TWI733247B - Detection device and detection method for obstructive sleep apnea - Google Patents

Abstract

A detection device and a detection method for obstructive sleep apnea (OSA) are provided. The detection device includes a process, a storage medium, and a transceiver. The transceiver obtains a first ECG, wherein the first ECG includes a first data set corresponding to non-OSA and a second data set corresponding to the OSA. The processor accesses and executes a plurality of modules in the storage medium, wherein the plurality of modules include a training module and a detection module. The training module uses the first data set and the second data set as training data to train a machine learning model. The detection module obtains, via the transceiver, a second ECG of a subject, and determines, according to the machine learning model and the second ECG, whether the subject has the OSA.

Description

Device and method for detecting obstructive sleep apnea symptoms

The present disclosure relates to a detection device and detection method, and in particular to a detection device and detection method for obstructive sleep apnea (OSA) symptoms.

OSA symptom is a common sleep disorder, which is a symptom of apnea or weakened due to complete or partial upper airway obstruction due to the collapse of the pharynx during sleep. At present, the main basis for the diagnosis of obstructive sleep apnea is polysomnography (PSG). During the PSG examination, the subject must go to the sleep laboratory or sleep center to sleep for one night. Under the supervision of the nursing staff, stick electrode patches on the head, corners of the eyes, chin, heart, and legs, and place them on the chest. Wear a sensor belt on your abdomen, put a blood oxygen meter on your fingers, put a breathing sensor on your nose and mouth, and put a sphygmomanometer on your arm to record your sleep physiological data throughout the night. However, not all patients have time to spend the night in a sleep laboratory or sleep center, and patients with OSA symptoms do not suffer from apnea events every night. Therefore, even if the patient goes to the sleep laboratory or sleep center overnight to measure the sleep physiology data, the sleep physiology data measured that night may not help the doctor to diagnose the patient’s OSA symptoms. Based on this, how to propose a simplified and convenient method for diagnosing sleep disorders is one of the goals of those skilled in the art.

The present disclosure provides an OSA symptom detection device and detection method, which can diagnose whether a subject suffers from OSA symptoms.

The device for detecting obstructive sleep apnea symptoms disclosed in the present disclosure includes a processor, a storage medium, and a transceiver. The transceiver obtains a first electrocardiogram, where the first electrocardiogram includes a first data set corresponding to symptoms of non-obstructive sleep apnea and a second data set corresponding to symptoms of obstructive sleep apnea. The storage medium stores multiple modules. The processor is coupled to the storage medium and the transceiver, and accesses and executes a plurality of modules, wherein the plurality of modules include a training module and a detection module. The training module uses the first data set and the second data set as training data to train the machine learning model. The detection module obtains the subject's second electrocardiogram through the transceiver, and determines whether the subject suffers from obstructive sleep apnea symptoms according to the machine learning model and the second electrocardiogram.

In an embodiment of the present disclosure, the above-mentioned first electrocardiogram and second electrocardiogram correspond to non-apnea events.

In an embodiment of the present disclosure, the above-mentioned machine learning model is a convolutional neural network model.

In an embodiment of the present disclosure, the above-mentioned training module generates a corresponding cardiogenic breathing signal according to the first electrocardiogram, measures a plurality of RR intervals according to the first electrocardiogram, and according to at least the plurality of RR intervals and the cardiogenic respiratory signal One of them generates second training data, and trains the support vector machine model based on the second training data.

In an embodiment of the present disclosure, the aforementioned detection module determines whether the subject suffers from obstructive sleep apnea symptoms based on the support vector machine model, the machine learning model, and the second electrocardiogram.

In an embodiment of the present disclosure, the aforementioned training module extracts multiple features from multiple RR intervals or cardiogenic breathing signals, and generates second training data based on the multiple features.

In an embodiment of the present disclosure, the above-mentioned multiple features are associated with at least one of the following: the average value of the RR interval, the correlation coefficient of the second or third sequence of the RR interval, the number of RR interval pairs, wherein Including two adjacent RR intervals, and the time interval between the two RR intervals exceeds 50 milliseconds, the standard deviation of the two adjacent RR intervals, the normalized extremely low frequency range power of the RR interval, and the cardiogenic respiratory signal The normalized very low frequency range power of the normalized low frequency range power of the cardiogenic respiratory signal, and the normalized high frequency range power of the cardiogenic respiratory signal.

In an embodiment of the present disclosure, the above-mentioned detection module sends a warning through the transceiver in response to determining that the subject is suffering from obstructive sleep apnea symptoms.

The method for detecting obstructive sleep apnea symptoms disclosed in the present disclosure includes: obtaining a first electrocardiogram, wherein the first electrocardiogram includes a first data set corresponding to non-obstructive sleep apnea symptoms and a first data set corresponding to obstructive sleep apnea symptoms The second data set; use the first data set and the second data set as training data to train the machine learning model; obtain the subject's second electrocardiogram; and determine whether the subject suffers from obstruction based on the machine learning model and the second electrocardiogram Symptoms of sleep apnea.

Based on the above, the ECG of the subject under normal conditions (ie, no non-apnea events) can be used to diagnose whether the subject suffers from OSA symptoms.

In order to make the content of this disclosure more comprehensible, the following embodiments are specifically cited as examples on which this disclosure can indeed be implemented. In addition, wherever possible, elements/components/steps with the same reference numbers in the drawings and embodiments represent the same or similar components.

FIG. 1 shows a schematic diagram of an OSA symptom detecting device 100 according to an embodiment of the disclosure. The detection device 100 includes a processor 110, a storage medium 120 and a transceiver 130.

The processor 110 is, for example, a central processing unit (CPU), or other programmable general-purpose or special-purpose micro control unit (MCU), microprocessor, or digital signal processing Digital signal processor (DSP), programmable controller, application specific integrated circuit (ASIC), graphics processing unit (GPU), arithmetic logic unit (ALU) , Complex programmable logic device (CPLD), field programmable gate array (FPGA) or other similar components or a combination of the above components. The processor 110 may be coupled to the storage medium 120 and the transceiver 130, and access and execute multiple modules and various application programs stored in the storage medium 120.

The storage medium 120 is, for example, any type of fixed or removable random access memory (RAM), read-only memory (ROM), or flash memory. , Hard disk drive (HDD), solid state drive (SSD) or similar components or a combination of the above components, which are used to store multiple modules or various application programs that can be executed by the processor 110. In this embodiment, the storage medium 120 can store multiple modules including a training module 121 and a detection module 122, the functions of which will be described later.

The transceiver 130 transmits and receives signals in a wireless or wired manner. The transceiver 130 may also perform operations such as low noise amplification, impedance matching, frequency mixing, up or down frequency conversion, filtering, amplification, and the like. The transceiver 130 can be used to receive the first electrocardiogram as training data or receive the second electrocardiogram measured from the subject. For example, the transceiver 130 may be communicatively connected to electrodes attached to the subject, wherein the electrodes may sense the heartbeat of the subject to generate a second electrocardiogram of the subject. For example, the transceiver 130 can be implemented through global system for mobile communication (GSM), personal handy-phone system (PHS), code division multiple access (CDMA) System, wideband code division multiple access (WCDMA) system, long term evolution (LTE) system, worldwide interoperability for microwave access (WiMAX) system, wireless fidelity (wireless) Fidelity (Wi-Fi) systems or communication technologies such as Bluetooth (Bluetooth) receive the first electrocardiogram as training data, or receive the second electrocardiogram measured by electrodes attached to the subject's chest.

The first electrocardiogram as training data may include a first data set corresponding to non-OSA symptoms and a second data set corresponding to OSA symptoms. : The electrocardiogram generated by the person’s heartbeat under no apnea event), and the second data set is, for example, measuring the heartbeat of a patient with severe OSA symptoms under normal conditions (ie: the patient has not had an apnea event). The generated electrocardiogram. In this embodiment, the first data set may include a plurality of one-minute electrocardiograms corresponding to non-OSA symptoms, and no apnea event occurs during each electrocardiogram measurement period. The second data set may include a plurality of one-minute electrocardiograms corresponding to OSA symptoms, and no apnea event occurred during each electrocardiogram measurement period.

The training module 121 can use the first data set and the second data set as training data to train the machine learning model, wherein the trained machine learning model can be used to determine whether the subject suffers from OSA based on the second electrocardiogram measured from the subject symptom. It is worth noting that the training data in this embodiment may be a time-domain signal that has not been converted by wavelet, rather than a time-frequency signal that has been converted by wavelet. Since the dimensionality of the time-domain signal is lower than that of the time-frequency signal, the training module 121 that uses the time-domain signal instead of the time-frequency signal to train the machine learning model will consume less computing power.

In one embodiment, the first electrocardiogram as training data further includes a third data set representing the electrocardiogram interfered by noise. The training module 121 can use the first data set, the second data set, and the third data set as training data to train the machine learning model. The machine learning model trained from the first data set, the second data set and the third data set can not only judge whether the subject suffers from OSA symptoms based on the second electrocardiogram measured from the subject, but also judge based on the second electrocardiogram The data of the second ECG received by the transceiver 130 may be interfered by noise. Therefore, when determining the severity of the OSA symptom of the subject, the computing module 122 can filter out the data interfered by noise first.

After training the machine learning model, the detection module 122 can determine whether the subject suffers from OSA symptoms according to the machine learning model and the second electrocardiogram measured from the subject.

In one embodiment, if the detection module 122 determines that the subject is suffering from OSA symptoms, the detection module 122 can send a warning via the transceiver 130 to remind the subject that the subject has suffered from OSA symptoms. For example, the detection module 122 may send a warning message to the mobile device of the subject through the transceiver 130 after determining that the subject suffers from OSA symptoms.

The above-mentioned machine learning model is, for example, a convolutional neural network (CNN) model. The biggest difference between convolutional neural networks and traditional multi-layer perception networks is that convolutional neural networks have more convolutional layers and pooling layers. These two layers allow convolutional neural networks to extract the characteristics of the input signal. The design of the convolutional layer has many characteristics. The first feature is local perception. In traditional neural networks, each neuron must be connected to each sampling point, so a large number of weights are required, which makes it extremely difficult to train the network. In a convolutional neural network, the number of weights of each neuron is the same as the size of the convolution kernel, so it is equivalent to that each neuron is only connected to the corresponding part of the sampling points, which can greatly reduce the number of weights. . A smaller number of weights can reduce the risk of overfitting. The second feature is the weight sharing mechanism. The convolutional neural network trains and updates the best convolution kernel weight through the back propagation error algorithm, but during the convolution process, the weight of the convolution kernel does not change. The third feature is multiple convolution kernels. If only one convolution kernel is used, only partial features of the signal can be extracted. If multiple convolution kernels are used, multiple features of the input signal can be extracted. The more the number of convolutional layers, the more features the convolutional neural network can extract.

After the input signal undergoes nonlinear conversion by the convolutional layer and activation function, a feature map can be generated. The most important function of the activation function is to introduce the nonlinearity of the neural network, because if the activation function is not added, the convolutional layer and the fully connected layer are just simple linear operations, and there is still no solution to the linear inseparable problem.

In order to reduce the dimensionality of the features extracted by the convolution operation and increase the speed of the learning process, a pooling layer is followed by the convolution layer. The pooling layer is a method of compressing feature maps and retaining important information. The sampling method adopted by the pooling layer may include max pooling or mean pooling. The maximum pooling method is to select the maximum value in the pooling window as the sampling value. The average pooling method is to add all the values in the pooling window and average them as the sampling value. The feature map after pooling still retains the greatest possibility of local area comparison. In other words, the pooled information can be more focused on whether there are matching features in the feature map instead of focusing on the location of these features. Therefore, compared with the traditional neural network, the convolutional neural network can determine whether a feature is included in the feature map without considering the location of the feature. Therefore, even if the feature of the input signal is shifted, the convolutional neural network can recognize the feature. After the pooling layer, the fully connected layer integrates the highly abstracted features after multiple convolutions and pooling. Then the output layer outputs a corresponding probability for each classification, where the sum of the probability of all classifications is 1.

FIG. 2 illustrates a schematic diagram of a convolutional neural network model 200 according to an embodiment of the disclosure, where the convolutional neural network model 200 is one aspect of the machine learning model trained by the training module 121. The convolutional neural network model 200 may include an input layer 220, a convolution layer 231, a pooling layer 232, a convolution layer 241, a pooling layer 242, a fully connected layer 251, a fully connected layer 252, and an output layer 260. The input data 210 of the convolutional neural network model 200 as shown in FIG. 2 is, for example, a second electrocardiogram measured from the subject, and the output data 270 of the convolutional neural network model 200 represents the judgment result of whether or not suffering from OSA symptoms.

的卷積核。經過卷積層231的輸入資料210會轉變為128個尺寸為

的特徵圖。接著，池化層232使用尺寸為

的滑動視窗對卷積層231輸出的特徵圖進行取樣以產生128個尺寸為

的特徵圖。卷積層241包括64個尺寸16

的卷積核。卷積層241進一步地對池化層232輸出的特徵圖進行卷積運算以產生64個尺寸為

的特徵圖。接著，池化層242使用尺寸為

的滑動視窗對卷積層241輸出的特徵圖進行取樣以產生64個尺寸為

的特徵圖。而後，池化層242輸出的特徵圖依序地輸入至具有128個神經元的全連接層251以及具有64個神經元的全連接層252。輸出層260可根據Softmax激活函數來計算全連接層252的輸出的對應於罹患OSA症狀的第一機率以及對應於未罹患OSA症狀的第二機率。若第一機率大於第二機率，則偵測模組122可判斷輸入資料210對應於罹患OSA症狀的患者。反之，若第一機率小於或等於第二機率，則偵測模組122可判斷輸入資料210對應未罹患OSA症狀的人員。 In this embodiment, the input data 210 is a 1-minute second electrocardiogram, and the input data 210 includes 6,000 sampling points sampled at a sampling frequency of 100 Hz. The convolutional layer 231 includes 128 sizes of

The convolution kernel. The input data 210 after the convolutional layer 231 will be transformed into 128 sizes as

Characteristic map. Next, the size of the pooling layer 232 is

The sliding window samples the feature map output by the convolutional layer 231 to generate 128 sizes of

Characteristic map. Convolutional layer 241 includes 64 sizes 16

The convolution kernel. The convolutional layer 241 further performs a convolution operation on the feature map output by the pooling layer 232 to generate 64 sizes of

Characteristic map. Next, the size of the pooling layer 242 is

The sliding window samples the feature map output by the convolutional layer 241 to generate 64 sizes of

Characteristic map. Then, the feature map output by the pooling layer 242 is sequentially input to the fully connected layer 251 with 128 neurons and the fully connected layer 252 with 64 neurons. The output layer 260 can calculate the first probability of suffering from OSA symptoms and the second probability of not suffering from OSA symptoms of the output of the fully connected layer 252 according to the Softmax activation function. If the first probability is greater than the second probability, the detection module 122 can determine that the input data 210 corresponds to a patient suffering from OSA symptoms. Conversely, if the first probability is less than or equal to the second probability, the detection module 122 can determine that the input data 210 corresponds to a person who does not suffer from OSA symptoms.

It is worth noting that the activation function used by the output layer 260 may be, for example, a softmax function, a sigmoid function, a hyperbolic tangent function, or a rectified linear unit (ReLU) function, and the disclosure is not limited thereto.

In an embodiment, the training module 121 can further generate a support vector machine (SVM) model. The detection module 122 can determine whether the subject suffers from OSA symptoms according to the support vector machine model, the machine learning model, and the second electrocardiogram measured from the subject. For example, if at least one of the support vector machine model and the machine learning model determines that the subject suffers from OSA symptoms, the detection module 122 may respond to at least one of the support vector machine model and the machine learning model to determine that the subject is affected The test subject suffers from OSA symptoms and the output represents the judgment result that the subject suffers from OSA symptoms.

The training module 121 can train the aforementioned support vector machine model according to the first electrocardiogram. The training module 121 can generate a corresponding ECG-derived respiration (EDR) signal according to the first electrocardiogram as the training data. On the other hand, the training module 121 can measure multiple RR intervals according to the first electrocardiogram. Then, the training module 121 can generate second training data for training the support vector machine model based on the cardiogenic breathing signal and at least one of the multiple RR intervals, and then train the aforementioned second training data based on the second training data Support vector machine model. Specifically, the training module 121 can extract multiple features from cardiogenic breathing signals or multiple RR intervals, and generate the second training data according to the features. The multiple characteristics are, for example, related to the average value of the RR interval, the correlation coefficient of the second or third sequence of the RR interval, the number of RR interval pairs (the RR interval pair includes two adjacent RR intervals, and the interval between the two RR intervals The time interval exceeds 50 milliseconds), the standard deviation of two adjacent RR intervals, the normalized very low frequency power (VLFP) of the RR interval, and the normalized very low frequency range of the cardiogenic respiratory signal Power, the normalized low frequency power (LFP) of the cardiogenic respiratory signal or the normalized high frequency power (HFP) of the cardiogenic respiratory signal, but the present disclosure is not limited to this. VLFP is approximately between 0.003-0.04 Hz. The aforementioned LFP is approximately between 0.04-0.15 Hz and the HFP is approximately between 0.15-0.4 Hz.

FIG. 3 illustrates a flowchart of a method for detecting OSA symptoms according to an embodiment of the present disclosure, wherein the detecting method is implemented by, for example, the detecting device 100 shown in FIG. 1. In step S301, a first electrocardiogram is obtained, where the first electrocardiogram includes a first data set corresponding to non-obstructive sleep apnea symptoms and a second data set corresponding to obstructive sleep apnea symptoms. In step S302, the first data set and the second data set are used as training data to train the machine learning model. In step S303, a second electrocardiogram of the subject is obtained. In step S304, it is determined whether the subject suffers from obstructive sleep apnea symptoms according to the machine learning model and the second electrocardiogram.

In summary, the present disclosure can use ECGs of patients with non-OSA symptoms and patients with OSA symptoms under normal conditions (ie, no apnea events) as training data to train machine learning models. The trained machine learning model can determine whether the subject suffers from OSA symptoms under normal conditions. In other words, even if the subject suffering from OSA symptoms does not have an apnea event, the present disclosure can still accurately predict that the subject has OSA symptoms based on the subject’s electrocardiogram, and further prompt the subject to go to Hospital inspection. In this way, the OSA symptoms of the subject can be detected and treated early.

100: Detection device 110: Processor 120: storage media 121: Training Module 122: Detection module 130: Transceiver 210: Input data 220: Input layer 231, 241: Convolutional layer 232, 242: Pooling layer 251, 252: Fully connected layer 260: Output layer 270: output data S301, S302, S303, S304: steps

FIG. 1 illustrates a schematic diagram of an OSA symptom detection device according to an embodiment of the disclosure. FIG. 2 illustrates a schematic diagram of a convolutional neural network model according to an embodiment of the disclosure. FIG. 3 illustrates a flowchart of a method for detecting OSA symptoms according to an embodiment of the disclosure.

100: Detection device 110: Processor 120: storage media 121: Training Module 122: Detection module 130: Transceiver

Claims (6)

A device for detecting obstructive sleep apnea symptoms includes: a transceiver to obtain a first electrocardiogram, wherein the first electrocardiogram includes a first data set corresponding to non-obstructive sleep apnea symptoms and a first data set corresponding to the obstructive sleep apnea. A second data set of sleep apnea symptoms; a storage medium, which stores a plurality of modules; and a processor, which is coupled to the storage medium and the transceiver, and accesses and executes the plurality of modules, wherein the The multiple modules include: a training module, which uses the first data set and the second data set as training data to train a machine learning model; and a detection module, which obtains the subject’s first data set through the transceiver 2. An electrocardiogram, and judging whether the subject suffers from the obstructive sleep apnea symptom according to the machine learning model and the second electrocardiogram, wherein the training module generates a corresponding heart source according to the first electrocardiogram Respiratory signal, measure multiple RR intervals according to the first electrocardiogram, extract multiple features from the multiple RR intervals or the cardiogenic respiratory signal, generate second training data based on the multiple features, and according to The second training data trains the support vector machine model, wherein the multiple features are associated with at least one of the following: RR interval average value, second or third sequence correlation coefficient of RR interval, number of RR interval pairs, Wherein the RR interval pair includes two adjacent RR intervals, and the time interval between the two RR intervals exceeds 50 milliseconds. The standard deviation of the two adjacent RR intervals, the normalized very low frequency range power of the RR interval, the normalized very low frequency range power of the cardiogenic respiratory signal, the normalized low-frequency range power of the cardiogenic respiratory signal, and the heart source The normalized high-frequency range power of the sexual breathing signal. The detection device according to claim 1, wherein the first electrocardiogram and the second electrocardiogram correspond to non-apnea events. According to the detection device described in claim 1, wherein the machine learning model is a convolutional neural network model. The detection device according to claim 1, wherein the detection module determines whether the subject suffers from the support vector machine model, the machine learning model, and the second electrocardiogram. Symptoms of obstructive sleep apnea. The detection device according to claim 1, wherein the detection module issues an alert through the transceiver in response to determining that the subject suffers from the obstructive sleep apnea symptom. A method for detecting obstructive sleep apnea symptoms, comprising: obtaining a first electrocardiogram, wherein the first electrocardiogram includes a first data set corresponding to non-obstructive sleep apnea symptoms and a first data set corresponding to the obstructive sleep apnea The second data set of symptoms; the first data set and the second data set are used as training data to train the machine learning model; the second electrocardiogram of the subject is obtained; the corresponding heart source is generated according to the first electrocardiogram Sexual breathing signal, according to the The first electrocardiogram measures multiple RR intervals, extracts multiple features from the multiple RR intervals or the cardiogenic breathing signal, generates second training data based on the multiple features, and trains based on the second training data A support vector machine model; and judging whether the subject suffers from the obstructive sleep apnea symptom according to the machine learning model and the second electrocardiogram, wherein the multiple features are associated with at least one of the following: RR interval average value, RR interval second or third sequence correlation coefficient, RR interval pair number, wherein the RR interval pair includes two adjacent RR intervals, and the time interval between the two RR intervals More than 50 milliseconds, the standard deviation of two adjacent RR intervals, the normalized extremely low frequency range power of the RR interval, the normalized extremely low frequency range power of the cardiogenic respiratory signal, and the normalized low frequency of the cardiogenic respiratory signal Range power and normalized high-frequency range power of cardiogenic breathing signals.

Images