404 results about "Cardiac sounds" patented technology

The invention belongs to the technical field of medical signal processing, and particularly relates to a heart sound and electrocardio combined coronary heart disease detection method based on time-frequency fusion, which comprises the following steps: S1, performing data preprocessing on PCG and ECG original signals, including noise reduction, baseline drift elimination, power line interference elimination and the like, and performing normalization processing on the PCG and ECG signals to obtain a normalized PCG and ECG signal; a Welch method is adopted to generate power spectrum signals of heart sound and electrocardio, and time domain signals and power spectrum signals of PCG and ECG are segmented into equal-length fragments. In order to effectively solve the problem of inaccurate diagnosis caused by poor single-modal data quality, a heart sound and electrocardio dynamic fusion module is constructed based on a time-frequency fusion result of two signals, dynamic weight distribution is carried out, a proper classifier is selected for classification, and the diagnosis accuracy is improved by supplementing frequency domain information. And the weight is distributed according to the contribution of the heart sound and the electrocardiosignal, so that the detection and classification result of the coronary heart disease is more accurate.

The invention relates to the technical field of biomedical signal processing and intelligent health monitoring, and particularly discloses an intelligent heart rate monitoring method and system based on heart sound and electrocardiosignals, heart sound signals and electrocardiosignals of a user are collected in real time, quality evaluation is conducted on the collected signals through signal quality indexes, and accurate signals are extracted; extracting time domain and frequency domain features of the first heart sound and the second heart sound of the heart sound signal, fusing multiple features by adopting wavelet transform and a main frequency analysis algorithm, and calculating a heart sound abnormal coefficient to evaluate the abnormal degree of the heart sound signal; the vibration amplitude and frequency characteristics of P waves in the electrocardiosignals are extracted, and an electrocardiograph abnormal coefficient is calculated through a vibration amplitude abnormal coefficient and a vibration frequency fluctuation coefficient and used for evaluating the stability of the electrocardiosignals; based on the heart sound abnormal coefficient and the electrocardiogram abnormal coefficient, a decision tree model is constructed, a heart rate abnormal index is dynamically calculated, dynamic monitoring of the heart rate is achieved, a dynamic early warning mechanism is conducted, and heart rate abnormal information is output.

The invention relates to the technical field of heart sound signal processing, and discloses a heart sound feature recognition method based on wavelet spectral analysis. The method comprises the following steps: collecting heart sound signal data and carrying out signal filtering and segmentation processing to obtain segmented cardiac cycle signals; performing n-level discrete wavelet transform decomposition to obtain an approximation coefficient and n detail coefficients; performing multi-dimensional statistical feature calculation to obtain a multi-dimensional feature vector; performing principal component analysis dimension reduction processing to obtain a dimension reduction feature vector; fitting feature distribution by using a Gaussian mixture model and introducing a dynamic adjustment factor to obtain an adaptive classification threshold; the hidden Markov model is constructed through the dimension reduction feature vectors, the likelihood probability is calculated through a forward algorithm, the likelihood probability is compared with a self-adaptive classification threshold value, and the heart sound feature recognition result is output. The time-frequency features of the heart sound signals are fully utilized, the feature distinction degree and the recognition accuracy are improved, and the recognition accuracy is improved. And meanwhile, the method has good robustness and generalization ability.

The invention discloses a heart sound-electrocardio multi-modal fusion detection method based on federal learning. According to the method, collaborative fusion of multi-modal information of heart sound and electrocardio data is achieved through a cross-modal collaborative attention mechanism, the comprehensive performance of federal learning is effectively improved, and the detection accuracy of a federal model is remarkably improved; meanwhile, by introducing a weight distribution mechanism, the adaptability of the model to the imbalance of each modal data is enhanced; a differential privacy protection technology is introduced to enhance the safe aggregation of model parameters; and the method has better robustness and privacy protection in practical application.

The invention discloses an intelligent electronic stethoscope system based on wireless transmission and a data processing method thereof, and belongs to the technical field of electronic stethoscopes, the intelligent electronic stethoscope system comprises a heart sound acquisition module, a signal conditioning module, a power supply module, a wireless transmission module, an upper computer module and a database module; the microprocessor module serves as a core control unit, and the system can collect heart sound signals of a patient in real time and conduct amplification, filtering and spectrum analysis processing on the heart sound signals. The processed heart sound data is transmitted to a mobile device or a computer in real time through a wireless transmission module, so that doctors and users can check heart sound waveforms and related data conveniently; the system also supports the uploading of the data to a cloud database for remote diagnosis and query; the method not only can meet the requirements of modern medical treatment for high efficiency and convenience, but also can play an important role in improving the diagnosis precision and the remote diagnosis capability and reducing the infection risk.

The invention discloses a heart sound signal noise detection method based on a multi-scale convolutional network. The method comprises the following steps: collecting heart sound signal information; cutting into segments with fixed length as network input; then, the cut heart sound fragments are preprocessed; evaluating the quality of the heart sound fragment based on the spectral density ratio, and correcting the label of the low-quality fragment; generating a spectrogram; effective features in spectrograms of different scales are extracted through a convolutional neural network of a hierarchical multi-scale structure, and accurate recognition and classification of heart noise are achieved; training the multi-scale convolutional neural network; and according to an output result of the multi-scale convolutional neural network, outputting a diagnosis result of the cardiac murmur. According to the method, the hierarchical multi-scale convolutional neural network is introduced, so that multi-scale features are effectively fused, and the accuracy and robustness of heart sound signal noise detection and clinical result classification are improved. And the quality of the heart sound fragments is evaluated based on the spectral density ratio, and the labels of the low-quality fragments are corrected, so that the classification performance is further improved.

The invention discloses a sleep apnea two-stage detection method, and relates to the field of respiration detection, and the sleep apnea two-stage detection method comprises the following steps: collecting a trachea sound signal of sleep respiration of a user, and generating a Mel spectrum of the trachea sound signal; wherein the trachea sound signal comprises a breathing audio signal and a heart sound signal; constructing a sleep-awakening detection model based on a VGG16 network, and calculating the sleep duration of the user; performing sleep apnea section detection on the Mel spectrum by adopting a longitudinal box algorithm AV-Box; extracting S1 peaks and S2 peaks of the potential sleep apnea section, and calculating an RR interval according to the time interval of the adjacent S1 peaks; acquiring statistical characteristics of the RR interval, and performing apnea detection by using a support vector machine (SVM) according to the statistical characteristics; and calculating the sleep apnea hypopnea index AHI of the user according to the sleep apnea times and the sleep duration counted according to the apnea detection result. Aiming at low diagnosis precision in a sleep apnea detection method based on a sound signal in the prior art, the detection precision is improved.

The invention discloses a heart sound feature extraction, decomposition, reconstruction and quantification method which is characterized by comprising the following steps: S1, collecting heart sound signals; s2, carrying out noise reduction pretreatment; s3, carrying out wavelet decomposition and reconstruction on the heart sound signals; s4, calculating signal power spectral density and normalizing an energy spectrum; s5, calculating a signal energy entropy value; and S6, analyzing the correlation between the energy entropy and the defect size. According to the method, visualization of the heart sound digitized signals is achieved through multiple steps, the heart sound digitized signal decomposition and reconstruction method is provided, the waveform distribution condition of the recombined heart sound is visualized, the accuracy is high, and the reliability is high.

The invention discloses a multi-mode-based flexible wearable heart function monitoring method, system and device, and relates to the technical field of health monitoring, and the method comprises the steps: outputting a multi-mode signal through a flexible patch sensing module when the flexible patch sensing module is arranged in a detection area; wherein the flexible patch sensing module comprises an electrocardio sensor, a pulse wave sensor, a heart sound sensor, a strain sensor and a microfluidic biosensor. According to the invention, high-precision synchronous signal acquisition is realized through flexible patch integrated multi-modal sensing, heart multi-source signal features are adaptively extracted in combination with a multi-scale convolutional neural model, and timing sequence relevance is captured by using a bidirectional long-short-term memory network, so that the accuracy and reliability of heart disease diagnosis are remarkably improved, and the accuracy and reliability of heart disease diagnosis are improved. Meanwhile, non-invasive and continuous monitoring and early pathological screening are supported.

The invention discloses a heart sound and electrocardio acquisition and analysis system, which relates to the technical field of biomedical engineering, and comprises a time-frequency analysis module for acquiring a heart sound and electrocardio original data set through an electrocardio and heart sound lead suction ball and an electrocardio and heart sound simulation complete machine and transmitting the data set to an upper computer to execute wavelet decomposition and short-time Fourier transform, the system comprises a heart sound time-frequency spectrum matrix and an electrocardio time-frequency energy matrix output module, a chaotic feature extraction module, the heart sound time-frequency spectrum matrix and the electrocardio time-frequency energy matrix are subjected to phase-space reconstruction, a high-dimensional dynamic track is formed, an improved wolf algorithm is applied to conduct dynamic index calculation on the high-dimensional dynamic track, and a dynamic parameter set is obtained. According to the method, through the improved wolf algorithm and the constructed heart sound space propagation model, the multi-modal fusion capability and analysis discrimination between the heart sound and the electrocardiosignal are improved, and the intelligent level of heart sound and electrocardiosignal collection and analysis is also improved.

An electronic device includes: a transmission unit that transmits a transmission wave; a receiving unit that receives a reflected wave of the transmission wave from the target; and a signal processing unit that detects the distance, direction, and speed of the target on the basis of a converted signal obtained by Fourier-converting a beat signal of the transmission wave and the reception wave. The signal processing unit extracts a signal component corresponding to heart sound vibration accompanying the heartbeat of the subject from the converted signal using a plurality of window functions, and performs frequency analysis on an envelope signal obtained by performing envelope processing on the signal component corresponding to the heart sound vibration. And performing calculation processing of a statistical result on a frequency analysis result, and outputting time sequence data of the heartbeat frequency of the object based on a calculation result.

The invention relates to the technical field of data processing, in particular to a heart and lung sound data monitoring method and system based on big data. The method comprises the following steps: acquiring a curve oscillogram of a heart sound signal and / or a lung sound signal, wherein the curve oscillogram comprises a plurality of data points; taking any extreme point as a target extreme point, acquiring data points on a curve segment between the target extreme point and the single-side adjacent extreme point, and taking the maximum value in absolute values of difference values of slopes of the data points and the adjacent data points as a first deviation index of the data points; determining a first deviation of the target extreme point; obtaining an abnormal score of the target extreme point through an isolated forest algorithm, and determining a second deviation of the target extreme point; determining a filtering window of the target extreme point; and the target extreme point is processed based on the filtering window, and the monitoring result of the heart sound signal and / or the lung sound signal is determined, so that the accuracy of heart and lung sound data monitoring can be effectively improved.

The invention provides a blood pressure detection and classification system and method based on a neural network, a medium and a terminal. Mel-frequency cepstral coefficient feature extraction, first heart sound time domain feature extraction and second heart sound time domain feature extraction are performed on acquired heart sound signal real-time data; the extracted Mel frequency cepstral coefficient feature data, the first heart sound time domain feature data and the second heart sound time domain feature data are used as model features to be input, time domain features and frequency domain features are included, and comprehensive coverage of time domain information and frequency domain information is achieved. The multi-input time domain convolutional neural network model is used for processing the three feature inputs and fusing the three feature inputs, the current blood pressure state of the user can be accurately detected, convenience and rapidness are achieved, and the result is accurate. The algorithm for extracting the frame where the heart sound peak point is located based on the MFCC0 is achieved, the influence of noise can be effectively avoided, the algorithm is simpler and more accurate, meanwhile, higher fault tolerance is achieved, and the first heart sound and the second heart sound can be accurately distinguished.

The invention provides an equipment detection method and device, an upper computer and a system, and the method comprises the steps: playing a noise signal through a loudspeaker under the condition that a bone conduction sensor of wearable equipment collects a heart sound signal of a wearer; determining an environmental noise signal according to a noise signal collected by a microphone at the wearable device; according to the heart sound signals collected by the bone conduction sensor, the interference degree of the environment noise signals on the wearable device is determined.

The invention discloses a heart sound, electrocardio and echocardiogram multi-modal data phase synchronization method, which belongs to the technical field of medical image processing, and comprises the following steps: A1, obtaining an original echocardiogram video and independently recorded heart sound audio data information; a2, preprocessing the acquired data and performing sample automatic construction; a3, a double-path neural network architecture is constructed, electrocardiogram waveform extraction is carried out, and a digital electrocardiogram signal sequence is obtained; a4, periodic feature points of the heart sound and the electrocardio are recognized based on the digital electrocardio signals; and A5, aligning the asynchronous data based on linear phase resampling. According to the method, the video frames, the electric signals and the heart sound signals under different sampling rates are forcibly mapped into the unified phase coordinate system through a linear interpolation algorithm, absolute synchronization of the video frames, the electric signals and the heart sound signals in cardiac pulsation physiological logic is finally achieved, and the influence of individual heart rate differences on data alignment is eliminated.

The invention discloses a thoracic cavity heart sound signal non-contact extraction system and method based on structured light. The system comprises a structured light projection module, an image acquisition module, an image processing and displacement extraction module and a signal processing and trigger point identification module, structured light is projected on the surface of the chest of a human body to form a regular pattern, and a high-speed image acquisition means is utilized to record micro displacement change generated by light spots under cardiac drive. Thoracic cavity vibration information is extracted through an image processing technology, periodic feature points corresponding to a cardiac rhythm are recognized in combination with a time sequence analysis strategy, a thoracic cavity heart sound signal is extracted, rhythm analysis is conducted, and a cardiac trigger point is output to serve as a trigger signal for synchronous control of medical equipment. The technical problems of signal distortion, equipment interference, potential safety hazards, inconvenience in use and the like in a strong electromagnetic environment in a traditional contact mode are solved, and a high-precision synchronous trigger signal is provided for dynamic imaging.

A non-invasive, passive, fully automated heart sound-based system and method is provided that provides estimation of blood velocity, tissue movement, and lumen size parameters for heart assessment. The system uses a computer processor and software to receive PCG acoustic signals from one or more sensors and simultaneously receive electrocardiogram (ECG) signals from one or more sensors attached to the patient. A phonocardiogram (PCG) processing system and method computes metrics of indices of echocardiogram parameters of heart tissue movement and valve blood flow for evaluation.

The invention relates to the technical field of audio signal processing and biomedical signal analysis, and still has a further optimized space for the recognition of anti-noise requirements, signal individual differences and complex pathological modes in a noise environment. The invention provides a double-path CNN heart sound classification method based on time-frequency and double-spectrum fusion features, and the method comprises the steps: carrying out the preprocessing of an original heart sound signal of a data set which is classified into a normal heart sound and an abnormal heart sound, and obtaining a to-be-recognized heart sound signal; based on dynamic continuous wavelet transform, adaptively selecting parameters to extract time-frequency characteristics, introducing bispectrum analysis, capturing nonlinear characteristics, generating a dual-channel characteristic pattern, and efficiently storing the dual-channel characteristic pattern in an HDF5 format; and based on a designed double-path convolutional neural network structure, respectively processing the extracted time-frequency and double-spectrum features, performing classification after fusion, and training a model in combination with category weighted loss and an optimization strategy to obtain a heart sound classification result. The heart sound recognition accuracy can be improved.

The invention relates to a heart health state detection method and device and electronic equipment, and the method comprises the steps: determining a plurality of first feature sets and a plurality of second feature sets in a short-time feature set based on a to-be-detected heart sound signal, each first feature set being determined based on the signal amplitude of the heart sound signal in a corresponding window at each sampling point, and each second feature set being determined based on the signal amplitude of the heart sound signal in the corresponding window at each sampling point; each second feature set is determined based on the duration of a first heart sound interval, a systolic period, a second heart sound interval and a diastolic period included in each cardiac cycle in the heart sound signal in the corresponding window; based on the heart sound signal to be detected, a time domain feature set, a frequency domain feature set and a Poincare map related parameter set in the long-time feature set are determined, the time domain feature set is determined based on the duration of each cardiac cycle, the frequency domain feature set is determined based on signal power, and the Poincare map related parameter set is determined based on a Poincare map generated based on the duration of each cardiac cycle; and inputting the short-time feature set and the long-time feature set into a set heart health state detection model to obtain a heart health state result.

The invention provides a non-contact heart sound monitoring method and system, a terminal and a medium, and particularly relates to the technical field of biomedical engineering.The non-contact heart sound monitoring method comprises the steps that lasers of different wavebands are used for conducting multi-point synchronous monitoring on the heart, and a defocus speckle image is obtained; performing signal extraction on the out-of-focus speckle image to obtain heart movement signals corresponding to multiple wavebands; and filtering the heart movement signals corresponding to the multiple wavebands to obtain heart sound signals. According to the scheme, the characteristic of out-of-focus independent interference imaging of the multi-band laser is utilized, multi-point synchronous monitoring is carried out on micro movement of the heart position in the thoracic cavity through the laser of different bands, the heart sound can be monitored in a non-contact mode, the requirement for measuring environment noise can be lowered, meanwhile, the detection precision is guaranteed, and the detection efficiency is improved. And the comfort of the target living body can be improved.

The invention belongs to the technical field of signal analysis, and discloses an electrocardio and heart sound signal cross-modal bidirectional coupling coronary heart disease detection method and system, and the method comprises the steps: obtaining an electrocardio signal and a heart sound signal of a testee, carrying out the preprocessing, and extracting the time-frequency domain features of the obtained signals; the signal-to-noise ratio of the electrocardiosignal and the heart sound signal is calculated, the signal-to-noise ratio sensing weight is calculated according to the signal-to-noise ratio, cosine similarity calculation is carried out based on the signal-to-noise ratio sensing weight, and the instantaneous form similarity of the time domain features of the electrocardiosignal and the heart sound signal is quantified; calculating mutual information of the frequency domain characteristics of the electrocardiosignal and the heart sound signal, and generating a fusion weight according to the instantaneous form similarity of the mutual information and the time domain characteristics; and inputting the fusion weight into a classifier for coronary heart disease detection. By constructing the bidirectional cross-modal coupling model, a dynamic interaction mechanism between electrical activity and mechanical motion is disclosed, the limitation of traditional single-modal analysis is overcome, and the sensitivity and specificity of early diagnosis of the coronary heart disease are remarkably improved.

The invention provides a heart sound anomaly detection method and device, electronic equipment and a storage medium. An original heart sound signal is acquired and preprocessed into a standard heart sound signal, a plurality of cardiac cycles are divided to form a periodic heart sound signal, and corresponding Mel filter bank coefficient features and Mel frequency cepstral coefficient features are extracted; fusing the Mel filter bank coefficient features and the Mel frequency cepstrum coefficient features, inputting the fused Mel filter bank coefficient features and Mel frequency cepstrum coefficient features into an encoder in a pre-trained MobileNetV2 network, capturing a long-distance dependency relationship in the periodic heart sound signals by using a self-attention mechanism, and outputting a potential space representation vector; and inputting the potential space representation vector into a classifier of the MobileNetV2 network, and outputting a prediction result through a Softmax function. According to the method, the model complexity can be greatly reduced while the model precision is guaranteed, effective features can be extracted while low calculation overhead is kept, and the recognition capability and generalization performance of the model on abnormal heart sounds can be enhanced.

The invention provides a heart sound classification method based on sample amplification and a noise attention network, which relates to the technical field of medical signal processing and artificial intelligence and comprises four steps of data set construction, data processing, heart sound classification model construction and classification result output. Noise outside the heart sound frequency range is filtered out through a band-pass filter and then input into a generative adversarial network to generate heart sound data, then a channel and space attention mechanism is introduced to facilitate extraction of meaningful features by a model, the model is enabled to converge more quickly, and then Gaussian noise with proper standard deviation is injected to obtain heart sound data. The robustness and generalization ability of the noise attention network with the channel and space attention mechanism are improved; along with gradual convergence of training, indexes such as accuracy, specificity and precision of the output cardiovascular disease types gradually tend to be stable, so that auxiliary diagnosis is provided during heart sound auscultation, the cardiovascular disease types are output, and a reliable basis is provided for heart sound auscultation results.

The invention discloses a heart sound data processing method and system based on a time-frequency characteristic graph, and relates to the technical field of electric digital data processing. The heart sound data processing method based on the time-frequency characteristic graph comprises the following steps: steadily dividing heart sound, optimizing and identifying, and identifying and verifying noise. According to the method, feature stationary difference division is carried out on the time-frequency feature map obtained based on the heart sound signals to obtain the heart sound stationary difference feature result, then heart sound optimization processing is carried out in combination with the heart sound stationary difference feature result, the noise level recognition result is obtained, finally, the noise level recognition result is verified, and if verification succeeds, the noise level recognition result is recognized. If yes, feeding back the noise level recognition result, otherwise, updating the noise level recognition result, so that the effect of more accurately recognizing the noise of the heart sound signal is achieved, and the problem that the stability difference of the heart sound signal is not fully considered in the heart sound signal noise recognition process based on the time-frequency feature map in the prior art is solved.

The invention discloses an intelligent heart sound recognition method, system and device and a medium, and the method comprises the steps: obtaining a heart sound signal, and carrying out the preprocessing of the heart sound signal; filtering noise of the heart sound signal; extracting local features and time-frequency features of the heart sound signals; inputting the heart sound signal features into a heart sound recognition model, wherein the heart sound recognition model comprises a first neural network, a second neural network and a Transform model; the first neural network is used for identifying the time-frequency features to obtain a first identification result; the second neural network is used for processing the local features to obtain a second recognition result; the Transform model is used for processing the heart sound signal to obtain a third recognition result; and fusing the first recognition result, the second recognition result and the third recognition result to obtain a heart sound recognition result. According to the method, the heart sound signals are efficiently and accurately recognized based on the heart sound recognition model of machine learning.

The invention discloses an intelligent necklace, and relates to the technical field of intelligent wearable device.The intelligent necklace comprises a necklace body and a pendant, the necklace body defines a wearing hole, and the necklace body can be worn outside a wearer through the wearing hole; the pendant comprises a pendant body, a heart sound sensor and a main control chip, the pendant body is connected to the chain body, a decorative face and a binding face are formed on the two opposite sides of the pendant body respectively, a containing cavity is formed in the pendant body, the heart sound sensor and the main control chip are both contained in the containing cavity, and the detection end of the heart sound sensor faces the binding face. According to the necklace, the heart sound sensor is arranged in the pendant of the necklace, so that the position aligned to the heart sound sensor does not need to be manually adjusted during heart sound detection, the accuracy of heart sound signal detection is improved, meanwhile, the heart sound sensor is perpendicular to the body surface of a wearer, all detected heart sound signals can be fully utilized, and the detection efficiency is improved. Therefore, the detection effect of the heart sound sensor and the accuracy of the detection result are improved.

The invention provides an MEMS heart sound and electrocardio detection system, method and equipment, and relates to the technical field of medical instruments, the MEMS heart sound and electrocardio detection system comprises a bionic cilium heart sound sensor, an electrocardio acquisition circuit and a signal fusion model; the bionic cilia heart sound sensor comprises hollow bullet type bionic cilia and a cantilever beam microstructure; the hollow bullet type bionic cilia and the cantilever beam microstructure are mechanically coupled, and the cantilever beam microstructure comprises a Wheatstone bridge formed by arranging a plurality of piezoresistors and is used for collecting heart sound signals; the electrocardio acquisition circuit and the bionic cilium heart sound sensor are packaged in an acoustic impedance matching mode, and the electrocardio acquisition circuit comprises an electrocardio acquisition electrode plate, a power supply voltage stabilization module, an instrument amplification module, a band-pass filtering module and a notch filter and is used for acquiring and conditioning electrocardio signals; the signal fusion model comprises a Resnet18 feature extractor, an encoder and a mutual cross attention mechanism, and is used for performing feature extraction, linear transformation and feature deep fusion on the collected electrocardiosignals and electrocardiosignals and outputting a signal classification result.

A system and method for monitoring heart function based on heart sounds (HS) is provided. The system includes electrodes configured to sense electrical cardiac activity (CA) signals over a period of time. An HS sensor is configured to sense HS signals over the period of time. The system includes memory to store specific executable instructions and includes one or more processors that, when executing the specific executable instructions, is configured to: identify a characteristic of interest (COI) of a heartbeat from the CA signals. The processors overlay a HS search window onto an HS segment of the HS signals based on the COI from the CA signals and calculate a center of mass (COM) for at least one of S1 or S2 HS based on the HS segment of the HS signals within the search window to obtain a corresponding at least one of S1 COM or S2 COM. The processors calculate at least one of an electromechanical activation time (EMAT) or a systolic interval (SI) based on the at least one of S1 COM or S2 COM and record the at least one of the EMAT or SI.

The invention provides a fetal hypoxia early risk identification method and system based on a large pulse pressure phenotype, and relates to the technical field of biomedical signal processing and heart sound signal analyse.The method comprises the steps that fetal heart sound signals and body surface vibration signals at the abdominal wall of a pregnant woman are collected through a multi-channel sensing array; a preprocessing means is adopted to improve the signal-to-noise ratio of a target signal and suppress an interference signal; in the sliding time window, a large pulse pressure phenotypic feature set and heart sound time phase parameters are extracted; based on the combined criterion, fetal hypoxia risk judgment is carried out; and outputting the graded early warning information, and supporting the playback of the risk event fragment. According to the method, fetal real heart sound mechanical vibration serves as an information source, hypoxia early warning is achieved based on a large pulse pressure phenotype, individualized base lines (including day and night layering) and gestational week self-adaptive correction are supported, the anti-interference capacity is high, the 7 * 24-hour continuous monitoring requirement in a hospital / family can be met, and the sensitivity and specificity of hypoxia recognition are improved.

This document discloses a computer-implemented method for non-invasive assessment of pulmonary arterial hypertension PH from heart sound signals, the method comprising the steps of: receiving sound signals acquired from a beating heart of a subject for a predetermined period of time; generating one or more 2D feature maps, the feature maps including a 2D feature map having the received sound signals, where a first axis of the map is arranged in time and a second axis of the map is arranged on each heartbeat; a pre-trained neural network is applied to associate the generated one or more 2D feature maps with a previously acquired training dataset and to generate training 2D feature maps for a set of PH subjects and a set of non-PH subjects, thus obtaining an indication of the presence of pulmonary arterial hypertension. The invention also discloses a training method and system of the neural network.