164 results about "Cell space" patented technology

The invention provides a circulating tumor cell recognition model training method and system based on deep learning, and the method comprises the steps: obtaining a multi-channel pathological sample image set of circulating tumor cells, carrying out the cross-modal feature extraction of a multi-channel pathological sample image, generating a fusion feature vector containing the morphological features and microenvironment features of the cells, and carrying out the recognition of the circulating tumor cells through the fusion feature vector. A deep convolutional network with a multi-stage training strategy is constructed, network parameters of the deep convolutional network in the training process are verified and adjusted by adopting a hierarchical cross validation mechanism, and finally a final recognition model is generated based on the verified and adjusted network parameters. The final recognition model is configured to receive the clinical pathology image stream and output localization coordinates and classification confidence of circulating tumor cells. According to the method, a circulating tumor cell recognition result can conform to a cell space distribution rule and has clinical pathological diagnosis interpretability, and the problems of high false positive rate and insufficient cell subtype discrimination caused by neglecting of microenvironment interaction in a traditional method are solved.

The invention belongs to the technical field of image processing, and relates to a KAN-based space-frequency double-domain implicit guide sampling remote sensing image fusion method, which comprises the steps of constructing a training sample, creating a KAN-based space-frequency double-domain implicit guide sampling network, training the network and inputting an image to be fused into the trained network. And a fused image with high spatial resolution and high spectral resolution after fusion is generated through mapping. The network comprises a coding unit, a space implicit guiding unit, a frequency implicit guiding unit and a decoding unit, the coding unit is composed of parallel KANs and maps hyperspectral and multispectral remote sensing images to a depth potential space, and the space implicit guiding unit and the frequency implicit guiding unit conduct guiding implicit sampling on HSI in the space domain and the frequency domain respectively. And the decoding unit fuses the space-frequency domain sampling features and generates a final fused image. According to the invention, a pure KAN-driven architecture is adopted, the dependence on a multi-layer sensor in the prior art is replaced, and the lightweight degree is relatively high.

The invention belongs to the technical field of tumor micro-immune environments, and particularly relates to a kit for detecting a tumor micro-environment, a detection method and application, the kit comprises rare metal element labeled monoclonal antibodies, the rare metal element labeled monoclonal antibodies comprise 115In-CD20, 139La-Histone H3, 142Nd-CD56, 144Nd-CD14, 146Nd-Vimentin, 148Nd-Pan-CK, 151Eu-CD31, 156Gd-CD4, 158Gd-E-Cadrin, 159Tb-CD68, 162Dy-CD8a, 166Er-CD45, 167Er-CD15, 170Er-CD3, 171Yb-CD11c and 209Bi-alpha-SMA, and the rare metal element labeled monoclonal antibodies comprise The invention can be used for simultaneously detecting and researching the quantity, positioning and spatial topological relation of each cell component in a tumor microenvironment, and provides a multi-immunofluorescence detection kit and method for ubiquitous epithelial differentiation-derived tumors based on an imaging mass spectrum flow technique. A plurality of biomarkers can be analyzed at the same time by using multiple immunohistochemical staining and rare earth metal tag signal detection technologies, so that the aim of performing high-dimensional analysis on a tumor microenvironment is fulfilled according to cell composition, functional state and cell-cell space interaction relationship.

The invention relates to the technical field of pilates fitness equipment processing equipment, and discloses special pilates bed accessory grinding equipment with a dust separation and collection function, which comprises a frame body, a grinding area and a dust collection area. The grinding area is provided with a movable workbench and a cylindrical grinding wheel, and the dust collecting area comprises a first processing area, a second processing area and a third processing area. A multi-stage separation device is adopted and comprises a first isolation assembly, a dust collection pipeline, a second isolation assembly and a dust scraping structure. The penetrating plate of the first isolation assembly is of an inner-outer double-layer composite honeycomb mesh structure and is integrally arranged in an arch shape. The inner-layer screen cloth adopts a small-size honeycomb unit, the outer-layer screen cloth is correspondingly provided with a large-size honeycomb unit, and the upper boundaries of the two layers of screen cloth coincide to form a continuous filtering channel. According to the structure, dust is temporarily contained through the honeycomb unit space, the lower unit is used for receiving overflowing dust to form multi-stage filtration, and the eddy current effect is generated through vibration and resonance of the two layers of screen cloth.

The invention relates to the technical field of power system load prediction, and discloses a power load prediction method based on a multi-scale space-time diagram attention network, which comprises a space-time load information diagram construction unit, a diagram attention modeling unit and a gating feature fusion unit. According to the power load prediction method based on the multi-scale space-time diagram attention network, the regional space is divided by using cells, a topological graph considering cellular spatial correlation is constructed, a prediction architecture of three-dimensional feature fusion is constructed, spatial topological correlation is analyzed through the diagram attention network, a load feature deep mode is mined through a one-dimensional convolutional network, and the power load prediction method based on the multi-scale space-time diagram attention network is realized. The bidirectional gating cycle unit captures dynamic characteristics of a time sequence, and a gating mechanism performs multi-scale feature weighted fusion, so that complex space-time dependence can be better modeled, the generalization ability of the model is improved, the method adapts to a high-frequency fluctuation and multi-source heterogeneous data environment, the accuracy of power load prediction is improved, and the power load prediction efficiency is improved. And powerful support is provided for building a clean, low-carbon, safe and efficient modernized electric power system.

The invention relates to a probe design of a high-throughput single cell space transcriptome based on multi-round imaging coding. The invention provides a design method of a readout probe for a high-throughput single-cell space transcriptome, which is simultaneously suitable for more than two species tissues, and comprises the following steps: generating a candidate DNA sequence library with a sequence length of n, filtering out a sequence capable of being compared to a genome of a target species, filtering out a candidate DNA probe containing a short sequence appearing on the genome for more than one time, and obtaining a read-out probe for the high-throughput single-cell space transcriptome. The method comprises the steps of filtering candidate DNA probes capable of being compared to transcriptomes of target species through local BLAST, repeating the steps S20-S40 for all target species, performing probe internal mutual binding screening through the local BLAST, performing probe secondary structure inspection and probe internal mutual binding inspection through two-sequence comparison, performing online BLAST inspection on the probes, and determining the target species according to the BLAST. The final DNA probe sequence is obtained.

The invention provides a tumor microenvironment cell spatial distribution characteristic quantitative analysis method and system, and the method comprises the steps: obtaining a pathological section image, and obtaining cell segmentation and classification information in the pathological section image through a pre-trained cell recognition model; obtaining cell space distribution information based on the cell segmentation and classification information; constructing a cell topology network according to the cell space distribution information, and obtaining a first cell space distribution characteristic; constructing a spatial weight matrix, and obtaining a second cell spatial distribution feature; and judging a tumor state by combining the first cell spatial distribution characteristic and the second cell spatial distribution characteristic so as to realize typing of the target object. The method has good cell segmentation and classification precision, fully pays attention to the adjacency relation, the action intensity and the space structure between cells, integrates multi-dimensional space distribution characteristic quantitative analysis, realizes a good tumor microenvironment cell analysis result, is used for accurately realizing the typing of a target object, and has good universality.

A semiconductor device includes a first buried power rail (BPR) disposed through etch stop layers and a second BPR disposed in direct contact with the first BPR, where the first BPR has a larger critical dimension (CD) than the second BPR. A bottom surface of the first BPR directly contacts a via-to buried power rail (VBPR) contact. Source / drain contacts (CA) are disposed adjacent the VBPR contact and source / drain regions collectively defining middle-of-line (MOL) components. Back-end-of-line (BEOL) components are then constructed adjacent to the MOL components, and the MOL and BEOL components bond to a carrier wafer. The second BPR is then constructed on the carrier wafer.

The invention discloses a battery charging and discharging system fault diagnosis method based on multi-cell spatial filtering, and belongs to the field of state estimation and fault detection, and the method comprises the steps: building a discrete state space model considering system time lag and fault, and then carrying out the state augmentation, and obtaining a discrete state space reconstruction model; a Lyapunov function and a # imgabs 0 # performance index are introduced to construct a # imgabs 1 # observer; performing multi-cell space expression on an augmented state estimation value output by the # imgabs2 # observer, and then constructing a target # imgabs3 # observer based on a multi-cell space Kalman filter and a real-time observation residual error; separating system state estimation and system fault estimation; and fault diagnosis is carried out after an envelope interval of system fault estimation is calculated. By constructing a triple filtering structure including state augmentation, ZKF optimization and dynamic residual error generation, the limitation of a traditional single-stage filter in a time delay system and fault detection is broken through, and the state estimation precision, the fault detection sensitivity and the system robustness are comprehensively improved.

3D modelling and robotic tool systems and methods are provided that enable more efficient spatial processing of an object and its surroundings. The robotic tool system comprises: a scanner, configured to capture spatial data of an object and its surroundings; a processor, configured to: allocate the spatial data to cells of a cellular space; determine one or more characteristics of a cell according to data of adjacent cells; and configure a robotic tool to operate on the object at least in part according to the one or more characteristics.

The invention relates to the field of photovoltaic technology, and provides a photovoltaic module, a film coating device and a film coating method thereof, and the photovoltaic module comprises at least one group of adjacent first battery piece and second battery piece; the first welding strip extends along a set direction, is positioned on the surfaces of the first battery piece and the second battery piece and on a gap between the first battery piece and the second battery piece, and is used for connecting the first battery piece and the second battery piece; the first thin film covers the surface of the first battery piece and part of the first welding strip, and the edge of the first thin film at least exceeds the edge of the first battery piece on one side close to the second battery piece; the second thin film covers the surface of the second battery piece and part of the first welding strip, and the edge of the second thin film at least exceeds the edge of the second battery piece on the side close to the first battery piece. According to the photovoltaic module, the edge of each film exceeds the edge of the battery piece, so that the gaps between the battery pieces are covered by the films, the solder strip is prevented from displacement or deformation, and the reliability of the photovoltaic module is improved.

The invention belongs to the technical field of breast cancer treatment, and particularly relates to a patient layering method based on a breast cancer single cell space proteome multicellular enrichment mode, which comprises the following steps: S1, breast cancer single cell space proteome image data cell segmentation and cell type identification; s2, constructing a cell network diagram, and simulating a tumor microenvironment; s3, detecting closely connected cell communities, and extracting cell community characteristics; s4, carrying out cell community feature clustering, and representing a tumor microenvironment distribution mode of the patient; and S5, calculating the similarity of the tumor microenvironment distribution modes of the patients, and layering the patients. By means of the method, the patients can be layered, and prognosis of the patient groups has significant difference.

The invention relates to an event-driven tactile sensing method and system based on a pulse graph neural network, and belongs to the technical field of tactile sensing and robot control. The method comprises the following steps: S1, extracting main features from data collected by an event-driven touch sensor; s2, encoding the main features extracted from the tactile sensor data; s3, using a spiking neural network to further learn feature embedding; s4, the encoded features are processed through a graph neural network; and S5, identifying the category or state of the target object by adopting a voting mechanism. According to the scheme, through the graph neural network, the graph based on the space structure of the tactile sensing unit can be automatically constructed, and local and global features are efficiently extracted; the pulse neural network is used for processing event-driven data, so that the energy consumption is remarkably reduced and the calculation efficiency is improved; the method has strong generalization ability, can adapt to different tactile sensor configurations and application scenes, and has excellent performance when processing real-time tactile perception tasks.

The utility model relates to a water floating platform, which comprises a working platform, a floating platform, a floating platform and a floating platform, the first floating body unit and the second floating body unit are arranged below the working platform and used for providing buoyancy for the working platform; the one or more connecting pieces are connected with the first floating body unit and the working platform; wherein the connecting piece is connected with the first floating body unit, a floating body unit space is defined, and the second floating body unit is arranged in the floating body unit space. According to the overwater floating platform, the buoyancy of the platform is improved through the combination of the multiple floating body units, and the requirement for supporting overwater photovoltaic equipment with large weight can be met.

The invention relates to the technical field of biological information analysis, in particular to a cell annotation method based on a single cell space transcriptome and a related device. According to the cell annotation method based on the single-cell spatial transcriptome, single-cell transcriptome data, spatial transcriptome data and a plurality of branch annotation models need to be acquired firstly; converting the single cell transcriptome data into a single cell annotation file according to a first format constraint condition; and converting the spatial transcriptome data into a spatial annotation file according to a second format constraint condition. Inputting the single cell annotation file and the space annotation file into corresponding branch annotation models for branch annotation to obtain a branch annotation result output by each branch annotation model, and finally performing comprehensive annotation prediction based on a plurality of branch annotation results to obtain a comprehensive annotation result of the target single cell space transcriptome. According to the method, a more accurate annotation result can be obtained while good efficiency of cell annotation is ensured.

The invention discloses a cell space composition inference method, which belongs to the field of bioinformatics, and comprises the following steps: carrying out feature extraction on single cell RNA (Ribonucleic Acid) sequencing (scRNA-seq) data and space transcriptomics (ST) data, then simulating a cell mixture on each sequencing site (spot) by utilizing an scRNA-seq data set to construct synthetic ST data called'pseudo ST ', and finally obtaining the cell space composition inference method. The method comprises the following steps: using an IntegriteData method of Seurt to integrate manufactured pseudo ST data and real ST data to eliminate a batch effect between the pseudo ST data and the real ST data and improve data consistency and comparability, then using a random projection forest to construct a weighted adjacency matrix to accurately represent a topological relation between the pseudo ST data and the real ST data, and importing features and the weighted adjacency matrix into a GAT model together to obtain a GAT model; the cell type composition of each position in space transcriptomics can be inferred; according to the method, the topological relation between the scRNA-seq data and the ST data is fully utilized, so that the cell space composition can be more accurately deduced.

The invention provides a cell data processing method and device, equipment, a storage medium and a program product. The method comprises: obtaining cell space group data to be processed, the cell space group data comprising an initial expression profile and a space coordinate set; performing feature extraction processing on the initial expression spectrum and the space coordinate set to obtain splicing features; performing cell type prediction processing based on the splicing features to obtain a cell type prediction result; the cell type prediction result is subjected to feature dimension splitting, a cell type proportion set and an expression profile weight set are obtained, and the expression profile weight set comprises sub-weight values of multiple features corresponding to each cell type in the cell type proportion set; and determining a cell type expression profile based on the initial expression profile and the expression profile weight set. According to the application, the accuracy of determining the proportion and type of the cells in the cell tissue can be improved.

The invention relates to a single cell space transcriptome method based on multi-round imaging coding. The single cell space transcriptome method comprises the following steps: designing a readout probe for gene coding according to species of a target tissue sample; aiming at the reverse complementary sequences of the target gene set and the read probe group, designing a target probe oligonucleotide pool for targeting a target gene RNA (Ribonucleic Acid) sequence; constructing a targeted probe library for hybridization by using the targeted probe oligonucleotide pool; treating the tissue slice and hybridizing the target probe to prepare an imaging sample; acquiring sequence images of a plurality of rounds of readout probes, capturing RNA original signals, and dyeing the images to obtain original position information of cells in tissues; codes of multiple rounds of imaging signals are combined with a cellular morphology image to be analyzed into an RNA spatial expression matrix of a single cell.

The application provides a GPU hardware acceleration unit full-pipeline design method, solves the problem that a conventional single draw configuration broadcast issuing mode is easy to cause unit space and low rendering efficiency, and mainly comprises the following technical scheme: S1, configuration information and data information in each draw are uniformly formatted and end marks are respectively agreed, and then the information is uniformly stored in a memory; S2, a configuration flow channel and a data flow channel are designed in the memory, a common channel, a local channel and a judgment module are designed for each hardware acceleration unit; S3, the transmission configuration of all draws in the hardware acceleration unit is repeated until the rendering of a frame of image is completed, the previous single draw configuration broadcast issuing mode is changed into a pipeline issuing mode, so that the whole pipeline unit can work simultaneously with different configurations, and the image rendering efficiency is greatly improved.

The application discloses a semi-implicit neural map construction method based on grid-like cell space coding. The method uses grid-like cell space coding to abstractly encode a three-dimensional space, inputs the coding result into a neural network for decoding, and generates a new visual view through rendering. The grid-like cell space coding on the three-dimensional space can improve the correlation between data, eliminate redundant information in the neural map, realize compression sensing of the environment, maximize the use of information, and thus improve the map reconstruction quality. The semi-implicit neural map construction method can be applied to technical fields such as medical imaging, automatic driving, game development or indoor design, and can automatically generate a three-dimensional scene model according to an input two-dimensional image.

The invention discloses a multi-construction-unit thermal storage conceptual model construction method, a multi-construction-unit thermal storage conceptual model construction device, multi-construction-unit thermal storage conceptual model construction equipment and a computer readable storage medium, and the method comprises the steps: carrying out the formation unit division processing of a standardized logging data set based on the logging response characteristics of formation units; obtaining a stratigraphic unit division result; fault structure unit identification is carried out based on the standardized logging data set and the stratigraphic unit division result, and a fault structure unit data set is obtained; based on the standardized logging data set, the stratigraphic unit division result and the well temperature logging curve, thermal reservoir identification is conducted, and a thermal reservoir parameter set is obtained; and based on the stratigraphic unit division result, the fault structural unit data set and the thermal reservoir parameter set, multi-structural unit space relation modeling is carried out, and a structural unit space relation model is obtained. The accuracy of constructing the multi-construction-unit heat storage conceptual model is improved, and accurate characterization of the heat storage system under the complex construction condition is achieved.

This utility model relates to the technical field of kitchen utensil storage, proposing a modular combined storage structure, including a support component for expanding the storage space of a storage unit, and an assembly component positioned and installed on the storage unit. The support component carries items and connects with the assembly component to form a storage expansion module that cooperates with the storage unit. The modular combined storage structure achieves flexible expansion of the storage unit space through the combined design of the support component and the assembly component. This structure allows retailers to freely assemble according to actual usage needs, increasing or decreasing the number of storage expansion modules in the storage unit. A storage area for consuming condiment containers of different heights is formed between the support component and the storage unit. The storage unit can accommodate taller condiment containers, while the support component can accommodate shorter condiment containers, effectively solving the problems of wasted space and unsightly arrangement caused by mixing short condiment jars with tall containers.

The present application discloses a method for quickly solving a 5-degree-of-freedom space motion equation. Among them, an iteration manager is used to trigger an algorithm calculation unit to perform iterative calculations; a data path dynamic configuration module in the algorithm calculation unit is provided with a plurality of arithmetic units. The data path dynamic configuration module is used to configure the paths between the plurality of arithmetic units according to the state output by the state manager, and input the operation results into the space motion equation calculation unit; the space motion equation calculation unit is used to respectively input the calculated derivative values into the algorithm calculation unit and input the calculated eigenvalue into the iteration manager to enable the state manager to perform state conversion; the iteration manager is further used to send an iteration termination instruction to the algorithm calculation unit to output the operation results when it is determined that the iteration termination condition is satisfied. Through the technical solution in the present application, the acceleration effect of solving the space motion equation is improved by means of hardware acceleration.

The invention provides a multi-target path planning method and a related device, and belongs to the technical field of path planning. The method comprises the following steps: in a reference cell space, carrying out integral operation on a central point of the reference cell space to obtain an image point of the central point of each reference cell; in the real cell space, performing interpolation calculation on the image point of the central point of each reference cell obtained by the reference cells around each real cell in the real cell space in the reference cell space to obtain a final cell mapping relation; based on the final cell mapping relation, solving the multi-target path planning model to obtain an optimal solution set; and performing path planning on multiple targets according to the obtained optimal solution set to obtain a multi-target path planning result. According to the method, the problems of low accuracy and calculation efficiency of multi-target path planning are solved.

Provided is a water electrolysis stack capable of improving durability. The water electrolysis stack includes a cell stack that is formed by stacking a plurality of water electrolysis cells, an inter-cell space is formed between each adjacent ones of the water electrolysis cells in the cell stack, and gas flows into the inter-cell spaces in water electrolysis.

Embodiments of the present disclosure provide a method for displaying a virtual scene, a device for displaying a virtual scene, a computer-readable medium, and an electronic device, which relate to the field of computer technology. The above method can, when it is detected that a target model generates wind force in a virtual scene, establish a fluid field in the virtual scene, where the fluid field includes multiple unit spaces; obtain the force field value obtained by each unit space under the action of the wind force; perform force field injection processing on the velocity of the corresponding unit space according to the force field value corresponding to each unit space to obtain the updated target velocity of each unit space; update the material information of the vertices in the corresponding unit space according to the target velocity of each unit space; and render the virtual scene according to the updated material information of the vertices. This can improve the authenticity of the interaction between models in the virtual scene, enhance the correlation of the pose changes between models, and further improve the authenticity of the virtual scene.

The invention relates to the technical field of microfluidics, in particular to a high-flux approximate single cell space transcriptional library building method based on a micro-fluidic chip, which comprises the following steps: carrying out earlier stage treatment on a tissue slice in a single micropore, including fixing, permeating and carrying out reverse transcription on the tissue in situ to obtain DNA and RNA molecular hybrid chains; the method comprises the following steps: pasting a micro-fluidic chip on a tissue slice to capture mRNA at a high throughput in a spatial dimension, so that an mRNA molecule of a single pixel site carries an identification code capable of identifying the identity of the mRNA molecule; carrying out chain conversion on the hybrid chain to obtain double-chain cDNA, and then carrying out a round of amplification on the cDNA to enlarge the molecular weight of the cDNA; and then carrying out Tn5 interruption on the quantified cDNA, carrying out two-round amplification, and finally purifying the library with the sequencing linker and then sending the sample for sequencing. According to the present invention, the sufficient gene number and the sufficient mRNA category number for subsequent analysis can be captured under the high throughput in space and the high resolution of the similar single cell.

The invention provides a multi-target path planning method and a related device, and belongs to the technical field of path planning. The method comprises the following steps: dispersing a working space of a multi-target path planning model to obtain a real cell space; the multi-objective path planning model takes the minimum driving path and the maximum safety as objective functions; the constraint conditions of the multi-objective optimization model comprise kinematics constraint, boundary condition and obstacle constraint; generalized cell mapping is established on the real cell element space for each state variable and control variable of the constraint condition and the target function, and a cell mapping relation based on a transition probability matrix of all controllable cells in the real cell element space is obtained; the cell mapping relation is stored by adopting a hash table; based on the cell mapping relation, solving the multi-target path planning model to obtain an optimal solution set; and performing path planning on multiple targets according to the obtained optimal solution set to obtain a multi-target path planning result. According to the method, the problems of low accuracy and calculation efficiency of multi-target path planning are solved.

The present disclosure provides a display panel and an array substrate. The array substrate includes a plurality of pixels forming a plurality of pixel columns, and one pixel includes a plurality of sub-pixels. The array substrate includes: a substrate; a pixel electrode layer, including a plurality of pixel electrodes in one-to-one correspondence with the plurality of sub-pixels; a touch electrode layer, disposed at a side of the pixel electrode layer, where the touch electrode layer includes a plurality of touch electrode cells spaced apart and mutually insulated; where there is a first gap between two t touch electrode cells that are adjacent in a direction of the pixel columns, and the first gap is located between both ends of a pixel electrode along the direction of the pixel columns.

The invention discloses a chloride ion concentration driven corrosion evolution probability type cellular automaton simulation method, and relates to the field of metal corrosion evolution simulation, and the method comprises the following steps: S1, setting a cellular space size and a cellular size, and constructing a three-dimensional cellular space; s2, cell types are defined, and initial distribution of various types of cells is completed in a three-dimensional cell space; s3, setting a reaction mechanism of a model iteration process, and judging a conversion rule among various types of cells according to a corrosion probability Pa, a passivation probability Pb and a passivation film fracture probability Pd; s4, a chloride ion concentration matrix, a probability enhancement coefficient and an attenuation coefficient are introduced, the cellular transformation probability in the iteration process is corrected through the real-time concentration value of chloride ions, and aggravation of the local corrosion process is achieved; and S5, comparing test and simulated corrosion morphology results under the same mass loss rate. According to the method, local corrosion difference can be effectively reduced on the basis of comprehensive corrosion, and support is provided for durability evaluation of the steel structure.