53 results about "Tissue characterization" patented technology

The invention, which belongs to the technical field of medical image processing and computer vision, discloses a multispectral imaging-assisted post-radiotherapy skin injury early detection system comprising a multispectral data acquisition module, a deep tissue feature extraction module, a three-dimensional lesion segmentation module and a space-time tracking evaluation module. The subcutaneous 2.5 cm depth tissue information is obtained through multispectral imaging at the wave band of 400-1350nm, blood perfusion, melanin concentration, collagen structure and other physiological parameters are inversed based on the radiation transfer theory, a three-dimensional medical image segmentation algorithm is adopted to achieve three-dimensional accurate segmentation of an injury area, a spatio-temporal evolution model is established to predict the injury development trend, and the damage development trend is predicted. The radioactive skin injury can be detected in the subclinical period, the detection time window is advanced by 5.2 days on average, the occurrence rate of severe dermatitis is reduced by 65%, and a basis is provided for clinical timely intervention.

This application relates to the field of medical image processing technology, and discloses a multimodal image recognition method and system for dynamic assessment of diabetic wound healing. The method includes acquiring infrared thermal imaging images, hyperspectral images, and visible light color images of a diabetic wound collected at the same time of visit; identifying the wound boundary between the diabetic wound and surrounding tissues in the visible light color image, and determining the central reference point of the visible light color image based on the wound boundary; establishing a virtual polar coordinate system based on the central reference point and preset anatomical reference markers; extracting tissue feature information characterizing the pathological state of the wound from the infrared thermal imaging image, hyperspectral image, and visible light color image, and mapping the spatial position of the tissue feature information in its corresponding image to the virtual polar coordinate system; normalizing the mapped tissue feature information, and calculating wound healing assessment indicators based on the normalized tissue feature information in the virtual polar coordinate system.

Changes in tissue stiffness have long been associated with disease. Systems and methods for determining the stiffness of tissues using ultrasonography may include a device for inducing a propagating shear wave in tissue and tracking the speed of propagation, which is directly related to tissue stiffness and density. The speed of a propagating shear wave may be detected by imaging a tissue at a high frame rate and detecting the propagating wave as a perturbance in successive image frames relative to a baseline image of the tissue in an undisturbed state. In some embodiments, sufficiently high frame rates may be achieved by using a ping-based ultrasound imaging technique in which unfocused omni-directional pings are transmitted (in an imaging plane or in a hemisphere) into a region of interest. Receiving echoes of the omnidirectional pings with multiple receive apertures allows for substantially improved lateral resolution.

Systems and techniques are disclosed for improving real-time tissue characterizations during endovascular procedures. In some implementations, a method involves characterizing biological tissue for an endoluminal procedure. A first electrode is positioned within an endoluminal space adjacent to a region of biological tissue. A second electrode is positioned spatially separated from the first electrode. A variable-frequency input electrical signal is applied across the first electrode and the second electrode to generate an electrical field corresponding to the region. A voltage response is measured between the first electrode and the second electrode based on the electric field. A voltage phasor is determined based on the voltage response relative to the variable-frequency input electrical signal. A characteristic of the region is determined based on the voltage phasor.

This invention discloses a deep learning-based method for predicting orthodontic outcomes, belonging to the field of artificial intelligence-assisted technology. The method includes acquiring multimodal initial data of the target patient, comprising craniofacial CBCT 3D images, intraoral 3D scan models, and a digital design scheme for orthodontic appliances containing target tooth displacement data. Based on the CBCT 3D images, macroscopic geometric features of the jawbone are extracted using a macroscopic morphological encoder, and microscopic tissue features of the periodontal region are extracted using a microscopic tissue feature encoder. The macroscopic morphological encoder is a 3D convolutional neural network, and the microscopic tissue feature encoder is a local window attention network based on a visual Transformer. This invention can predict the dynamic effects of the interaction between tooth movement and adaptive skeletal remodeling, and is particularly suitable for predicting orthodontic outcomes in adolescents during their growth and development period, addressing the industry pain point of large systematic biases in static models in such scenarios.

The invention discloses a cardiac magnetic resonance report generation method and system based on a multi-dimensional tissue feature knowledge graph. The method comprises the following steps: performing motion correction and multi-view feature extraction on a cardiac magnetic resonance multi-sequence image to obtain an original numerical attribute of an instance node; constructing a dynamic knowledge graph containing tissue characteristics and medical priori, and generating a target reasoning chain with a clinical basis in combination with neural symbol reasoning; then, in a controlled generation stage, an inference chain is injected into a prefix tree, in a decoding step, mask physical shielding logic conflict description is utilized, and a pointer network is synchronously started to directly copy original numerical value attributes into a text sequence; and finally, executing consistency logic verification through a text restoration technology, and identifying and intercepting factual errors. Through the atlas constraint and numerical unvarnished transmission mechanism, the factual illusion problem of a generative model is effectively solved, and the accuracy and reliability of a heart diagnosis report are remarkably improved.

Provided herein are methods of analyzing tissue using a magnetic resonance imaging (MRI) compatible tissue analysis device that includes radio-frequency (RF) tracking and imaging elements. Related kits, systems, and computer program products are also provided.

To detect early changes in the characteristics of disease-related tissues within the subject's body. [Solution] The medical information processing device according to this embodiment includes an acquisition unit, a comparison unit, a tissue change estimation unit, and a display control unit. The acquisition unit acquires tissue information indicating the state of tissues related to diseases in the body of a subject, and surface information indicating the state of the body surface of the subject. The comparison unit compares the tissue information and the surface information and calculates the comparison result between the tissue information and the surface information. The tissue change estimation unit estimates the change in the characteristics of the tissue based on the change in the comparison result over time. The display control unit displays the estimated result of the change in the characteristics of the tissue on a terminal.

The present disclosure relates to an apparatus for optoacoustic sensing of tissue features, such as tissue architecture, morphological information, function, physiology and / or molecular information about biomolecules, in particular metabolites, in a biological tissue (1) and a corresponding method. The apparatus comprises at least one illumination device (2) configured to illuminate the tissue (1) with first electromagnetic radiation (3) exhibiting one or more wavelengths or wavelength ranges in a first spectral range, at least one detection device (5) configured to detect, in response to illuminating the tissue (1) with the first electromagnetic radiation (3), at least one first optoacoustic signal corresponding to ultrasonic waves (6) emanating from a region of interest (ROI) in the tissue (1), and a processor (7) configured to determine, based on the first optoacoustic signal, at least one first parameter characterizing a first tissue feature of interest in the region of interest (ROI).

According to one embodiment, a medical diagnostic apparatus includes processing circuitry and display circuitry. The processing circuitry estimates a position of a structure in a subject based on data obtained by scanning with respect to the subject and analyzes tissue characterization in the subject. The display circuitry displays an analysis result of the tissue characterization obtained by the processing circuitry with respect to a plurality of positions in the subject except for the estimated structure position.

The invention provides a laparoscope data processing method and system, relates to the technical field of data processing, and aims at outputting and interacting laparoscope data based on forms such as natural language processing, interacting according to the requirements of doctors, realizing preoperative identification of operators, and improving the accuracy of laparoscope data processing. The system is automatically converted into image modes for habits of an operator, including brightness, saturation, focal length and the like. The blood vessel or the ureter can be recognized based on the fact that the special spectrum can be combined with typed cells or tracers in the blood vessel or the ureter, blood vessel damage can be avoided as much as possible by recognizing the blood vessel, intraoperative bleeding is reduced, and the ureter damage probability can be reduced by recognizing the ureter. The method can obtain the abdominal cavity image collected by the laparoscope, determines the tissue characteristics of each tissue catheter based on the abdominal cavity image, the tissue characteristics comprise the tissue trend and the tissue diameter, determines the organ information of the recognition object, can perform point location recommendation and determine the tissue type according to the data collected by the laparoscope, and improves the judgment efficiency and accuracy of the user.

The present application relates to pathological analysis technical field, specifically to a kind of pathological section intelligent analysis method and system based on deep learning, including the following steps: based on tissue section image, utilize multi-channel analysis tissue structure and dye distribution, adjust collection parameter and optimize image definition and cell structure performance, combine spatial distribution and gray layering relationship, automatically distinguish mutation area and structure boundary, output demarcation structure positioning mark.The present application is through collection sequence optimization, parameter dynamic adjustment and spatial distribution automatic discrimination, microstructure change, dye distribution and cell arrangement relationship are mapped as structure response index, realizes the efficient linkage of collection parameter and tissue characteristics, supports the synchronous identification of spatial mutation, dye layering and demarcation relationship in complex section, greatly improves the objectivity and coherence of area identification, structure annotation and demarcation positioning, avoids feature omission, enhances the integrity of data processing link and the practical depth of medical image analysis.

The invention discloses a duodenum mucosa remodeling system based on pulse field ablation and a control method, and the system comprises a multi-modal sensing ablation catheter which is used for collecting multi-modal data related to duodenum ablation through a multi-modal sensor; the multi-sensing data fusion unit is used for extracting key features after preprocessing the multi-modal data, and fusing the key features according to a preset algorithm to obtain a multi-modal feature vector; the real-time tissue feature analysis module is used for recognizing the tissue type of the current ablation point location in real time according to the multi-modal feature vector and analyzing the dynamic change of tissue features; the self-adaptive parameter adjustment engine is used for determining corresponding ablation parameters according to the tissue type of the current ablation point location and the dynamic change of the tissue characteristics; and the motion compensation and prediction system is used for predicting the motion trail of the duodenum in a future preset time period. The problem that in the prior art, a duodenum mucosa remodeling system is poor in remodeling effect is solved.

A medical system includes first and second opposing jaws each having an internal surface, at least one light emitter and at least one light sensor each disposed on the internal surface of one of the first and second jaws, and at least one of the first and second jaws having an external surface opposite the internal surface, the external surface having at least one marking disposed thereon aligned with the at least one light sensor. The medical system also includes at least one visual display, and a controller coupled to the light sensor and the visual display. The controller is configured to determine a position of a tissue relative to the light sensor, and control the visual display to display a graphical interface comprising at least one marking corresponding to the marking on the external surface in combination with an image corresponding to the tissue.

This invention discloses a deformable capsule endoscope probe and image fusion system based on multimodal imaging, belonging to the field of medical endoscopy technology. This invention solves the problems of existing capsule endoscope technology, which has relatively limited functionality and a fixed shape, making it difficult to adapt to the complex structure of the digestive tract and resulting in poor imaging effects. By using flexible silicone material to make the capsule endoscope probe's shell and deformable structure, it can automatically adjust its shape according to the shape of the digestive tract lumen, ensuring clear and accurate multimodal images are acquired from different locations. A neural network model accurately identifies potential lesion areas and adaptively adjusts the fusion weights based on the importance of the lesion and tissue characteristics to optimize image quality. Furthermore, a deep learning fusion network generates a comprehensive image, which, after contrast optimization and automatic labeling of lesion areas, provides doctors with clearer and more detailed diagnostic prompts, aiding in the early diagnosis and precise treatment of digestive tract diseases.

A system is provided that includes a processor circuit in communication with an intravascular imaging catheter. The processor circuit is operable to: receive an intravascular imaging signal; perform, tissue characterization to identify a plaque component of the blood vessel; generate an intravascular image; receive an extravascular image of a blood vessel; co-register the intravascular image and the extravascular image to associate the intravascular image with a location of the blood vessel in the extravascular image; determine, if a spatial distribution of the plaque component within the intravascular image satisfies a criterion associated with the spatial distribution; and when the spatial distribution satisfies the criterion, output a screen display comprising: the extravascular image; and a first indicator at the location of the blood vessel in extravascular image. The first indicator is representative of the spatial distribution of the plaque component. Associated methods and devices are also provided.

In some aspects, the described systems and methods provide for a method for training a deep learning model to assess liver pathology, including accessing annotated liver pathology images associated with a group of patients in one or more randomized controlled clinical trials of nonalcoholic steatohepatitis therapy, each of the annotated liver pathology images including at least one annotation describing one or more tissue characteristic categories for a portion of the image, and training the deep learning model based on the annotated liver pathology images to predict the tissue characteristic categories, selected from a group comprising steatosis, lobular inflammation, hepatocyte ballooning, and fibrosis stage.

The application discloses a postpartum pelvic floor function evaluation system based on three-dimensional image reconstruction, and belongs to the technical field of medical image processing and biomechanical analysis, and the system comprises an ultrasonic image acquisition module, a three-dimensional reconstruction module, a tissue elasticity analysis module, a rehabilitation scheme generation module and a visual display module, three-dimensional space registration and feature point extraction are carried out through a multi-angle ultrasonic image sequence, a multi-scale fusion network is used to realize three-dimensional reconstruction of pelvic floor anatomical structures with a precision of 0.2mm; a deep learning neural network is used to extract tissue features, and the elastic modulus distribution is calculated through a strain analysis model; an individualized rehabilitation training plan is automatically generated according to the evaluation result, and the rehabilitation effect is improved by 75%; three-dimensional visual display and rehabilitation process tracking functions are provided, the application realizes accurate evaluation and scientific rehabilitation guidance of postpartum pelvic floor function, the screening efficiency is improved by 320%, and the application has significant clinical application value.

Fetal oximetry information may be determined using a plurality of different types of measurement systems and / or calculation techniques. In some instances, a first measurement system may be used to interrogate maternal abdominal tissue to determine one or more characteristics thereof. Then, a second measurement system may be used to interrogate the pregnant mammal's abdominal tissue as well as fetal tissue and the characteristics of the maternal tissue determined using measurements of the first system may be used to process the measurements obtained using the second system to, for example, increase the accuracy of the processing of the measurements from the second system.

The application provides a physical information and organization double-driven metal material performance prediction method and system, and relates to the cross technical field of material informatics and machine learning. The method comprises the following steps: collecting metal material data samples, and constructing an original data set comprising an organization annotation set and a non-organization annotation set; performing physical information-driven feature engineering conversion and standardization processing on the data set, and dividing the data set into a training set and a test set; adopting a Bayesian optimization and cross-validation strategy, training a first XGBoost model with the organization annotation set, and constructing a microstructure feature prediction module; cascading the first XGBoost model with a second XGBoost model, training the second model with the non-organization annotation set, and constructing a cascaded performance prediction model; inputting the composition and process parameters of the metal material to be predicted into the cascaded model after conversion processing, and obtaining the organization features and mechanical performance parameters. The application realizes high-precision prediction of non-organization annotation samples and traceable analysis of physical mechanisms.

The application discloses a kind of based on organization prediction transverse crack determination method and system, comprising the following steps: S1, real-time collection L2 system slab identification information, when finding slab cutting, from L2 system collection the production performance information of this slab;S2, according to production performance information calculation obtains cooling curve, i.e. The temperature-time curve in the process from liquid to solid state;S3, according to production performance information, cooling curve, through slab organization prediction model module carries out the prediction of slab solidification organization information;S4, based on production performance information, slab solidification organization information, through transverse crack determination model module carries out the prediction of slab transverse crack, for the slab predicted as transverse crack carries out offline processing.The application fully considers the organization characteristics and performance in production of transverse crack occurrence, can reduce misjudgment rate, is favorable for improving hot charging hot delivery rate.

This invention provides a method and apparatus for segmenting the inferior mesenteric artery. The method includes: converting the pixel values ​​of each peritoneal imaging image in a sequence of peritoneal imaging images into CT values ​​to obtain an original tensor matrix; inputting the original tensor matrix into a segmentation model to obtain an image of the inferior mesenteric artery output by the segmentation model. The method and apparatus for segmenting the inferior mesenteric artery provided by this invention can perform coarse-grained segmentation with relatively messy tissue features, and after refining the initially segmented tissue features, perform fine-grained segmentation to extract the vascular morphology of the inferior mesenteric artery separately, thereby improving the accuracy and efficiency of inferior mesenteric artery segmentation.

A catheter-based ultrasound imaging system configured to provide a full circumferential 360-degree view around an intra-vascular / intra-cardiac imaging-catheter-head by generating a three-dimensional view of the tissue surrounding the imaging-head over time. The ultrasound imaging system can also provide tissue-state mapping capability. The evaluation of the vasculature and tissue characteristics include path and depth of lesions during cardiac-interventions such as ablation. The ultrasound imaging system comprises a catheter with a static or rotating sensor array tip supporting continuous circumferential rotation around its axis, connected to an ultrasound module and respective processing machinery allowing ultrafast imaging and a rotary motor that translates radial movements around a longitudinal catheter axis through a rotary torque transmitting part to rotate the sensor array-tip. This allows the capture and reconstruction of information of the vasculature including tissue structure around the catheter tip for generation of the three-dimensional view over time.

The invention provides an instance segmentation method of a knee joint image, which comprises the following steps: adding a memory enhancement module behind a detection branch and a prototype mask prediction branch of a YOLACT instance segmentation model, and obtaining category features based on features of the prototype mask prediction branch and mask confidence, bounding boxes and category information of the detection branch; storing the instance characteristics of each tissue organ of the knee joint according to a set updating mechanism; the memory enhancement module can memorize previously seen tissue features, so that when similar features are encountered, the model segmentation decision can be better guided, the recognition and segmentation capabilities of the model on different tissues are improved, and more context information can be utilized to better understand the structures and relationships of the similar tissues in the image, so as to improve the recognition and segmentation capabilities of the model on different tissues. And the instance segmentation precision of the knee joint image is obviously improved.

The invention discloses an ultrasonic intervention puncture needle guiding, monitoring and early warning system and method, and particularly relates to the field of monitoring and early warning, the system comprises an ultrasonic image real-time acquisition module, a puncture path planning module, a motion tracking module, a dynamic risk early warning module, a real-time path correction module and a puncture result verification module; dynamic images are collected in real time through a high-frequency ultrasonic probe, and the position of a submillimeter-level tracking needle point is analyzed through improved Radon transformation and local phase symmetry; extending Kalman filtering to fuse electromagnetic sensor data, calculating tip offset and performing multi-dimensional feedback early warning: optimizing an obstacle avoidance path based on multi-modal registration, superposing an intelligent guide arrow and digital parameters by a real-time correction module, and starting multi-step correction during complex offset; and finally, the puncture accuracy is confirmed through multi-plane ultrasonic verification and tissue characteristic analysis, and a structured report containing three-dimensional coordinates, a deviation curve and an AI risk score is generated.

A theragnostic system includes a surgical station and a storage and processing device that contains data in a suitable storage in which patient data and treatment data, e.g. in form of electrical and optical features, are combined. The electrical features are derived from electrical parameters of the voltage and the current with which an instrument is supplied. The optical features are derived from light of the spark that is produced upon influencing the tissue. By combining electrical and optical features in a data collection, that even contains additional features, such as tissue features and patient characteristics, it can be determined whether the instrument influences benign or malign tissue. The prediction accuracy can be increased by machine learning by adding histological data to the data sets. These data can be collected in a cloud computing system that is connected with many surgical stations.

The invention relates to the field of biological information, in particular to a tissue space structure recognition method and device and application. The method comprises the following steps: determining a space in-situ image of a given tissue cell population according to space coordinate information of the given tissue cell population; generating a space structure of the given tissue based on the space in-situ image of the cell population of the given tissue; wherein the space coordinate information of the given tissue cell population is determined by the following steps: (a) based on tissue space transcriptome data, acquiring the tissue cell group information; and (b) annotating the tissue cell group information by using a given tissue characteristic gene set so as to determine space coordinate information of the given tissue cell group. Through the method, accurate self-service identification of the organization space structure without depending on professional expert knowledge can be realized.

This invention discloses a cervical cancer diagnostic method based on fuzzy anomaly perception. The method constructs a multi-granularity fuzzy perception module to model tissue features at different scales and utilizes learnable fuzzy membership functions to quantify abnormal responses, forming a fuzzy representation of potential lesion areas. Furthermore, a fuzzy anomaly perception modulator is designed to generate spatial anomaly prototypes through local structural gradients and feature distribution offsets. These prototypes are then used to back-evaluate channel anomaly responses, jointly modulating the dual-path convolution output to achieve refined anomaly perception of suspicious areas. Finally, an uncertainty-driven subnet consensus mechanism is proposed. This mechanism assesses diagnostic uncertainty and dynamically regulates the interaction and fusion of multiple subnets to form stable and consistent consensus features, thereby improving diagnostic reliability. Experiments demonstrate that this method has higher diagnostic performance for cervical cancer.

The invention provides a lung cancer pleural invasion auxiliary diagnosis system and method based on a multiphoton microscopy technology, and belongs to the technical field of medical artificial intelligence and intraoperative pathology assistion.The method comprises the steps that multiphoton images of unstained paraffin sections, frozen sections and fresh tissue are collected through the multiphoton microscopy system; constructing a multi-source data set and extracting 142-dimensional tissue features; after cross-domain data enhancement of the generative adversarial network, semantic features extracted by the deep learning pre-training model are fused, and a pleura infringement state classifier is constructed; in the operation, a real-time image is obtained through a portable multi-photon system and input into a classifier to rapidly predict VPI negative and positive. According to the method, rapid, accurate and label-free diagnosis in the operation of the VPI is realized, real-time support is provided for decision-making of the resection range of the lung cancer operation, insufficient treatment or excessive resection is avoided, and the prognosis of a patient is effectively improved.

The application provides a brown fat recognition method and recognition device based on plain CT images, and solves the technical problem that the recognition cost and recognition accuracy of the existing brown fat recognition cannot be matched. The method comprises the following steps: establishing a tissue feature recognition model of brown fat by using the image features of the brown fat in the plain CT image resources; identifying the brown fat in the to-be-detected plain CT image according to the tissue feature recognition model, and quantifying the physical properties of the brown fat. The tissue feature recognition model is mainly trained or optimized by using low-cost plain CT image data, so that the brown fat and white fat can be effectively distinguished. The application provides a basic tool for target point recognition and tissue quantification, so as to explore how to activate the brown fat tissue in the human body through drugs, nutrition and other means, and to expect to find new strategies for treating obesity and diabetes.