66 results about "G i endoscopy" patented technology

The invention discloses a digestive endoscopy structured diagnosis report generation method and system based on image recognition. The method comprises the following steps: obtaining video data collected in the digestive tract endoscope operation process; according to the video data, carrying out current part identification and focus identification based on each video frame; and according to the identified parts and lesions and through combination of a medical knowledge base, generating a corresponding description text, and adding the description text to a structured template to obtain a diagnosis report. According to the method, the natural language description text can be automatically generated based on the endoscopic examination video to obtain the structured diagnosis report, so thatthe standardization and normalization degree of the diagnosis report is improved, and the working efficiency of doctors is improved.

The invention provides a capsule endoscopy tablet and a detection method thereof. The capsule endoscopy tablet comprises an outer package structure body and a tablet core. A sealing cavity formed by the outer package structure body is filled with the tablet core. The outer package structure body comprises a time control part. The time control part comprises slow-releasing enteric-coating material and colon locating and dissolving material. The tablet core comprises material used for changing colors of urine and/excreta. By recognizing changes in colors of urine and/excreta, judgments can be made as to whether an examinee is suitable for examination of capsule endoscopy or not.

The invention discloses a method for intelligently identifying and recording alimentary canal tissues and foreign matters in endoscopic examination. Images detected by the gastrointestinal endoscopy are stored in a classified manner; a deep learning neural network model is established, image data stored in a classified mode are randomly divided into a training set and a test set for deep learningtraining, real-time enteroscope image information is obtained and subjected to vector gridding processing, then the real-time enteroscope image information is input into the deep learning neural network model for recognition, and the beneficial effects of being rapid, accurate and autonomous in learning are achieved.

The invention provides a pancreas ultrasonic endoscopy navigation method based on artificial intelligence. The pancreas ultrasonic endoscopy navigation method comprises the following steps: acquiringa pancreas ultrasonic endoscopy image; screening qualified pancreatic ultrasonic endoscopic images; performing site identification on the screened qualified images by using a pre-trained artificial intelligence network classification model; generating a guide line containing all stations in advance, displaying the stations and operation instructions of the corresponding stations on the guide lineaccording to the identified stations, and prompting the next endoscope operation, wherein the operation instructions are preset station operation instructions in one-to-one correspondence with different stations, and the guide line specifically comprises a gastral cavity line, a gastral cavity line, a duodenum ball line and a duodenum descending line. Endoscope doctors, especially beginners, can be well helped to ensure pancreas full coverage during pancreas ultrasonic endoscopy, a training system is perfected, and the training quality is improved.

The invention relates to the technical field of medical image processing, in particular to a method for automatic freezing of a digestive endoscopy image based on a perceptual hash algorithm. The method comprises the following steps: S1, analyzing an endoscopy video stream acquired by endoscopy equipment into image data; S2, calculating the similarity between a picture at a time point t and previous n frames of pictures to obtain the weighted similarity k of the picture; and S3, comparing the picture weighting similarity k at the time point t with a freezing boundary l, and triggering an imagefreezing instruction when the value k reaches the value l. After the method is applied, when an endoscopic doctor needs to carefully check a view image, the endoscopic doctor only needs to stop moving an endoscope body to keep the view unchanged, a frozen image can be automatically judged, and the doctor does not need to manually operate a freezing button, so the workload of the doctor is reduced. A system automatically executes a freezing instruction, so offset of a frozen image view or loss of effective information caused by slow response or unskilled operation of people can be avoided, anda clear image of the optimal view can be effectively obtained.

The invention provides a machine learning-based endoscopic auxiliary biopsy system and method, and belongs to the technical field of endoscopes, and the system comprises an image collection module which is used for obtaining a to-be-detected part video frame collected in real time in an endoscopic examination process; a lesion infiltration depth recognition module used for recognizing a lesion area of the video frame of the to-be-detected part by utilizing the lesion infiltration depth recognition model and carrying out infiltration depth scoring on different differentiation types of the lesion area to obtain mask images with scoring matrixes with different infiltration depths; wherein the lesion infiltration depth recognition model is obtained by training a plurality of groups of data, and each group of data comprises an endoscopic image containing a focus area and labeling information used for labeling different differentiation types of the lesion in the endoscopic image. An endoscopic lesion scoring matrix can be displayed in real time according to the lesion differentiation type and the infiltration depth, so that an endoscopic doctor is assisted in selecting the optimal biopsy part, and the optimal diagnosis and treatment decision is obtained clinically.

The invention relates to the technical field of medical assistance, in particular to an esophageal cancer preoperative management method and system based on artificial intelligence. The system comprises a client and a server. The server comprises a case typical picture library, a picture pattern recognition module, a high-risk lesion target recognition module, a surface blood vessel typing recognition module and an AVA segmentation recognition module. According to the method, a plurality of deep learning models (relating to image classification, target detection and semantic segmentation) areconstructed according to different tasks, a cascaded model fusion technology is adopted, and a full-process intelligent auxiliary diagnosis function is provided in the esophageal disease screening process of an endoscopic doctor. The esophagus high-risk focus is detected in the NBI distant view mode, and AB typing and AVA typing are observed according to endoscopic examination, so that doctors canbe helped to better and accurately evaluate the illness state of a patient, and more scientific treatment options are provided.

The invention relates to the technical field of telemedicine, in particular to a remote medical system and method for digestive endoscopy. The remote medical system comprises a primary hospital doctorend, an equipment end, a server end, a remote diagnosis doctor end and a patient terminal, wherein the primary hospital doctor terminal is used for uploading complete data of a patient; the equipmentterminal is used for automatically acquiring the iconographic data of the patient and sorting the data of the patient; the server is used for automatically allocating doctors to perform remote diagnosis and auditing, regularly reminding the doctors to complete diagnosis and treatment tasks on time, generating a personalized diagnosis and treatment report according to the specific condition of thepatient and sending the personalized diagnosis and treatment report to the patient terminal; and the remote diagnosis doctor terminal enables doctors participating in the remote diagnosis and auditing task to register information and complete the task. According to the invention, the purpose of remote diagnosis and treatment of the endoscopic examination result of the patient can be realized; andthe report content is intelligently adjusted according to the basic information of the patient, so personalized report service is realized, and diagnosis accuracy is improved.

The invention belongs to the technical field of medical equipment, and particularly relates to an endogastric biopsy Raman image auxiliary diagnosis method and system based on artificial intelligence. The artificial intelligence technology is applied to gastroscope stimulated Raman scattering endoscopic biopsy tissue image auxiliary diagnosis for the first time, and the method comprises the step: after histopathology image information is obtained by means of the stimulated Raman scattering microimaging technology, adopting the image classification and image omics data analysis based on a deep learning neural network and machine learning, and constructing an endogastric biopsy Raman image auxiliary diagnosis system. The system comprises a stomach tissue Raman image data preprocessing module, a neural network model, an algorithm module for training the neural network model, a neural network fine tuning module and a test module; compared with an existing traditional endoscope diagnosis and treatment system, the system has the advantages that real-time, rapid and intelligent diagnosis support in the endoscope examination process is achieved, and pathologists do not need to conduct explanation.

The invention provides a monitoring method and device for intestinal endoscopic examination. The monitoring method for the intestinal endoscopic examination comprises the steps that an intestinal endoscopic examination video shot by an intestinal endoscopic during the intestinal endoscopic examination and a plurality of intestinal left images obtained by left images when a doctor performs the intestinal endoscopic examination are obtained; extracting a plurality of intestinal endoscopic images based on the intestinal endoscopic examination video; performing three-dimensional reconstruction on the plurality of intestinal endoscopic images to obtain an intestinal three-dimensional model; the intestinal tract three-dimensional model is expanded, an intestinal tract inner wall two-dimensional expanded view is obtained, and the intestinal tract inner wall two-dimensional expanded view comprises an intestinal tract scanning area and an intestinal tract non-scanning area shot by an intestinal tract endoscope; determining the area of a reserved image region based on the plurality of intestinal tract reserved images; calculating a reserved image area ratio of the area of the reserved image area to the area of the intestinal scanning area; and determining an intestinal endoscopy monitoring result based on the left image area proportion. According to the application, the process of the intestinal endoscopy can be accurately monitored, and the quality of the intestinal endoscopy can be accurately evaluated.

The invention discloses a method and a system for automatically identifying biopsy parts and biopsy quantity under an endoscope. The method comprises the following steps: acquiring video frames acquired in real time in an endoscope examination process; detecting whether biopsy forceps exist frame by frame based on a pre-trained biopsy forceps detection model; if the biopsy forceps exist, based ona pre-trained endoscope part recognition model, determining an endoscope part for biopsy at present; and meanwhile, starting to track the biopsy forceps to obtain continuous video clips containing thebiopsy forceps, and determining whether biopsy is carried out or not according to a pre-trained biopsy taking action recognition model. The biopsy forceps can be accurately detected and tracked, thebiopsy taking part is determined, the biopsy taking scene is accurately inferred, the biopsy taking frequency is counted, and a guarantee is provided for a standard pathological report.

The invention discloses an endoscopic image lesion area detection method, device and equipment and a storage medium. The method comprises the steps of obtaining a to-be-detected image; preprocessing the to-be-detected image to obtain a preprocessed image; performing depth estimation processing on the preprocessed image to obtain a target depth image; based on depth features of the target depth image, performing semantic segmentation processing on the preprocessed image to obtain a target segmented image, wherein the target segmented image comprises a focus area and a background area; and fusing the target segmented image and the to-be-detected image to obtain a target detection image. According to the application, the semantic segmentation model is guided to segment the pre-processed image through the obtained depth image, so that the detection rate and the detection accuracy of a tiny lesion area are improved, and the segmentation detection effect of the lesion area edge is improved, the false positive probability is greatly reduced, and a good auxiliary effect is achieved on clinical endoscopy.

The invention relates to an intra-cavity lesion positioner in capsule endoscopy. The intra-cavity lesion positioner comprises a capsule endoscope for detecting and discovering a lesion portion and is characterized by further comprising a capsule positioner, the capsule positioner is connected with the capsule endoscope through a remote control disconnecting and connecting device, the capsule positioner comprises a capsule shell, an automatic parking and positioning device, a positioning signal transmitter, a signal transmitter remote control switch and a battery, the automatic parking and positioning device is used for fixedly parking the capsule positioner on the lesion portion, the positioning signal transmitter is electrically connected with the battery and the signal transmitter remote control switch, and the signal transmitter remote control switch receives an instruction and controls the positioning signal transmitter to transmit positioning signals. The intra-cavity lesion positioner is ingenious in structure and convenient to operate, an intra-cavity minimal lesion can be accurately positioned in a laparoscopic minimally invasive operation is performed, operation efficiency is improved, positioning errors are avoided, operation errors are reduced, the capsule endoscope and the capsule positioner are mutually dragged, and tumbling of the capsule endoscope in an intestinal tube and examination blind areas are decreased.

The invention provides an information processing method and device for an endoscopic examination report. The method comprises the steps: obtaining first report result information of a first patient; judging whether the first patient has a focus or not according to the first report result information; if the first patient has a focus, obtaining first focus information; obtaining first criticality level information of the first focus information; obtaining first association list information of the criticality level and the report printing priority; obtaining a first printing priority according to the first association list information; after a first report result is printed for the first patient according to the first printing priority, obtaining first personal information of a first doctor;obtaining a first skilled field of the first doctor; judging whether the first criticality level information is matched with the first skilled field or not; if the first criticality level informationis not matched with the first skilled field, obtaining second doctor information; and sending the second doctor information to the first patient, so as to achieve the technical effects of evaluatingthe focus of the patient in real time, printing a report in time and matching a proper diagnosis and treatment doctor for the patient.

The invention provides an endoscopic examination auxiliary system based on artificial intelligence. The system comprises: an inspection data acquisition module configured to obtain inspection data through an endoscope device, the inspection data at least comprising inspection time data and inspection image data; the inspection quality monitoring module, which is configured to perform data processing on the inspection data, wherein the data processing at least comprises data statistics and data classification, and then the inspection data of a specific type are respectively input into a corresponding quality monitoring model constructed based on a neural network to determine the inspection quality; an examination auxiliary module, which is configured to input the examination image data into a corresponding recognition model constructed based on a neural network to obtain an identifier for auxiliary operation based on a part or an object corresponding to the examination image data, and output the identifier and the corresponding examination image data according to a preset mode.

The invention provides a method for measuring the size of an optical target under an endoscope in real time. The method comprises the following steps: calibrating a lens of the endoscope; training an image semantic segmentation network suitable for the endoscope image target; carrying out target segmentation on an endoscope image in the endoscope examination, extracting size information of the object image, and recording corresponding pose data; and calculating the size of the to-be-measured target through a multi-view geometrical relationship by using the size information of the two groups of object images and the pose data of the two groups of object images. According to the method, the target in the endoscope image can be tracked in real time on the premise of not changing the current endoscope examination program, the size of the target is accurately measured, and the examination efficiency and accuracy are improved.

The invention relates to the technical field of digestive department examination, and discloses an endoscopic examination auxiliary device for the digestive department. The endoscopic examination auxiliary device comprises a frame body; a water receiving tank is fixedly connected to the bottom of the frame body; a supporting plate is movably connected to the top of the water receiving tank; a connecting shaft is fixedly connected to the top of the supporting plate; a top plate is fixedly connected to the outer portion of the connecting shaft; the exterior of the top plate is movably connected with a clamping mechanism; the exterior of the clamping mechanism is movably connected with a containing mechanism; and water spraying holes are formed in the containing mechanism. A servo motor drives a main gear to rotate, the main gear drives a transmission gear to rotate, the transmission gear drives a containing basket to rotate, thereby guaranteeing that spraying of the water spraying holes to the endoscope device is more uniform in the cleaning process, and a contact effect between cleaning liquid and an endoscope device is fully improved; in the rotating process of the containing basket, bristles rotate; meanwhile, due to the fact that the endoscope device is fixed by the clamping mechanism, the bristles and the endoscope device move relatively, so that the bristles brush the endoscope, thereby effectively guaranteeing the cleaning effect of the endoscope device.

The invention provides application of a red cabbage pigment in preparation of an endoscope coloring agent. When the endoscope coloring agent prepared by the red cabbage pigment is used, the definition of a mucosa in an endoscopy is enhanced, and the microstructure, texture and lesion boundary of the mucosa are all more distinct, so that a doctor can observe more conveniently; the application is especially suitable for the screening of early-stage cancer.

The invention relates to an endoscopic examination image acquisition method and device. The endoscopic examination image acquisition method comprises the steps of determining the sum of Hamming distances between adjacent frames of images of an endoscopic examination video according to a preset image acquisition frame number threshold value in an endoscopic examination process; and when the sum of the Hamming distances reaches a preset first Hamming distance threshold value, storing an image with the highest recognition probability value or meeting a preset recognition threshold value from various types of endoscopic examination images which are classified and recognized in advance. According to the method, the endoscopic examination efficiency is improved, the working efficiency of a user is improved, and the manpower burden of endoscopic examination is reduced.

The invention belongs to the technical field of medical auxiliary instruments, and particularly relates to a breathing endoscopy auxiliary device which comprises an upper connecting cover and a connecting block, a positioning boss used for positioning and connecting an endoscope device hose is arranged in the upper connecting cover, through the arrangement of the connecting block, a rectangular block at one end of the connecting block is arranged on a limiting sliding rod in a sleeving mode, and the other end is arranged in a rectangular groove in the square groove in a sleeved mode through a T-shaped block. The arrangement plays a role in facilitating connection and disassembly of the upper connecting cover and a lower connecting cover and facilitating positioning and installation of a hose of the endoscope device on the positioning boss. Through the arrangement of a clamping block, a T gear shaft drives a right-angle bevel gear to rotate through a tooth groove, the right-angle bevel gear drives a T-shaped rod to move up and down in a limiting mode through a rack mechanism, and the T-shaped rod fixes the endoscope device hose in the positioning boss through the clamping block. Through the arrangement, the positioning boss can be conveniently limited and fixed, the work labor intensity of medical workers is relieved, and the stability of the endoscope device hose is improved.

The invention discloses a fixing device for endoscopy. According to the device, a body position supporting block capable of moving left and right and rotating is installed at the upper end of a fixing plate, a control box capable of moving front and back is installed on the surface of a side plate on the right side, a lifting plate capable of moving up and down is slidably connected to the surface of the right end of the control box, and an adjusting box capable of swinging up and down is hinged to the upper end of the lifting plate; and the left side and the right side of the upper end of the adjusting box are slidably connected with moving wheels capable of synchronously and relatively moving correspondingly, and the two moving wheels can synchronously and relatively rotate. When the device is used for examination, an endoscope tube can be placed between the two moving wheels, the endoscope tube can be limited and prevented from shaking when the endoscope tube is used due to the fact that the endoscope tube is too long, the two moving wheels can rotate relatively, a user can manually insert the endoscope tube when inserting the endoscope tube, and the two moving wheels can also be driven to rotate relatively by rotating a control knob. The endoscope tube is pushed to move through the two moving wheels, and the functions are diversified.

The invention discloses an automatic assessment device for digestive endoscopy examination quality of children and adults based on artificial intelligence, in particular to the technical field of automatic assessment devices for digestive endoscopy examination quality. The automatic assessment device comprises an assessment device host machine and a placing cabinet, wherein the assessment device host machine is arranged at the top end of the placing cabinet; a plurality of partition boards which are distributed in an equal distance are fixed on the inner wall of the placing cabinet; two accommodating mechanisms are arranged at the front end of each of the partition boards; each accommodating mechanism comprises a cavity; each cavity is formed in the front end of the corresponding partitionboard; an accommodating rod is arranged in each cavity; one end of each accommodating rod penetrates through the corresponding cavity to extend out of the front end of the corresponding partition board; a circular plate is fixedly arranged at one end of each accommodating rod; and four springs are arranged in each accommodating rod. According to the automatic assessment device disclosed by the invention, long transmission wires of a plurality of electronic gastroscopes are separately sorted and accommodated, so that the situation that the transmission wires are in disorder in an assessment site is avoided, each person can well identify the corresponding digestive endoscopy, and the normal running of the assessment work is guaranteed.

The invention discloses novel negative pressure suction type biopsy forceps, and solves the problems that an existing biopsy forceps occlusion texture is difficult in operation. The novel negative pressure suction type biopsy forceps are characterized in that one end of a suction tube 4 is provided with a sliding handle 5; the middle of the sliding handle 5 is vertically provided with a sliding handle holder 6; the sliding handle holder 6 is in sliding insertion in the suction tube 4; the other end of the suction tube 4 is connected with a biopsy tube 2; the other end of the biopsy tube 2 is connected with a biopsy cap 1; the tip of the biopsy cap 1 is provided with a step type groove 7; the step type groove 7 is provided with a steel wire rope head fixator 8 and a steel wire rope positioning hole 9; one end of a steel wire rope 3 is fixed on the steel wire rope head fixator 8, and the other end of the steel wire rope 3 penetrates through the steel wire rope positioning hole 9, the biopsy cap 1, the biopsy tube 2 and the suction tube 4 to be connected with the sliding handle holder 6; and a steel wire loop 10 is formed in the step type groove 7 of the biopsy cap 1. The novel negative pressure suction type biopsy forceps have the advantages that a biopsy forceps head is small, a target area is not blocked, biopsy can be accurately carried out in a target area, accuracy is high and efficiency is good. The novel negative pressure suction type biopsy forceps are especially suitable for applying to endoscopy and pathological specimen fetching.

The invention relates to an intracavity lesion positioner for the examination of an integrated capsule endoscope. The intracavity lesion positioner is characterized by including a capsule shell in which an automatic parking positioning device, a positioning signal transmitter, a signal transmitter remote control switch, a battery and a camera shooting and signal transmitting device are arranged, wherein the automatic parking positioning device is used for parking and fixing the intracavity lesion positioner at a lesion part, the positioning signal transmitter is electrically connected with the battery and the signal transmitter remote control switch, the signal transmitter remote control switch receives commands and controls the positioning signal transmitter to transmit positioning signals, and the camera shooting and signal transmitting device is used for shooting pictures and transmitting the pictures to an extrinsic picture recorder and camera shooting workstation. The intracavity lesion positioner for the examination of the integrated capsule endoscope is igneous in structure and convenient to operate, and can help surgical doctors quickly and accurately position the intracavity minimal lesion during a minimally invasive surgery, improve surgery efficiency and avoid positioning errors.