normalization layer, 1 maximum pooling layer and 1 full-connection layer; the model structure diagram is shown in fig. 3. Since 4 kinds of physiological signals of the blood oxygen saturation signal, the airflow signal, the chest respiration signal, and the abdominal respiration signal extracted for 7 hours are time-series data in nature, one-dimensional convolution is used.
The input of the convolutional neural network is training data, namely 4 time sequences of the blood oxygen saturation signal with the sampling frequency of 1 in the time length of 7 hours, the airflow signal, the chest respiration signal and the abdomen respiration signal, so the input dimension of the convolutional neural network is 25200 x 4; firstly, input normalization processing is completed through a batch normalization layer, then, the normalization processing is carried out through a one-dimensional convolution layer provided with 4 convolution kernels with the size of 1 x 8 and activated through a Relu function, then, maximum pooling down-sampling is carried out through a maximum pooling layer with the size of 1 x 2, then, the normalization processing is carried out through one-dimensional convolution layers provided with 128 convolution kernels with the size of 1 x 8, one-dimensional convolution layers provided with 64 convolution kernels with the size of 1 x 8 and one-dimensional convolution layers provided with 16 convolution kernels with the size of 1 x 8, and finally, the output is converted into the severity of the tested sleep apnea syndrome through a full connection layer and an output function softmax (x): and 4 normal, mild, moderate and severe classification probabilities, and taking the class with the highest probability as the prediction result of the model.
Defining a convolution operation:
y(l)represents the output of the first convolutional layer;
α(l)an activation function representing the first convolutional layer;
ω(l)representing the convolution kernel weight of the convolution layer of the first layer;
X(l)representing the input of the first layer convolution layer;
b(l)indicating the bias of the first layer convolution layer;
the batch normalization layer is used for converting data into data with a mean value of 0 and a standard deviation of 1, and the conversion formula is as follows:
where μ is the mean of all training samples and σ is the standard deviation of all training samples. The role of the batch normalization layer is to normalize and normalize the training data.
The convolutional neural network provided by the invention uses a maximum pooling method to reduce the dimension of data and reduce the calculation amount. As shown, the output result of the first convolutional layer is reduced in dimension through the maximum pooling layer.
The output of the 4 convolutional layers was converted to the severity of the test sleep apnea syndrome by the fully-connected layer and the output function softmax (x): and 4 normal, mild, moderate and severe classification probabilities, and taking the class with the highest probability as the prediction result of the model. Output function of convolutional neural network softmax (x):
the convolutional neural network proposed by the present invention uses an Adam optimizer and selects a cross entropy function as the loss function of the training, the cross entropy loss function L (y, f (x)) is as follows:
wherein y represents a real label, f (x) represents a prediction function of the model, p, q represent a real distribution of the data label and a distribution given by the model prediction, respectively, and p (y)i|xi) Represents a sample xiThe true distribution of the labels is such that,representing a given sample xiThe model predicts the probability distribution over the various classes.
And training the neural network by using a training set by adopting a random gradient descent algorithm, adjusting parameters of the neural network by combining a test set and a verification set, and storing the trained model as a trained sleep apnea syndrome recognition model.
The above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.
Claims (7)
1. A sleep apnea syndrome rapid diagnosis device based on a convolutional neural network is characterized by comprising the following components:
the data acquisition and acquisition module is used for monitoring night sleep monitoring data of the testee of the messenger by using the polysomnography or the portable sleep monitoring equipment and acquiring a corresponding medical diagnosis report;
the data preprocessing module comprises a data extraction unit, a signal down-sampling unit and a data acquisition unit, wherein:
the data extraction unit is used for extracting 4 physiological signals of a blood oxygen saturation degree signal, an airflow signal, a chest respiration signal and an abdomen respiration signal from the sleep monitoring data of each testee, extracting label data from a medical diagnosis report, and forming a group of original data by each group of the physiological signals and the label data; the signal downsampling unit is used for downsampling 4 physiological signals of the blood oxygen saturation signal, the airflow signal, the chest respiration signal and the abdomen respiration signal in each set of original data; the data acquisition unit is used for extracting data positioned in the middle from the original data of each testee after the downsampling processing as training data of the testee; constructing training data of different testees into a data set;
and the neural network module is used for inputting the sleep monitoring data of the person to be tested into the trained sleep apnea syndrome recognition model to predict the sleep apnea syndrome of the person to be tested and outputting the severity of the sleep apnea syndrome of the person to be tested.
2. The convolutional neural network-based sleep apnea syndrome rapid diagnosis device of claim 1, wherein the sleep apnea syndrome identification model is constructed by the following method:
constructing a convolutional neural network, wherein the convolutional neural network comprises 4 one-dimensional convolutional layers, 1 batch normalization layer, 1 maximum pooling layer and 1 full-connection layer; the input of the convolutional neural network is training data, firstly, the input normalization processing is completed through a batch normalization layer, then, the training data is activated through a one-dimensional convolutional layer provided with 4 convolutional kernels with the size of 1 × 8 and a Relu function, then, the maximum pooling down-sampling is performed through a maximum pooling layer with the size of 1 × 2, then, the training data is respectively passed through a one-dimensional convolutional layer provided with 128 convolutional kernels with the size of 1 × 8, a one-dimensional convolutional layer provided with 64 convolutional kernels with the size of 1 × 8 and a one-dimensional convolutional layer provided with 16 convolutional kernels with the size of 1 × 8, and finally, the output is converted into the severity of the tested sleep apnea syndrome through a full connection layer and an output function: and 4 normal, mild, moderate and severe classification probabilities, and taking the class with the highest probability as the prediction result of the neural network.
3. The convolutional neural network-based sleep apnea syndrome rapid diagnosis device of claim 1, wherein the construction method further comprises: and training the neural network by using a training set by adopting a random gradient descent algorithm, adjusting parameters of the neural network by combining a test set and a verification set, and storing the trained model as a trained sleep apnea syndrome recognition model.
4. The convolutional neural network-based sleep apnea syndrome rapid diagnosis device as recited in claim 1, wherein the night sleep monitoring data of the subject is collected for at least 7 hours; correspondingly, the data acquisition unit extracts data located in the middle 7 hours from the raw data of each subject after the down-sampling process.
5. The convolutional neural network-based sleep apnea syndrome rapid diagnosis device of claim 1, wherein the 4 physiological signals are down-sampled to 1 Hz.
6. The convolutional neural network-based sleep apnea syndrome rapid diagnosis device as recited in claim 1, wherein the data preprocessing module further comprises a data set partitioning unit for partitioning the data set into a training set, a validation set and a test set in a ratio of 6:2: 2.
7. The convolutional neural network-based sleep apnea syndrome rapid diagnosis apparatus as recited in claim 1, wherein the sleep monitoring data of the subject is monitoring data of 4 signals of blood oxygen saturation, nasal airflow, chest and abdomen movement in a sleep state of the subject.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110655249.6A CN113576401A (en) | 2021-06-11 | 2021-06-11 | Sleep apnea syndrome rapid diagnosis device based on convolutional neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110655249.6A CN113576401A (en) | 2021-06-11 | 2021-06-11 | Sleep apnea syndrome rapid diagnosis device based on convolutional neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113576401A true CN113576401A (en) | 2021-11-02 |
Family
ID=78243728
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110655249.6A Pending CN113576401A (en) | 2021-06-11 | 2021-06-11 | Sleep apnea syndrome rapid diagnosis device based on convolutional neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113576401A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114795133A (en) * | 2022-06-29 | 2022-07-29 | 华南师范大学 | Sleep apnea detection method, device, equipment and storage medium |
CN114869241A (en) * | 2022-07-11 | 2022-08-09 | 西南交通大学 | Sleep respiratory event prediction method, device, equipment and readable storage medium |
CN117064333A (en) * | 2023-08-01 | 2023-11-17 | 大连市中心医院 | Primary screening device for obstructive sleep apnea hypopnea syndrome |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN202723828U (en) * | 2012-01-12 | 2013-02-13 | 谢汝石 | Obstructive sleep apnea-hypopnea syndrome (OSAHS) patient primary screening system |
CN108091391A (en) * | 2017-12-27 | 2018-05-29 | 深圳和而泰数据资源与云技术有限公司 | Illness appraisal procedure, terminal device and computer-readable medium |
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 |
CN109620208A (en) * | 2018-12-29 | 2019-04-16 | 南京茂森电子技术有限公司 | Sleep Apnea-hypopnea Syndrome detection system and method |
EP3536225A1 (en) * | 2018-03-07 | 2019-09-11 | Koninklijke Philips N.V. | Sleep apnea detection system and method |
CN111248859A (en) * | 2019-12-27 | 2020-06-09 | 黄淮学院 | Automatic sleep apnea detection method based on convolutional neural network |
CN112190253A (en) * | 2020-09-17 | 2021-01-08 | 广东工业大学 | Classification method for severity of obstructive sleep apnea |
CN112426147A (en) * | 2020-10-21 | 2021-03-02 | 华南师范大学 | Sleep respiratory event detection model processing method, system and storage medium |
-
2021
- 2021-06-11 CN CN202110655249.6A patent/CN113576401A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN202723828U (en) * | 2012-01-12 | 2013-02-13 | 谢汝石 | Obstructive sleep apnea-hypopnea syndrome (OSAHS) patient primary screening system |
CN108091391A (en) * | 2017-12-27 | 2018-05-29 | 深圳和而泰数据资源与云技术有限公司 | Illness appraisal procedure, terminal device and computer-readable medium |
EP3536225A1 (en) * | 2018-03-07 | 2019-09-11 | Koninklijke Philips N.V. | Sleep apnea detection system and method |
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 |
CN109620208A (en) * | 2018-12-29 | 2019-04-16 | 南京茂森电子技术有限公司 | Sleep Apnea-hypopnea Syndrome detection system and method |
CN111248859A (en) * | 2019-12-27 | 2020-06-09 | 黄淮学院 | Automatic sleep apnea detection method based on convolutional neural network |
CN112190253A (en) * | 2020-09-17 | 2021-01-08 | 广东工业大学 | Classification method for severity of obstructive sleep apnea |
CN112426147A (en) * | 2020-10-21 | 2021-03-02 | 华南师范大学 | Sleep respiratory event detection model processing method, system and storage medium |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114795133A (en) * | 2022-06-29 | 2022-07-29 | 华南师范大学 | Sleep apnea detection method, device, equipment and storage medium |
CN114869241A (en) * | 2022-07-11 | 2022-08-09 | 西南交通大学 | Sleep respiratory event prediction method, device, equipment and readable storage medium |
CN114869241B (en) * | 2022-07-11 | 2022-09-23 | 西南交通大学 | Sleep respiratory event prediction method, device, equipment and readable storage medium |
CN117064333A (en) * | 2023-08-01 | 2023-11-17 | 大连市中心医院 | Primary screening device for obstructive sleep apnea hypopnea syndrome |
CN117064333B (en) * | 2023-08-01 | 2024-03-26 | 大连市中心医院 | Primary screening device for obstructive sleep apnea hypopnea syndrome |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Shen et al. | Multiscale deep neural network for obstructive sleep apnea detection using RR interval from single-lead ECG signal | |
CN113576401A (en) | Sleep apnea syndrome rapid diagnosis device based on convolutional neural network | |
Page et al. | Wearable seizure detection using convolutional neural networks with transfer learning | |
CN109273085B (en) | Pathological respiratory sound library establishing method, respiratory disease detection system and respiratory sound processing method | |
Wang et al. | Detection of sleep apnea from single-lead ECG signal using a time window artificial neural network | |
CN108670200A (en) | A kind of sleep sound of snoring classification and Detection method and system based on deep learning | |
CN109009102B (en) | Electroencephalogram deep learning-based auxiliary diagnosis method and system | |
CN111248859A (en) | Automatic sleep apnea detection method based on convolutional neural network | |
Wang et al. | An efficient method to detect sleep hypopnea-apnea events based on EEG signals | |
CN116098602A (en) | Non-contact sleep respiration monitoring method and device based on IR-UWB radar | |
CN115500843A (en) | Sleep stage staging method based on zero sample learning and contrast learning | |
Jiang et al. | A multi-scale parallel convolutional neural network for automatic sleep apnea detection using single-channel EEG signals | |
Liu et al. | Automatic sleep arousals detection from polysomnography using multi-convolution neural network and random forest | |
CN113925459B (en) | Sleep stage method based on electroencephalogram feature fusion | |
CN113974631A (en) | Autism computer-aided diagnosis system based on physiological signal recognition and analysis | |
Bashi et al. | Diagnosis of obstructive apnea disease AHI in chemical warfare veterans based on HRV signals analysis using the ANFIS neural network | |
Vyas et al. | Sleep Stage Classification Using Non-Invasive Bed Sensing and Deep Learning | |
He et al. | An approach to sleep apnea syndrome detection based on change characteristics of blood oxygen saturation and pulse rate | |
CN116189899B (en) | Emergency critical illness auxiliary evaluation system based on machine learning | |
Reddy | An Efficient Method to Detect Sleep Apnea | |
CN113143263B (en) | System for constructing sleep apnea discrimination optimal model | |
Cui et al. | MOTILITY OF THE UNPREPARED INTESTINAL TRACT IN HEALTHY HUMANS AND PATIENTS WITH SLOW TRANSIT CONSTIPATION REVEALED BY USING MICRO-ELECTRONIC CAPSULE | |
US20240008765A1 (en) | Establishing method of sleep apnea assessment program, sleep apnea assessment system, and sleep apnea assessment method | |
Kumar et al. | Sleep Track: Automated Detection and Classification of Sleep Stages | |
Huang et al. | Fingertip Pulse Analysis of an Anxiety Patient Based on Deep Learning |
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 |