375 results about "Endoscopic image" patented technology

The invention relates to the technical field of medical image processing, in particular to a digestive endoscopy image deblurring enhancement method and system.The method comprises the steps that firstly, an input digestive endoscopy original image is processed through a blurred region classification network, and a pixel-level blurred classification map capable of distinguishing an adhesion blurred region and a motion blurred region is generated; then, parallel processing is carried out according to the classification graph: for an adhesion fuzzy region, physical model restoration and color correction are carried out by estimating a transmissivity graph and an ambient light value; for a motion blur region, a self-adaptive non-blind deconvolution kernel is constructed to perform deconvolution sharpness. And finally, inputting the two processing results and the original clear area into a multi-scale feature fusion network together, carrying out adaptive feature weighted fusion and image reconstruction, and outputting a globally clear and detail-enhanced final image. According to the method, accurate identification and targeted enhancement of composite blurring are realized, and the visual quality and diagnosis availability of the digestive endoscopy image are effectively improved.

The invention relates to the technical field of endoscopic image analysis, in particular to a cystoscope tumor focus real-time positioning and boundary intelligent recognition system and method, and the system comprises a feature extraction module, a breakpoint detection module, a contour closing module, a boundary evaluation module and a smooth display module. According to the method, edge features are accurately captured by calculating pixel gradient vector data, homology matching is performed on end points of a fracture contour according to a gradient direction trend, connection pixels are automatically interpolated and filled between the breakpoints along a tangential direction, and non-closed edge gaps caused by illumination or shielding are repaired to form a complete closed contour. A boundary uncertainty evaluation model is constructed by combining gray uniformity and gradient intensity in a local neighborhood, a smoothing processing intensity radius is adaptively adjusted according to the boundary uncertainty evaluation model, depth smoothing is applied to a high-risk area, high-confidence area details are reserved, zigzag artifact interference is eliminated, and a high-risk area is obtained. And a continuous and high-precision self-adaptive focus boundary conforming to the real anatomical form of the tissue is generated.

The invention belongs to the technical field of medical robots and minimally invasive surgery, and provides a robot control system and method for blue laser vaporization surgery of prostatic hyperplasia. Mapping the nuclear magnetic image volume data to a deformation field under an ultrasonic acquisition coordinate system, and deforming the preoperative nuclear magnetic image to an intra-operative ultrasonic space by using the deformation field to complete image registration; fusing the registered image with a stereoscopic vision system; the spatial depth of the surface of the target tissue is obtained from the endoscopic image so as to supplement navigation information; according to the utility model, the prostate deformation and probe posture change adaptive capacity in an operation is improved, the real-time visual closed-loop regulation and control capacity is improved, and the characteristics of small light spots, shallow heat diffusion, excellent hemostasis and the like of blue laser are combined, so that the vaporization and hemostasis precision in a tiny blood vessel dense area is ensured, and the problems of large tissue trauma and the like are avoided.

The invention discloses an intelligent auxiliary diagnosis and grading system for gastrointestinal tumor endoscopic images, which belongs to the technical field of medical image processing and computer-aided diagnosis and comprises a multi-modal image preprocessing and segmentation module, a lesion feature extraction and diagnosis module, a grading and depth evaluation module and a closed-loop feedback optimization module. The system receives white light, a narrow band and an amplified endoscopic image, adaptive segmentation is performed to obtain a lesion area, mucous membrane morphology, capillary and gland features are extracted, Paris classification, Vienna classification and infiltration depth evaluation are realized, and segmentation parameters are subjected to closed-loop optimization according to classification confidence. And intelligent auxiliary support is provided for early diagnosis and treatment decision of gastrointestinal tumors.

One or more processors select a model from among a plurality of types of machine learning models according to a region whose image is being picked up by an endoscope, generate notification information of a type of the model selected, receive an instruction signal for switching the model, measure a time interval from a selection of the model to a reception of the instruction signal, select a model selected immediately previously when the time interval is less than a first predetermined time, and select a model scheduled to be selected immediately subsequently when the time interval is equal to or greater than a second predetermined time.

The invention discloses a self-adaptive intervention type single-array-element photoacoustic endoscopic imaging near-field noise removal method based on a spatial domain and morphological characteristics. The method comprises the following steps that 1, original data of photoacoustic endoscopic imaging are obtained; step 2, adaptively generating a patch mask area; 3, preprocessing the image in a frequency domain; step 4, executing strip interference detection based on adaptive spatial domain filtering on the non-plaque region to obtain a spatial domain noise mask; step 5, screening strip-shaped noise areas; step 6, performing interpolation restoration on the noise area; 7, calculating the convergence degree and the noise level in the iteration process; and step 8, outputting a final result after polar coordinate transformation. The intrusive single-array-element photoacoustic endoscopic imaging method can effectively inhibit strip-shaped interference signals occurring in intrusive single-array-element photoacoustic endoscopic imaging and reduce the covering influence of background noise on target signals, so that the imaging quality of a near-end area is improved, and the structural definition and contrast ratio of overall imaging are remarkably optimized.

The invention relates to the technical field of image processing, in particular to a small intestine capsule endoscope target detection method which comprises the following steps: acquiring a small intestine capsule endoscope image; the method comprises the following steps: constructing an improved RT-DETR network, inserting an SCSA module between a backbone network and a neck network of the RT-DETR, and introducing a DFM module behind a three-branch Fusion module of the neck network; and the improved RT-DETR network adopts a Focaler-IoU loss function. According to the method, the problem that the accuracy of recognizing the tumor target by using the small intestine capsule endoscope image through RT-DETR needs to be further improved is solved.

The invention relates to the technical field of image segmentation, and provides a gastrointestinal tract endoscope image segmentation method, which comprises the following steps: carrying out time sequence preprocessing on an obtained intestinal tract video, eliminating fuzzy and shielded invalid frames, constructing a rhythm signal based on a multi-scale time sequence visual feature, and estimating a creeping phase; outputting a phase time sequence and a phase label of the video; selecting a representative phase key frame, performing pixel-level focus segmentation on the key frame, and calculating a pixel-level uncertainty score to obtain a key frame segmentation result with high confidence; carrying out time sequence consistency recovery on the segmentation mask of the key frame based on phase consistency mapping and phase alignment, and recovering an uninterpretable section caused by shielding with the assistance of adaptive segmentation correction of motion perception; and carrying out time sequence aggregation on a cross-frame segmentation result, extracting fragment-level statistical representation and confidence, forming a fragment-level suspicious area report, and outputting structured data for clinical auditing and index retrieval.

The invention relates to the technical field of nondestructive testing, and discloses an aero-engine endoscopic eddy current testing system based on voice interaction and video recognition, which comprises an endoscopic eddy current integrated probe, a voice input module, a processor, an AI image recognition module, a parameter storage module, a display module and a video playing module. The integrated probe integrates an endoscopic optical unit and an eddy current detection unit, and synchronous acquisition of images and eddy current signals is realized. The AI image recognition module analyzes the real-time endoscopic image and automatically recognizes the type and the position of the workpiece; the processor intelligently calls matched detection parameters from the parameter storage module according to the recognition result, automatically configures the working mode of the probe and can control the video playing module to call and display corresponding operation guidance videos according to voice instructions. According to the invention, a traditional detection mode is changed, and automatic setting of detection parameters and intelligent guidance of a detection process are realized through deep fusion of voice interaction and AI visual identification.

The invention provides an end-to-end real-time analysis method for an endoscope image fluorescence brightness area based on double deep learning model joint training and a self-group iteration upgrading device. The end-to-end real-time analysis method comprises the following steps: acquiring an image of a biological tissue and processing the image by using a deep learning algorithm to obtain a training data set; sequentially constructing and training an image preprocessing deep learning model and an unsupervised image segmentation deep learning model based on the training data set; inputting image data acquired by the endoscope into an image preprocessing deep learning model to obtain a first output image; inputting the first output image into an unsupervised image segmentation deep learning model to obtain a second output fluorescence image; and carrying out brightness analysis on the second output fluorescence image, and visually outputting and storing the second output fluorescence image. The real-time performance and the accuracy of fluorescence region analysis can be improved. According to the self-group iteration upgrading device, model training iteration upgrading can be carried out in multiple modes, and the performance of the device is continuously improved.

Disclosed are systems and methods for determining target characteristics during a laser procedure, comprising (i) obtaining a relationship between (i) a number of pixels associated with a light beam reflected from a target or an object located in proximity to the target on an endoscopic image obtained from a video sensor coupled to an endoscope and (ii) a distance of the target from a tip of the endoscope. The method further comprising (ii) measuring the number of pixels associated with the light beam reflected from the target or the object located in proximity to the target during a procedure, and (iii) based at least in part on the relationship obtained in step (i) and the measured number of pixels in step (ii), determining at least one of a size of the target or a distance of the target from the tip of the endoscope.

The invention provides an endoscope sharing control method based on a human body signal and a game theory, and belongs to the field of medical instruments, and the endoscope sharing control method comprises the following steps: obtaining eyeball movement data; based on the eyeball movement data, determining eye movement control position data used for representing a manual control intention through gaze thermodynamic diagram analysis; based on the eyeball movement data, through quantitative analysis of sight, eye opening degree and pupil size, obtaining reference indexes used for representing intention integrating degree and manual control burden; machine control position data used for representing the expected moving distance of the endoscope are obtained by analyzing the position relation between a target object and the endoscope in an endoscope image, and the endoscope image is collected in real time through the endoscope; based on the machine control position data, the eye movement control position data and the reference index, determining shared control position data through a cooperative game strategy; movement of the endoscope is controlled based on the shared control position data.

The present embodiment relates to a training data creating device including one or more processors that perform processing of generating training data for machine learning. The one or more processors plot a plurality of endoscopic images of a training device in the feature space based on individual features, determine classes for the endoscopic images, and set a set region for each of the determined classes. The one or more processor newly generate training data so that the number of pieces of training data newly generated in a first region which is an intersection region of the set regions for each class is greater than the number of pieces of training data newly generated in a second region which is a region excluding the first region from a union region of the set regions for each class.

A processor acquires a three-dimensional image of a subject, acquires a real endoscopic image in a lumen structure of the subject, which is picked up by an endoscope inserted into the lumen structure of the subject, derives a virtual viewpoint in the three-dimensional image of the endoscope using the real endoscopic image and the three-dimensional image, acquires a radiation image of the subject in which the endoscope is inserted into the lumen structure, performs registration between the radiation image and the three-dimensional image, and superimposes and displays at least a part of the lumen structure included in the three-dimensional image, on the radiation image, based on the virtual viewpoint and a result of the registration.

The invention discloses an endoscope imaging optimization method, and belongs to the technical field of medical optical imaging and image processing. The method comprises the following steps: firstly, obtaining a super-structure lens distortion image, a high-definition truth value image and a semantic segmentation label; extracting multi-scale initial features through a multi-branch encoder; carrying out feature enhancement and self-adaptive downsampling by utilizing a dynamic super-surface feature extraction module; constructing robust features through a dynamic parameter generation module in combination with the high-halal truth value image; then, extracting contour guide features from the high-halal truth value image and fusing the contour guide features with robust features; and finally, image segmentation and recovery tasks are executed in parallel, and a segmentation mask and a distorted image are output. By fusing physical prior and data driving, the problems of chromatic aberration, information dispersion and the like of super-structure lens imaging are effectively solved, and the image definition, the small focus recognition precision and the model generalization ability are remarkably improved.

An endoscopic image processing apparatus includes a processor. The processor performs processing for acquiring lesion information including information indicating a position of a lesion region included in an endoscopic image, determines whether the lesion region is included in the endoscopic image, and performs processing for generating a display image for displaying one or more marks in at least one of N mark display regions set as regions as many as a maximum display number of the one or more marks, and generating the one or more marks indicating in which reference region among a predetermined plurality of reference regions set in the endoscopic image a present position of the lesion region or a position of the lesion region immediately before the detection is interrupted is included and displaying the one or more marks in at least one of the N mark display regions.

The present application provides an endoscope image data enhancement method and system based on a generative adversarial network, comprising: acquiring a multi-domain medical data set, importing a medical feature extraction model to obtain a sample feature vector, acquiring high-value rare samples based on the sample feature vector, and performing feature space enhancement; encapsulating the conventional samples and the high-value rare samples as a training data set; based on the training data set, constructing and training an anatomical domain encoder and a pathology instance encoder; constructing a meta-domain image enhancement model; based on the training data set, acquiring a meta task using the anatomical domain encoder and the pathology instance encoder; training the meta-domain image enhancement model based on the meta task; acquiring a new domain medical data set; and acquiring a new domain synthetic data set using the anatomical domain encoder, the pathology instance encoder and the meta-domain image enhancement model. The present application realizes an endoscope image data enhancement method that intelligently mines high-value rare samples to improve data distribution and efficiently adapts to new domain styles.

An endoscopic image capturing assembly is provided and includes a holder, an image sensing device, a light emitting component and a conducting track. The holder includes a first surface, a protruding end surface, a first body and a protruding body connected to the first body and protruding from the first surface. The first surface is formed on the first body. The protruding end surface is formed on an end portion of the protruding body away from the first surface. The image sensing device is mounted on the first surface. The light emitting component is mounted on the protruding end surface. The conducting track is at least partially embedded in the holder and electrically connected to one of the image sensing device and the light emitting component. Besides, an endoscopic device having the aforementioned endoscopic image capturing assembly is also provided.

A display processing unit causes a display device to display an endoscopic image of a living tissue in a somatic cavity captured by an endoscope. An operation reception unit receives a user operation that is performed so as to set an area of interest in the endoscopic image. An area-of-interest setting unit sets the area of interest in the endoscopic image based on the user operation. A three-dimensional information acquisition unit acquires three-dimensional shape information of the living tissue captured by the endoscope. A virtual surface derivation unit derives three-dimensional shape information of a virtual surface in the area of interest from three-dimensional shape information of an area different from the area of interest. A size information identification unit identifies information concerning a size of the virtual surface from the three-dimensional shape information of the virtual surface.

The application discloses an endoscope image processing method and device, electronic equipment and storage medium, and relates to the technical field of endoscopes. In the processing method, the pixels of a first a channel image are subjected to a mapping operation based on a preset mapping table, which can map a red region in the image into a preset color; each pixel of a first b channel image is subjected to color offset, which can reduce the value of each pixel of the first b channel image, thereby reducing the color of the background in the current frame image to obtain a second b channel. The first HSV image is adjusted by using historical HSV adjustment parameters corresponding to the previous frame image, which can improve the smoothness of the video when the video composed of the continuous multiple frames of images is displayed by the display module. The application realizes electronic dyeing of the endoscope image under white light irradiation, facilitates medical practitioners to observe the endoscope image, and can reduce the equipment cost and the occupied volume of the endoscope.

A method for guiding bone removal includes receiving at least one two-dimensional image of bony anatomy captured by a two-dimensional imaging system and at least one endoscopic image of the bony anatomy captured by an endoscopic imager. A three-dimensional model of the bony anatomy is registered with the endoscopic imager in a common reference frame using the two-dimensional image and a tracking system. A position and orientation of the three-dimensional model relative to the endoscopic image is determined based on registration of the three-dimensional model with the endoscopic imager in the common reference frame. The endoscopic image is displayed simultaneously with at least a portion of the three-dimensional model based on the determined position and orientation of the three-dimensional model relative to the endoscopic image.

Described herein are methods for tracking a location of a fixed point in endoscopic images. Endoscopic video data is received. A selection from a user of a location within a first two-dimensional image on which to place a graphical marker is received. The graphical marker is placed on the first two-dimensional image at the location within the first two-dimensional image. A first location of an interest point within the first two-dimensional image is detected by applying a machine learning model to the first two-dimensional image to identify the interest point. A second location of the interest point within a second two-dimensional image of the video data is detected by applying the machine learning model to the second two-dimensional image to identify the interest point. The graphical marker is placed on the second two-dimensional image in accordance with the detection of the second location of the interest point.

An AI-based endoscopic diagnostic aid system includes: an endoscope module providing an endoscopic image of internal organs of the body of a patient; an input module configured to be capable of inputting arbitrary medical information about the patient; a control module which analyzes the endoscopic image provided from the endoscope module through a pre-stored image processing program to detect lesion information, matches the detected lesion information with the medical information input from the input module through a pre-stored lesion diagnosis program while generating at least one diagnosis information of malignancy and malignancy probability corresponding to the matching result, and outputs a preset notification signal according to the lesion information and the diagnosis information; and a notification module which visually displays on an arbitrary screen according to the notification signal output from the control module.

The present application belongs to the technical field of endoscope and image processing, and in particular to an endoscope image anti-shake method based on motion vector compensation, comprising the following steps: collecting a sequence of continuous video frames, performing stability screening on the initial feature point set, eliminating unstable feature points located in the edge region, high light reflection region and low contrast region of the image to form a reliable feature point set, performing robust estimation processing on the original motion vector set based on the matched feature points in the reliable feature point set to generate a stabilized output frame, wherein the remapping adopts a bilinear interpolation algorithm; by constructing a closed-loop anti-shake system taking local feature motion vectors as input, spatio-temporal domain joint filtering as core and sub-pixel remapping as output, high-precision separation and compensation of high-frequency micro-shaking in endoscope imaging are realized without introducing additional mechanical structures.

In the endoscopic examination support apparatus, three three-dimensional model generation means generates a three-dimensional model of a luminal organ in which an endoscope camera is placed, based on endoscopic images acquired by imaging an interior of the luminal organ with the endoscope camera. The unobserved area detection means detects an area estimated not to be observed by the endoscope camera, as an unobserved area, based on the three-dimensional model. The display image generation means generates a display image including information indicating an observation achievement degree for each of a plurality of sites of the luminal organ, based on the detection result of the unobserved area. The endoscopic examination support apparatus may be used to support user's decision making.

1. Name of the product in this design: Full-touch picture-in-picture graphical user interface for electronic devices. 2. Purpose of this design: An electronic device. 3. The key design feature of this product is its graphical user interface. 4. The picture or photo that best illustrates the key design points: Design 1 main view. 5. Design 1 is designated as the basic design. 6. Purpose of the graphical user interface: This graphical user interface is used to simultaneously observe the images displayed in the main display area and the sub-display area (e.g., endoscopic images and ultrasound images), and to adjust the images displayed in the main display area and the sub-display area by touch. In the main view of Design 1, perform a zoom-in gesture on the sub-display area displayed on the left side of the main display area in the center to display the changing state diagram of Design 1. In the main view of Design 2, perform a zoom-in gesture on the sub-display area displayed on the left side of the main display area in the center to display the change state diagram of Design 2. In the main view of Design 3, perform a zoom-in gesture operation on the sub-display area displayed on the left side of the main display area in the middle, and display the change state diagram of Design 3 (Figure 1). In the design 3 change state diagram 1, the zoom gesture operation is performed again on the sub-display area to display the design 3 change state diagram 2. In the main view of Design 4, perform a zoom-in gesture operation on the sub-display area displayed on the left side of the main display area in the middle, and display the change state diagram of Design 4 (Figure 1). In Design 4 Change State Diagram 1, a zoom-out gesture is performed on the sub-display area, and Design 4 Change State Diagram 2 is displayed. In the main view of Design 5, perform a zoom-in gesture operation on the sub-display area displayed on the left side of the main display area in the center, and display the change state diagram of Design 5 as shown in Figure 1. In Design 5 Change State Diagram 1, a zoom-out gesture is performed on the sub-display area to display Design 5 Change State Diagram 2. In the main view of Design 6, perform a zoom gesture operation on the sub-display area displayed on the left side of the main display area in the middle, and display the change state diagram of Design 6 (Figure 1). In the design 6 change state diagram 1, the zoom gesture operation is performed again on the sub-display area to display the design 6 change state diagram 2. In the design 6 change state diagram 2, a shrink gesture operation is performed on the sub-display area to display the design 6 change state diagram 3. In the main view of Design 7, perform a zoom gesture operation on the sub-display area displayed on the left side of the main display area in the middle, and display the change state diagram of Design 7 (Figure 1). In the design 7 change state diagram 1, the zoom gesture operation is performed again on the sub-display area to display the design 7 change state diagram 2. In the design 7 change state diagram 2, a shrink gesture operation is performed on the sub-display area to display the design 7 change state diagram 3. In the design 7 change state diagram 3, the shrink gesture operation is performed again on the sub-display area, and the design 7 change state diagram 4 is displayed. In the main view of Design 8, click the "Main Display Switch" button to display the Design 8 change state diagram 1. The images displayed in the central main display area and the sub-display area to the left of the main display area are swapped. See Design 8 change state reference diagram 1. In Design 8 Change State Diagram 1, a zoom-in gesture is performed on the sub-display area to display Design 8 Change State Diagram 2. The main display area is shown in the middle of the main view of Design 9. Clicking the "Picture-in-Picture" button on the right side of the main display area will display the sub-display area on the left side of the main display area, showing the change status diagram of Design 9, Figure 1. In Design 9 Change State Diagram 1, a zoom-in gesture is performed on the sub-display area to display Design 9 Change State Diagram 2. The main display area is shown in the middle of the main view of Design 10. Clicking the "Picture-in-Picture" button on the right side of the main display area will display the sub-display area on the left side of the main display area, showing the change status diagram of Design 10 (Figure 1). In Design 10 Change State Diagram 1, a zoom-in gesture is performed on the sub-display area to display Design 10 Change State Diagram 2. In the design 10 change state diagram 2, the zoom gesture operation is performed again on the sub-display area to display the design 10 change state diagram 3. In all the designs, the dark gray and light gray overlays represent two content screens that are independent of the graphical user interface.

This invention discloses a method for identifying lesion regions in gastrointestinal images based on a visual language model. It combines the cross-modal feature alignment capability of the visual language model, a medical semantic guidance mechanism, and a data imbalance handling strategy. By constructing a matching relationship between images and corresponding text categories, it achieves the identification of gastrointestinal lesions. This method introduces a bidirectional contrastive learning loss, enhancing the discriminative ability of different lesion categories while maintaining consistency between image features and text semantics. Furthermore, to address the uneven distribution of lesion categories in the training data, this invention introduces a category weighting mechanism and sampling strategy, allowing minority class samples to obtain higher weights or higher frequencies during training, effectively improving the identification accuracy and model stability of minority class lesions. During the inference stage, classification is completed by calculating the similarity between the image and all categories of text features, achieving the identification and classification of various gastrointestinal endoscopic images without the need for additional labeled data or complex training.

The invention relates to the technical field of surgical navigation, in particular to a children urinary system endoscopic surgery visual auxiliary operation system which comprises a respiratory displacement field generation module, a tissue stiffness inversion module, a visual scale correction module, a tangential gradient calculation module and an anisotropic registration correction module. According to the method, a respiration displacement field is analyzed to invert tissue hardness parameters, a preoperative physical model with biomechanical characteristics is constructed, non-rigid deformation behaviors of kidneys and ureters of children in a stress state are simulated, the physical depth of pixels is calculated in combination with endoscopic image characteristics, sparse feature point cloud is converted into real-scale spatial data, and the spatial depth of the kidneys and ureters of the children is calculated. And according to the tangential gradient change of the path, constructing an anisotropic weight matrix to dynamically adjust the registration constraint intensity, and driving the virtual model to follow the movement track of the endoscope to carry out adaptive deformation correction, so that real-time three-dimensional visual guidance for soft tissue respiratory movement and contact deformation is realized.