29 results about "Magneto encephalography" patented technology

The invention discloses a multi-modal deep learning traceability method and system fusing magnetoencephalogram and electroencephalography, and relates to the technical field of artificial intelligence and neuroimage.Real magnetoencephalogram signals and electroencephalography signals are preprocessed and then input into a traceability model, the probability of occurrence of a source in a corresponding area is predicted, and the traceability of the source in the corresponding area is obtained by combining an imported source partition distance matrix. A final traceability result is obtained; the training process of the traceability model is as follows: constructing a generative adversarial network, and generating a multi-modal neural electrophysiological data set; inputting the multi-modal neural electrophysiological data set into a residual network of a double-branch structure, and performing stage hierarchical extraction and decoupling on magnetoencephalogram signals and electroencephalogram signals respectively; extracting features in different stages by using a multi-scale convolution module, fusing the extracted features, inputting the fused features into a classifier, defining a loss function, and updating trainable parameters of the traceability model; the traceability method improves the accuracy and generalization ability of traceability positioning.

The invention discloses a magnetoencephalogram-based epilepsy lesion positioning method and system and a medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: carrying out the preprocessing of a structure image of a patient, carrying out the registration with the magnetoencephalogram of the patient, carrying out the grid division of a cerebral cortex region based on the preprocessed structure image and a registration result, and obtaining a forward model of the patient; performing preprocessing and spine wave detection on the magnetoencephalogram to obtain a spine wave time point sequence; based on the forward model and the ratchet wave time point sequence, using a magnetic dipole algorithm to calculate a source coordinate and a source direction of a ratchet wave time point, and classifying the ratchet wave time point sequence according to the source coordinate and the source direction to obtain a clustering result; calculating the source coordinate and the source direction of the class center of each class in the clustering result by using a magnetic dipole algorithm; generating a clinical report of the patient; according to the positioning method, the problems of large epilepsy diagnosis difference and error proneness caused by level difference of different doctors are solved, and the epilepsy focus position can be quickly determined.

Aspects provide systems and methods for utilizing quantum computing systems for processing of data generated by quantum sensor. For example, magnetic field data may be captured from a brain using a quantum sensor array. This magnetic field data may then be processed using a quantum computing apparatus including a plurality of qubits in order to generate a model of the brain.

Aspects provide systems and methods for utilizing quantum computing systems for processing of data generated by quantum sensor. For example, magnetic field data may be captured from a brain using a quantum sensor array (132). This magnetic field data may then be processed using a quantum computing apparatus (110) including a plurality of qubits (114) in order to generate a model of the brain.

The application discloses a multi-modal deep learning source tracing method and system combining magnetoencephalogram and electroencephalogram, relates to the field of artificial intelligence and neuroimaging technology, and inputs real magnetoencephalogram signals and electroencephalogram signals into a source tracing model after preprocessing, predicts the probability of the occurrence of corresponding regional sources, combines an imported source partition distance matrix, and obtains a final source tracing result; the training process of the source tracing model is as follows: a generative adversarial network is constructed to generate a multi-modal neuroelectrophysiological data set; the multi-modal neuroelectrophysiological data set is input into a double-branch structure residual network to perform stage hierarchical extraction and decoupling on the magnetoencephalogram signals and the electroencephalogram signals respectively; a multi-scale convolution module is used to extract features at different stages, the extracted features are input into a classifier after being fused, and a loss function is defined to update trainable parameters of the source tracing model; and the source tracing method improves the accuracy and generalization ability of source positioning.

The invention relates to a method for inferring epileptogenicity of a brain region not observed to be recovered or not observed to be not recovered in a seizure activity of a brain of an epilepsy patient, comprising the steps of: providing a computerized model modeling individual regions of a primate brain and connectivity between said regions; providing said computerized model with a model capable of reproducing the dynamics of a seizure in a primate brain; providing structural data of a brain of an epilepsy patient and using said structural data to individualize the computerized model in order to obtain a virtual epilepsy patient (VEP) brain model; translating a state space representation of the virtual epilepsy patient (VEP) brain model into a probabilistic programming language (PPL) using probabilistic state transitions in order to obtain a probabilistic virtual epilepsy patient brain model (BVEP); and acquiring electroencephalogram or magnetoencephalogram data of the brain of the patient and fitting the probabilistic virtual epilepsy patient brain model against said data in order to infer the epileptogenicity of said brain region not observed.

This invention provides a deep learning model training method for classifying temporal lobe epilepsy lateralization, belonging to the fields of brain science signal processing, medical artificial intelligence, and deep learning technology. The method includes preprocessing and source reconstruction of magnetoencephalography (MEG) signals; obtaining a brain region time-series matrix by selecting brain regions of interest; constructing a multi-band adjacency matrix; constructing a node feature matrix based on the adjacency matrix; and defining and training a deep learning model based on the node feature matrix. The deep learning model is used to classify temporal lobe epilepsy lateralization. Based on MEG brain network characteristics, this invention automatically identifies whether a subject belongs to the healthy group, the left TLE group, or the right TLE group, thereby providing an objective basis for the clinical judgment of temporal lobe epilepsy lateralization.

The present disclosure provides an AI-based clinical decision support system comprising an input module configured to receive clinical information comprising brain scan data, an analysis module configured to parse the clinical information using statistical measures from functional connectivity analysis with counterfactual explanations to identify brain connectivity patterns associated with brain disorders, and an output module configured to present a recommended diagnosis and explanation comprising attribution information identifying connectivity features contributing to the diagnosis. The brain scan data comprises functional magnetic resonance imaging, electroencephalography, and magnetoencephalography data. The statistical measures comprise functional connectivity analysis and graph theory metrics including degree metrics, betweenness centrality measures, and clustering coefficients. The analysis module comprises a functional connectivity engine configured to process brain scan data and generate connectivity features, a feature bank configured to store connectivity features, and modeling backbones configured to analyze connectivity features using machine learning techniques.

The application provides a specific neurofeedback training system for improving anxiety based on multi-modal fusion, which forms specific electroencephalogram signals through the projection strategy of magnetoencephalogram signals to electroencephalogram, and is used for the regulation and improvement of anxiety emotion. The application belongs to the technical field of medical treatment, and comprises an electroencephalogram acquisition module, a real-time processing module and a visual feedback module which are connected with each other; a complete treatment closed loop from signal acquisition, signal processing, signal feedback and signal acquisition is realized. The application decodes and analyzes electroencephalogram signals in real time, extracts features, and uses a specific mapping model of multi-modal feature fusion to establish a specific electroencephalogram mapping signal reflecting the real-time activity of core brain regions related to emotion (such as amygdala), which has stronger spatial accuracy and symptom specificity; the specific mapping signal is applied to neurofeedback treatment, which can assist in formulating individualized and multi-course neuroregulation training, improving the anxiety emotion of users, and providing a non-invasive and convenient neuroregulation platform for users.

The utility model relates to a brain magnetic helmet structure and a brain magnetic mapping instrument. The utility model is suitable for the technical field of medical instruments. According to the technical scheme, the brain magnetic helmet structure comprises an equipment main body lower half part, and a lower helmet matched with the rear half part of the head of the human body is arranged at the intersection of the upper surface and the front end face of the equipment main body lower half part; an upper helmet matched with the front half part of the human head is arranged at the intersection of the lower surface and the front end face of the equipment main body upper half part; the sliding rail assembly is arranged between the equipment main body lower half part and the equipment main body upper half part, and the equipment main body upper half part can be installed on the equipment main body lower half part in a front-back moving mode relative to the equipment main body lower half part through the sliding rail assembly; the front end positioning mechanism can position the upper half part of the equipment main body at a first position where the equipment main body moves back and forth, and when the upper half part of the equipment main body is positioned at the first position, the upper helmet corresponds to the lower helmet in vertical position; and the rear end positioning mechanism can position the upper half part of the equipment main body at a second position moving back and forth.

The application provides a kind of child magnetoencephalogram personalized stimulation paradigm optimization method based on reinforcement learning, comprising: step S1, the multi-channel magnetoencephalogram signal of the child to be measured is continuously collected by using optically pumped magnetometer array, and behavior data and physiological data are synchronously collected;Step S2, the magnetoencephalogram signal feature, behavior feature and physiological feature are extracted, and real-time state feature vector is fused and constructed;Step S3, the real-time state feature vector is input into the reinforcement learning model to obtain the prediction Q value of stimulation adjustment action, and the parameters of magnetoencephalogram stimulation paradigm of the child to be measured are dynamically adjusted based on the prediction Q value;Step S4, based on the magnetoencephalogram stimulation paradigm after parameter adjustment, the new multi-channel magnetoencephalogram signal, behavior data and physiological data of the child to be measured are collected, and the comprehensive reward value is calculated to update the reinforcement learning model.The beneficial effect is that the application can intelligently and dynamically adjust the magnetoencephalogram stimulation paradigm, and improve the multi-channel magnetoencephalogram signal quality of the child to be measured.

The utility model provides a photostimulation assembly for treating brain function related diseases, which adopts a light transmission piece capable of circumferentially emitting near-infrared light, the light transmission piece is spirally arranged in an accommodating cavity of a bearing piece, and a light uniformizing layer is arranged on one side of a light transmission part of the bearing piece, so that the near-infrared light can be transmitted to the light transmission piece, and the light transmission piece can transmit near-infrared light to the bearing piece. The near-infrared light can be uniformly irradiated to the head of a patient, light spots with a certain area can be emitted to a head treatment area at a short distance, meanwhile, relatively uniform distribution of light power is guaranteed, and the treatment effect on patients with brain function related diseases in some special scenes is greatly improved. Meanwhile, the photostimulation assembly is not interfered by a high-intensity magnetic field in a treatment area of the head of the patient, and can be combined with magnetoencephalography, functional magnetic resonance imaging and other technologies, so that the change condition of the brain function state of the patient can be monitored in time while the brain function related diseases are treated by adopting photostimulation; the method can be used for evaluating the treatment effect of photostimulation, guiding the photostimulation treatment scheme and the like.

The invention provides a children magnetoencephalogram personalized stimulation normal form optimization method based on reinforcement learning, and the method comprises the steps: S1, continuously collecting multi-channel brain magnetic signals of a to-be-detected child through an optical pump magnetometer array, and synchronously collecting behavior data and physiological data; s2, brain magnetic signal features, behavior features and physiological features are extracted and fused to construct a real-time state feature vector; s3, inputting the real-time state feature vector into a reinforcement learning model to obtain a predicted Q value of a stimulation adjustment action, and dynamically adjusting parameters of a magnetoencephalogram stimulation normal form of the to-be-tested child based on the predicted Q value; and S4, acquiring new multichannel brain magnetic signals, behavior data and physiological data of the to-be-tested child based on the parameter-adjusted brain magnetic map stimulation normal form, and calculating to obtain a comprehensive reward value to update the reinforcement learning model. The method has the advantages that the magnetoencephalogram stimulation normal form can be intelligently and dynamically adjusted, and the multi-channel magnetoencephalogram signal quality of the child to be tested is improved.

The invention provides a method of determining whether a subject has a neurological dysfunction associated with a signal in a particular electroencephalogram (EEG) or magnetoencephalogram (MEG) frequency range, the method comprising: obtaining an EEG power spectrum from the subject; and obtaining a metric quantifying the magnitude of power in particular frequency range metric quantifying the power in the power spectrum in the particular frequency range, wherein the metric summarises the power in said frequency range corrected using an estimate of the power in said frequency range that is attributable to background signal that is specific to said frequency range, wherein the metric is indicative of the presence and / or severity and / or direction of a neurological dysfunction. Related methods and devices are also described.

The application discloses a magnetometer depth adjustment method of a wearable magnetoencephalograph, and comprises the following steps: obtaining two-part three-dimensional information of a subject's helmet and face based on an optical scanner, and obtaining the position and direction of the magnetometer relative to the scalp surface of an MRI image by two-step registration; calculating the minimum distance of the bottom surface of the magnetometer moving along the probe axial direction to the scalp based on the position and direction of the magnetometer and the scalp in the MRI coordinate system; and based on the distance, using a motor push rod or other control method to make the magnetometer travel a corresponding distance along the axial direction. The magnetometer depth adjustment method has the characteristics of automation, high precision and high efficiency, can automatically calculate the optimal distance of magnetometer depth adjustment, assist in helmet design, improve the magnetoencephalic source positioning precision and the helmet scalp adhesion, simplify the depth adjustment process, facilitate the use of medical staff, and further promote the medical application of magnetoencephalography.

1. Name of the product in this design: Horizontal magnetoencephalography (MEG) instrument. 2. Purpose of this design: To detect weak magnetic field signals generated by active intracranial nerves in a non-invasive manner, and to analyze the characteristics and location of active intracranial nerve sources. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the design's key points: a 3D model.

This invention discloses a method for assessing the quality of magnetoencephalography (MEG) based on multi-band features, belonging to the field of MEG processing technology. The invention acquires a reference signal and the patient's MEG signal, dividing them into five frequency bands in the frequency domain. After merging the signals from each band and performing iFFT transformation, a frequency band fused time-domain signal is obtained. This fused signal is then linearly mixed with the patient's MEG signal to generate a mixed signal. Next, each frame of the mixed signal is subjected to FFT transformation to extract five sets of depth and spectral features, and wavelet features are extracted through wavelet transform. These three types of features are then concatenated and dimensionally aligned to form a feature map. Finally, the feature map is input into a U-Net-structured deep learning network to obtain the signal quality assessment result of the mixed signal. This method effectively improves the accuracy of MEG quality assessment through the fusion of multi-band features and processing by a deep learning network.

The invention provides a magnetoencephalography sensor array calibration method based on a large-size composite coil, relates to the field of brain science and neural engineering, and aims to solve the problem that an existing calibration method is difficult to capture an OPM real space pose, so that forward model errors are caused, and the brain magnetic traceability precision is reduced. According to the invention, a large-size composite coil containing a plurality of uniform and gradient coils is arranged in a zero magnetic environment, the response of the sensor in a uniform / gradient magnetic field generated by the sensor is obtained, and the spatial position, the measuring axis direction and the gain of the sensor are solved through an algorithm, so that calibration is completed. According to the method, the problems that the real OPM pose is difficult to capture, the forward model is inaccurate and the traceability precision is low in the existing method are solved, 12 OPM parameters can be completely calibrated, the measurement noise is reduced, and the calibration pose stability and the target area magnetic field error lt are improved; the traceability error is obviously reduced, and the accuracy of the MEG forward model is improved.

The application discloses a kind of based on magnetoencephalogram and functional magnetic resonance imaging auditory nerve representation device, device includes: auditory stimulation system is used to generate magnetoencephalogram signal and functional magnetic resonance signal;Head-mounted array sensor system is used to collect generated magnetoencephalogram signal;Magnetoencephalogram signal acquisition system is used to receive the magnetoencephalogram signal;Functional magnetic resonance imaging signal acquisition system is used to receive the functional magnetic resonance signal;Magnetoencephalogram data preprocessing system is used to preprocess magnetoencephalogram signal, obtain magnetoencephalogram processing data;Functional magnetic resonance imaging data preprocessing system is used to preprocess the functional magnetic resonance signal, obtain function processing data;Dual-mode data joint analysis system is used to receive and fuse magnetoencephalogram processing data and function processing data, using support vector machine classifier constructs stimulation two two mutual classification diagonal symmetric representation dissimilarity matrix, according to representation dissimilarity matrix obtains auditory nerve representation result.

The invention discloses a brain magnetic signal deep learning classification method and system based on spatial-temporal feature fusion, and belongs to the technical field of biomedical signal processing. The method comprises the following steps: acquiring and preprocessing a multi-channel magnetoencephalogram (MEG) signal, particularly an MEG signal acquired by adopting an optical pump magnetometer (OPM); performing feature extraction and classification on the preprocessed MEG signals through a spatial-temporal feature fusion deep learning model, and extracting local time features and preliminary spatial interaction features from a time sequence of the MEG signals through a parallel multi-scale one-dimensional convolutional neural network (1D-CNN) module; flattening the extracted feature map, and inputting the flattened feature map into a Transform encoder module; and outputting a classification result through the full-connection classifier. According to the method, the problem of poor processing of high-dimensional and low-signal-to-noise-ratio characteristics of MEG signals in a traditional method is solved, and the accuracy and robustness of classification are remarkably improved through deep fusion of spatial-temporal characteristics.

The application discloses a visual real-time six-degree-of-freedom head pose estimation framework for a wearable optical pumped magnetometer-magnetoencephalography (OPM-MEG) system, aiming at solving the motion pose error problem caused by non-rigid head-helmet coupling. The framework includes two stages of initialization and real-time tracking: in the initialization stage, a hybrid differentiable rendering strategy is adopted, combined with soft rasterization and hard rendering technology, to realize high-precision initial pose alignment between the head and the helmet through multi-view contour matching; in the real-time tracking stage, based on the checkerboard marker, an improved PnP algorithm is used to estimate the real-time change of the head relative to the initial pose, and a multi-candidate solution screening and nonlinear optimization mechanism is introduced to suppress drift. The application does not need to rely on structured light or laser scanning hardware, can realize low-delay and high-robust real-time pose tracking, effectively reduces the registration error caused by motion, and improves the reliability and accuracy of the neural magnetic signal source positioning.

A portable magnetoencephalography (MEG) system (100) for non-invasive recording of magnetic fields of the brain during natural behaviors, the system (100) comprising: an adaptable non-ferromagnetic helmet (1) configured to be worn on the head of a test subject; a plurality of interchangeable sensor capsules (2) attached to the helmet (1), each sensor capsule (2) comprising: a triaxial quantum magnetometer (3) in the form of an optically pumped magnetometer configured to measure magnetic fields along three orthogonal axes, and a local inertial measurement unit (IMU) (4) configured to provide motion signals with six degrees of freedom; at least one additional IMU (5) attached to the helmet (1) and configured to measure head movements; an active field control subsystem (6) comprising a plurality of magnetic field generating coils (7) embedded in the helmet (1) and at least one reference quantum magnetometer (8) positioned to detect magnetic fields in the environment; an external electronic module (9) containing at least one clock generator (25), at least one digital processor (26), at least one power supply circuit and a wireless communication interface (19); a weak magnetic connection (10) between the sensor capsules (2) and the external electronics module (9), wherein the weak magnetic connection (10) comprises at least one optical wiring and / or one non-ferromagnetic electrical wiring; and a control unit (11) that is connected to the sensor capsules (2), the reference quantum magnetometer (8), the IMUs (4, 5), the active field control subsystem (6) and the external electronics module (9), wherein the helmet (1) and the sensor capsules (2) are configured to maintain a sensor-to-scalp distance of less than or equal to 10 mm over a range of head sizes, and the control unit (11) is configured that they: (i) to synchronize the magnetometer signals from the triaxial quantum magnetometers (3) and the IMU signals from the IMUs (4, 5) with a synchronization error of no more than 0.5 ms, (ii) continuous updating of the conduction fields that map the brain sources to the sensor capsules (2) based on the IMU signals from the IMUs (4, 5), and (iii) to reconstruct the magnetic activity of the brain in real time using the updated conduction fields while the test subject performs voluntary movements.

The invention provides a method for suppressing errors of a brain magnetic multichannel array. The method comprises the following steps: acquiring a multichannel magnetic field signal matrix which is acquired by a magnetoencephalography and is based on three orthogonal annular exciting coils; calculating a covariance matrix, performing singular value decomposition, and extracting a noise subspace matrix; a laser scanner is used for obtaining position and direction information of a sensor and an exciting coil, and a guiding field matrix and a signal transmission model are constructed based on the Bios-Savart law. And constructing a cost function, reducing the sampling space range through iterative optimization, gradually approaching the accurate position of the coil, solving a sensor array error parameter matrix, and further correcting the array error in the multi-channel magnetic field signal matrix. According to the method, the calibration precision can be improved, and the spatial positioning accuracy of the multi-channel brain magnetic signals is enhanced.

The invention relates to a brain signal analysis method based on cross-frequency information fusion. The method is technically characterized by comprising the steps that a measurement normal form based on a magnetoencephalogram is established, and brain magnetic signals are collected; preprocessing the brain magnetic signals; calculating a weighted phase lag index between different channel signals in the preprocessed data, and constructing intra-layer connection of the cross-frequency multi-layer network; phase-amplitude coupling between channel signals under different frequency bands is calculated by adopting a modulation index measurement method, and interlayer connection of a frequency-crossing multi-layer network is constructed; and constructing a cross-frequency multi-layer brain function network according to the intra-layer connection and the inter-layer connection, and performing multi-dimensional feature extraction on the cross-frequency multi-layer brain function network. According to the method, the same-frequency and different-frequency connection characteristics are simultaneously integrated in the cross-frequency domain, so that the hierarchical organization structure of the brain function network can be more comprehensively described, the important topological characteristics of the multi-layer network can be stably extracted, the utilization efficiency of the cross-frequency coupling information in the brain magnetic signal is effectively improved, and the method can be widely applied to brain signal detection and analysis.

The application discloses a method and system for locating an epileptogenic focus of epilepsy based on magnetoencephalography and a medium, and relates to the technical field of artificial intelligence, and comprises the following steps: registering a structural image of a patient after preprocessing with a magnetoencephalogram of the patient, dividing a cerebral cortex region into a grid based on the preprocessed structural image and the registration result, and obtaining a forward model of the patient; preprocessing the magnetoencephalogram and detecting a spike wave, and obtaining a spike wave time point sequence; calculating source coordinates and source directions of the spike wave time points based on the forward model and the spike wave time point sequence, classifying the spike wave time point sequence based on the source coordinates and the source directions, and obtaining a clustering result; calculating source coordinates and source directions of a class center of each class in the clustering result by using a magnetic dipole algorithm; and generating a clinical report of the patient; the locating method solves the problem that epilepsy diagnosis is greatly different and prone to errors due to differences in the levels of different doctors, and can quickly determine the location of an epileptogenic focus.

The invention relates to a magnetoencephalogram analysis method and device based on a dynamic brain function track and disease subspace projection, electronic equipment and a storage medium, and relates to the technical field of neural image data processing, and the method comprises the steps: employing a wavelet packet decomposition algorithm to extract a multi-scale time-frequency feature set of magnetoencephalogram time sequence data, fitting the multi-scale time-frequency feature set into a continuous low-dimensional brain function trajectory; calculating a multi-dimensional anomaly score of the low-dimensional brain function trajectory by using a hierarchical Gaussian process trajectory model, and calling a disease feature subspace based on the multi-dimensional anomaly score; and projecting the low-dimensional brain function trajectory to the disease feature subspace to calculate a projection coefficient and a contribution degree to reconstruct a high-dimensional abnormal vector, and decomposing and mapping the high-dimensional abnormal vector back to an original brain region space to generate a space-time abnormal map. The method is helpful for capturing the instantaneous recombination process of the brain function network and reducing the cognitive deviation of the brain function state.

Various embodiments disclosed herein comprise a Magnetoencephalography (MEG) apparatus. The MEG apparatus comprises MEG sensors and a rigid support member. The MEG sensors measure magnetic fields. The rigid support member stabilizes the MEG apparatus to maintain the location and orientation of the MEG sensors. In some examples, the MEG apparatus comprises an interior shell with sensor slots. The MEG sensors are mounted in the sensor slots and conform to a user's head. The rigid support member is coupled to the interior shell to stabilize the interior shell to maintain the locations and orientations of the MEG sensors when conformed to the head. In some examples, the MEG apparatus comprises sensor mounts that house the sensors and mount to the sensor slots of the interior shell. In some examples, the MEG apparatus comprises an exterior shell coupled to the rigid support member to conceal the internal components of the MEG apparatus.