179 results about "Sleep staging" patented technology

A method and system for automated sleep staging are disclosed. The method comprises determining at least one physiological signal during a predetermined time period, extracting at least one feature from the at least one physiological signal, and classifying the at least one feature using a machine learning classifier to output at least one sleep stage. The system includes a sensor to determine at least one physiological signal during a predetermined time period, a processor coupled to the sensor, and a memory device coupled to the processor, wherein the memory device includes an application that, when executed by the processor, causes the processor to extract at least one feature from the at least one physiological signal and to classify the at least one feature using a machine learning classifier unit to output at least one sleep stage.

A method for diagnosis of a sleep-related condition of a patient having a thorax. The method includes receiving physiological signals from sensors coupled to the thorax of the patient, and analyzing the physiological signals, independently of any electroencephalogram (EEG) or electro-oculogram (EOG) signals, in order to identify sleep stages of the patient.

This invention relates to systems and methods for treating sleep apnea, which include a first dry electrode for detecting EEG signals of a user, positioned at or near a head of a user; a sleep stage processor for determining a sleep stage of the user based, at least in part, on the EEG signals detected by the first dry electrode, and a pressure delivery device for delivering a controllable stream of air to at least one of a nose and a mouth of the user, the stream of air having a pressure selected based, at least in part, on the sleep stage determined by the sleep stage processor.

The invention discloses an automatic sleep staging method of a single-lead electroencephalogram. A training model comprises a feature extraction module and a staging optimization module, the feature extraction module comprises CNNs (convolutional neural networks) (1) and an Softmax layer (2), the staging optimization module comprises bi-directional LSTM (long-short term memory) recurrent neural networks (3) and a CRF (corticotropin releasing factor) conditional random field model (4), and the CNNs (1), the Softmax layer (2), the LSTM recurrent neural networks (3) and the CRF conditional random field model (4) are sequentially connected. The method only needs the single-lead sleep electroencephalogram, portable and comfortable sleep monitoring requirements are met, temporal and spatial characteristics of the electroencephalogram are sufficiently excavated according to the convolutional neural networks and the recurrent neural networks, the method has dynamic learning capacity and can adapt to great changed environments of diseases, the staging optimization module sufficiently considers relation between the front and the back of N 30s of electroencephalogram data, and the staging accuracy and the generalization ability of the model are improved.

The invention provides an acousto-optic sleep intervention system and method based on electroencephalogram signals. The method includes the steps: acquiring, processing and analyzing the electroencephalogram signals in real time during sleep through a universal electroencephalogram acquisition system; staging sleep by a data mining method; intervening through the sound and light in the REM (rapid-eye-movement) period of sleep and shallow sleep to achieve the purpose of improving sleep quality. The system includes a sleep electroencephalogram acquisition module, a sleep electroencephalogram staging module, and a sleep intervention module; the sleep electroencephalogram acquisition module is used to acquire the electroencephalogram signals that have continuously changed during sleep, the electroencephalogram signals after the output, and the extracted electroencephalogram characteristic information; the sleep electroencephalogram staging module is used to staging the sleep electroencephalogram characteristic information with trained sleep staging models; the sleep intervention module is used for selecting a suitable sleep intervention program based on the staging results, and the subject undergoes double intervention of sound and light.

The invention discloses a sleep quality judging method and a sleep instrument. The method comprises the following steps: according to the sleep stage of a tested subject, determining a sleep structure score of the tested subject; according to the sleep state of the tested subject, determining a sleep state score of the tested subject; according to the sleep habit of the tested subject, determining a sleep habit score of the tested subject; according to the sleep environment of the tested subject, determining a sleep environment score of the tested subject; according to one or more of the sleep structure score, the sleep state score, the sleep habit score and the sleep environment score, calculating a sleep quality score of the tested subject; and according to the sleep quality score, determining the sleep quality of the tested subject.

The invention relates to a single lead electroencephalogram sleep automatic staging method based on Stacking, and belongs to the field of machine learning algorithms. The method comprises the steps that S1: sleep electroencephalograms are preprocessed; S2: multi-feature extraction and screening are carried out on the sleep electroencephalograms; S3: machine learning classification is carried out;S4: sleep automatic staging is carried out. The method can acquire a filtering method combining a wavelet function of a self-adaption threshold value and an IIR filtering function to conduct noise reduction processing on the electroencephalograms and effectively improve the signal to noise ratio of the electroencephalograms; a feature algorithm can be optimized and screened to acquire a new feature parameter set so as to take the new feature parameter set as a feature of sleep staging; a new multi-feature and integrated learning algorithm composition with high accuracy can be acquired and taken as the sleep staging method.

The invention discloses a real-time sleep staging detection method based on a piezoelectric sensor and a device for realizing the method. According to the method and the device, a sleep piezoelectric sensing mattress is adopted for collecting BCG (Ballistocardiographic, namely, the ballistocardiography) signals of long-term sleep, and the signals are mixed with the information including the heartbeat, the breath and the body motion during the sleep process. The sleep staging algorithm mixed with multiple signals is adopted, the hidden markov model is applied to the research for the sleep breathing data, by utilizing the advantages of the hidden markov model mixed with the heartbeat and the breath rate at different stages, and in combination with the body motion detecting algorithm, the influences of the body motion for the staging based on the heartbeat and the respiratory signals are reduced, meanwhile, the discernibility of the body motion detecting algorithm for the waking state and the sleep state is enhanced, and further, the long-time automatic sleep staging accuracy rate based on the piezoelectric sensing signals is improved; the algorithm is deployed to a server, so that the automatic sleep staging is realized.

The invention discloses a multi-sensor data decision level fusion-based sleep staging method. The method comprises the steps of firstly collecting radar and audio data of all night by using a radar sensor and an audio sensor, and extracting radar and audio signal features; secondly according to a feature number, performing data classification on the features, and according to the data classification, building a radar residual fragment model and a radar and audio fragment model; thirdly performing identification classification on radar features in the radar residual fragment model to obtain a model prediction result 1, and performing identification classification on radar and audio features in the radar and audio fragment model by using a classifier to obtain model prediction results A andB; fourthly performing decision making on the model prediction results A and B by using a naive Bayesian model to obtain a model prediction result 2; and finally performing time sequence splicing on the model prediction results 1 and 2 to obtain an all-night sleep staging result. The method is simple and feasible, is high in accuracy and meets an actual condition.

The invention discloses a multi-sensor feature optimization algorithm-based sleep staging method. The method comprises the steps of firstly performing signal acquisition on a continuous wave radar sensor and an audio sensor at the same time; secondly performing digital signal processing on signals to obtain vital sign signals including breathing, heartbeat and body movement, and a snore signal; thirdly performing feature extraction, adjusting weights of features in a fused feature set by utilizing a feature adjustment optimization algorithm, and training a classifier; and finally performing sleep staging by using the trained classifier. The method is effective and feasible, is reliable in performance, slightly influences sleep of a user, and accurately performs assessment on the sleep of the user.

The invention provides an electroencephalogram sleep staging method based on a deep convolutional neural network. The electroencephalogram sleep staging method comprises the following steps that S1, sleep signals of a subject are collected, and multi-lead electroencephalogram signals in the sleep signals are extracted; S2, performing data preprocessing on the electroencephalogram signals; S3, constructing and training an end-to-end deep convolutional neural network classifier; and S4, performing electroencephalogram sleep staging by using the deep convolutional neural network classifier. Compared with the conventional CNN electroencephalogram sleep staging method, the electroencephalogram sleep staging method provided by the invention has the advantages that under the condition of the sameiteration times and learning rate, each batch of the model adopts higher data, and the obtained output result is more stable. In terms of accuracy and F score, the method provided by the invention has better classification performance.

The invention discloses a non-contact sleep staging method. The non-contact sleep staging method comprises the following steps of 1, filtering radar echo signals, and obtaining a respiration signal and a heartbeat signal separately; 2, according to the respiration signal, obtaining respirations per minute (RPM), variance of the respiration per minute (RPM Variance), according to the heartbeat signal, obtaining heartbeats per minute (BPM), variance of the heartbeat per minute (BPM Variance), and according to the respiration signal, obtaining a body movement signal; 3, conducting cluster analysis on the RPM, the RPM Variance, the BPM, the BPM Variance, and the body movement signal separately; 4, according to a sound signal, a video signal and obtained signals after the cluster analysis in the step 3, acquiring a division result of sleep states. According to the non-contact sleep staging method, medical workers can be helped to take treatment measures in time according to real sleep situations of patients. The non-contact sleep staging method has the advantages of being high in safety, high in precision, small in size and intelligent.

The invention belongs to the technical field of sleep monitoring, and relates to a single-lead electroencephalography (EEG) automatic sleep staging method. The method comprises the following steps: (1) performing signal pre-processing; (2) extracting classification features; (3) carrying out sleep staging. The method provided by the invention has the advantages that 1, a pre-processing algorithm is designed to obtain single-lead EEG signals with better quality; 2, the method extracts multiple features from the time domain, the frequency domain and the nonlinear domain, and screens out the representative features; 3, a random forest model is adopted, so that over-fitting does not need to be worried about; the method has a good anti-noise ability, and the staging results of all decision trees can be obtained, so that the confidence probabilities of the random forest for all sleep stages are obtained; 4, a D-S evidence theory is combined, so that the accuracy rate of the sleep staging isfurther improved.

The invention discloses a non-contact type sleep stage classification and sleep breathing disorder detection method. The non-contact type sleep stage classification and sleep breathing disorder detection method includes the detection steps that 1, when a person is detected, the detected person lies in bed, antennae of a wireless reception and transmission machine are arranged right above the detected person or askew above the detected person, and a reception antenna and a transmitting antenna are arranged in parallel; 2, the wireless reception and transmission machine performs digital signal processing and mode recognition on received signals, and finally records and reports disorder time. Wireless signals transmitted by using the non-contact type sleep stage classification and sleep breathing disorder detection method are low in power, within 20mw, and harmless to human bodies. Due to the fact that non-contact type measurement is used in the non-contact type sleep stage classification and sleep breathing disorder detection method, the non-contact type sleep stage classification and sleep breathing disorder detection method is convenient to use, and facilitates long time dynamic monitoring and testing.

The invention provides a sleep staging method based on a BCG (ballistocardiogram) signal. The method comprises the steps as follows: extracting a heart rate signal, a pulse signal and a respiration signal from the BCG signal acquired in advance, and calculating heart rate variability (HRV) and cardiopulmonary coupling (CPC) power spectrum according to the heart rate signal, the pulse signal and the respiration signal; acquiring change trend signals of frequency domain characteristics of the HRV and the CPC power spectrum, converting the change trend signals of the frequency domain characteristics of the HRV and the CPC power spectrum into multichannel image characteristics, performing characteristic extraction of the HRV and the CPC power spectrum, and extracting characteristic vectors according to time domain signals of the extracted HRV and CPC power spectrum; and training sleep staging results corresponding to HRV and CPC power spectrum density signals according to the multichannelimage characteristics and characteristic vectors. The sleep staging method combines the HRV and the CPC power spectrum to distinguish sleep stages and has higher adaptability and accuracy.

The invention relates to a method and device for determining sleep staging. The method includes the steps that a current-moment sleep characteristic signal of a user is collected; according to the current-moment sleep characteristic signal, the current-moment sleep characteristic parameter is obtained; whether the current-moment sleep characteristic parameter belongs to the range of the sleep characteristic parameter corresponding to the that-night history sleep stage or not is judged, wherein the that-night history sleep stage includes the awakening stage, the light sleep stage and the deep sleep stage; when the current-moment sleep characteristic parameter belongs to the range of the sleep characteristic parameter corresponding to the first sleep stage, it is determined that the current moment of the user is located in the first sleep stage, wherein the first sleep stage is the light sleep stage or the deep sleep stage. By means of the method and device for determining sleep staging, judgment of sleep staging can be carried out in real time, and different from a traditional method that sleep staging judging is carried out after sleep is completed, sleep intervention can be more conveniently carried out.

The invention discloses a sleep staging detection method. The sleep staging detection method comprises the steps that human's heart rate variability signals and wrist pulse triaxial acceleration dataare collected; the heart rate variability signals and the triaxial acceleration data are partitioned according to time spans, a plurality of feature parameters are extracted from the heart rate variability signals and the triaxial acceleration data in each time span; the feature parameters in one time span are input into a sleep staging detection model, and a sleep stage in one time span is obtained; the sleep stages in all the time spans are counted so as to obtain the sleep stages in the whole sleep time. The invention further discloses a wearable sleep staging detection device. In the sleepstaging detection method and the wearable sleep staging detection device, the wrist pulse acceleration information and the heart rate variability signals in the sleep process of the human body are integrated, the sleep staging accuracy can be improved, and the sensitivity can be enhanced.

The invention provides an automatic sleep staging method, device, computer equipment and computer readable storage medium. The method comprises the following steps: acquiring sleep data to be processed, wherein the sleep data includes electroencephalogram (EEG) signals, ocular electrical signals and mandibular muscle electrical signals; and inputting the sleep data into an automatic sleep stagingmodel, and outputting sleep stages of the sleep data, wherein, the automatic sleep staging model is obtained by using known sleep data and the sleep stages of the known sleep data as a sample for training a classifier. Because the EEG signals, ocular electrical signals and mandibular muscle electrical signals are used, compared with a method of automatic sleep staging based on single-channel EEG in the prior art, the automatic sleep staging process of the scheme can help to improve the accuracy of automatic sleep staging.

The invention discloses an intelligent sleep aid system based on electroencephalogram monitoring and a sleep headphone thereof, and belongs to the field of intelligent sleep aid. According to the intelligent sleep aid system, on the basis of monitoring the electroencephalogram situation when a user sleeps, electroencephalogram signals are collected, and the electroencephalogram signals are specially classified; the energy ratio of beta wave to alpha wave to theta wave to delta wave serves as a characteristic value, incremental learning is conducted by using a support vector machine, the sleepof the user is staged, and while the classification accuracy is ensured, the time complexity of a sleep staging algorithm is reduced as much as possible. The sleep headphone of the intelligent sleep aid system can accurately and effectively achieve staging, according to requirements of different users, the music type of the headphone can be set, the volume size can be adjusted, so that the electroencephalogram (EEG) of the user enters a deep sleep stage or a REM stage, that is, the user enters the optimal sleep state.

An automatic sleep staging method based on LSTM and using multiple physiological signals comprises the following steps: step 1, signal acquisition: acquiring an ECG (electrocardiogram) signal, a respiration signal and an acceleration signal of a testee; step 2, signal processing; step 3, classification feature extraction; step 4, model construction: inputting artificially extracted features in a first-layer LSTM model, taking an output probability of the first-layer LSTM model as new features, and inputting the new features into a second-layer LSTM model together with the artificially extracted features to construct classifiers for different classification tasks; and step 5, applying trained models to classification of sleep staging. Multiple physiological signals, including the ECG signal, the thoracic-abdominal respiration signal and a head acceleration signal but not including an EEG (electroencephalogram) signal, are adopted in the invention, and defects caused by application of EEG to sleep staging are overcome; and meanwhile, the LSTM models are adopted and suitable for big samples and big data, besides, the temporal correlation between sleep events is considered, and the accuracy and reliability of sleep staging are improved.

The invention provides a random forest-based sleep staging method and system. The method is as follows: acquiring a plurality of EEG signal samples and performing sleep staging on each EEG signal sample; preprocessing each EEG signal sample; segmenting each pre-processed EEG signal sample into a plurality of segments having a certain duration; extracting a plurality of characteristic parameters oneach segment, wherein the plurality of characteristic parameters include time domain characteristic parameters, frequency domain characteristic parameters, and non-linear characteristic parameters; training a plurality of decision trees in a random forest classifier with the extracted characteristic parameters as the characteristic vectors to form a random forest model; acquiring an EEG signal tobe analyzed; after preprocessing, segment segmentation and characteristic extraction, inputting the EEG signal to be analyzed into the random forest model to obtain sleep staging of the EEG signal tobe analyzed.