25 results about "Blood volume pulse" patented technology

The invention provides an iPPG blood pressure detection method based on deep learning. According to the method, a transfer learning strategy is adopted, and the problem that deep learning training cannot be completed due to scarcity of iPPG data is solved. According to the method, the model is pre-trained by using existing PPG data with abundant types and sufficient data volume, so that the model preliminarily understands the implicit relationship between the blood volume pulse wave and the blood pressure, and then the pre-trained model is finely adjusted by using self-collected iPPG data, so that the model is more adaptive to an iPPG blood pressure detection task. Before the data are input into the model, the data are classified according to a volume pulse wave waveform characteristic quantity K value and divided blood pressure intervals, and each interval is independently trained to adapt to different types of blood pressure detection. During actual deployment and use, a personalized model is automatically allocated to carry out subsequent blood pressure detection on a subject by taking a true value measured for the first time of the subject as a benchmark.

The invention provides a non-contact physiological and psychological health detection device based on AI. The non-contact physiological and psychological health detection device comprises a cabinet, a host, a liquid crystal display, a high-precision camera, a multi-wavelength RGB camera, a light supplement lamp, an identity card swiping module, a social security card swiping module, an IC card swiping module, a printer and a power supply. A high-precision camera is adopted to collect a face image of a human body, and face parameters are obtained based on the face image; a multi-wavelength RGB camera is adopted to detect subtle color changes caused by human skin surface pulses, an RPPG technology is utilized to extract blood volume pulse signals, then physiological parameters are obtained, and a deep learning network is utilized to analyze and process the facial parameters and the physiological parameters in real time so as to evaluate the health state of a user. And data visualization is carried out on the result, the facial parameters and the physiological parameters output by the AI analysis module, and the data are displayed in a liquid crystal display, so that a user can better know the health state of the user.

An advanced cardiovascular health monitoring system utilizes an optical pulse wave velocity detection device in combination with a system for analyzing data waveforms structures in combination with personalized patient data integration to determine specific physiological parameters for monitoring and assessing specific health conditions. Using acceleration pulse waves, vascular health is calculated by the ratio of peak to the second derivative of blood volume pulse wave, which can detect various health indices such as vascular age, vascular health index, and arteriosclerosis. The device collects separate pulse points within a time interval. It connects them into a line to form the PPG Image, which is then analyzed through data filtering and fitting methods. The periodic cycle acquisition method excludes abnormal cycle data and ensures correct data acquisition. Artificial Intelligence and data analytics are aiding the software to improve and optimize underlying data ranges continuously; the algorithms are summarized in figures. The invention may be used as part of non-invasive screening testing.

The invention discloses a face video high-quality blood volume pulse signal prediction method based on end-to-end deep learning. The face video high-quality blood volume pulse signal prediction method comprises the following steps: 1) acquiring and preprocessing a face video and a reference signal; 2) building a blood volume pulse signal prediction network, and processing the preprocessed face image sequence to obtain a predicted blood volume pulse signal; and 3) setting a loss function training network to obtain an optimal blood volume pulse signal prediction model for predicting a high-quality blood volume pulse signal. According to the method, the face video is mapped into the high-quality blood volume pulse signal through the end-to-end deep learning network model, so that high-quality physiological signal guarantee is provided for a non-contact physiological parameter detection technology.

The invention discloses a non-contact heart rate data acquisition method. The non-contact heart rate data acquisition method comprises the following steps: reading a video image of a latest frame in a face video stream of a person to be monitored; performing face detection and key point positioning on the read video image by adopting a face detection and key point positioning model to obtain a plurality of face key points of the video image; according to a preset screening rule, rejecting invalid face key points, and constructing at least one irregular region of interest for the video image by using the valid face key points; in each irregular region of interest, extracting and calculating RGB data of the video image; repeating the previous steps to obtain RGB data of continuous multiple frames of video images, and reconstructing the blood volume pulse wave of the person to be monitored according to the RGB data of each frame of video image; and analyzing the blood volume pulse wave, and calculating heart rate data of the person to be monitored. According to the non-contact heart rate data acquisition method, motion artifacts can be effectively inhibited, and the positioning reliability of the region of interest is improved. The invention further discloses a fatigue driving monitoring method and device based on the non-contact heart rate data acquisition method.

A sensor device or system for performing a blood pressure measurement comprises a finger abutment surface, a first sensor operative to sense blood volume pulses in a finger pressed on the finger abutment surface, a second sensor operative to sense a force or pressure with which the finger is pressed on the finger abutment surface, and processing circuitry operative to process amplitudes of the sensed blood volume pulses and the sensed force or pressure with which the finger is pressed on the finger abutment surface to determine an absolute blood pressure.

The present invention proposes an AI-based non-contact physiological and psychological health detection device, comprising a cabinet, a host, a liquid crystal display, a high-precision camera, a multi-wavelength RGB camera, a fill light, an ID card swiping module, a social security card swiping module, an IC card swiping module, a printer, and a power supply; a high-precision camera is used to collect facial images of a human body, and facial parameters are obtained based on the facial images; a multi-wavelength RGB camera is used to detect subtle color changes caused by pulses on the surface of human skin, and RPPG technology is used to extract blood volume pulse signals to obtain physiological parameters, and a deep learning network is used to perform real-time analysis and processing of facial parameters and physiological parameters to evaluate the health status of a user; and the results, facial parameters, and physiological parameters output by the AI ​​analysis module are visualized and displayed on the liquid crystal display, so that the user can better understand his or her own health status.

The invention relates to a remote heart rate detection method based on a multispectral imaging technology. The method comprises the following steps of: acquiring a multi-spectral image sequence of the face of a testee by using a multi-spectral camera; separating and extracting the multispectral image sequence into a plurality of single-channel image sequences; extracting a target face region of interest of the testee from the plurality of single-channel image sequences by using a face detection algorithm, and performing face region clipping on the image sequences according to the target face region of interest; preprocessing each cut single-channel image sequence, and constructing standardized spectrum data; extracting independent components from the standardized spectrum data by using an independent component analysis method; selecting a signal corresponding to the maximum power spectral density from the independent components as a preliminary BVP signal; and filtering the preliminary BVP signal to obtain a final blood volume pulse signal. According to the non-contact detection method, interference of ambient light can be better eliminated, and the accuracy and robustness of blood volume pulse signal extraction are improved.

The invention relates to the technical field of sleep monitoring, and discloses a sleep diagnosis device for detecting sleep events by monitoring blood volume pulse waves, which comprises a host and a bandage for attaching the host to the skin of a patient, wherein the host comprises a power supply module, an optical sensor, an acceleration sensor, a control circuit and a wireless communication interface. And the power supply module is used for providing power for the host. The optical sensor is arranged on the side, attached to the skin of a patient, of the host and comprises at least one light emitter and at least one light receiver, and the light emitters and the light receivers are located on the same side of the host and used for detecting light reflected by tissue of the patient and outputting corresponding blood volume pulse wave signals. The device is compact in structure, skin reflected light of a patient is directly monitored through the reflective optical sensor, interference of electrode or circuit connection in a traditional method on sleep of the patient is reduced, and the wearing comfort and sleep quality of the patient are improved.

An optical vital signs sensor (100) is provided, which comprises a PPG sensor (102), a force transducer (130) configured to measure a force applied via the PPG sensor (102) to a skin (1000) of the user and a processing unit (140) configured to extract information regarding a blood volume pulse from the output of the PPG sensor (102) and to map the extracted information to the blood pressure value at a specific force applied to the PPG sensor (102).

The invention provides a non-contact heart rate measurement method based on frequency domain learning. The method comprises the following steps: firstly, extracting spatio-temporal features containing physiological rhythm information from an input video through a feature extraction network; then spatial redistribution is carried out on the spatial-temporal features, and spatial dimension information is embedded into channel dimensions; the method comprises the core steps that a frequency spectrum adaptive learning module is adopted to convert a redistributed feature map into a frequency domain, adaptive learning is carried out on a signal, key frequency components related to a real physiological rhythm are enhanced, and noise interference is suppressed; finally, a prediction signal is output through a differential blood volume pulse prediction network, a blood volume pulse signal is reconstructed through integration, and the heart rate is accurately estimated. According to the method, the accuracy of heart rate estimation and the stability in a complex environment are remarkably improved, the calculation cost is low, continuous and non-inductive health state monitoring in daily life is facilitated, and the method has wide application prospects in the fields of remote medical treatment and personalized health management.

A portrait image processing method, applying to a portrait image processing device, comprising: (a) acquiring a BVP (Blood Volume Pulse) signal of a user; (b) acquiring blood flow distribution of a specific portion of a user; (c) capturing an image of a specific portion of the user; and (d) adjusting the image of the specific portion according to the BVP signal and the blood flow distribution. A portrait image processing device, which comprises a camera and a processing circuit, is also disclosed. The processing circuit can acquire BVP signals from an image captured by the camera or from a PPG sensor.

A system is configured to diagnose sleep comprising an apparatus configured to be attached to a patient; the apparatus comprises: i) an optical sensor configured to measure a blood volume pulse of the patient; and ii) a wireless communication interface configured to wirelessly transmit the measured blood volume pulse. The system further comprises means for performing: i) obtaining the wirelessly transmitted measured blood volume pulse; and ii) deriving from the blood volume pulse the peripheral arterial tone; iii) determining occurrences of sleep events from changes in the peripheral arterial tone.

A wearable device is wearable on an appendage of a user and includes a force-sensitive display screen to generate force data when a force is applied to the force-sensitive display screen, a photoplethysmogram (PPG) sensor, and one or more processors. The PPG sensor is positioned below an external surface of the force-sensitive display screen and generates PPG sensor data when the force is applied to the force-sensitive display screen. The one or more processors are configured to: determine, from the PPG sensor data and the force data, a variation of PPG blood volume pulse amplitude with a variation in pressure applied to the appendage, and calculate an estimate of blood pressure of the appendage from the variation of PPG blood volume pulse amplitude with the variation in the pressure applied to the appendage.

The invention relates to the technical field of sleep monitoring, and discloses a blood volume pulse wave-based sleep diagnosis method, which comprises the following steps of: 1, measuring PPG waves of a patient through an optical sensor; step 2, wirelessly transmitting the PPG wave signal obtained by measurement to an electronic terminal; step 3, applying a preset signal processing algorithm on the electronic terminal to process the PPG wave signal, wherein the signal processing algorithm comprises the following steps: eliminating 50 / 60Hz power frequency interference by using an LMS filter; baseline drift correction is carried out through morphological filtering to remove respiratory rhythm interference, and the length of a structural element is larger than or equal to 2 seconds; the motion artifact elimination technology is adopted, independent components of PPG wave signals are separated, and components with the acceleration correlation larger than 0.8 are removed. According to the invention, the signal-to-noise ratio is improved by 15dB based on multi-stage anti-interference processing of PPG signals, and the data reliability is guaranteed. High-precision event identification is realized by innovatively fusing multiple parameters; the autonomic nerve wakefulness is identified by utilizing the PPG pulse interval, and electroencephalogram dependence is replaced.

A high intraocular pressure discrimination method based on eye video dynamic fusion double-branch complementary features belongs to the technical field of physiological signal detection, and the method comprises the following steps: firstly, collecting eye videos and synchronously acquiring intraocular pressure labels to construct a training data set; secondly, using a trained semantic segmentation model to respectively segment the iris-pupil region and the sclera region from the eye video frames, and extracting the RGB time sequence signals of each region; thirdly, further extracting the blood volume pulse (BVP) signals from the two RGB signals respectively, segmenting by using a sliding window and converting into a two-dimensional space-time feature map; and finally, constructing and training a high intraocular pressure discrimination model DB-CFFNet, which respectively inputs the two space-time feature maps, realizes the classification and discrimination of the intraocular pressure state through double-branch feature extraction and fusion. The method realizes low-cost and continuous dynamic intraocular pressure state monitoring based on a common camera.

A heart rate detection method for compressed video, comprising the following steps: 1) estimating skin reflection model parameters; 2) synthesizing a noise-free observation signal; 3) extracting a blood volume pulse (BVP) signal. Based on the skin reflection model, the model parameters are estimated using the observation signal, and finally a noise-free observation signal is generated, followed by a POS pulse signal extraction algorithm, and finally a pulse signal output is obtained. The present application can reduce the cost of video storage and transmission under the premise of ensuring the high quality of rPPG signal, and provide technical possibility for remote home health care, cloud video data processing and other remote application scenarios, greatly expand the practical application range of rPPG.

The invention discloses a near-infrared video-based driver heart rate monitoring method and electronic equipment, and belongs to the field of non-contact heart rate monitoring, and the method comprises the following steps: collecting a face near-infrared image of a driver in real time; face feature points are recognized from the face near-infrared image, and a plurality of ROIs are segmented from the face near-infrared image based on coordinates of the face feature points; averaging the pixels in each ROI so as to obtain an original single-channel rPPG signal; the original single-channel rPPG signal is converted into a multi-channel rPPG signal through time dimension extension; extracting a blood volume pulse (BVP) signal from the multi-channel rPPG signal through an orthogonal image decomposition method; obtaining a heart rate result corresponding to each ROI according to the BVP signal; and screening and fusing the heart rate results through a heart rate clustering method based on a Gaussian mixture model (GMM) to obtain the real-time heart rate of the driver. According to the invention, the accuracy and robustness of monitoring the heart rate of the driver in the driving scene can be improved.

A method, system, device and medium for contactless heart rate detection of a person's side face based on rPPG. The method captures a person's side face with a camera, detects key points of the person's side face with MediaPipe, and obtains all key points of the person's side face; then determines an ROI (region of interest) for heart rate detection, extracts signals of the GREEN channel and RED channel in the RGB image from the image data within the ROI using an rPPG algorithm, and weightedly combines them as a BVP (blood volume pulse) signal; processes the BVP signal to obtain the person's heart rate information, and displays it on an interface, thereby realizing contactless measurement of the person's heart rate by detecting the person's side face; the system, device and medium realize contactless heart rate detection of the person's side face based on the rPPG method; the present invention can still detect the person's heart rate in real time even when the front of the person's face may be blocked or the angle is not conducive to detection, and has the advantages of convenience, high adaptability and high accuracy, while avoiding the inconvenience and infection risk that may be caused by traditional contact detection.

The application discloses a video heart rate detection method for removing irregular motion artifacts, which is characterized by the following steps: firstly, extracting the time sequence of green channel pixel mean value in the face region of interest in the video, mainly including the positioning, division and screening of the region of interest, tracking the region of interest and extracting the time sequence of green channel pixel mean value; secondly, extracting the heart rate signal, including signal preprocessing, using the method combining non-negative matrix factorization (NMF) and independent vector analysis (IVA) to perform blind source separation on the regular heart rate source signal, selecting the blood volume pulse signal; finally, calculating the heart rate, after performing Fourier transform on the blood volume pulse signal, the heart rate value is obtained through screening and calculation. The application extracts the heart rate signal from the video image, is used for heart rate detection under the condition of irregular motion artifacts, can improve the accuracy of heart rate detection under the condition of irregular motion, and thus provides a new method for non-contact heart rate detection research.

The present application relates to a blood pressure measuring device based on multi-frequency modal fusion, comprising: a data preprocessing module for extracting a space-time feature map from a face video; a multi-frequency modal signal separator for decomposing the space-time feature map into an initial multi-frequency band signal, the initial multi-frequency band signal comprising n signal components; a signal refining network for performing multi-scale feature extraction on the n signal components respectively, thereby corresponding to generate n high-dimensional intrinsic modal signals; a signal reconstruction network for performing waveform reconstruction on the n high-dimensional intrinsic modal signals respectively, thereby obtaining n high-dimensional signals; and a multi-frequency modal signal fusion device for selecting effective components from the n high-dimensional signals for fusion, and determining the positions of wave crests and troughs to obtain a blood volume pulse wave signal. The present application is based on a lightweight short-time heart rate real-time monitoring network of multi-modal fusion, which can refine effective signal features with a small number of parameters.

The present invention relates to a pilot physiological state monitoring and control system based on brain-computer interaction, belonging to the field of brain-computer interaction, and solves the problems of the instability of signal acquisition, the complexity of real-time processing, and the practicality of wearable monitoring devices in the existing brain-computer interaction for aviation applications. It includes a head-mounted physiological monitoring module and a flight control center module; the head-mounted physiological monitoring module is used to monitor the physiological state of the pilot in real time, including blood volume pulse wave PPG signal, regional cerebral oxygen saturation rScO2, and six-axis acceleration signal data, and conduct risk early warning and flight state assessment based on the real-time monitoring data; the flight control center module is used to receive the risk early warning data of the pilot for flight control, store the physiological state data and physiological monitoring logs, provide visualization of the real-time risk early warning and flight state assessment results, and early warning voice broadcast to ensure flight safety. Safe flight is achieved.

The present invention discloses a contactless heart rate detection method, system, and medium based on facial feature point detection. The method comprises: obtaining a facial video image and determining a facial region based on the facial video image; inputting the facial region into a facial feature point detection network model to obtain facial feature points; the facial feature point detection network model comprises a feature extraction portion and a multi-scale fusion prediction portion, wherein the feature extraction portion comprises a standard convolutional layer and multiple block layers, wherein the block layers are depthwise separable convolution modules proposed by MobileNetV3; the multi-scale fusion prediction portion enhances the expressiveness of the model by fusing multiple images of different scales; determining a region of interest based on the facial feature points, thereby obtaining an observation signal; and using a pulse wave extraction algorithm to separate a blood volume pulse wave signal from the observation signal, thereby calculating the heart rate value. The present invention utilizes facial feature point detection to determine the region of interest, thereby improving detection accuracy.

The present invention belongs to the field of monitoring the health status of flight crews in the cockpit, and particularly relates to a monitoring system for the incapacitation status of flight crews based on contact sensors, including a wearable wrist sensor, a wearable chest strap sensor, a seat pressure sensor, a data synchronization module, an intelligent computing platform, a voice acquisition and recognition module, and a status warning module. Through contact multi-sensor fusion technology and algorithms, the present invention can accurately monitor and analyze the ECG electrocardiogram signal, PPG blood volume pulse signal, SPO2 blood oxygen saturation signal, SKT skin temperature signal, and HR heart rate signal of pilots, realizing real-time monitoring of the incapacitation status of flight crews and improving the safety of the single-pilot driving mode.

The invention discloses a human side face non-contact heart rate detection method, system and device based on rPPG and a medium, and the method comprises the steps: capturing the side face of a person through a camera, carrying out the key point detection of the side face of the person through Mediape, and obtaining all key points of the side face of the person; the method comprises the following steps: firstly, determining an ROI (region of interest) for heart rate detection, extracting signals of a GREEN channel and an RED channel in an RGB image from image data in the ROI through an rPPG algorithm, weighting and combining the signals as bvp (blood volume pulse) signals, processing the bvp signals to obtain heart rate information of a person, and displaying the heart rate information on an interface. The non-contact measurement of the heart rate of a person is realized by detecting the side face of the person; the system, the device and the medium realize human side face non-contact heart rate detection based on the rPPG human side face non-contact heart rate detection method. Under the condition that the front face of the human face is possibly shielded or the angle is not beneficial to detection, the heart rate of the person can still be detected in real time, the method has the advantages of being convenient, fast, high in adaptability and high in accuracy, and meanwhile inconvenience and infection risks possibly brought by traditional contact type detection are avoided.