55 results about "Motion artifacts" patented technology

This invention discloses a preprocessing method for highly polluted pediatric EEG data, belonging to the field of EEG signal processing technology. The invention proposes a robust motion artifact detection method based on a multi-channel voting mechanism. This method overcomes the sensitivity of single-channel detection to transient noise through multi-channel joint decision-making, automatically identifying time periods requiring restoration. A comprehensive preprocessing workflow integrating PCHIP interpolation restoration, adaptive Bayesian wavelet denoising, bandpass filtering, and independent component analysis is designed. This workflow is validated using real pediatric continuous task EEG data. This invention demonstrates excellent performance in preserving frequency band information and improving the weighted signal-to-noise ratio, achieving an average accuracy of 96.4% (205 test cases) and a maximum of 100% based on permutation entropy feature classification. This invention effectively solves the preprocessing challenge of pediatric EEG data in high-noise environments, significantly improving the accuracy and reliability of subsequent classification tasks.

The present invention relates to a method for detecting and removing a motion artifact of a functional near-infrared spectroscopy signal. Disclosed are a method and a device for detecting and removing a motion artifact of a near-infrared spectroscopy signal in real time on the basis of an artificial neural network model, the method comprising: a conversion step of converting time series data of functional near-infrared spectroscopy signals measured through a plurality of channels from a target into image data; a detection step of detecting whether noise is present in the converted image data using a pretrained detection model; and a removal step of removing noise from image data where noise has been detected using a pretrained noise removal model.

This invention discloses a high-precision phase recovery method for tooth surface fringe images during gear rotation. The method first establishes an N-step static phase-shift image dataset of the gear at subdivided rotational positions. When constructing network training data pairs, three image combinations with different position indices and phases satisfying the standard three-step phase shift are selected to generate an input sequence simulating unidirectional rotation and motion errors. The high-precision sine and cosine terms calculated using the N-step static images are used as labels. Secondly, an encoder-decoder deep neural network with a bottleneck layer dual attention mechanism and global residual connections is constructed and trained using an adaptive loss function. Finally, the dynamic image to be tested is input into the network to directly predict the sine and cosine terms, calculating the wrap-around phase without motion artifacts. This invention effectively suppresses motion errors caused by gear rotation, contributing to the high-precision reconstruction of the three-dimensional morphology of the tooth surface.

Disclosed in the present invention are a motion management method and system, and a related device, which relate to the technical field of radiotherapy. The system comprises: a real-time imaging module used for acquiring a body surface contour of a patient in real time and acquiring a respiratory curve of the patient according to the body surface contour, the body surface contour being used for auxiliary positioning of the patient; and a 4D acquisition module used for sending a time signal to a sliding gantry CT according to the respiratory curve of the patient to establish a respiratory phase synchronization timing sequence, wherein when a treatment couch is at a scanning position of the sliding gantry CT, according to a real-time respiratory signal of the patient, 4DCT images or breath-hold images of the patient in different respiratory phases are acquired using the sliding gantry CT. The 4DCT images or the breath-hold images are used for registration with a positioned 4DCT image, thereby overcoming the drawbacks that previously only tumor images at a certain respiratory phase are obtainable and the images are prone to motion artifacts, resulting in errors in registration. As a result, the registration accuracy of a tumor that is greatly affected by respiratory motion is improved, and the clinical efficiency is enhanced.

The present application belongs to the technical field of inertial measurement, and discloses a soft-attached inertial measurement unit artifact compensation method, comprising: acquiring soft-attached inertial measurement unit signals and rigid reference inertial measurement unit signals under the same motion process, performing time alignment, normalization processing and sliding window segmentation to obtain training samples; constructing an attachment state perception residual time sequence neural network, including a time feature extraction branch, an attachment hidden state coding branch, a feature gating and translation modulation module and a residual output module, and obtaining a compensation model after training with the training samples; inputting real-time collected soft-attached inertial measurement unit signals into the compensation model to output compensated inertial measurement results. The present application also discloses a soft-attached inertial measurement unit artifact compensation device. The method and device can quickly compensate for the motion artifacts of the soft-attached inertial measurement unit under different motion conditions and restore it to the output of the hard-attached inertial measurement unit.

This invention discloses an AI-assisted diagnostic system for CT rib fractures based on image recognition, belonging to the field of smart healthcare technology. It includes recognizing the patient's contour using thermal imaging data, constructing a respiratory model to analyze respiratory status, and automatically triggering CT image acquisition of the ribs at the optimal inhalation moment. Subsequently, artifact detection is performed; if qualified, the system is handed over to AI for intelligent fracture diagnosis, achieving a fully automated process from image acquisition to diagnosis. This invention uses non-contact thermal imaging to monitor respiration and body movement, constructing a respiratory model that integrates chest contour and airflow temperature changes, accurately identifying the optimal inhalation moment to trigger CT acquisition, reducing motion artifacts at the source and improving image quality. The system is highly robust, automatically determining the optimal acquisition time even under conditions of clothing or weak breathing, reducing reliance on patient cooperation, and is particularly suitable for rib fracture patients, improving examination success rate and efficiency.

This invention discloses a method for determining exponential distribution delay based on light intensity and event threshold control, comprising the following steps: Based on the fact that DVS image sensor events exhibit exponential distribution characteristics in the time domain when image information changes rapidly in a scene, an exponential distribution model is established that correlates the time constant with light intensity and a threshold; according to the exponential distribution model and the parameters calculated by the model, the exponential distribution delay of the DVS image sensor event to be analyzed in the time domain is determined, which is characterized by the event trigger rate, including the delay magnitude and its distribution in the spatial domain. This invention, based on the exponential distribution delay model, can quickly and conveniently process and calculate the exponential distribution delay of pixels, providing a basic explanation for the origin of motion artifacts in low-light conditions, and assisting researchers in developing artifact suppression methods based on delay mechanisms.

According to an embodiment of the present disclosure, an image generation device providing a motion artifact correction function is disclosed. The device comprises a light generation unit configured to emit light irradiated onto a target object, a light receiving unit configured to receive optical signals reflected from the target object, a driving unit configured to control a light propagation path, and at least one processor. The processor is configured to irradiate light onto the target object to generate at least one scanning image, estimate a motion speed, generate a corrected scanning image by correcting the at least one scanning image based on the estimated motion speed, and generate a final corrected image based on the at least one corrected scanning image.Also disclosed is a method for operating the image generation device.

This invention relates to the field of image processing, specifically to a method for correcting motion artifacts in multimodal optical images. The method includes: obtaining observation event time points based on the motion trajectories of scanned images at different time intervals and the target physiological texture of the scanned images; correcting the temporal motion trajectory using an arbitrary deviation vector, and recording the time point at which the corrected temporal motion trajectory passes through the target physiological texture in the global image as the predicted event time point; performing Hungarian algorithm matching on the predicted event time point and the observed event time point, wherein the matching cost of the Hungarian algorithm matching is obtained by the difference between the predicted event time point and the observed event time point and a penalty weight; and using the deviation vector with the minimum matching cost to correct motion artifacts. This invention effectively avoids motion artifact problems caused by complex motion instability factors.

The application belongs to the field of brain-computer interaction, and particularly relates to an earphone type brain-computer and a use method thereof, wherein the brain-computer comprises an ear cover, an earphone module installed on the ear cover, and an integrated module installed on the ear cover; the earphone module comprises a microphone for audio input and installed on the ear cover and a loudspeaker for audio output and installed on the ear cover. The application integrates EEG collection, PPG, body position monitoring and earphone functions in a single device, avoids the cumbersome of users wearing multiple devices, improves the use convenience and comfort, corrects signals through EEG, PPG and body position fusion processing, reduces motion artifacts and environmental interference, realizes efficient wireless data transmission through low-power Bluetooth transmission, supports real-time monitoring and remote control, optimizes power management, prolongs the endurance time, and meets the daily continuous use demand.

The application discloses a pre-processing method based on high-pollution children's electroencephalogram data and belongs to the technical field of electroencephalogram signal processing. The application proposes a robust motion artifact detection method based on a multi-channel voting mechanism, overcomes the sensitivity of single-channel detection to transient noise through multi-channel joint decision, and automatically identifies the repair period. A set of comprehensive preprocessing procedures are designed, which integrate PCHIP interpolation repair, adaptive Bayesian wavelet denoising, band-pass filtering and independent component analysis. The procedures are verified based on real children's continuous task electroencephalogram data. The application performs excellently in preserving frequency band information and improving weighted signal-to-noise ratio, and the average accuracy rate based on permutation entropy feature classification reaches 96.4% (205 test sets), and the highest is 100%. The application effectively solves the pre-processing problem of children's electroencephalogram data in a high-noise environment, and significantly improves the accuracy and reliability of subsequent classification tasks.

The present application relates to the technical field of remote photoplethysmography, and in particular to a rPPG physiological index estimation method and system based on a multi-scale time sequence Mamba architecture; a parallel signal processing architecture is first constructed, a confidence degree guided optical flow motion compensation module is used to improve ROI time sequence consistency and relieve motion artifacts, a differential denoising module is combined to weaken light disturbance, and two enhanced spatial features are obtained; then a gating weighting module is used to map the spatial features to the channel dimension for adaptive fusion, information conflicts in spatial fusion are avoided, and effective inputs are provided for time sequence modeling; finally, a bidirectional time sequence collaborative module that integrates time sequence separable convolution and state space model is constructed, local short-term dynamics and long-term periodic rules are collaboratively captured, and the robustness of time sequence feature representation is improved while maintaining computational efficiency. The present application can efficiently and accurately extract physiological indexes such as heart rate in complex scenes, has strong robustness, and is suitable for digital health scenes such as precision medicine and telemedicine.

The application relates to the technical field of flexible electronics, and specifically discloses a liquid metal-based flexible hybrid electronic device for monitoring human pulse rate and a preparation method. The device comprises an intrinsically stretchable circuit formed by patterning liquid metal on a rubber substrate, and micro functional elements integrated in the form of a patch, which are collectively encapsulated by an elastomer. The preparation method comprises six steps of circuit printing, hot pressing, chemical etching, selective infiltration, stripping transfer and integrated encapsulation. The key is to utilize the selective infiltration of liquid metal to copper for patterning, and utilize the characteristics of semi-solid polymer for soft transfer. The device of the application benefits from its intrinsic stretchable characteristics, can conformally attach to human skin and synchronously deform, effectively suppresses motion artifacts, and thus realizes accurate, real-time and long-term monitoring of pulse rate in various states such as sitting, walking and running. The preparation method is simple in process and low in cost, and is suitable for large-scale production.

This invention discloses a visual early warning device and method for cardiovascular diseases in obese children, relating to the field of early warning measurement technology. The method includes the following specific steps: S1: Simultaneously collecting cardiovascular-related parameters and status data of children in both clinical and home modes using a flexible wearable device; S2: Correcting the collected physiological parameters through motion artifact filtering and establishing a correlation model between skin conductance response and cardiovascular parameters to correct for emotional interference in diagnosis; S3: Constructing visual interactive interfaces for children and adults to display parameters and provide emotional guidance; S4: Constructing an early warning indicator system based on children's baseline indicators and metabolic-related indicators, classifying early warning levels, and displaying, transmitting, and storing the data through the adult visual interface. This invention improves the convenience and success rate of early warning measurement by determining the emotional state during measurement through motion artifact filtering and emotional interference correction.

This invention discloses a differential interferometric phase-contrast imaging system, method, and time-difference incubator device, belonging to the fields of optical imaging and medical device technology. The system includes an illumination source module, a microscopic imaging module, a polarization interferometry module, and a quantitative phase image reconstruction module. By placing all polarization interferometry components, such as the polarizer and polarization beam splitter, on the image side of the microscopic imaging module, the imaging beam does not need to pass through the plastic culture dish before passing through the polarization elements, thus completely avoiding interference from plastic birefringence on the imaging and achieving high-quality observation of samples in cell biology or materials science fields within plastic culture dishes. Simultaneously, the system uses a polarization camera to simultaneously acquire multiple phase-shifted interferograms in a single exposure, combined with a quantitative phase reconstruction algorithm, achieving high-speed, high-precision quantitative phase imaging, overcoming the shortcomings of traditional mechanical phase-shifting methods, such as slow speed and susceptibility to motion artifacts. Furthermore, the imaging system of this invention can also be integrated into a time-difference incubator.

The application discloses a floor flatness detection method and system based on a laser robot, and relates to the technical field of flatness detection. Multi-source pose information is used to evaluate the motion state of the laser robot, the body disturbance features caused by the floor fluctuation are extracted, and the dynamic reference reconstruction is performed on the three-dimensional point cloud data, so that the target point cloud data eliminating the motion artifacts of the laser robot itself is determined, and the accuracy of the target point cloud data is improved. The target point cloud data is globally spliced and error homogenized, so that the overall topographic features of the floor are generated. The overall topographic features of the floor are dynamically identified, and the corresponding continuous detection section is determined in combination with the flatness constraint relationship of the floor. The multiple flatness deviation contents are determined according to the detection of the continuous detection section, and the flatness detection report of the floor is determined in combination with the corresponding flatness deviation area and the topological relationship between each grid, so that the accuracy of the flatness detection report of the floor is improved.

The application provides a video de-field line method and device, the method comprising: obtaining a reference image frame of a previous frame processing completion, and an odd field reconstruction frame and an even field reconstruction frame corresponding to a current frame; dividing the odd field reconstruction frame and the even field reconstruction frame into a plurality of non-overlapping image blocks according to the same position; respectively performing block matching and block calculation on the image blocks at the same position in the odd field reconstruction frame and the even field reconstruction frame with the reference image frame, determining a reconstruction frame to which the best-performing image block belongs, and adding one to the count of the reconstruction frame, wherein the reconstruction frame to which the best-performing image block belongs meets the object motion law in the video and has no motion artifacts; after completing the matching and calculation of all image blocks, taking the reconstruction frame with the most counts as the processing completion frame of the current frame, and taking the processing completion frame as the reference image frame required in the next frame of the video de-field line processing. The application improves the dynamic fluency after the video de-field line.

This invention provides an automatic exposure sampling control method for an image sensor, an image sensor, and a storage medium, comprising: responding to incident light, driving a pixel array to be divided into exposure rows with different exposure durations according to a preset period to generate an image charge signal; within the readout period of the same row signal, quantizing the charge signal using at least two analog gain parameters based on column position information to obtain an energy sample set formed by row-column cross-distribution; converting the comparison result of the sample set and the target brightness into driving parameters to drive the register to update global parameters during the vertical blanking period. This invention eliminates the linear dependence of automatic exposure convergence on multi-frame temporal iteration by implementing spatial multiplexing sampling in the pixel row and column dimensions. While effectively avoiding inter-frame decoherence and motion artifacts, it significantly shortens the latency overhead of the first frame image output and reduces the bandwidth load of the underlying readout link.

This application relates to a method, apparatus, and wearable device for detecting physiological parameters. It can simultaneously acquire initial physiological signals and multiple inertial measurement signals, then fuse and analyze the multiple inertial measurement signals to obtain a synthetic inertial measurement signal. Next, adaptive filtering analysis is performed on the initial physiological signal and the synthetic inertial measurement signal to obtain a pure physiological signal, which is then used for extraction and analysis to determine physiological parameters. This scheme, by simultaneously acquiring initial physiological signals and multiple inertial measurement signals, and fusing the multiple inertial measurement signals during the data processing stage to obtain a synthetic inertial measurement signal with a higher signal-to-noise ratio, is then input into subsequent adaptive filtering analysis. By combining hardware denoising through multi-inertial measurement unit fusion with software denoising through adaptive filtering analysis, motion artifact noise generated by user movement is removed from the initial physiological signal, significantly improving the accuracy of physiological parameter acquisition.

The application discloses a VR real scene teaching model big data teaching knowledge mining method and system, belongs to the technical field of big data analysis, and first clock drift correction and motion artifact removal are carried out on collected multi-modal data to obtain synchronous and pure data flow; then, based on the data flow, a multi-dimensional teaching behavior characteristic vector containing a cognitive load index and a behavior trajectory is extracted; then, based on the characteristic vector, learning behavior mode clustering is carried out, and a cognitive load weighted PrefixSpan algorithm is used to mine an optimal operation path sequence; finally, the mining result is visualized. Through fine pretreatment of multi-modal data and cognitive load weighted path mining, the application solves the problems of poor data quality and ignoring learning burden in the prior art, can accurately depict learner behavior, recommend scientific and efficient learning paths, and improves the VR teaching effect.

A non-contact method and system for monitoring and warning of abnormal vital signs in firefighters includes: transmitting frequency-modulated continuous waves into the chest cavity using a millimeter-wave radar module integrated into the firefighter's helmet and receiving the echoes; collecting body surface temperature using an infrared thermal imaging module and acquiring environmental and inertial parameters using sensors; extracting chest cavity micro-motion signals from the echoes to estimate heart rate and respiratory rate; extracting body surface temperature features from temperature images; compensating for temperature and smoke attenuation using environmental parameters and adaptively suppressing motion artifacts using motion parameters to obtain pure physiological signals; inputting heart rate, respiratory rate, and body surface temperature features into a cross-modal attention fusion network to generate a fused feature vector; dynamically calibrating based on individualized physiological baselines according to motion intensity and calculating physiological deviation; simultaneously inputting the feature vector sequence into an LSTM to predict the trend; if the deviation exceeds the limit or the predicted trend enters the danger zone, outputting graded warning information with abnormality type and confidence level.

The application provides a fusion noise reduction intelligent massage evaluation method and device and electronic equipment, and relates to the technical field of biological signal detection. In the application, in response to a first massage operation of a massage device on a target object, an original surface electromyogram of the target object and a pressure signal of the massage device acting on the target object are acquired in real time; the pressure signal is subjected to signal decoupling processing through a first signal filter to obtain a first signal component related to a motion artifact generated by the operation of the massage device; noise components caused by the motion artifact in the original surface electromyogram are eliminated based on the first signal component to obtain a pure surface electromyogram; and a first relaxation index for evaluating the massage effect is determined based on a first median frequency determined based on a power spectral density of the pure surface electromyogram and a second median frequency of a resting surface electromyogram of the target object in a resting state. In this way, the massage effect of the massage device can be accurately evaluated.

There is a need for techniques to avoid or reduce motion artifacts in medical images. According to the invention, there is provided control circuitry for a medical imaging system. The control circuitry is configured to control the medical imaging system to provide synchronized display and nudging to provide a patient with information on the progress of a scan. In this way, patient anxiety may be reduced by providing the patient with a sense of time during the scan, leading to a reduction in motion artifacts.

This invention discloses a dynamic light source adaptive camera system, including a multispectral sensor array, a spectral quality assessment module, a cross-spectral feature fusion unit, a scene detection module, an intelligent control unit, a light source driving module, and an image optimization module. Through simultaneous acquisition of three spectra and dynamic weight fusion, it perceives scene illumination and target information. Combined with a customized multi-task deep learning model optimized through mixed-precision quantization and structured pruning, it dynamically adjusts the brightness, emission angle, and spectral adaptation mode of multi-zone LED fill lights. It also coordinates multi-frame fusion, noise suppression, and local HDR fusion algorithms to achieve performance under adverse weather and low-light conditions. This system effectively solves problems of overexposure, underexposure, severe noise, and motion artifacts. It features low power consumption, high stability, and wide adaptability, making it suitable for various applications such as vehicle-mounted devices, smart door locks, security monitoring, drone aerial photography, medical endoscopes, and outdoor live streaming equipment.

The application discloses an image registration method and device, a medical image motion correction method and system, equipment and a medium. The image registration method comprises the following steps: acquiring medical image data of a plurality of time points containing a target object; determining a motion artifact degree of the medical image data, and determining a plurality of frames of candidate image data according to a balanced position of the target object in the medical image data; selecting medical image data with the lowest motion artifact degree from the plurality of frames of candidate image data as registration reference image data; and performing image registration on the medical image data by using the registration reference image data. According to the two dimensions of the motion artifact degree and the balanced position, the application automatically selects the registration reference image, can avoid selecting an image with poor image quality as the registration reference image, can reduce the registration difficulty and cost, and can improve the quality and efficiency of image registration, and further proposes a quality control scheme for medical image motion correction.

This invention relates to the field of bio-optical imaging technology, specifically to an organ status assessment method and organ perfusion system based on laser speckle imaging. The method utilizes high-frame-rate laser speckle imaging technology combined with a physiological signal gating algorithm. It extracts the original speckle image only during the diastolic window when organ motion is minimal. The original speckle image is then preliminarily corrected using a vibration-speculiar distortion model. Subpixel-level registration is then performed on the preliminarily corrected speckle image using optical flow, generating a high signal-to-noise ratio microcirculation perfusion distribution map free of motion artifacts. Based on the microcirculation perfusion distribution map, perfusion heterogeneity indices for each target assessment region are calculated to determine whether local blood flow abnormalities exist in the transported organ, thus achieving status assessment and real-time monitoring of the transported organ. This assessment method is applicable not only to organ transport but also to organ quality evaluation and immediate post-transplant assessment scenarios.

This application discloses a method and apparatus for determining myocardial strain parameters based on 4DCTA images. The method includes: using a Mamba-based neural network to separate a first myocardial region image sequence from an original image sequence; performing motion correction on the myocardial region images in the first myocardial region image sequence to eliminate motion artifacts, resulting in a second myocardial region image sequence; obtaining the deformation field of the left ventricular myocardium region between consecutive myocardial region image frames in the second myocardial region image sequence through motion tracking; performing layering and block segmentation on the myocardial region images in the second myocardial region image sequence; and determining myocardial strain parameters using the deformation field of the left ventricular myocardium region between consecutive myocardial region image frames in the second myocardial region image sequence and the layering and block segmentation results of the myocardial region images in the second myocardial region image sequence. This application solves the technical problem of low efficiency in myocardial strain processing in related technologies.

This invention discloses a detail-enhanced EEG denoising method based on diffusion residual learning, comprising: 1. acquiring noisy single-channel EEG signals and their clean labels, constructing a first-stage reconstructive denoising network to coarsely denoise the noisy EEG signals, obtaining an initial estimate that preserves the dominant waveform structure; 2. constructing residual detail labels based on the difference between the clean labels and the initial estimate, and performing forward diffusion and backward denoising learning in the residual space; 3. constructing a conditional diffusion residual network, extracting residual variables and conditional information features through a bi-branch structure, and fusing them to generate residual details; 4. adding the initial estimate and the residual details to obtain a detail-enhanced denoised EEG signal. This invention can alleviate oversmoothing phenomena, improve the recovery ability of high-frequency and low-amplitude EEG details, and is applicable to the removal of EEG artifacts in electrooculography, electromyography, electrocardiography, and motion artifacts.

The present application relates to the field of medical image processing, and provides a method for correcting MRI motion artifacts based on physical guided k-ViT and implicit decoupling. The method first decomposes the phase disturbance component caused by motion from the original signal containing motion artifacts based on the k-space physical model; then, the k-ViT network is used to suppress the phase disturbance through its phase-aware attention calculation, and the preliminary corrected k-space signal is obtained; then, the preliminary corrected signal is decoupled into static tissue structure and dynamic motion field through the implicit neural representation network, and the artifact-free k-space signal is reconstructed only according to the static structure; finally, the reconstructed signal is inverse Fourier transformed to obtain the corrected MRI image. Through the physical guided decomposition at the data source, the global suppression in the frequency domain and the decoupling reconstruction of the signal components, the present application can accurately and completely remove the motion artifacts and effectively improve the image quality, providing a more reliable image basis for clinical diagnosis.