58 results about "Tumor margin" patented technology

The invention discloses a tumor space occupying brain function partition nerve influence image segmentation method based on a deep learning model, and particularly relates to the technical field of artificial intelligence and medical cross. Comprising the following collaborative operation steps: multi-modal data fusion acquisition, tumor edge feature enhancement, anatomy-functional feature collaborative modeling, occupation deformation compensation, adversarial boundary optimization, dynamic loss regulation and control, cascade segmentation strategy, clinical interaction verification, and construction of a U-Net + + based improved three-dimensional feature fusion network. Utilizing a multi-scale cavity convolution module to extract edge gradient features of the tumor infiltration area in a layered manner; in order to solve the problem of multi-modal fusion and dynamic adaptation mechanism deficiency in the prior art, a three-dimensional cross-modal fusion network of a channel competition mechanism is constructed, multi-modal key features are dynamically screened through spatial pyramid pooling and soft attention weight, and a tumor volume self-adaptive dynamic loss function is combined, so that multi-modal fusion and dynamic adaptation are realized. The problems of feature confusion and poor generalization of a traditional method are solved.

The invention relates to the technical field of image recognition, in particular to a urinary system CT image tumor benign and malignant identification method based on deep learning, which comprises the following steps: acquiring a urinary system CT image, constructing a map band sequence and extracting gray level distribution, identifying a heterogeneous edge and a texture mutation region, aggregating perturbation map blocks to form an abnormal structure, and identifying the benign and malignant tumors. And fusing multiple types of image layers to complete label integration, and generating a feature recognition image layer. According to the method, the extension recognition capability of the tumor edge external expansion region is enhanced by combining a graph band gray scale aggregation and sequence construction mode, the judgment precision of local heterogeneous change is improved by fusing gray scale kurtosis and migration analysis, and the texture disturbance trend is extracted based on direction vector included angle change. The block gray level fluctuation and gradient relationship supports abnormal structure aggregation identification, spatial coincidence and boundary difference combined screening realizes multi-feature region unified coverage, abnormal region expression definition and structure positioning accuracy are enhanced, and image layer consistency and identification stability are improved.

Apparatuses and methods for using augmented / mixed reality in surgical settings are provided, and, in particular, apparatuses and methods for intraoperative integration of augmented / mixed reality 3D models with near real-time pathology results. Real-time pathology results are generated using a machine learning analysis of biopsy pathology images in conjunction with a predictive model based on the patient's clinical and demographic data and radiographic imaging data. The apparatuses and methods can be used to provide more accurate intraoperative visualization, tracking, and determination of tumor margins to improve resection extent and any intraoperative or postoperative adjuvant therapy.

The application discloses a brain tumor infiltration degree prediction method, device, storage medium and equipment, and belongs to the technical field of medical image processing. The application innovatively takes tumor segmentation probability and fiber density features of a tumor edge region as infiltration analysis features, constructs a brain tumor infiltration degree prediction model, device, storage medium and equipment, and is used for predicting the infiltration degree of a brain tumor. Experiments show that the brain tumor infiltration degree predicted by the model has a relationship (P<0.05) with survival, and the survival time of the high-infiltration degree is obviously lower than that of the low-infiltration degree, which has important clinical significance.

The invention discloses a cross-modal lightweight colorectal tumor segmentation method and device, and aims to solve the problems that an existing medical image segmentation model depends on large-scale annotation data, small lesions and fuzzy boundaries are inaccurately recognized, the calculation complexity is high, clinical deployment is difficult, and the like. According to the method, on the basis of an improved U-shaped network architecture, adaptive histogram equalization and Gamma correction are introduced in a data preprocessing stage to enhance boundary information of a low-contrast CT image; the encoder captures a multi-level context by adopting a multi-scale convolution and cavity convolution fusion module, and strengthens tumor edge response in combination with a boundary attention module; and meanwhile, the calculation amount is reduced by introducing rapid spatial pyramid pooling. The decoder effectively recovers details and avoids artifacts through lightweight jump connection and bilinear interpolation up-sampling. The parameter quantity of the whole model is only about 4.25 M, the reasoning efficiency and the cross-equipment generalization ability are remarkably improved while the high segmentation precision is guaranteed, and the method is suitable for a resource-limited clinical environment.

A multi-axis tumor excision marker system is configured to deposit markers along multiple axes radially about a tumor to define a boundary of margin tissue for excision. A catheter may deploy a marker extension that extends to a margin tissue boundary for depositing the marker. The marker extension may be a shape memory material that extends radially outward from catheter opening or may be directed radially outward by a guide head. The radiopaque marker may be a phase change marker, such as an adhesive that changes from a liquid to a solid. A marker may be a discrete marker and may have an extension portion that extends from an anchor portion. An envelopment marker has a plurality of marker extensions that each extend along an offset axis from each other. A spiral maker includes a maker extension that spirals around the margin boundary tissue.

Medical software tools platforms utilize a surgical display to provide access to specific medical software tools, such as medically-oriented applications or widgets, that can assist surgeons or surgical team in performing various procedures. In particular, an endoscopic camera may register the momentary rise in the optical signature reflected from a tissue surface and in turn transmit it to a medical image processing system which can also receive patient heart rate data and display relevant anomalies. Changes in various spectral components and the speed at which they change in relation to a source of stimulus (heartbeat, breathing, light source modulation, etc.) may indicate the arrival of blood, contrast agents or oxygen absorption. Combinations of these may indicate various states of differing disease or margins of tumors, and so forth. Also, changes in temperatures, physical dimensions, pressures, photoacoustic pressures and the rate of change may indicate tissue anomalies in comparison to historic values.

The invention discloses a positioning simulation system based on tumor marginal particles, and relates to the technical field of tumor marginal particles. The system comprises an image generation unit, a particle generation and injection unit is arranged at the output end of the image generation unit, a radiation transmission unit is arranged at the output end of the particle generation and injection unit, a detector unit is arranged at the output end of the radiation transmission unit, and a data processing unit is arranged at the output end of the detector unit. The output end of the data processing unit is provided with a particle positioning unit, and the output end of the particle positioning unit is provided with a visualization unit. The image generation unit comprises a medical image type selection module, a tumor model definition module, a tumor edge generation module and a tumor internal feature simulation module, so that it is ensured that the generated image meets the research or training requirement of a user, and the user can input related information of a tumor in a parameterized mode; therefore, a tumor model meeting user requirements is generated.

The invention provides a method for determining a tumor boundary based on a mechanical atlas, which comprises the following steps: characterizing the Young modulus of an in-vitro tumor tissue formed by rigid tumor cells, processing the data obtained by characterization, constructing the mechanical atlas, and determining the specific position of the tumor boundary by analyzing the Young modulus distribution curve of the tumor tissue. And finally, calculating the stiffness gradient by using the Young modulus data of each point, calculating the maximum stiffness gradient direction of the tumor edge perpendicular to the maximum stiffness gradient, calculating the maximum stiffness gradient of the tumor tissue edge area, assisting to obtain a predicted tumor boundary line, and further confirming the boundary of the glioblastoma. According to the method, the mechanical property distribution of in-vitro tumor tissues is represented through a micro-nano mechanical representation technology, the tumor boundary is determined by constructing a mechanical map, and the safe resection range can be determined to the maximum extent while a normal functional area is protected.

The invention relates to a multi-modal ultrasonic image-based knowledge graph construction method and device, and a storage medium, and the method comprises the steps: obtaining a tumor gray-scale ultrasonic image, obtaining a tumor edge image and a tumor internal image through target segmentation, respectively extracting corresponding multi-modal candidate image omics features, and carrying out the target segmentation; forming a sample set comprising a plurality of samples in combination with a tumor type corresponding to the tumor gray-scale ultrasonic image; performing synthetic minority oversampling processing on a sample set, performing Z scoring processing on each candidate radiomics feature in each sample, screening radiomics features with strong correlation through principal component analysis, and performing recursive feature elimination to obtain finally reserved radiomics features corresponding to each tumor type; and establishing a knowledge graph and / or a machine learning model based on the finally reserved radiomics features corresponding to each tumor type. The method has the advantages that the association between the tumor type and the selected iconography features is fully established, and the abundant information of the tumor edge is fully considered.

Devices, systems, and methods to generate a plan for an interstitial laser ablation procedure are disclosed. The systems may be configured as an interstitial optical mapping system including a catheter, an emitter optical fiber, an imaging optical fiber, a light source, and a processing unit. The emitter optical fiber and the imaging optical fiber are used to interstitially image a fluorescent dye associated with a tumor, including the tumor margin, at discrete imaging positions along a length of the catheter. The processor calculates a location of the fluorescent dye at each discrete position and creates an optical map representing the tumor. The optical map is used to generate an interstitial laser ablation plan that includes laser fiber pull-back positions.

This invention provides a histologic system and method for rapidly and accurately assessing tumor margins for the presence or absence of tumor using machine learning algorithms. This affords a rapid and accurate histologic tumor readout and increase process efficiency and decreases the chance for human error. Advantageously and uniquely, the system and method allows for analyzing the tissue section as complete or incomplete as the first criteria to determine whether a tissue section is clear of tumor. A machine learning process receives whole slide images (WSI) of tissue and determines (a) if each image of the WSI contains complete / incomplete tissue samples and (b) if each image of the WSI contains tumorous tissue or an absence thereof. A reconstruction process generates a model of the tissue that maps types of tissue therein, and a display process provides results of the model or report for use and manipulation by a user.

The invention discloses a brain tumor automatic segmentation method, which is applied to the field of medical image processing, and aims to solve the problem of low accuracy of the existing brain tumor segmentation, the whole 3D image is compressed from the angle of three axes through a three-axis strip-shaped pooling method in combination with three learnable weight vectors, and the segmentation accuracy of the brain tumor is improved. And finally, the information of the three axes is used for mutual complementation, so that the spatial information is reserved to the greatest extent, and the aim of reducing the calculation amount and the internal memory burden is fulfilled. An information entropy attention module is designed, and an area containing the most information amount is obtained by comparing information entropy, so that enhancement is performed. Noise interference of an intracranial complex environment is greatly reduced, so that the segmentation performance of the model is further improved; in addition, curvature filtering attention is designed, and the tumor edge shape is extracted through learnable convolution, so that in final segmentation, the edge of the model is depicted clearest.

The present invention discloses a scalpel structure capable of marking positions and a forming method thereof. The method includes: forming a stent that can be expanded or closed; detachably installing a column that can rotate relative to the stent on the stent; installing a laser pen at the lower end of the column for confirming the edge of a tumor; installing a measuring scale near the lower end of the column for measuring the expanded resection range of the tumor edge when confirming the tumor edge; and installing a scalpel at one end of the measuring scale for excising the tumor and the tumor peripheral tissue within the expanded resection range while the measuring scale measures the expanded resection range of the tumor edge. The scalpel structure of the present invention can not only mark and measure the position of the tumor edge, but also determine the expanded resection range at the same time. Meanwhile, it can also excise the tumor and its peripheral tissue. One set of equipment has multiple functions, liberating the doctor's hands, reducing the doctor's workload, making the doctor time-saving and labor-saving during the operation, and improving the operation efficiency.

A system for automated tumor margin determination in radiological images, comprising: an image acquisition interface configured to receive radiological image data from one or more imaging modalities; a preprocessing unit operationally coupled to the image acquisition interface and configured to perform intensity normalization, spatial resampling, and noise reduction on the received image data to generate standardized image inputs; a feature extraction unit comprising a plurality of convolutional layers arranged in a hierarchical structure and configured to extract spatial features of different orders of magnitude from the standardized image inputs;a transformer coding unit operationally coupled to the feature extraction unit and configured to generate context-sensitive feature representations by applying self-attention operations over spatial areas of the extracted features; a fusion unit operationally coupled to both the feature extraction unit and the transformer coding unit and configured to combine features derived from convolutions and context representations derived from transformers into a unified feature map; a decoding unit operationally coupled to the fusion unit and configured to generate a segmentation map corresponding to the tumor regions by incrementally increasing and reconstructing the spatial resolution; a boundary refinement unit configured to improve the delineation of tumor margins in the segmentation map;a processing unit that is operationally coupled with the preprocessing unit, the feature extraction unit, the transformer coding unit, the fusion unit, the decoding unit, and the boundary refinement unit, wherein the processing unit executes instructions stored in a memory unit to perform automated tumor boundary delineation; and a display interface configured to overlay the delineated tumor boundaries onto the radiological image data.

The invention provides a precursor compound capable of targeting FAP. The structure of the precursor compound is shown as a formula (I) or a formula (II). Based on the precursor compound, the invention further provides a bimodal FAP targeting probe which is a complex formed by the precursor compound as shown in the formula (I) or the formula (II) as a ligand and radionuclide. The radionuclide is a radionuclide capable of forming stable complexation with a DOTA group, preferably 68Ga. The invention also provides an application of the bimodal FAP targeting probe in preparation of a living body imaging agent. The bimodal FAP targeting probe disclosed by the invention can accurately describe the tumor edge, and has the potential of guiding intra-operative decision in future clinical application.

This application provides a method and system for image recognition and diagnosis of brain tumors in neurosurgery. First, the brain tumor image is preprocessed. The preprocessed brain tumor image is then decomposed into at least two feature component maps that complement each other in terms of structural and texture information. Next, each feature component map is sequentially subjected to a first-stage adaptive nonlocal mean filtering based on local statistical characteristics and a second-stage filtering based on structural correlation similarity measurement. The filtered feature component maps are then fused and reconstructed to obtain a feature-enhanced brain tumor image. This feature-enhanced brain tumor image is input into a pre-trained brain tumor recognition model, and the neurosurgical diagnosis results are obtained through the model. Using the scheme of this application, a two-stage adaptive filtering based on local scattering statistics and structural correlation measurement can be achieved, thereby enhancing tumor edge and texture features while suppressing ultrasound speckle, thus improving the accuracy of segmentation and diagnosis.

A method for automating margin measurement for mass removal that utilizes one or more machine learning models is disclosed. The one or more machine learning models may assist in identifying and segmenting each biopsy in an image, classifying the biopsy as a clear biopsy or an ambiguous biopsy, rotating the image of the biopsy to a proper orientation, and identifying and segmenting the skin and tumor found in the image of the biopsy. The method may then determine a lateral and deep margin value and display the values on the image.

The invention provides a double-net-bag cooperative percutaneous puncture integral tumor excision system. The double-net-bag cooperative percutaneous puncture integral tumor excision system comprises net bags, the puncture coaxial guide system is used for establishing a percutaneous working channel from the body surface to the tumor edge; a fixing anchor is arranged at the far end of the inner core fixing rod and is used for fixing tumor tissues; the double-net-bag cutting system comprises an outer net bag and an inner net bag; the outer net bag is made of a high-rigidity material and is used for firstly expanding to create a protected operation space in tissues around the tumor. The whole core cutting component is double-net-bag cooperative cutting, no circuit or sensor or hydraulic system is adopted, the production cost is extremely low, large-scale manufacturing is easy, operation is visual, the whole operation process only comprises several simple actions of pushing, rotating and pulling, the learning cost of doctors is low, the surgical operation intuition is well met, and popularization is easy.

A method for characterizing a tissue sample is disclosed, which includes placing at least one nanostructured sensor in contact with said tissue sample, wherein said nanostructured sensor includes at least one response characteristic exhibiting dependence on a refractive index of said tissue sample when said nanostructured sensor is in contact with said tissue sample; directing interrogating optical radiation at a plurality of frequencies to said nanostructured sensor; detecting, for each of said optical radiation frequencies, radiation redirected by or transmitted through said nanostructured sensor in response to said interrogating optical radiation; analyzing said detected redirected or transmitted radiation to determine said at least one response characteristic of said nanostructured sensor at each of said plurality of interrogating optical radiation frequencies; and determining at least one attribute of said tissue sample based on said determined at least one response characteristic at said interrogating optical radiation frequencies.

A multi-axis tumor excision marker system is configured to deposit markers along multiple axes radially about a tumor to define a boundary of margin tissue for excision. A catheter may deploy a marker extension that extends to a margin tissue boundary for depositing the marker. The marker extension may be a shape memory material that extends radially outward from catheter opening or may be directed radially outward by a guide head. The radiopaque marker may be a phase change marker, such as an adhesive that changes from a liquid to a solid. A marker may be a discrete marker and may have an extension portion that extends from an anchor portion. An envelopment marker has a plurality of marker extensions that each extend along an offset axis from each other. A spiral maker includes a maker extension that spirals around the margin boundary tissue.

This invention provides a histologic system and method for rapidly and accurately assessing tumor margins for the presence or absence of tumor using machine learning algorithms. This affords a rapid and accurate histologic tumor readout and increase process efficiency and decreases the chance for human error. Advantageously and uniquely, the system and method allows for analyzing the tissue section as complete or incomplete as the first criteria to determine whether a tissue section is clear of tumor. A machine learning process receives whole slide images (WSI) of tissue and determines (a) if each image of the WSI contains complete / incomplete tissue samples and (b) if each image of the WSI contains tumorous tissue or an absence thereof. A reconstruction process generates a model of the tissue that maps types of tissue therein, and a display process provides results of the model or report for use and manipulation by a user.

Provided are Raman systems and methods for detection of cancerous tissues in tumor margins in real time during surgical procedures. The systems include a laser excitation source, a probe, a spectrometer and a camera specifically designed for use of Raman spectroscopy in real time during surgery. The inventive methods pertain to quality assessment of Raman data, comparison of obtained Raman data against known biomarkers and classification of tissue as cancerous or non-cancerous on the basis of corresponding Raman data. The systems and methods allow for previously unavailable detection and classification of tissue with Raman spectroscopy in real time surgical applications, including the identification of cancer sub-types.

The invention relates to the technical field of medical image processing, in particular to a tumor tissue segmentation method and system based on an MR image and a storage medium, and the method comprises the following steps: constructing a tumor edge segmentation module and a tumor non-edge region segmentation module through a neural network; a mutual learning module is arranged between the tumor edge segmentation module and the tumor non-edge region segmentation module; and training a model structure formed by combining the tumor edge segmentation module, the tumor non-edge region segmentation module and the mutual learning module to obtain a tumor tissue segmentation model. The tumor edge segmentation module, the tumor non-edge region segmentation module and the mutual learning module are constructed, so that edge segmentation and non-edge region segmentation are subjected to mutual learning and mutual guidance, the tumor tissue main body region can be accurately segmented, and the complex boundary of the tumor tissue can be accurately segmented; and the precise segmentation effect of the tumor tissue is improved.

The application discloses a cross-modal lightweight colorectal tumor segmentation method and device, aiming at solving the problems of existing medical image segmentation models, such as dependence on large-scale labeled data, inaccurate identification of small lesions and fuzzy boundaries, and high computational complexity difficult to be clinically deployed. The method is based on an improved U-shaped network architecture, and adaptive histogram equalization and Gamma correction are introduced in the data preprocessing stage to enhance the boundary information of low-contrast CT images. The encoder adopts a multi-scale convolution and a hollow convolution fusion module to capture multi-level context, and combines a boundary attention module to strengthen the tumor edge response. At the same time, a fast spatial pyramid pooling is introduced to reduce the computational amount. The decoder recovers details and avoids artifacts through lightweight skip connection and bilinear interpolation up-sampling. The overall model parameter amount is only about 4.25M, which can ensure high segmentation accuracy, significantly improve inference efficiency and cross-device generalization ability, and is suitable for resource-constrained clinical environment.

This invention provides a hepatocellular carcinoma (HCC) image classification method based on interference quantification and pathological priors, belonging to the field of medical technology. The method includes: acquiring HCC tumor images and dividing them into multi-region segmentation data, including a tumor core region, a tumor edge transition region, and a distal cirrhotic background region; calculating an interference index based on the gray-level and texture-related features of the HCC tumor images; performing adaptive feature selection and graded specific correction using a pre-constructed graded correction parameter library to obtain a corrected anti-interference tumor feature set; constructing a regularized loss function containing pathological constraints based on predefined cirrhosis and case association rules; training a model based on the anti-interference tumor feature set; and outputting the HCC image classification result. This invention solves the problem of difficult HCC image classification due to the blurred background boundaries and mixed features of tumors in HCC tumor images caused by varying degrees of cirrhosis in patients, as well as the scarcity of HCC tumor subtype samples.

A method and system for determining the presence or absence of cancerous cells within subject tissue. The method includes: providing a material that includes a peptide component configurable in a non-binding form when disposed in a neutral pH environment, and in a binding form when disposed in an acidic pH environment, wherein the peptide component is configured to produce a first Raman spectrum when subjected to one or more predetermined wavelengths of light; administering the material to a subject tissue; interrogating the subject tissue with light; sensing the subject tissue for light emitted from the subject tissue, and producing signals representative of the sensed emitted light; analyzing the signals to determine a presence or absence of the first Raman spectrum; and determining the presence or absence of cancerous cells based on the presence or absence of the first Raman spectrum within the sensed light emitted from the subject tissue.

Disclosed herein is a method for determining tumor margins relative to non-tumorous tissue using a deep-learning platform. Also disclosed herein are methods for removing a tumor from a tumor biopsy.

A multi-axis tumor excision marker system is configured to deposit markers along multiple axes radially about a tumor to define a boundary of margin tissue for excision. A catheter may deploy a marker extension that extends to a margin tissue boundary for depositing the marker. The marker extension may be a shape memory material that extends radially outward from catheter opening or may be directed radially outward by a guide head. The radiopaque marker may be a phase change marker, such as an adhesive that changes from a liquid to a solid. A marker may be a discrete marker and may have an extension portion that extends from an anchor portion. An envelopment marker has a plurality of marker extensions that each extend along an offset axis from each other. A spiral maker includes a maker extension that spirals around the margin boundary tissue.

Deep-ultraviolet scanning microscopy uses a first imaging apparatus arranged on a first side of a sample and a second imaging apparatus arranged on a second side of the sample. The first imaging apparatus includes a first ultraviolet light source to illuminate the first side of the sample and a first camera to receive light emitted from the first side of the sample. The second imaging apparatus includes a second ultraviolet light source to illuminate the second side of the sample and a second camera to receive light emitted from the second side of the sample. The first and second sides can be imaged in parallel, and can be sparsely sampled to increase imaging speed. A machine learning model can be used to generate images from the acquired signals. Signals can be detected from intrinsic sources (e.g., tryptophan) and extrinsic sources (e.g., propidium iodide and / or eosin Y) at the same time.