69 results about "Brain ct" patented technology

The invention relates to a method for calculating brain blood volume. The method is used for calculating the brain blood volume on the stable status CT (Computed Tomography) brain perfusion imaging principle. The method comprises the following steps of (1) carrying out common whole brain CT scanning to obtain a brain tissue structure image before contrast agent injection; injecting the contrast agent to the vessel exciting area to reach the stable status, and obtaining the brain tissue structure image of the same position; (3) deleting an overhigh brain blood volume value with a vessel pixel deleting method, and deleting the interference of the artery in the existing area to the brain blood content measuring value; (4) and calculating the brain blood flow value of the brain tissue by calculating the ratio of the brain tissue to the signal change in the vessel according to a formula. With the method, all brain scanning can be carried out and the radial dosage of the perfusion scanning can be reduced, thus the method for calculating brain blood volume can be widely applied into the clinical trial.

The invention provides a brain focus target puncture positioning method and a puncture positioning device. The puncture positioning device comprises a vertical positioning ruler, a first sliding baseand a horizontal positioning ruler, wherein the vertical positioning ruler is provided with first scales in the vertical direction; the first sliding base is vertically and slidably arranged on the vertical positioning ruler; the horizontal positioning ruler is arranged on the first sliding base. The puncture positioning method comprises the steps that a middle-line sagittal view surface of the brain of a patient is determined; brain CT scanning is carried out; CT scanning data is analyzed, and a target transverse-cutting layer image of a brain focus of the patient is determined; a middle-linesagittal image and the target transverse-cutting layer image are modeled in proportion, and simulated puncture is carried out; then solid puncture is carried out on the brain of the patient. The focus target puncture positioning method and the puncture positioning device thereof are used so that multi-point puncture positioning can be realized, the influence of human factors is eliminated, the puncture accuracy of a focus target and the success rate of an operation are improved, the operation difficulty is reduced, the operation time and the pain of the patient are reduced, and the method andthe device have high application and popularization significance.

The application discloses a method and device for extracting skeletons of a brain CT (computerized tomography) image. The method comprises the steps of: carrying out grey mapping on pixel points of an original brain CT image to obtain a corresponding grey column diagram, and acquiring a first preset division threshold according to the grey column diagram; calculating a background pixel grey threshold according to the first preset division threshold, and removing background pixel points in the original brain CT image according to the background pixel grey threshold to obtain a pre-processed brain CT image; calculating a skeleton threshold based on the pre-processed brain CT image; and searching all pixel points of which the grey values are larger than or equal to the skeleton threshold in the pre-processed brain CT image, and extracting the image corresponding to the pixel points, wherein the image represents brain skeletons of the brain CT image. By applying the invention, the skeleton extraction error caused by carrying out empirical value setting on the skeleton threshold in advance is avoided; in addition, the method and device disclosed by the application can be used for extracting the image of brain skeletons within short time and do not bring negative influence to operation speed of integral algorithm.

The invention discloses a cerebral hematoma segmentation method and system based on deep learning. The method comprises the following steps: constructing a neural network model, the neural network model comprising a plurality of image information compression modules which are connected in sequence and a plurality of image information fusion modules which are connected in sequence, wherein the image information compression module comprises a first self-attention convolution unit, a second self-attention convolution unit and a pooling layer which are connected in sequence, and the image information fusion module comprises an up-sampling unit, a feature map splicing unit and a third self-attention convolution unit which are connected in sequence; obtaining a brain CT sample image; training the neural network model by taking the brain CT sample image as input and the bleeding condition of each pixel point in the brain CT sample image as a label; and adopting the trained neural network model to perform cerebral hemorrhage identification on the to-be-segmented brain CT image. According to the method, the bleeding area in the brain CT image can be accurately and efficiently segmented.

The invention belongs to the field of image processing, and particularly relates to a CT image segmentation method based on an improved AU-Net network, and the method comprises the steps: obtaining a to-be-segmented brain CT image, and carrying out the preprocessing of the obtained brain CT image; inputting the processed image into the trained improved AU-Net network for image recognition and segmentation to obtain a segmented CT image; identifying a cerebral hemorrhage area according to the segmented brain CT image. The improved AU-Net network comprises an encoder, a decoder and a hopping connection part. Aiming at the problem of low segmentation precision caused by large size and shape difference of the hemorrhage part of the cerebral hemorrhage CT image, the invention provides a coding-decoding-based structure, and a residual octave convolution module is designed in the structure, so that the model can segment and identify the image more accurately.

The invention implements a method and a device for detecting ischemic stroke focus based on brain CT images. In this method, brain CT images were preprocessed and region growing based on region growing method, which increased the difference of image characteristics between lesion region and normal brain. Then, based on the symmetry of the brain, The images of the left and right brain regions are divided into a plurality of pixel squares, and the characteristic differences of two pixel squares with symmetry positions in the left and right brain are compared one by one, so that it is possible todetermine which pixel square contains the focus of ischemic stroke according to the characteristic differences. Through the above steps, the purpose of effectively detecting ischemic stroke lesions by brain CT images is realized.

The invention provides a method for automatically generating a brain CT medical report based on weak supervised attention, relates to the three fields of medical images, computer vision and natural language processing, and designs a weak supervised attention mechanism WGAM to clearly guide an attention model to focus on a focus region so as to improve the accuracy of medical report generation. WGAM is a hierarchical structure and comprises two attention mechanisms of space attention and sequence attention, and the space attention is weakly supervised by a gradient weighted class activation mapping algorithm to obtain a better attention effect. A keyword-driven interactive loop network KIRN is designed as a language generation module to generate a brain CT medical report, a hidden layer state of the language generation module is activated through keyword information containing focus position information, and the accuracy of generating a brain CT image report is improved through dynamic interaction of LSTMword and LSTMsen. According to the method, the work of automatically generating the brain CT medical report is explored for the first time, and effectiveness is achieved.

The invention provides a brain centerline detection method and system, a terminal and a storage medium. The method comprises the steps of obtaining a brain CT image and corresponding annotation data of a brain centerline offset patient; preprocessing the brain CT image and the corresponding annotation data, and dividing the preprocessed brain CT image and the corresponding annotation data into a training set and a test set according to a preset proportion; respectively inputting the training set into a preset attitude correction network, a context-aware feature optimization network and a neutral line regression network for training to obtain an initial brain neutral line detection model; performing optimization test on the initial brain centerline detection model through the test set to obtain a brain centerline detection model. The invention can automatically detect the brain centerlines with different offset degrees, can be used for subsequent measurement of brain centerline offset to obtain the coordinate of the offset brain centerline, and realizes efficient and accurate detection of the brain centerline.

The invention relates to a brain auxiliary diagnosis system which effectively assists primary doctors in diagnosing specialized brain diseases. The system comprises a medical image browsing module which is responsible for storing and displaying a brain CT image file and providing a paintbrush tool for a doctor to mark a suspected focus point area, a data preprocessing module responsible for splicing and establishing a unique feature fingerprint of the image file and marking coordinate points of an area, a network transfer module responsible for sending a similar data set C to the medical imagebrowsing module, a data analysis module responsible for analyzing a received information carrier to obtain a suspected lesion file fingerprint instruction and a coordinate instruction, a matching model module responsible for establishing a feature fingerprint database, a medical information database and an image database, calculating similarity and determining a similar target set B, and a database searching and mapping module used for obtaining illness state data and an image file, and determining a similar target set C relative to the B and taking the C as the input of the network transfermodule.

The invention discloses a surgical navigation and mechanical arm device for craniocerebral puncture. The device comprises a three-dimensional CT scanning device, a handheld scanner, a computer system,a mechanical arm control system and a mechanical arm execution system, wherein the three-dimensional CT scanning device, the handheld scanner, the mechanical arm control system and the mechanical armexecution system are connected with the computer system; and the mechanical arm control system is connected with the mechanical arm execution system. Moreover, the invention also provides a positioning method of the device. The method mainly comprises the following steps of obtaining a target point needing a minimally invasive surgery through preoperative brain CT, performing fixed point markingon the brain, performing scanning by using a probe, planning a surgery track, and registering the probe position of the mechanical arm and a fixed point mark in an image space. After registration, a doctor can obtain coordinates of any feature point of a human-computer interface by clicking the feature point, and the mechanical arm moves the surgical tool after receiving the coordinates of the feature point, so that the doctor can accurately find target features.

The invention discloses a cerebral hemorrhage automatic detection method based on a brain auxiliary image and an electronic medium, and is applied to the technical field of image processing, and the method comprises the steps: carrying out the CT scanning of the head of a patient, and obtaining a brain CT image; generating a brain auxiliary image according to the brain CT image; and inputting thebrain CT image and the brain auxiliary image into a pre-trained neural network, automatically detecting a cerebral hemorrhage area in the brain CT image in combination with the brain auxiliary image,and outputting a mask including the cerebral hemorrhage area and type. According to the invention, the brain auxiliary image and the brain CT image scanned by the patient are combined for use, the cerebral hemorrhage area and type identification accuracy of the neural network is improved, the technical requirements of doctors in the diagnosis process are reduced, and the diagnosis efficiency is greatly improved.

The invention discloses a brain arteriovenous malformation detection method and detection system based on CT images. The method comprises the steps of preprocessing the brain CT images to extract thebrain effective area; scanning the brain effective area on the basis of a three-dimensional convolution neural network algorithm, and automatically positioning all the lesion areas with brain arteriovenous malformation lesions; carrying out image segmentation on the edges of the lesion areas on the basis of the three-dimensional convolution neural network algorithm to automatically obtain lesion outline areas, and accurately distinguishing the lesions from normal brain tissue around; and automatically measuring the average density of the brain arteriovenous malformation lesions on the basis ofthe lesion outline areas. With the brain arteriovenous malformation detection method and detection system based on the CT images, the function of carrying out automatic positioning, edge automatic segmentation, average density automatic measurement and the like on the brain arteriovenous malformation lesion areas can be implemented; and furthermore, various image characteristics output when the detection is in use can be provided for doctors as the basis for determination so as to help the doctors do the clinical classification diagnosis work on the brain arteriovenous malformation lesions better.

The invention discloses a medical report generation method based on hierarchical self-attention sequence coding. The method comprises the following steps: (1) acquiring a brain CT image and corresponding medical report data, and preprocessing the brain CT image and the corresponding medical report data; (2) constructing a feature extractor; (3) constructing a sequence processor, and obtaining an image feature code VNSA containing information of each adjacent fault block and a three-dimensional brain CT image feature code VSA based on the whole case after the sequence processor passes through the sequence processor; (4) constructing a decoder; and (5) performing model training. The application and development of deep learning in intelligent medical treatment are rapid, and the automatic generation technology of the medical report for the lung is mature, but the research and invention for the automatic generation of the medical report for the brain CT are vacant; therefore, according to the method the model realizes the coding of the three-dimensional brain CT data, and combines the coding with a language model in the field of image description to realize the automatic generation of the medical report of the CT image.

The invention discloses a method, a device and a system for obtaining a brain CT perfusion imaging parameter graph and a computer storage medium. The method comprises the following steps: obtaining a CT perfusion image, carrying out the pretreatment of the CT perfusion image, and obtaining a discrete tissue density change curve C (n); reading collection time information of the CT perfusion image to obtain an collection time array T (n); intercepting an collection time array T (n) to obtain a relative collection time array t (n); combining the tissue density change curve C (n) with the corresponding relative collection time array t (n) to obtain a tissue density time curve C (tn) of each pixel point; and after fitting or interpolating the discrete tissue density time curve C (tn), re-dispersing according to the same time interval to obtain a discrete tissue density time curve C (n)'; and performing the same processing on the artery input change curve AIF (n) and the vein output change curve VOF (n). According to the method, the availability of the tissue density time curve is improved, the solving difficulty is reduced, and the solving accuracy is improved.

The invention discloses an auxiliary diagnosis system based on a Mask R-CNN network and an auxiliary diagnosis information generation method, and belongs to the technical field of medical image processing and segmentation, and the system comprises: a data uploading end which is used for uploading a CT image of a patient and corresponding medical auxiliary information, and the CT image carries the cerebral hemorrhage information of the patient; the image processor that is connected with the data uploading end and is used for enhancing and graying the CT image to obtain a grayscale image; the target detection module that is connected with the image processor, and is used for detecting a grayscale image by using a trained Mask R-CNN network model so as to identify and extract lesion feature information; and the diagnosis analysis module that is connected with the target detection module, and is used for matching the focus characteristic information with the structured medical record in the medical record database, and synthesizing medical auxiliary information to generate auxiliary diagnosis information. According to the invention, the trained Mask R-CNN network model is utilized to scan the brain CT image to generate the auxiliary diagnosis information, so that the efficiency of cerebral hemorrhage screening in emergency treatment is improved.

The invention provides an early acute cerebral infarction detection method and device. The early acute cerebral infarction detection method comprises the following steps of: acquiring a plurality of diagnosed early acute cerebral infarction brain CT images; constructing an early acute cerebral infarction detection model based on a 2.5 D segmentation network; training the early acute cerebral infarction detection model through the early acute cerebral infarction brain CT images; and inputting a to-be-detected brain CT image, performing early acute cerebral infarction probability prediction on each voxel point of the to-be-detected brain CT image through the trained early acute cerebral infarction detection model, and realizing early acute cerebral infarction detection according to a prediction result. According to the invention, early acute cerebral infarction can be automatically detected, and the method has high efficiency and accuracy.

The invention discloses a brain CT image classification method fusing multi-scale superpixels, and belongs to the field of medical image research. The method has the following characteristics: 1) multi-scale superpixels and a brain CT image are fused, image redundant information is removed, and the gray similarity between a focus and surrounding brain tissue pixels is reduced; 2) a multi-scale superpixel encoder based on regions and boundaries is designed, and focus low-level information contained in multi-scale superpixels is effectively extracted; 3) a fusion model fusing multi-scale super-pixel features is designed, and comprehensively utilizing of high-level features extracted by a residual neural network and low-level features of multi-scale super-pixels is carried out to realize classification of brain CT; and 4) compared with the traditional deep learning algorithm, the method provided by the invention can effectively utilize the lesion information contained in the multi-scale superpixels, thereby more accurately classifying the diseases existing in the brain CT image, and the method is reasonable and reliable, and can provide powerful help for the classification of the brain CT image.

The invention relates to a brain image segmentation method and device, computer equipment and a readable storage medium. The method comprises the steps that a brain CT image is acquired, wherein the brain CT image comprises adjacent multi-layer sectional images; the brain CT image is input into a preset segmentation model to obtain an intracranial hemorrhage segmentation result corresponding to the brain CT image, wherein the segmentation model is obtained by training according to the sample image, a segmentation gold standard and a classification gold standard, and the classification gold standard is used for representing a bleeding subtype of intracranial bleeding in the sample image. According to the method, the classification loss is used for guiding the segmentation model to carry outtraining, so that the precision of the segmentation model obtained by training can be greatly improved, and the precision of a brain CT image segmentation result is further improved. Meanwhile, the brain CT image is an adjacent multi-layer sectional image, so that the context information among all layers of images can be fully considered, and the precision of the obtained brain CT image segmentation result is further improved.

The invention relates to an intracranial hemorrhage parameter acquisition method and device based on self-supervised learning and M-Net. Comprising the steps of obtaining brain CT sequence images, and preprocessing the brain CT sequence images, where the preprocessed brain CT sequence images at least comprise brain CT marked images and brain CT unmarked images, and the brain CT marked images are used for marking intracranial hemorrhage areas; inputting the brain CT unmarked image into a self-supervised learning network model for pre-training to obtain a first model; modifying model parameters of the first model based on the brain CT marker image data to obtain a second model; inputting the to-be-tested brain CT image into the second model, obtaining an intracranial hemorrhage area segmentation image, obtaining hemorrhage parameters of the to-be-tested brain CT image based on the segmentation image, where the hemorrhage parameters comprise a hemorrhage area and a hemorrhage area. According to the invention, a method of combining self-supervised learning and deep learning is adopted, and rapid identification of cerebral hemorrhage features and accurate calculation of the hemorrhage area are realized.

The invention discloses a cerebral hemorrhage CT image classification method based on image sequence analysis. The method belongs to the technical field of medical image analysis, and the operation process comprises the steps of brain CT image data acquisition, data imaging processing, model construction and classification method implementation. The method comprises the following specific operation steps: acquiring a brain CT scanning result to obtain brain CT image original data; performing imaging processing on the original data of the brain CT image to form a three-channel image sequence as the input of a subsequent model; constructing a model comprising a channel attention module and a space attention module to extract features of the three-channel image sequence; according to the method, the model is trained to optimize parameters, cerebral hemorrhage positive is recognized, cerebral hemorrhage subtypes are distinguished, cerebral hemorrhage CT images are classified, CT scanning image information can be fully utilized by the model, attention to key information in an image sequence is improved, the judgment process of the whole method conforms to the diagnosis process of doctors in a real scene, and the diagnosis efficiency is improved. And the cerebral hemorrhage classification effect can be improved.

The embodiment of the invention discloses a method, device and equipment for segmenting a ventricular region in a brain CT image, and the method comprises the steps: recognizing a brain tissue regionin the brain CT image, carrying out the two-group clustering of the brain tissue region, and obtaining a background region and a non-background region of the brain tissue region; identifying connecteddomain areas in the non-background are, and sorting all the connected domain areas according to the number of included pixel points; determining the number of reserved connected domains, and selecting the first N connected domain regions including a large number of pixel points as candidate ventricular regions; judging whether the current candidate ventricular region can be determined as the ventricular region or not by utilizing the first target ratio and the first threshold value; and if not, adjusting the candidate ventricle. Therefore, the ventricular region obtained by segmentation is more complete and accurate. The ventricular region in the brain CT image can be automatically divided, so that the efficiency of dividing the ventricular region in the brain CT image is improved.

The invention provides a method, a device and equipment for determining a bleeding area based on a brain image, and a medium. The method comprises the following steps: determining a brain tissue area of each image in a brain CT image sequence; according to each brain tissue area, determining a ventricular edge corresponding to each brain tissue area; performing Hough transform straight line detection on each ventricular edge to obtain a target line segment of each ventricular edge; and determining a target area according to the target line segment of each ventricle edge, performing parabola fitting on pixel points of the target area to obtain a fitting result, and determining whether a bleeding area exists in the target area or not according to the fitting result. According to the method, the bleeding area is automatically determined based on the brain CT image, manual participation is not needed, the recognition speed is high, and the accuracy rate is high.

The present application discloses an automatic positioning system and method for human body transcranial ultrasound stimulation based on a mechanical arm. The system is used to automatically perform ultrasonic stimulation to a preset position of the head and comprises an input module, a spatial positioning module, an ultrasonic stimulation module, a safety ellipsoid determination module and a control module; the input module is used to input brain CT or magnetic resonance scan data of subjects; the spatial positioning module is used to obtain real-time three-dimensional coordinate informationof the head and to obtain real-time three-dimensional coordinate information of the ultrasound stimulation module; the safety ellipsoid determination module is used to determine a safety ellipsoid that surrounds the entire head and forms a predetermined safe distance from an outer surface of the head; and the control module is used to control a multi-degree-of-freedom mechanical arm to drive a ultrasound transducer coupling device to move, so that the ultrasonic transducer coupling device and the head form a fit contact at a preset position of the scalp. The automatic positioning system and method have advantages of high spatial accuracy and high reliability.

The invention provides an old cerebral infarction detection method and device. The method comprises the following steps: acquiring a plurality of diagnosed old cerebral infarction brain CT images; constructing an old cerebral infarction detection model based on a 2.5 D segmentation network; training the old cerebral infarction detection model through the old cerebral infarction brain CT image; inputting a to-be-detected brain CT image, carrying out the prediction of the old cerebral infarction probability of each voxel point of the to-be-detected brain CT image through the trained old cerebralinfarction detection model, and achieving the detection of the old cerebral infarction according to a prediction result. According to the method, the old cerebral infarction can be automatically detected, and the method has relatively high efficiency and accuracy.