64 results about "Cerebral structure" patented technology

The invention discloses a novel genotype analysis method based on multimodal brain images. With utilization of information, including a structural image and a functional image, of two kinds of modes, regression on the rs429358 locus genotype linked to the Alzheimer's disease closely is realized. The method comprises: extracting voxel features from a brain structure images and extracting connecting network features from a brain function image; carrying out feature extraction on the two kinds of features by using an elastic network and keeping brain area and network connection with a high association degree; and then carrying out fusion and genotype regression on selected nodes and side attributes by using a multi-kernel support vector. The method has the following advantages: for the brain image with the complicated structure and functions and genotype data, a novel regression analysis method for a single locus genotype based on structural and functional multimodal brain images is put forward; the analysis effect is good; and complementary information of the two modes is utilized fully.

The invention relates to an individualized brain covariant network construction method based on three-dimensional textural features, which comprises the following steps of: 1) segmenting a brain structure image into brain tissue component concentration graphs by using tissue segmentation, and registering the brain tissue component concentration graphs to a standard space template to obtain a standardized brain structure image; 2) extracting three-dimensional texture features corresponding to the standardized brain structure image at a voxel level through at least two gray feature extraction modes, and obtaining a spatial distribution diagram of each texture feature; and 3) defining a brain region map as a network node, extracting the texture feature of each brain region of the individual subject from the grayscale matrix texture feature data set, calculating the Pearson's correlation of the texture feature vectors of any two brain regions, and constructing a covariant matrix of the texture features between the brain regions. According to the method, brain image data of an individual subject can be utilized, the similarity of brain region texture feature vectors serves as measurement of a brain network edge, and then a brain covariant network of the individual subject is constructed.

The invention discloses a comprehensive magnetic resonance imaging device. According to the invention, the cerebral function imaging, the MRI perfusion weighted imaging and the magnetic resonance diffusion imaging are united, so that the assessment for the cerebral injury caused by motor neuron disease and Parkinson disease is assisted, a beneficial judging basis is supplied for illness state judgment, and the accuracy rate of disease judgment is increased; an ALS (amyotrophic lateral sclerosis) patient is accompanied with FTD (frontal temporal dementia), so that the detection means, such as cortex thickness, also is beneficial to the disease diagnose and research; functional magnetic resonance imaging (fMRI) and diffusion tensor imaging techniques have strong complementarity; the combination of the functional magnetic resonance imaging and diffusion tensor imaging techniques has a wide application prospect in neurosciences research; unique brain structure and function information can be supplied.

The invention discloses an abnormal brain connection prediction system, method and device and a readable storage medium, and the method comprises the steps: automatically extracting high-order correlation features in different modes and high-order complementary features between different modes through a deep learning method; and realizing the analysis of abnormal connection of the multi-modal brain network and prediction of different cognitive diseases through an adversarial training method. The method solves the problem that an existing method cannot accurately evaluate the change rule of brain structural morphology and functional connection. According to the method, a prior knowledge guide model is used for learning interpretable characterization, the consistency of different modal characterization distribution is restrained through a paired collaborative discriminator, and then brain graph data is reconstructed for feature codes through a reverse generator and a decoder; and finally, inter-modal and intra-modal high-order correlation features are extracted through a hypergraph perception fusion module, and an adversarial loss function, a reconstruction loss function and a classification loss function are set to guide model learning so as to achieve the purpose of mining the abnormal brain connectivity of the Alzheimer's disease.

The invention discloses a graph convolutional neural network evolution method for a dynamic brain structure. A graph convolutional neural network is adopted to simulate an evolution process of evolving a normal human brain structure network into depression. A direction vector is introduced in the evolution process, the vector not only contains brain structure network information of a normal person, but also contains brain structure network information of a depression patient, the characteristics of the normal person and the depression patient can be extracted at the same time through graph convolution operation, and the evolution direction and the evolution degree can be controlled. The invention provides a graph convolutional neural network model of brain structure network evolution. A deep learning method based on a tensorflow framework is utilized, the cross entropy of a first evolution result and a real brain network of a depression patient is calculated, and a gradient descent optimization method is utilized to enable the evolution of the network to always face the direction of the brain network of the depression patient. And finally, a brain structure network close to a realdepression patient is outputted, and an evolution model closer to the real network is obtained.

The invention relates to an establishment of a conversion model of the intracranial hemorrhage of a dog after autologous thrombus embolic infarction thrombolysis by using an intervention method and anapplication thereof. The method comprises the following steps of: preparation of thrombus, preparation of a cerebral infarction model, thrombolysis and determination of hemorrhage result. The dog hemorrhage conversion model has the advantages that: first, the cerebral structure of a beagle is similar to that of a human being, the variation of the cerebral vascular diameter is small, compared withsmall animals, such as a mouse and the like, the related experimental result can achieve better clinical conversion. Second, in small animal experiments, a laser Doppler blood flow meter is used to monitor the changes of cerebral blood flow, thus judging the occlusion and recanalization of blood vessels. DSA is a gold standard for judging whether the blood flow is smooth or not, the injection ofthe thrombus can be guided, accurate embolism can be carried out, and direct evidence for judging whether the thrombolysis is valid or not is provided. Thirdly, an intervention procedure is used as aminimally invasive technique, other unnecessary operation damage can be reduced to the utmost extent, so that the accuracy and reliability of the experimental result can be improved.

The present invention relates to the determination of brain deformation. The invention relates in particular to a device for tracking brain deformation, an imaging system for tracking brain deformation, a method for tracking brain deformation, a method of operating a device for tracking brain deformation, as well as to a computer program element and a computer readable medium. For providing enhanced information about the brain deformation, a first 3D representation (112) of a cerebrovascular vessel structure of a region of interest of an object is provided (110) and (114) a second 3D representation (116) of the cerebrovascular vessel structure. Then, at least a part of the first 3D representation is elastically three dimensionally registered (118) with at least a part of the second 3D representation. A deformation field (122) of the cerebrovascular vessel structure is determined (120) based on the elastic registration. The determined vessel deformation is applied (124) to a brain structure representation to determine a deformation (126) of the cerebral structure. For example, planning data (130) for an intervention of the cerebral structure is provided (132) and the determined deformation of the cerebral structure is applied (134) to the planning data to generate deformation adjusted planning data (136).

The invention relates to a resting-state electroencephalogram rTMS curative effect prediction and intervention closed-loop feedback diagnosis and treatment method, multi-source heterogeneous data such as EEG, brain structure and psychological behavior detection data are fused, a curative effect prediction model is introduced, and through EEG collection which is low in price and good in portability, curative effect prediction and classification are conducted on insomnia patients; besides, the treatment effect is evaluated and fed back before treatment and during treatment, so that an operator can conveniently adjust a treatment scheme, and dynamic optimization of the prediction model is also facilitated; meanwhile, individualized scheme making and accurate regulation and control under navigation are carried out on patients insensitive to treatment, and reasonable medical suggestions are made for patients with poor effects after 20 times of accurate treatment; the invention relates to a closed-loop feedback diagnosis and treatment system covering evaluation, classification, treatment, reevaluation and retreatment, which greatly saves social resources and medical cost while reducing the blindness of rTMS treatment, improving the accuracy of rTMS treatment and realizing high efficiency of rTMS intervention, and has important application value.

The invention provides a brain structure three-dimensional reconstruction method and device and terminal equipment. The method comprises the following steps: acquiring a 2D image of a brain, inputtingthe 2D image of the brain into a trained 3D brain point cloud reconstruction model for processing, and outputting to obtain a 3D point cloud of the brain, wherein the 3D brain point cloud reconstruction model comprises a ResNet encoder and a graph convolutional neural network, the ResNet encoder is used for extracting an encoding feature vector of a 2D image of the brain, and the graph convolutional neural network is used for constructing a 3D point cloud of the brain according to the encoding feature vector. Based on a brain structure three-dimensional reconstruction method provided by the invention, the 2D image of the brain can be converted into the 3D point cloud of the brain, more visual information is provided for doctors, and the doctors can better diagnose and treat.

The invention discloses an Alzheimer's disease auxiliary diagnosis system. The system comprises a preprocessing module which carries out preprocessing of an original MRI structure image, obtains a preprocessed image, converts the preprocessed image into an MNI space, obtains a preprocessed image, and carries out resampling of the image at an isotropic resolution; an ROI positioning module which performs rigid registration on the image obtained after preprocessing and a standard brain template to obtain a deformation field of non-rigid registration, and selects a plurality of ROIs on the preprocessed individual image according to the deformation field; and an analysis module which constructs, trains, verifies and tests a deep learning model, calculates the characteristic spectrum and the probability value of each ROI in the plurality of ROI individuals by using the deep learning model, and judges whether a subject is an Alzheimer's disease patient according to the probability values. According to the system, the deep learning technology is adopted to learn features of brain structure changes, whether a subject suffers from the Alzheimer's disease or not is predicted, the prediction accuracy is high, and an auxiliary function is provided for diagnosis of doctors.

The invention provides a task state functional magnetic resonance scanning data denoising method, and a device thereof, electronic equipment and a storage medium, and the method comprises the steps: determining N non-neuron voxel scanning data in task state functional magnetic resonance scanning data corresponding to the execution of a target task by a subject according to brain structure magnetic resonance scanning data of the subject; performing sparse learning based on the N non-neuron voxels to obtain a noise dictionary; for voxels in the task state functional magnetic resonance scanning data, using each noise dictionary element as a regression parameter, performing linear regression on a BOLD signal sequence corresponding to the voxels, and obtaining a noise sparse coefficient; and generating de-noised task state functional magnetic resonance scanning data based on the noise sparse coefficient corresponding to each voxel in the task state functional magnetic resonance scanning data and each noise dictionary element. The noise in the task state functional magnetic resonance scanning data can be removed in a targeted manner.

The invention relates to the technical field of electronic informatization of biological indexes, in particular to an anxiety trait quantification method based on multi-dimensional internal perceptionfeatures. The anxiety trait quantification method comprises the following steps: obtaining multi-dimensional data of tested ethology, electroencephalogram physiology and nuclear magnetic resonance image; preprocessing ethological heartbeat perception sensitivity, heartbeat consciousness potential, brain structure and task state data; and constructing a multi-dimensional deep learning network andestablishing an automatic quantization system. The multi-dimensional features of ecology, electroencephalogram physiology and nuclear magnetic resonance image under an internal perception normal formare combined, feature learning is carried out through deep network learning by means of automatic learning and nonlinear hierarchical system advantages, feature value extraction modeling is carried out, individualized anxiety trait level scoring results are obtained, a tool is provided for quantitatively and objectively evaluating the anxiety trait level, the anxiety level is accurately quantified, the anxiety disorder can be effectively recognized in the ultra-early stage, the biological objective diagnosis effect is huge, and diagnosis and treatment of the anxiety disorder can be assisted.