84 results about "Blood smear" patented technology

The invention provides a blood cell image recognition method based on a multi-scale Vision Transform model, and the method comprises the steps: constructing a multi-scale feature extraction architecture, combining the global modeling capability of a Transform network, carrying out the parallel processing of Patches of different sizes, and fusing the multi-scale information, thereby achieving the recognition of a blood cell image. The cells in the blood smear can be comprehensively perceived from microscopic details to macroscopic contexts, so that not only is the adaptability of the model to cell size difference and morphological diversity enhanced, but also the recognition stability under the conditions of cell overlapping and dyeing difference in a complex visual field is improved; the accuracy and calculation efficiency of cell classification and counting are effectively improved, and full-process automation from visual field quality evaluation to automatic cell recognition and counting is achieved.

The invention discloses a reticulocyte recognition and grading system based on a blood smear, and particularly relates to the field of medical cell morphology examination, and the system comprises an image acquisition module, a staining normalization module, an instance segmentation module, an erythrocyte classification module, a reticulocyte grading module and a statistics output module. The image acquisition module adopts an automatic microscope platform to continuously scan the blood smear under a 100-time visual field, acquires a high-quality bright field channel image through an automatic focusing technology, and monitors image definition, contrast and illumination uniformity; the statistical output module is used for calculating the proportion and bluish violet intensity statistics of reticulocytes of each level, and generating a visual quality control result and a diagnostic report; according to the invention, full-automatic processing from samples to reports is realized, interpretability and high classification precision are both considered, subjective difference and omission ratio of manual microscopic examination are significantly reduced, and efficiency and consistency of clinical detection are improved.

The invention discloses a blood smear image classification method and system based on knowledge distillation and a graph neural network, and the method comprises the steps: obtaining a training data set, and carrying out the preprocessing, and obtaining a training feature matrix; inputting the training feature matrix into a pre-trained teacher model, outputting teacher features and a teacher prediction result, inputting the training feature matrix into a student model, and outputting student features and a student prediction result; calculating characteristic distillation loss, classification loss and time sequence consistency loss, generating student model loss, and optimizing a student model; and when a to-be-processed blood smear image and a test period are obtained, preprocessing the to-be-processed blood smear image and the test period, inputting the to-be-processed blood smear image and the test period into the trained student model and outputting a classification result. According to the method, the training difficulty is reduced through knowledge distillation, so that the stability and effect of a blood smear image processing result are improved, deployment can be realized in an environment with limited resources, and the practicability of the model is improved.

The present invention relates to the field of spectral detection technology, and more specifically to an intelligent blood smear quality detection method for a laboratory. The method comprises: obtaining overall spectral data of a blood smear and spectral signals of various detection positions in different regions of the blood smear; determining the baseline drift fluctuation segment and normal fluctuation segment of the spectral signal at each detection position; performing a primary correction on the spectral signal at the detection position and the overall spectral data to obtain a corrected spectral signal and first corrected overall spectral data; performing a secondary correction on the first corrected overall spectral data using the peak values of the baseline drift fluctuation segments immediately before and after the current baseline drift fluctuation segment and the peak value of the current baseline drift fluctuation segment in the corrected spectral signals at each detection position to obtain final second corrected overall spectral data; and detecting the quality of the blood smear using the second corrected overall spectral data. Thus, the present invention improves the detection accuracy and reliability of blood smear quality detection.

The invention discloses a multispectral microscope blood cell automatic classification and counting system. The system is composed of a multispectral illumination and microscopic imaging module, a spectrum and geometric calibration module, a multispectral preprocessing and cell segmentation module, a multispectral discrimination index and spectral band weight adaptive updating module, a cell graph structure classification module, a man-machine interaction module, an online adaptive learning module and the like. The method comprises the following steps: acquiring a blood smear image by a system under a narrow-band multi-spectrum condition, constructing cellular spectrum-morphological characteristics by combining a multi-scale segmentation result after noise suppression, flat field correction and background deduction, generating a discrimination index with a self-adaptive spectrum band weight, and completing joint classification and counting on a cell map; and meanwhile, carrying out constrained increment updating on the spectral band weight and the classification model by utilizing an artificial correction result of the low-confidence-coefficient cells. Compared with a traditional single-channel microscopic imaging and static classification method, the method has higher classification accuracy and counting stability under different dyeing and imaging conditions, and the workload of manual recheck can be reduced.

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

The present invention relates to a computer-implemented method for producing a trained model for detecting one or more defined cell types and / or cellular indicators, such as biomarkers or cellular abnormalities, in particular genomic aberrations, from single-cell images obtained from a sample of a body liquid, bone marrow, cytology smear or slide preparations, in particular a blood smear sample. It relates also to a computer-implemented method for detecting one or more defined cell types and / or cellular indicators, such as biomarkers or cellular abnormalities, in particular genomic aberrations, from single-cell images obtained from a sample of a body liquid, bone marrow or cytology smear, in particular a blood smear sample.

The application relates to a blood slice sample pretreatment device, which comprises a box body, a punching and liquid distribution assembly, a sample suction and drying assembly, an oscillator, a first reagent pump, a second reagent pump and a nitrogen generation assembly; the box body is connected with the nitrogen generation assembly; the punching and liquid distribution assembly, the sample suction and drying assembly, the first reagent pump and the second reagent pump are arranged in the box body; the punching and liquid distribution assembly is connected with the first reagent pump; the sample suction and drying assembly is connected with the second reagent pump; the oscillator is arranged on the punching and liquid distribution assembly and the sample suction and drying assembly. The application can realize batch punching and batch liquid adding at the same time, the box body can complete different functions in different areas, different areas can be used simultaneously or separately, different functions can be completed, personnel intervention can be reduced, errors can be reduced, and manpower can be saved.

The utility model discloses a structure capable of realizing urgent detection of a single blood smear, which comprises a support table, wherein an X-axis linear moving module is arranged on the support table; the slide urgent cabin is connected to the X-axis linear moving module and is driven by the X-axis linear moving module to move along the X-axis direction; a jacking hole is formed in the bottom surface of the slide urgent cabin; a lifting plate is arranged on the supporting table and located below the slide urgent cabin, jacking supporting columns are arranged on the top face of the lifting plate, a lifting mechanism is arranged on the supporting table, and the driving end of the lifting mechanism is connected to the lifting plate to drive the lifting plate to ascend and descend in the direction close to or away from the slide urgent cabin. The slide conveying table is close to or far away from the upper surface of the slide urgent cabin in the Y-axis direction. According to the utility model, the automatic urgent detection of a single blood smear can be realized without manually stopping or waiting for the scanning work of a normal blood smear by medical staff, so that the working intensity of the medical staff is reduced while the working efficiency is improved.

A blood smear device for examination belongs to the technical field of medical instruments and comprises a smear mechanism, a hydraulic pusher, a tube opening, an isolation door, a smear platform and a blood smear pushing plate, the hydraulic pusher is located at the top midpoint of the smear mechanism, and the tube opening is annularly mounted at the edge of the top midpoint of the smear mechanism. When a glass slide is fixed in blood smear operation, a pressing claw synchronously descends through a linkage shaft II and an equipment disc to press and fix one side of the glass slide above a smear platform, so that the glass slide can be prevented from displacing when being pushed by stress, and the angle and the force of a slide pushing plate can be more accurately controlled when the slide is pushed; meanwhile, the blood track above the glass slide is prevented from deviating, the smear platform automatically enters and exits in the smear mechanism through up-down movement of the pusher, the interior of the smear mechanism is isolated through the isolation door, the safety and quality of the blood sample are guaranteed, the blood sample is kept in the physiological state, and inspection errors are avoided.

The invention discloses a blood smear manufacturing device and a manufacturing method thereof.The blood smear manufacturing device comprises a rack, a conveying component, a cleaning component, a moving component and a slide pushing piece, the conveying component, the cleaning component and the moving component are all arranged on the rack, and a slide pushing area is arranged on the conveying component; the cleaning part is positioned on the side edge of the sheet pushing area; the moving part passes through the upper part of the cleaning part and extends to the upper part of the sheet pushing area; the slide pushing piece is arranged on the moving part; the moving part is used for driving the glass slide to a position above the slide pushing area so as to push away dropping liquid on the glass slide; the moving part is further used for driving the glass pushing piece to the position right opposite to the cleaning part; the cleaning part is used for cleaning and blow-drying the slide. By moving the glass pushing piece to move back and forth between the cleaning part and the piece pushing area, the glass pushing piece is automatically cleaned, and glass slides can be continuously placed on the conveying part for piece pushing operation, so that the use continuity of the device is improved, the piece pushing process is integrated, the automation degree is increased, manual intervention is reduced, and the detection efficiency is improved.

The utility model discloses a blood cell smear auxiliary device. The blood cell smear auxiliary device comprises a blood filling pool and an adjusting part, the blood filling pool is provided with a containing groove used for containing blood, the groove bottom face of the containing groove is opened to be used for being matched with a glass slide, and the adjusting part comprises a first adjusting piece and a second adjusting piece. The first adjusting piece and the second adjusting piece are oppositely arranged in the blood filling pool, and the first adjusting piece and the second adjusting piece can move so as to adjust the volume of the containing groove. According to the blood cell smear auxiliary device, the preparation efficiency and quality of the blood smear can be remarkably improved, and a more reliable sample is provided for hematological examination, so that the accuracy and efficiency of medical diagnosis are promoted.

Blood-based biomarkers involving evaluation of membrane protein clustering (MPC) on peripherally circulating lymphocytes can identify individuals experiencing depression, schizophrenia, or bipolar disorder, while differentiating those that will respond to treatment from those that will not respond to treatment within the depressed populations. These markers can help diagnose and treat depression, schizophrenia, or bipolar disorder. Further, this diagnostic tool can identify individuals with treatment-resistant depression earlier. The disclosed automated method improves the throughput of blood-based biomarkers and can be standardized across multiple sites of blood collection / analysis. The automated approach performs similarly to human scorers for each method of blood smear collection in accurately obtaining cluster size means for each sample.

The invention relates to a cell image analyzer, a sample analysis system and a cell image analysis method. The cell image analyzer comprises: an imaging device for shooting a cell image; the bearing device is used for bearing the blood smear on which the blood film is applied; the control device is used for controlling the blood smear and the imaging device to move relative to each other, so that a plurality of shooting points in a target area of the blood film are respectively positioned in a visual field of the imaging device, and controlling the imaging device to shoot images of the shooting points in the visual field of the imaging device so as to obtain target cell images of the plurality of shooting points in the target area; the image analysis device is used for identifying platelets in the multiple target cell images and acquiring actual platelet distribution information of the blood membrane based on the platelets identified in the multiple target cell images and the shooting positions of the multiple target cell images in the smearing direction; and based on the actual platelet distribution information, acquiring platelet count and / or judging whether the to-be-detected blood sample is a platelet aggregation sample or not.

This invention relates to a sialylated glycoprotein detection reagent, its preparation method, and its application, belonging to the field of biodetection technology. The detection reagent uses purified water as a solvent and comprises the following components: surface-modified titanium dioxide nanomaterials, metal ions, a pH adjuster, and a solvent. The surface-modified titanium dioxide nanomaterials are obtained by dispersing titanium dioxide nanoparticles in an organic solvent and reacting them with a silane coupling agent. The pH value of the sialylated glycoprotein detection reagent is ≥ 8.0. The reagent of this invention utilizes the high affinity of modified titanium dioxide nanomaterials for sialic acid to specifically aggregate sialylated glycoproteins in blood into particles, facilitating observation. This reagent exhibits good stability, high sensitivity, and strong specificity, making it suitable for rapid tumor screening of blood smear samples.

The application relates to the technical field of medical image processing, in particular to a blood smear evaluation method, device and system based on medical examination. The method adopts a traversal scanning mode to obtain blood cell images of all regions of a blood smear, and constructs an image set in column units. By screening a red blood cell central region and determining a first blood cell region and a second blood cell region, analysis is jointly performed, so that the cell aggregation degree of each blood cell image can be accurately and effectively quantified. A cell aggregation degree sequence is constructed, and uniformity in the sequence is analyzed, so that the observation difficulty of each image set is obtained, an observable region can be screened, and accurate blood smear quality evaluation is realized. The application effectively quantifies the cell aggregation degree in the image through the image acquisition mode of traversal scanning, divides a correct observable region through the special morphology of the blood smear, and then realizes accurate blood smear quality evaluation.

The invention discloses puncture blood sampling equipment for blood sampling, and belongs to the field of medical instruments, the puncture blood sampling equipment comprises an operation bin and a smear mechanism, the operation bin comprises a storage bin and a positioning bin, the positioning bin is arranged at the top of the storage bin, the smear mechanism is slidably arranged in the positioning bin, and the smear mechanism comprises a blood drawing cylinder, an auxiliary cylinder and a blood transfusion ring; the blood drawing barrel is arranged above the auxiliary barrel, the blood transfusion ring is arranged between the blood drawing barrel and the auxiliary barrel, the storage bin comprises a first bin body, a bottom plate and a first spring, the top of the first bin body is open, the bottom plate is arranged in the first bin body in a sliding mode, the first spring is arranged between the bottom plate and the bottom wall of the first bin body, and clamping edges are symmetrically arranged on the top wall of the first bin body. The blood smear making device can quickly help a patient to make blood smears more conveniently and repeatedly.

The present invention pertains to the computer implemented method for identifying a microorganism in a blood comprising: receiving (101) a microscopic image of a blood smear; detecting (102) plurality of microorganism cells in the microscopic image using a deep learning cellular segmentation model; extracting (103) from the segmented microscopic image a patch of size HxW centred around a centroid of the detected microorganism cell for each detected microorganism cell; encoding (104) of each patch by a deep neural network encoding model into a patch representation vector u n of size 1×D; aggregating (105) the patch representation vectors un using a pooling method to generate a sample representation vector u of size 1×D; and indicating (106) a species of microorganism present in the microscopic image by classifying the sample representation vector u by a neural network classifier to class representing specific species. In another aspect the invention relates to the data processing system for carrying out the steps of the method according to the invention.

Apparatus and methods for use with blood samples are described. Using a microscope (24), three images of a microscopic imaging field of a blood sample are acquired, each of the images acquired using a respective different imaging condition, and a first of the three images acquired under violet light brightfield imaging. Using at least one computer processor (28), an artificial color microscopic image of the microscopic imaging field is generated by mapping the first of the three images to a red color channel of the artificial color microscopic image, mapping a second of the three images to a second color channel of the artificial color microscopic image, and mapping a third of the three images to a third color channel of the artificial color microscopic image. Other applications are also described.

This invention relates to a constant-temperature blood smear drying device, comprising a housing, a fan, a heater, a temperature sensor, and a constant-temperature control module. The housing contains a sealed drying chamber with an open top. A slide holder is placed at the top opening of the housing, and slides are mounted on the slide holder. The fan is positioned on one side of the upper part of the housing, and the heater is located at the fan's outlet. The temperature sensor is installed outside the housing and extends into the drying chamber to collect the measured temperature in real time. The constant-temperature control module dynamically adjusts the heater power and fan speed based on the difference ΔT between the set target temperature and the measured temperature within the drying chamber, thereby correcting the set target temperature within the drying chamber. The beneficial effects of this invention are: 1. Solving temperature unevenness; 2. Cell protection; 3. Reduced power consumption.

According to the automatic detection method and device for the blood cells in the blood smear image, a local teacher model and an external teacher model are combined to generate high-quality pseudo labels for original extremely sparse labeled data, the models are guided to learn a complete foreground concept, and the problem of a large amount of missing detection caused by insufficient labeling is effectively relieved; the text feature of the blood cell category name is used as a domain invariant anchor point, a detector is guided to generate visual representation with strong generalization, and excellent generalization performance is achieved in multiple visual domains; inter-text comparative learning and inter-visual comparative learning are introduced to promote clearer alignment between text features and visual features from multiple dimensions, and generalization is improved while discrimination is guaranteed.

The present invention discloses a malaria blood smear cell detection and segmentation method based on Mask R-CNN, which realizes the detection and segmentation of malarial parasites and red blood cells in thin blood smear images, including: using VoVNet-57 as the backbone feature extraction network to extract richer thin blood smear image features; improving the feature pyramid network structure to make full use of the cell classification information rich in high-level features to enhance the representation ability of the underlying features; using the attenuation function to improve the traditional non-maximum suppression algorithm to reduce the cases of missed detection due to cell stacking; reconstructing the cell classification loss function to improve the problem of low detection accuracy due to imbalanced cell categories. The method proposed in the present invention can effectively improve the accuracy of the Mask R-CNN network model for cell detection and segmentation in thin blood smear images, providing a scientific basis for malaria diagnosis.

The invention discloses a smear pushing device for blood smear manufacturing, and belongs to the technical field of blood smear processing. Comprising a fixing box, a movable circular groove is formed in the upper surface of the fixing box, a storage cylinder is rotationally connected to the lower surface of the interior of the movable circular groove through a rotating shaft, a plurality of storage grooves are formed in the upper surface of the storage cylinder, and blood smear bodies are arranged in the storage grooves; a multi-station circulating pushing mechanism is constructed by forming a plurality of storage grooves in the storage cylinder and cooperating with a storage cylinder rotating structure driven by a second motor, after a blood smear body in a single storage groove is pushed, the second motor can rapidly drive the storage cylinder to rotate, the next storage groove in which the blood smear is pre-stored is precisely in butt joint with a pushing station, and the blood smear body is pushed to the next storage groove. The next round of pushing can be started without shutdown, meanwhile, workers can supplement materials to the unoccupied storage grooves in the pushing process, and pushing and material supplementing are carried out in parallel.

The application discloses a blood pushing machine and method. The blood pushing machine comprises a pushing device, an image processing device and a dyeing device. The pushing device is used for pushing and processing a sample to be detected on a blood smear to form a blood film sample on the blood smear. The image processing device is used for acquiring a detection image of the blood film sample and determining whether the blood smear meets a qualified condition based on the detection image. The dyeing device is used for dyeing and processing the blood film sample on the blood smear meeting the qualified condition. Through the above implementation, whether the blood smear meets the qualified condition is determined based on the detection image, and then the dyeing device only dyes and processes the blood film sample meeting the qualified condition, so that the quality of the blood film is greatly improved, and the problem of low microscopic examination efficiency caused by unqualified smears is solved.

The application discloses a blood smear white blood cell classification counting method and system based on image analysis, and belongs to the technical field of medical image processing. The method comprises the following steps: after collecting a peripheral blood smear image, performing adaptive background correction and color standardization pretreatment; detecting and segmenting each white blood cell through an instance segmentation network; extracting three-dimensional features of nuclear morphology, cytoplasmic staining and nuclear-cytoplasmic ratio and adaptively fusing; adopting a cascade classifier to perform five-class white blood cell classification and abnormal morphology identification; and counting and labeling suspected abnormal cells for review. The classification coincidence rate of the application reaches more than 95%, can effectively identify abnormal morphologies such as atypical lymphocytes, immature cells and toxic granules, and provides technical support for blood tests at the grassroots level.

The present invention relates to a computer-implemented method for producing a trained model for detecting one or more defined cell types and / or cellular indicators, such as biomarkers or cellular abnormalities, in particular genomic aberrations, from single-cell images obtained from a sample of a body liquid, bone marrow, cytology smear or slide preparations, in particular a blood smear sample. It relates also to a computer-implemented method for classifying a sample of a body liquid, bone marrow or cytology smear, in particular a blood smear sample, as indicative or non-indicative of one or more defined cell types and / or cellular indicators, such as biomarkers or cellular abnormalities, in particular genomic aberrations, from single-cell images obtained from the sample.

The invention provides a deep learning-based blood smear whole red blood cell abnormal morphology evaluation system. The system comprises a dual classification module, a fitting comparison module and an evaluation output module, the method comprises the following steps: collecting a blood smear full-slide image containing various erythrocyte morphological abnormality lesion features, carrying out classification labeling on the image, constructing an abnormal erythrocyte database, realizing double classification of erythrocytes by utilizing a deep convolutional neural network combined with a multi-feature extraction algorithm, and carrying out fitting comparison on double classification results. And generating a comprehensive full-slide blood smear erythrocyte abnormality assessment report. The innovation of the invention lies in that a set of complete deep learning detection process is constructed, through deep learning and multi-feature fusion, the accuracy and interpretability of abnormal red blood cell recognition are significantly improved, accurate evaluation of abnormal red blood cell classification under the full-slide scale is realized, and an objective quantitative evaluation basis is provided for clinical diagnosis.

The utility model belongs to the technical field of medical tools, and particularly relates to an angle-fixed smear pushing plate for blood smear manufacturing, which comprises a U-shaped operation frame, sliding grooves are symmetrically arranged on the inner walls of two sides of the U-shaped operation frame, a clamping groove is arranged on the bottom surface of the inner wall of the U-shaped operation frame, a supporting plate is inserted into the inner sides of the sliding grooves, and the supporting plate is connected with the U-shaped operation frame. An inclined flat barrel is inserted into one side of the supporting plate, a scraping plate penetrates through the inner side of the inclined flat barrel, and a glass slide is clamped to the inner side of a clamping groove. The glass slide is clamped on the inner side of the operation frame, blood drops on the right side of the top face of the glass slide, then the scraping plate is inserted into the inner side of the flat barrel, the flat barrel is moved to make the scraping plate make contact with the blood, then an operator pushes the flat barrel at a constant speed, the scraping plate can scrape the blood drops evenly, the angle of the scraping plate is fixed, and manual operation errors are eliminated; and meanwhile, experience dependence is reduced, the training period is shortened, the success rate of one-time smear is high, and in addition, the whole tool is simple in structure and low in use cost.

The invention discloses a sepsis patient diagnosis and prognosis system based on a neutrophil toxicity index. According to the sepsis patient diagnosis and prognosis system based on the neutrophil toxicity index provided by the invention, only a small amount of blood is required to be subjected to sample suction, sheet pushing, dyeing, drying and sheet reading according to a standardized process, and an optical microscopic image of a blood smear is obtained for cell morphology analysis. The number of the neutrophils containing the poisoning particles can be clearly quantified, the toxicity index is generated according to the neutrophils, and the method is rapid, simple and visual. The significance of TI, CRP and PCT in prompting diagnosis and prognosis of sepsis patients is analyzed through ROC curve comparison, the area of TI under the curve is larger than that of CRP and PCT, and the method is high in sepsis early-stage risk prediction and diagnosis efficacy.

The application provides a sample analysis system, an image analysis system and a method for processing a sample image, the method comprising: detecting a sample to be tested to obtain various detection parameters of the sample to be tested; preparing a blood smear based on the sample to be tested; moving the blood smear relative to an imaging device so that the blood smear is located in a field of view of the imaging device; capturing a specific region of the blood smear in the field of view to obtain a sample image; and processing the sample image by using a trained neural network to obtain an enhanced target image for output to a user. The application uses a neural network to perform image enhancement, which can accurately identify various cells or other tangible substances and significantly improve the quality of the image.