257results about "Blood vessel patterns" patented technology

The invention relates to the technical field of image processing, and discloses a pulmonary embolism focus segmentation method and system based on image recognition, and a storage medium. The method comprises the following steps: extracting a multi-level blood vessel topological structure of a CTPA image through blood vessel diameter gradient analysis; modeling blood vessel density distribution by using a Weibull mixed model to obtain embolism characteristic parameters; the pixel embolism probability is estimated through variational Bayesian reasoning, and a focus distribution diagram is generated; performing multi-scale feature fusion on the lesion probability graph to obtain a segmentation boundary; and obtaining a final embolism focus segmentation result based on the vascular connectivity constraint optimization boundary. The problems that blood vessel level differentiation processing cannot be achieved, and accurate probability modeling and anatomical constraint verification are lacked are solved. The accuracy of pulmonary embolism focus segmentation is improved.

The disclosure herein relates to systems, methods, and devices for medical image analysis, diagnosis, risk stratification, decision making and / or disease tracking. In some embodiments, the systems, devices, and methods described herein are configured to analyze non-invasive medical images of a subject to automatically and / or dynamically identify one or more features, such as plaque and vessels, and / or derive one or more quantified plaque parameters, such as radiodensity, radiodensity composition, volume, radiodensity heterogeneity, geometry, location, and / or the like. In some embodiments, the systems, devices, and methods described herein are further configured to generate one or more assessments of plaque-based diseases from raw medical images using one or more of the identified features and / or quantified parameters.

The invention relates to the technical field of ultrasonic detection, and discloses a robot ultrasonic detection method, device and equipment based on three-dimensional perception and a medium, and the method comprises the steps: obtaining three-dimensional point cloud data of the surface of a target object; generating a scanning path for detecting the target object based on the three-dimensional point cloud data; controlling a motion execution device to drive the detection probe to move along the scanning path to obtain ultrasonic image data of the internal structure of the target object; in the moving process of the detection probe, the contact force between the detection probe and the target object is maintained within a preset force value range; analyzing the ultrasonic image data, identifying a lumen contour and determining a narrowest position point; and adjusting the posture of the detection probe at the narrowest position point and executing secondary detection to obtain hydrodynamic data. The self-adaptive scanning path is generated through the three-dimensional point cloud data, the movement and the contact force of the detection probe are controlled, the ultrasonic image and the space position are associated, and the problems that the detection path is not attached to the curved surface, the contact force fluctuates, and the image space is not synchronous are solved.

The invention discloses an OCT image segmentation method and system based on cross-domain attention and multi-level feature fusion, and relates to the technical field of image segmentation. The method comprises the steps that an image segmentation model based on cross-domain attention and multi-level feature fusion is trained based on an obtained OCT blood vessel image training set, and the operation mechanism of the model is that an OCT blood vessel image enters an encoder to be subjected to feature coding and then enters a multi-scale feature extraction module to be subjected to self-adaptive multi-scale feature extraction; the features output by each down-sampling module in the encoder and the features output by the multivariate scale feature extraction module enter a multi-layer dense fusion module together for feature fusion; the features output by the multi-layer dense fusion module and the features output by the multivariate scale feature extraction module enter a decoder together for feature decoding; and then entering a segmentation head to generate a final segmentation image. The method can be used for automatically identifying the blood vessel plaque in the OCT image, and the overall quality and the detail integrity of a segmentation result can be remarkably improved.

The invention discloses a dynamic holographic reconstruction system for CTA and DSA coronary artery image fusion. An image acquisition module transmits acquired image data to a data storage unit for storage; the dynamic registration module is connected with the image acquisition module, a dynamic feature extraction unit and a space-time calibration unit are arranged in the dynamic registration module, and a time synchronization mechanism is established; the image fusion module adopts a fusion algorithm based on deep learning to perform deep fusion on the preprocessed images to obtain fused coronary artery image data; the holographic reconstruction module is connected with the image fusion module, performs three-dimensional holographic reconstruction according to the fused coronary artery image data, and constructs a holographic three-dimensional model of the coronary artery by adopting a three-dimensional reconstruction technology; the display module is connected with the holographic reconstruction module and is used for displaying the coronary artery image after dynamic holographic reconstruction; the control module is used for controlling working parameters and operation processes of all the modules. According to the invention, accurate dynamic fusion and dynamic holographic reconstruction of coronary images can be realized, and a more comprehensive and visual image basis is provided for diagnosis and treatment evaluation of coronary diseases.

The embodiment of the invention provides a pulmonary artery image anomaly classification method and device, equipment and a medium. The pulmonary artery image anomaly classification method comprises the steps that a target CT image and a target ultrasonic image of a pulmonary artery are acquired; respectively coding the target CT image and the target ultrasonic image, and respectively extracting first global and local features of the target CT image and second global and local features of the target ultrasonic image after coding; performing weighted fusion on the first global and local features and the second global and local features through a preset multi-head attention mechanism to obtain multi-modal fusion features; performing decoding processing on the multi-modal fusion features, and performing reconstruction after decoding processing to obtain a target fusion image of the pulmonary artery; and performing anomaly classification based on the target fusion image to obtain an anomaly classification result of the pulmonary artery, thereby improving the accuracy of anomaly classification of the pulmonary artery image.

The invention discloses a novel graph contrast learning framework with mutual information maximization, and relates to the technical field of graph contrast learning, multi-modal learning and mutual information maximization. The framework comprises the following steps: constructing a nearest neighbor graph, spreading similar samples to data of a missing mode, optimizing a loss function, and performing clustering optimization. The existing method has two main problems: (1) the existing method is mostly suitable for complete data and has a poor processing effect on data of a missing mode; and (2) the current palmprint and palm vein fusion recognition technology does not completely capture the complex relationship between the palmprint and palm vein features, resulting in inaccurate clustering result. According to the method, the modal loss in the incomplete palm print and palm vein fusion recognition task is effectively reduced, and the clustering performance is enhanced. According to the method, the mutual information maximization loss function is introduced, the adaptability and robustness of the model to incomplete data are effectively enhanced, and the method has high recognition precision and clustering performance when processing incomplete multi-modal data.

The invention relates to the field of medical image processing, in particular to a gastric cancer precancerous lesion progress risk assessment system based on time sequence image analysis, which comprises a time sequence gastroscope image registration module, a lesion curved surface dynamic representation module, a multi-dimensional feature manifold construction module, a risk dynamic prediction module and a virtual pigment endoscope enhancement module, according to the method, a differential geometry theory is introduced, the gastric mucosa surface is modeled as a Riemannian curved surface, and Gaussian curvature, average curvature and other characteristics are extracted; constructing a feature manifold space by adopting a manifold learning method, and analyzing a lesion state evolution trajectory through geodesic distance; and predicting the time for the lesion to reach a high-risk state based on the probability density distribution and the evolution vector field on the manifold. According to the method, accurate characterization of gastric cancer precancerous lesion morphological characteristics, quantitative analysis of dynamic evolution and accurate prediction of risk progress are realized, a scientific basis is provided for clinical follow-up visit decisions, and the method has the remarkable advantages of improving the early intervention rate, optimizing medical resource allocation and the like.

The invention discloses a non-contact palm vein multi-mode identification system based on deep learning. The system comprises a multi-mode image acquisition and processing terminal, a mode matching and identification decision terminal, a safety protection and anti-counterfeiting terminal and an iterative optimization terminal. The multi-modal image acquisition and processing terminal is used for acquiring palm print and palm vein images, performing dynamic calibration, image quality enhancement and preprocessing, and outputting multi-modal images; the mode matching and identification decision terminal is used for carrying out feature extraction, multi-modal feature fusion, mode matching and identification decision and outputting a final matching result; the safety protection and anti-counterfeiting terminal is used for carrying out living body detection, anti-counterfeiting monitoring and data encryption; and the iterative optimization terminal is used for realizing model lightweight, knowledge distillation, end-to-end optimization and continuous learning. According to the invention, through deformation invariant feature extraction and cross-modal attention fusion, the recognition accuracy and environmental adaptability in a complex state are improved.

The invention provides a palm vein high-security identity authentication system based on spectral analysis and deep learning, and relates to the technical field of biological recognition, the system comprises a hardware layer, a data processing layer, an algorithm layer and an application layer, through an antagonistic living body detection module in the algorithm layer, in combination with physical modeling and a GAN (Generative Adversarial Network), the high-security identity authentication of a complex imitation is realized. The method comprises the following steps: carrying out effective identification of a high-biomimetic material, carrying out physical modeling to analyze the absorption characteristics of palm veins in a hyperspectral range, such as the difference between oxyhemoglobin and oxyhemoglobin and periodic spectrum micro-interference caused by heart rate, and providing living body evidence based on physiological dynamics; meanwhile, a high-simulation template sample generated by the GAN is used for training a discriminator, the insight ability of the system to a template means is further enhanced, and compared with single traditional texture or static signal detection, the system combines physical characteristics with deep learning, and the confidence coefficient of in-vivo detection reaches up to 99.99%.

The invention provides a registration method and device in a cardiac surgery and computer equipment, which are used for improving the safety and accuracy of the surgery. The registration method comprises the following steps: acquiring an electrocardiosignal and multi-modal physiological data of a surgical patient, and performing cardiac time phase division on a cardiac cycle based on the electrocardiosignal and the multi-modal physiological data; collecting multi-angle image data of the heart under each cardiac time phase, and synchronously collecting physiological parameter data; performing three-dimensional reconstruction based on the multi-angle image data corresponding to each cardiac time, and mapping the physiological parameter data corresponding to each cardiac time to a corresponding three-dimensional reconstruction result to obtain a multi-temporal three-dimensional heart model library; the method comprises the following steps: acquiring real-time electrocardiosignals and real-time multi-modal physiological data of a surgical patient in a surgical process to identify a current cardiac time phase, selecting a current time phase three-dimensional heart model from a multi-time phase three-dimensional heart model library according to the current cardiac time phase, and performing surgical registration by using the current time phase three-dimensional heart model.

The invention relates to a biometric enrollment method comprising the following steps: - obtaining a video stream showing the vicinity of a biometric data capture device (2) for a hand (3); - a user (4) presenting their hand to the capture device (2), obtaining a virtual mask of the hand (3) from the video stream; - generating a virtual hand (13) from the virtual mask; - integrating the generated virtual hand (13) into the video stream in a position indicating a theoretically optimal positioning of the user's (4) hand (3); - displaying the video stream including the integrated virtual hand (13) on a screen (7) visible to the user (4); - once a positioning of the user's (4) hand (3) has been validated, capturing the biometric data of the user's hand. Figure 1 for the abstract

The application provides a blood vessel lesion determination method and device, electronic equipment and storage medium. The determination method comprises: inputting acquired no-label images and labeled images as input images into a blood vessel lesion analysis model in a single-alternating input manner; if the input image is a no-label image, training the blood vessel lesion analysis model according to a reconstruction loss function; if the input image is a labeled image, determining whether the reconstruction loss function, a lesion type loss function and a lesion degree loss function simultaneously satisfy a target condition, and if the target condition is satisfied, obtaining a trained blood vessel lesion analysis model; and determining a lesion type result and a lesion degree result of a blood vessel image according to the trained blood vessel lesion analysis model. The technical solution provided by the application can reduce manual labeling operations, reduce the workload of doctors, and ensure the accuracy of blood vessel lesion determination.

The present application belongs to the technical field of palm vein recognition, and particularly relates to a palm vein recognition data enhancement method and system based on adversarial learning, which utilizes a conditional deep convolutional generative adversarial network composed of a generator and a discriminator, and synchronously trains the generator and the discriminator; step 2 selects a palm vein classifier and the trained generator to form an adversarial network, step 3 performs multiple batches of adversarial training on the palm vein classifier based on the adversarial network, generates adversarial samples for increasing the loss of the classifier and training samples based on real palm vein recognition data to form an augmented data set in each batch of training, trains the classifier, and updates the parameters of the classifier; and updates the noise vector set required for the next batch of adversarial training. The present application reduces the requirements for computing and storage resources in data enhancement, and provides more high-quality training samples for palm vein recognition model training.

A deep learning model for the detection of obstructive coronary artery disease (CAD) can take a set of polar maps and patient information as input, then output obstructive CAD scoring data, such as probabilities of obstructive CAD associated with various cardiac territories, as well as an attention map and a CAD scoring map. The model can operate agnostic of camera type used to capture the set of polar maps. The attention map indicates regions of the polar maps important to the deep learning process for that particular set of polar maps. The attention map and obstructive CAD scoring data can be used to generate a CAD scoring map showing CAD probability by segment on a standard 17-segment model of a left ventricle. The attention map and / or CAD scoring map can act as easily explainable tools for interpreting the results of a myocardial perfusion imaging study.

The present application relates to the technical field of physical analysis, in particular to a blood smear quality intelligent detection method for clinical laboratory, comprising the following steps: extracting pixel direction and building path block, marking turning connection point as contour chain, splicing boundary to generate closed area, extracting gray gradient to label structure change, matching contour to judge offset overlap, extracting channel barycenter to compare axial difference, and positioning direction abnormal area to generate detection scheme. In the present application, the structural bending points in the image are recognized and the continuous path is constructed, the spatial expression of the shape turning area is enhanced, the accuracy of the closed boundary is improved by combining the edge head-tail coordination judgment, the structure partition identification is constructed relying on the gray gradient and the direction difference, the classification labeling of the structure disturbance area is carried out, the abnormal area is integrated based on the spatial overlap and the direction consistency, the color axial difference is judged by the channel barycenter offset, the comprehensive recognition ability for the shape variation, the boundary offset and the dyeing abnormality is improved, and the multi-dimensional intelligent judgment of the image quality is realized.

The embodiment of the invention provides a fundus artery and vein segmentation method and device, electronic equipment and a storage medium, and belongs to the technical field of image processing. The method comprises the following steps: acquiring a fundus blood vessel topological structure labeling sample set, and acquiring a fundus artery and vein labeling sample set; acquiring an arteriovenous segmentation loss function; based on the fundus blood vessel topological structure labeling sample set, performing model training to obtain a target fundus blood vessel topological structure classification model; based on the target fundus blood vessel topological structure classification model, performing function adjustment on the arteriovenous segmentation loss function to obtain an initial total loss function; based on the fundus artery and vein labeling sample set and the initial total loss function, performing model construction to obtain a target fundus artery and vein blood vessel segmentation model; and based on the target fundus artery and vein segmentation model, performing artery and vein segmentation on preset fundus artery and vein image data to obtain a fundus artery and vein segmentation image. According to the embodiment of the invention, the accuracy of fundus artery and vein segmentation can be improved.

This invention discloses an OCT image segmentation method and system based on cross-domain attention and multi-level feature fusion, belonging to the field of image segmentation technology. The method includes: training an image segmentation model based on cross-domain attention and multi-level feature fusion using an acquired OCT vascular image training set. The model's operating mechanism is as follows: the OCT vascular image enters an encoder for feature encoding, then enters a multi-scale feature extraction module for adaptive multi-scale feature extraction; the features output by each downsampling module in the encoder and the features output by the multi-scale feature extraction module are combined and fused together in a multi-layer dense fusion module; the features output by the multi-layer dense fusion module and the features output by the multi-scale feature extraction module are combined and decoded together in a decoder; finally, the image is processed by a segmentation head to generate the final segmentation map. This invention can be used for the automatic identification of vascular plaques in OCT images, significantly improving the overall quality and detail integrity of the segmentation results.

A blood vessel image segmentation method and device, electronic equipment and storage medium are disclosed. The method comprises: acquiring an initial blood vessel image, determining a coronary skeleton image of the initial blood vessel image; determining each skeleton point on the skeleton line in the coronary skeleton image, and determining at least one partition skeleton point in each skeleton point based on the degree of each skeleton point; determining the partition skeleton line segment corresponding to each partition skeleton point, and performing vein identification on each partition skeleton line segment to determine the vein branch line segment in the coronary skeleton image; and determining the coronary segmentation image of the initial blood vessel image based on the initial blood vessel image and the vein branch line segment. The technical solution disclosed in the present application solves the problem of low accuracy of coronary segmentation of blood vessels in the prior art, and improves the accuracy of coronary segmentation.

User enrollment in a biometric identification system begins with a pre-enrollment process on a selected generic input device (GID), such as a smartphone. The user enters identifying data, such as their name, and may use the GID's camera to capture first image data, such as of their hand. The first image data is processed to determine a first representation. Upon presenting the hand at the biometric input device, second image data is captured. The second image data is processed to determine a second representation. If the second representation is deemed associated with the first representation, the enrollment process may be completed by saving the second representation for later use.

The application provides a kind of blood vessel anastomosis skill automatic evaluation method based on computer vision technology, it is related to video processing technical field, the method is to utilize unlabelled blood vessel anastomosis operation video data, visual feature extraction model is trained, obtains pre-trained visual feature extraction model, blood vessel anastomosis operation video data is extracted, and blood vessel anastomosis skill visual feature is obtained;Blood vessel anastomosis skill visual feature is up-sampled using high-resolution pyramid, and high spatial resolution feature map is obtained;Using a variety of downstream task models, high spatial resolution feature map is operated action class identification, target segmentation and tip position identification;Action time consumption is calculated using operation action class, motion trajectory is obtained using instrument tip position, suture binary mask is identified using target segmentation, multidimensional index is calculated, and blood vessel anastomosis skill automatic evaluation result is obtained.The application solves the problem that the prior art is difficult to improve small target recognition accuracy.

This computer program causes a computer to execute processing of: acquiring, on the basis of a signal detected by a catheter inserted into a hollow organ, a plurality of first medical images generated along an axial direction of the hollow organ; extracting, from the acquired plurality of first medical images, a plurality of second medical images which are at a prescribed interval along the axial direction of the hollow organ; identifying a type and a region of an object included in the extracted plurality of second medical images, by inputting the plurality of second medical images into a learning model which outputs, in response to input of a medical image, information related to an object included in the medical image; determining a reference image from among the plurality of second medical images on the basis of the type and the region of the object; and inputting, to the learning model, the first medical images within a prescribed range in the axial direction of the hollow organ from the reference image.

The vessels of a biological object are to be segmented more reliably. To this end, a method for training a machine learning algorithm to segment such vessels is proposed. First, a 3D reconstruction (8) of the vessels is provided. A starting vessel region (7) is identified in the 3D reconstruction (8). Sub-regions representing a starting vessel segment are extracted from the starting vessel region (7). The algorithm is trained using these extracted sub-regions. Subsequently, the trained algorithm is applied to a first neighboring region (10) immediately adjacent to the starting vessel region (7). This identifies a first vessel segment in the first neighboring region (10). Finally, the algorithm is retrained using the first vessel segment identified in the first neighboring region (10).

The invention discloses an identification method suitable for retinal artery occlusion. The method comprises the following steps: S10, acquiring a CFP image and an OCT image; s20, carrying out differential preprocessing, and carrying out standardization, enhancement and normalization processing on the obtained image; s30, performing feature fusion by using a double-model freezing mode, including the steps of performing general feature extraction by using a basic model, performing local feature enhancement by using a small model, and performing feature fusion; and S40, carrying out multi-modal dynamic decision making, predicting the fused data through a classifier to obtain a prediction result, and integrating multi-image prediction results by adopting a maximum probability selection strategy to obtain an identification result of the retinal artery occlusion. According to the method, the advantages of the multi-modal image can be effectively fused, and the powerful deep learning model is utilized to realize the new method of accurate and automatic identification, so that the defects of the prior art are overcome, and efficient and reliable technical support is provided for early identification of the RAO.

The invention relates to the technical field of intelligent injection, and particularly discloses an intelligent anesthesia injection system based on an electronic booster, which comprises a local blood vessel thickness feature extraction module for extracting local blood vessel thickness features of an anesthesia injection part; the initial injection speed determination module determines an initial injection speed based on local blood vessel thickness characteristics of an anesthesia injection part; the first injection propulsion module obtains pressure record data of the syringe needle in the piercing process, controls an electronic booster to conduct propulsion injection based on the initial injection speed, and meanwhile obtains pressure record data of the syringe needle in the injection process; the injection speed optimization module determines an optimized injection speed based on the pressure record data of the syringe needle in the piercing process, the pressure record data of the syringe needle in the injection process and the initial injection speed; the second propulsion injection module controls the electronic booster to carry out propulsion injection based on the optimized injection speed; the speed of anesthesia injection is intelligently determined, and the precision of anesthesia injection is improved.

The present disclosure provides systems and methods for multi-modal analysis of patient-specific anatomical features from medical images. The system can receive a medical image of a patient and metadata associated with the medical image indicative of a selected pathology, and automatically classify the medical image using a segmentation algorithm. The system can use an anatomical feature identification algorithm to identify one or more patient-specific anatomical features within the medical image by exploring an anatomical knowledge dataset. A 3D surface mesh model representing the one or more classified patient-specific anatomical features can be generated, such that information can be extracted from the 3D surface mesh model based on the selected pathology. Physiological information associated with the selected pathology for the 3D surface mesh model can be generated based on the extracted information.

According to an aspect, a detection device includes: a sensor comprising a plurality of optical sensor elements arranged two-dimensionally; and an acquirer configured to acquire a pattern of a blood vessel of a human finger that is included in a light intensity pattern of light detected by the sensor. The acquirer is configured to acquire information on force from the finger toward the sensor based on the light intensity pattern.

A biometric recognition device for palm prints and palm veins includes: a light source being used to emit illuminating beams in a near-infrared band; a polarizing unit being used to convert the illuminating beams to polarized beams which are emitted to a palm; an imaging unit receiving retroreflected beams from a palm and being used to converge the retroreflected beams to a focal plane; a polarizing beam splitter forming transmitted beams and reflected beams based on different polarization directions of beams from the imaging unit, wherein the transmitted beams propagate after pass through the polarizing beam splitter, and the reflected beams propagate after reflected by the polarizing beam splitter. An identity authentication method is further provided. Based on the non-contact technical solution, only one light source is used to obtain the palm print images and palm vein images at the same time and in the same space.

The application discloses a living body attendance anti-fake method and system based on vein and facial feature hierarchical fusion, relates to the field of living body anti-fake identification, collects multiple palm images, optimizes the images through a vein texture enhancement algorithm, acquires a bifurcation point set through an improved U-Net network, completes vein living body inspection and feature matching to obtain a vein user hash; generates random instructions and collects multiple facial images, completes facial living body inspection and feature matching, and the facial user hash needs to be consistent with the vein user hash; dynamically determines a confidence weight ratio, screens optimal palm and facial images for consistency inspection, and integrates the user hash, a time period and the confidence weight ratio into an attendance record to upload a database, thereby guaranteeing the authenticity and security of the attendance result.