101 results about "Digestive endoscopy" patented technology

The invention relates to the field of endoscope image processing, and particularly discloses an automatic lesion recognition system for endoscopy of the digestive system department, which is characterized in that after a preprocessed original endoscope image is acquired, a double-branch parallel processing architecture is used to acquire characteristics with low resolution and rich semantic information through a deep context branch, and the characteristics of the original endoscope image are acquired. The potential area of the focus is accurately deduced; meanwhile, the fine texture of the mucous membrane is captured in a lossless manner through shallow detail branches which keep high resolution in the whole process. Furthermore, through a context-guided asymmetric enhancement mechanism, a global view of a deep branch is utilized to generate an uncertainty perception attention map as a reference, and weak detail features corresponding to a potential focus area in a shallow branch are accurately irradiated and adaptively enhanced. Thus, a conservative enhancement strategy is adopted in an uncertain focus area, background noise is effectively inhibited, and therefore the detection sensitivity and robustness of low-contrast and flat focuses are fundamentally improved.

The invention relates to the technical field of medical care, in particular to a digestive endoscopy postoperative intelligent nursing system based on deep learning. The system comprises an endoscopic postoperative data acquisition and processing module, an endoscopic postoperative feature analysis module, an endoscopic postoperative recovery risk assessment module and a postoperative recovery risk intelligent nursing module, and can acquire physiological data of a patient after a digestive endoscopic surgery and digestive tract state image data after the endoscopic surgery; the long and short time memory network is used for obtaining the long and short time sequence characteristics of the physiological indexes after the endoscopic surgery, the convolutional neural network is used for identifying and obtaining the healing abnormal lesion characteristics after the endoscopic surgery, the characteristics are input into the preset deep neural network model for characteristic fusion, and postoperative recovery risk assessment calculation and postoperative intelligent nursing analysis are carried out. And generating digestive endoscopy postoperative patient recovery nursing suggestions so as to execute corresponding digestive endoscopy postoperative risk recovery nursing operation. According to the invention, postoperative data of the patient can be accurately analyzed, so that nursing decision-making work can be intelligently guided.

The invention relates to the technical field of artificial intelligence, and provides a digestive tract tumor non-invasive evaluation system and method based on an artificial intelligence technology, and the method comprises the steps: carrying out the recognition of a lesion position and a digestive tract part of a digestive endoscopy image; guiding a confocal laser microendoscope to perform targeted biopsy on the central position and / or the edge position of the focus; lesion classification identification is carried out on the confocal endoscope image; when it is determined that the lesion exists, an ultrasonic endoscope is guided to conduct ultrasonic scanning on the lesion position; performing focus size and focus infiltration depth identification on the ultrasonic endoscope image; generating a question text based on multi-modal data according to the lesion position, the alimentary canal part, the lesion category, the lesion size and the lesion infiltration depth, retrieving an answer from a preset vector database according to the question text, filling a prompt template with a retrieval result and the question text, and inputting the retrieval result and the question text into an intelligent question and answer language model to generate an alimentary canal tumor assessment result, and non-traumatic evaluation and auxiliary diagnosis of digestive tract tumors are realized.

The invention relates to the technical field of computer vision and medical image processing, and discloses a digestive endoscopy report generation method and device based on a multi-modal large language model. The method specifically comprises the following steps: for a plurality of upper gastrointestinal endoscopy images, extracting a first image feature through a visual encoder to perform anatomical part identification; extracting a second image feature from the first image feature by using a self-attention mechanism to perform background gastric mucosa classification; generating description about digestive tract diseases through a large language model, and under the guidance of the description, extracting third image features from the first image features to perform digestive tract disease classification; the constructed text instruction about report generation and the third image feature are combined and then input into a multi-mode decoder, and a digestive endoscopy report is generated. According to the method, a plurality of upper gastrointestinal endoscopy image analysis tasks can be processed at the same time, and a digestive endoscopy report which does not depend on a fixed template and has rich context content can be generated with high efficiency and high accuracy.

The invention relates to the technical field of medical instruments, in particular to an oral cavity closing prevention device for digestive endoscopy. Comprising a circular tube, an inflation device and a drainage tube, an elastic piece is attached to the exterior of the inflation device, longitudinal supporting mechanisms are rotatably arranged at the two ends of the elastic piece and used for opening the oral cavity and the lower jaw of a patient, and telescopic plates are slidably arranged between the supporting mechanisms and the circular tube; a connecting structure used for applying pressure to the expansion assembly is arranged between the longitudinal supporting mechanism and the transverse supporting mechanism, and a filling area is formed between every two adjacent supporting mechanisms. The movable plate is used for applying pressure to the air bags in the transverse supporting plate, the air bags expand and extend outwards firstly, and when the air bags on the two sides make contact, the connecting baffle is matched with the movable plate to indirectly apply pressure to the transverse supporting plate through the air bags, so that the transverse supporting plate expands the mouth of a patient in the transverse direction, and the longitudinal supporting plate and the transverse supporting plate are prevented from moving at the same time; and the comfort of the mouth in the transverse direction is improved.

The application relates to the field of medical masks, in particular to a mask with a gastrointestinal mirror channel, which comprises a mask main body, an oxygen supply assembly, a gastrointestinal mirror channel assembly, an anti-aspiration assembly and a sealing adjustment assembly. The oxygen supply assembly provides clean and controllable breathing support through a multi-layer filter membrane and a flow adjustment function; the gastrointestinal mirror channel assembly is designed with a flexible sealing sleeve and an elastic support strip, thereby ensuring air tightness and adaptability during gastrointestinal mirror operation; the anti-aspiration assembly effectively deals with vomiting reflex or gastric content reflux through an arc-shaped baffle and a drainage groove; and the sealing adjustment assembly improves the adhesion and stability of the mask by using an air bag. The application can simultaneously realize breathing support and gastrointestinal mirror operation, and takes into account the safety and comfort of patients, thereby providing a perfect solution for digestive endoscopy.

The utility model provides a clamping structure for a digestive endoscopy accessory, relates to the technical field of clamping structures for digestive endoscopy accessories, and adopts the following technical scheme that the clamping structure for the digestive endoscopy accessory comprises two supporting legs which are symmetrically arranged, the tops of the two supporting legs are fixedly connected with clamping heads, and the two supporting legs are fixedly connected with the clamping heads. The outer surfaces of the two supporting legs are rotatably connected with a fixing shaft, a pipeline is arranged outside the two supporting legs, the bottom of the pipeline is fixedly connected with a sliding rail, the inner surface of the sliding rail is slidably connected with symmetrical first sliding blocks, and the tops of the two first sliding blocks are fixedly connected with moving plates. According to the device, the clamping head can deviate towards one side for use in the moving process through the arrangement of the bent sliding groove, meanwhile, the clamping head can be rotated for use through the arrangement of the sliding rail at the bottom of the pipeline, and the situation that an original clamping head needs to stretch back and forth to determine the direction is avoided; and the working efficiency of medical staff and the feeling of a user are improved.

The invention relates to the technical field of artificial intelligence, and provides an eosinophilic gastroenteritis typing system and method based on multi-modal data, and the method comprises the steps: obtaining a digestive endoscopy image, an ultrasonic endoscopy image, a CT examination image and clinical manifestation data of a to-be-detected person collected in the same time period; the gastrointestinal mucosa lesion state is recognized through a digestive endoscopy image and is represented by numerical features; identifying the abnormal thickening condition of each film layer structure of the gastrointestinal wall through an ultrasonic endoscope image, and representing the abnormal thickening condition by numerical features; lesion hierarchical distribution is judged through an ultrasonic endoscope image, and numerical feature representation is carried out; determining seroperitoneum and intestinal wall thickening conditions through a CT examination image, and performing numerical characteristic representation on the seroperitoneum and intestinal wall thickening conditions; clinical symptom features are obtained through the clinical manifestation data and are represented by numerical features; generating a feature vector based on multi-modal data by using each numerical value feature representation; and inputting the feature vectors into a dynamic typing prediction model to realize the typing prediction of eosinophilic gastroenteritis, thereby improving the accuracy of auxiliary diagnosis and evaluation.

The application belongs to the field of medical image recognition and relates to an esophageal lesion image recognition method based on a deep neural network, which comprises the following steps: step 1, processing actual digestive endoscopy video data to construct training data; step 2, constructing a deep neural network model and training the deep neural network model through the training data to obtain an early esophageal cancer recognition model; step 3, inputting digestive endoscopy video data to be detected into the early esophageal cancer recognition model, and the early esophageal cancer recognition model judging whether there is a suspected esophageal cancer lesion and outputting a lesion judgment result; the lesion judgment result comprising a spatial position and a progression stage of the lesion; the method models the time sequence dynamics of lesion characteristics in the endoscopy recognition process, and combines a graph neural network to depict the feature correlation of early esophageal squamous cell carcinoma under multi-view conditions, so as to improve the recognition accuracy, consistency and robustness of the model in a real clinical application scenario.

The invention belongs to the field of medical instruments, and provides a gastroendoscopic gastric calculus crushing device which comprises a control assembly, the control assembly comprises a shell and a handle, and the handle is fixedly mounted on the right side surface of the shell; the adjusting assembly is mounted on the surface of the bottom side of the shell; the stretching-in mechanism is installed on the surface of the bottom side of the shell, and the stretching-in mechanism is used for stretching into the stomach of a patient; the crushing assembly is mounted in the bottom end of the stretching mechanism, and the crushing assembly is used for crushing the gastric calculi; under the action of a vibration motor, the vibration motor acts on a tool apron to move, the tool apron drives a smashing cutter to conduct smashing, a steel wire rope drives a net basket to move, the net basket conveniently drives broken stones to move, continuous material supplementing is facilitated, the stone smashing efficiency is improved, the stone smashing effect is met, and after the attached broken stones are squeezed away through a cleaning plate, it is guaranteed that the surface of the smashing cutter is clean; the crushing cutter is convenient to operate and use.

The invention discloses a hollow organ three-dimensional model construction method, a model, electronic equipment and a storage medium. The method comprises the following steps: acquiring a two-dimensional plane graph by using digestive endoscopy to determine the contour of a target hollow organ in an expanded state, performing three-dimensional modeling by using a CT image to obtain an orthogonal projection drawing, obtaining the contour of a target hollow organ in a non-expanded state by combining the contour of the target hollow organ in the expanded state, and calculating an expansion coefficient through perimeter, expanding the hollow viscera in the non-expanded state according to the expansion coefficient to obtain an expanded target hollow viscera three-dimensional model; the digestive endoscopy image and the CT image are both original examination habits and techniques, the enlargement condition of the gastral cavity after inflation can be known under the condition that the original examination habits and techniques are not separated, and perception of the relation between the gastral cavity and adjacent tissue is recovered.

The invention discloses a YOLO-based lightweight attention-guided digestive endoscopy abnormal region detection method. The method comprises the following steps: preprocessing data by adopting an image enhancement strategy, and inputting the preprocessed data into a backbone network to extract features; and the extracted features are subjected to up-sampling, fusion and enhancement on the neck layer and then are input into the detection layer to complete position regression and category prediction of the target area. According to the method, an MDR-YOLO model is developed and is used for abnormal region detection of a digestive endoscopy image, and a proposed MSC2f module comprises a multi-scale parallel convolution attention module used for extracting multi-scale feature information; the provided D4f module comprises a deep semantic expression structure constructed by a multi-level convolution residual path, so that high-quality feature fusion and enhancement are realized; a proposed RSC2f module comprises a channel attention residual module which is used for improving the ability of the model to extract all-channel features. The experimental result verifies the effectiveness of the MDR-YOLO model in the detection of the abnormal region in the digestive endoscopy image, and especially shows higher accuracy and robustness in the detection of a small target.

The present invention discloses a controllable transparent cap and method for assisting digestive endoscopy controlled by a brushless motor, and belongs to the field of digestive endoscopy medical devices. Specifically, a brushless motor and a track fixing part are provided on the outer sleeve of a coupling connector, and the clockwise and counterclockwise rotation and rotation angle of a movable rotor are controlled by the brushless motor; a plurality of movable red plates are provided between the movable rotor and the track fixing part, and the movable red plates are evenly spaced along the circumference of the movable rotor; one end of the movable red plate is fixedly connected to the movable rotor; a plurality of sliding grooves are provided on the surface of the track fixing part in contact with the movable rotor, and a positioning column is provided on the side of the movable red plate in contact with the track fixing part, and the positioning column forms a sliding connection in the sliding groove; when the movable rotor and the track fixing part rotate relative to each other, the positioning column on the movable red plate slides back and forth in the sliding groove, realizing the opening and closing of the movable red plate and thus adjusting the inner diameter of the transparent cap.

A large-opening sliding groove hemostatic clamp for digestive endoscope comprises: a receiving tube, a clamping part, a sliding part and a traction part; the clamping part, the sliding part and the traction part are slidably arranged inside the receiving tube; the clamping part is connected to the sliding part, and the sliding part is connected to the traction part; the number of the clamping parts is two and they are symmetrically arranged up and down; the clamping part comprises an arc-shaped clamping piece, an elastic bent rod, a limiting hemisphere and a coupling hole; when the clamping part extends out of the receiving tube, it is in an open state; when the clamping part enters the receiving tube, it is in a closed state; the traction part is used to drive the sliding part, and the sliding part drives the movement of the clamping part; the arc-shaped clamping piece in the clamping part cooperates with the setting mode of the limiting hemisphere to facilitate the extension and retraction of the clamping piece, and the limiting hemisphere abuts against the inner wall of the receiving tube to ensure that the hemostatic clamp clamps the bleeding point more tightly; the connecting tube on the sliding part is equipped with a locking function to further limit the clamping part, thereby ensuring the stability and reliability of the hemostatic clamp.

The invention relates to the technical field of fruit picking, in particular to a digestive gastroscope inspection auxiliary device which comprises a fixing bracket, a plurality of locking bolts are arranged on the fixing bracket, the fixing bracket is detachably mounted on the bedside of a sickbed through the locking bolts, a jacking supporting block is fixedly mounted at one end of the fixing bracket, and a jacking block is fixedly mounted at the other end of the fixing bracket. A groove for placing a gastroscope tube is formed in the jacking support block; in the examination process, a doctor holds the operation board with one hand and pinches the gastrointestinal endoscope tube with the thumb and the index finger with the other hand to push the gastrointestinal endoscope tube, at the moment, the gastroendoscope tube can move freely, smoothness of the pushing process is ensured, when the next section of gastroendoscope tube is pushed, the thumb and the index finger retreat and open the driving opening and closing mechanism, and the gastrointestinal endoscope tube is pushed out. According to the gastroscope tube locking device, the locking mechanism can lock the gastroscope tube, the phenomenon that the gastroscope tube retreats due to factors such as the gravity of the gastroscope tube is effectively avoided, the continuity and accuracy of examination are guaranteed, when the gastroscope tube is clenched again to be pushed, the locking mechanism can drive the opening and closing mechanism to reset, and circulation and stability of operation are achieved.

The application discloses a method for predicting biological age based on enteroscopy pictures, and belongs to the field of digestive endoscopy image analysis. The method first screens enteroscopy pictures without abnormalities, and completes quality control and edge cutting; then constructs an image package with multiple pictures in a single report, extracts features and weighted aggregation by using a multi-instance learning model based on attention; and converts age prediction into a classification task and introduces a Gaussian soft label to realize high-precision biological age prediction. The application can mine aging-related features from enteroscopy images without obvious abnormalities, does not need additional equipment and invasive operation, has strong generalization, and can be used for risk stratification of enteroscopy-negative people and precise prevention and treatment of colorectal cancer.

This invention discloses a highly flexible appendiceal stent, comprising a proximal anchoring section, a middle flexible section, and a stent body. The outer wall of the stent body is provided with one or more anti-displacement rings. The stent is composed of multiple layers; preferably, the outer layer material of a three-layer stent is one or more combinations of polyurethane, modified nylon Pebax, and nylon. The outer layer materials of the stent body and the flexible section are more flexible than the proximal outer layer material. The middle imaging-enhancing layer is a spring mesh or braided mesh made of metal wires; the inner layer is a super-slippery polytetrafluoroethylene film. Compared to currently used appendiceal stents released via digestive endoscopy, the stent body of this invention is more flexible, causing less damage to the appendix. The stent with the flexible section is less prone to bending and provides smooth drainage. Furthermore, the appendiceal stent of this invention shows good imaging under X-rays, facilitating surgery. The anti-displacement rings and barbs placed in the stent body reduce the risk of displacement.

The utility model belongs to the field of medical equipment, and particularly relates to a digestive endoscope gas supply device which comprises a supporting column, and a clamping assembly is arranged on one side of the supporting column. And the clamping assembly comprises a sliding block, the sliding block is slidably connected to the front side of the supporting column, the interior of the sliding block is hollowed out, and a partition plate is fixedly connected to the middle of the sliding block. Through the synergistic effect of the sliding blocks, the partition plates, the telescopic rods, the springs and the surrounding-shaped clamping plates, the gas storage tank can be stably clamped. The elastic effect of the spring enables the clamping plate to adapt to the gas storage tanks with different diameters, and the compatibility of the device to the gas storage tanks with various specifications is effectively enhanced. In the moving or using process of the equipment, the problems of connection looseness and the like caused by shaking and displacement of the gas storage tank can be effectively prevented, the continuity and the stability of gas supply are ensured, a reliable gas source foundation is provided for digestive endoscopy operation, and smooth proceeding of the medical process is ensured.

The utility model relates to the technical field of minimally invasive surgery specimen processing, and discloses a fixed measuring plate for a full-thickness resection specimen in a digestive endoscopy super minimally invasive surgery, which comprises a measuring plate body and a fixing piece, the measuring plate body is used for supporting and observing a surgery specimen, the edge of the measuring plate body is provided with scale values, the measuring plate body is a transparent plate, and the fixing piece is arranged on the measuring plate body. The bottom of the measuring plate body is provided with an illuminating part for illuminating the measuring plate body, the plurality of fixing parts are detachably arranged on the measuring plate body, and the edge of the measuring plate body is provided with scale values, so that a specimen can be directly measured without an additional measuring tool; and the specimen is fixed through the arranged fixing piece, so that puncture damage to the specimen is avoided, the physical damage risk of the specimen in the fixing process is greatly reduced, and the integrity of the specimen is guaranteed.

The invention discloses a digestive endoscopy postoperative liquid diet guidance model based on swallowing function evaluation, which belongs to the field of diet guidance and comprises an evaluation mechanism, a driving mechanism arranged on the surface of the evaluation mechanism, a guidance mechanism mounted on the surface of the driving mechanism and an adjusting mechanism mounted on the surface of the guidance mechanism. The guiding mechanism comprises a mounting disc mounted on the surface of the driving mechanism, and the upper surface of the mounting disc is fixedly connected with a storage barrel used for storing liquid diet through bolts; through cooperative use of all the devices, accompanying personnel or nursing personnel are guided by adopting the liquid diet viscosity corresponding to the grade, an air pump is started according to the grade of the swallowing function of a patient, so that air enters a storage barrel, and through mutual cooperation of a second electromagnetic valve and an air conveying pipe, the swallowing function of the patient is achieved. The liquid diet divided into different grades by the partition plate is discharged through the corresponding discharge pipes and is conveyed into the oral cavity of a patient through the feeding pipe, so that medical personnel can guide accompanying personnel or nursing personnel.

The application provides a kind of intelligent biopsy area prompting method and system for digestive endoscopy, it is related to artificial intelligence auxiliary diagnosis and treatment field, including: obtaining real-time digestive endoscopy video stream;Through multi-scale visual neural network, multi-scale feature extraction is carried out frame by frame to video stream, candidate biopsy area detection is carried out based on multi-scale feature;Biological histology level multi-feature fusion analysis is carried out to candidate biopsy area, and the biopsy value of candidate biopsy area is evaluated;Based on biopsy value, generate prompt element, and real-time superimposed in original video stream;Wherein, the biological histology level multi-feature fusion analysis is from color, texture, blood vessel, morphology, semantic multi-feature of candidate biopsy area is constructed, and the confidence of multi-feature and candidate biopsy area is used to score biopsy value.The application can accurately prompt potential biopsy area in the process of examination, so as to assist doctors to improve sampling scientificity and diagnostic accuracy.

The utility model relates to the technical field of medical instruments, in particular to a digestive endoscopy operation auxiliary platform which comprises a first pipe body and a second pipe body. Wherein the outer side of the far end of the first tube body is fixedly sleeved with a cap; a guide pipe is arranged on the cap in a penetrating mode, the guide pipe is arranged on one side of the first pipe body, the guide pipe comprises a first arc-shaped plate-shaped structure and a second arc-shaped plate-shaped structure, and the two sides of the second arc-shaped plate-shaped structure are fixedly connected with the two sides of the first arc-shaped plate-shaped structure respectively; wherein the first arc-shaped plate-shaped structure and the second arc-shaped plate-shaped structure are enclosed to form a cavity. According to the auxiliary platform for operation of the digestive endoscopy, the diameter between the catheters can be changed through the arrangement of the catheters, the first arc-shaped plate-shaped structure and the second arc-shaped plate-shaped structure, so that blood clots or part of intestinal tract solid feces can be cleaned through different endoscope operation instruments, and the operation time of an endoscopic surgery is shortened.

The invention belongs to the field of medical image recognition, and relates to an esophageal lesion image recognition method based on a deep neural network, which comprises the following steps: step 1, processing actual digestive endoscopy video data, and constructing training data; 2, constructing a deep neural network model, and training the deep neural network model through the training data to obtain an early esophageal cancer recognition model; step 3, inputting to-be-detected digestive endoscopy video data into the early esophageal cancer recognition model, judging whether a suspected esophageal cancer focus exists or not by the early esophageal cancer recognition model, and outputting a focus judgment result; the lesion judgment result comprises a spatial position and a progress stage of the lesion; modeling is carried out on time sequence dynamic states of focus features in the endoscope recognition process, the feature incidence relation of the early esophageal squamous carcinoma under the multi-view condition is depicted in combination with a graph neural network, and the recognition accuracy, consistency and robustness of the model in a real clinical application scene are improved.

The invention relates to the technical field of medical instruments, in particular to a pharyngeal comfort nursing device for a patient after digestive endoscopy and a using method.The pharyngeal comfort nursing device comprises a U-shaped silica gel pad, two medicine carrier grooves are symmetrically formed in the inner side wall of the U-shaped silica gel pad, and sponge carriers are arranged in the medicine carrier grooves; negative pressure drainage channels are formed in the two sides of the U-shaped silica gel pad, a secretion retention area is formed in the U-shaped silica gel pad, and the negative pressure drainage channels are communicated with the secretion retention area; the double functions of medicine slow release and active drainage are achieved, through the symmetrical medicine carrier grooves of the U-shaped silica gel pad, the sponge carrier can continuously release analgesic and repairing medicine for 6-8 hours, and repeated medication is reduced; meanwhile, the negative pressure drainage channel, the secretion retention area and the one-way valve form a closed-loop drainage system, retention secretions are actively removed, the bacterium breeding risk is reduced, and the comfort level of a patient is improved.

The present invention discloses an auxiliary device of a digestive endoscope and a method of using the same, belonging to the technical field of medical auxiliary equipment, and comprising an endoscope connector and a guard plate, wherein a light guide plug is connected to the bottom of the endoscope connector, a video interface is provided on the front of the endoscope connector, and an angle knob is provided on the side of the endoscope connector. In the present invention, a splint, a sliding rod, a first limiting rod, a first limiting groove, a connecting bearing, and a rotating cylinder are adopted. When the first elastic component quickly resets the first limiting rod to the first limiting groove, precise control of the rotation of the connecting cable is achieved, ensuring that the angle of the connecting cable is effectively limited after it is rotated to a specified angle, avoiding arbitrary rotation, and greatly improving the stability of the angle control. This improvement not only significantly reduces the difficulty of operating the digestive endoscope, but also achieves significant optimization in the observation accuracy in the patient's body, thereby improving the overall diagnosis and treatment efficiency and the safety and comfort of the patient.

The invention provides a bite block capable of monitoring the breathing state in real time, relates to the technical field of medical monitoring equipment, and solves the problem that an existing bite block cannot accurately monitor abnormal breathing in real time and cannot give an early warning in time. According to the bite block, a bite block body extending forwards from the lower end face of an occlusal surface is integrated with a dual-mode combined sensor, namely a pressure sensor and a temperature sensor, pressure and temperature changes of respiratory airflow are detected, signals are converted into electric signals through a signal conversion module, data fusion and analysis are carried out through a signal processing module, and respiratory mode waveforms are generated; the feedback system judges the breathing state according to the waveform characteristics and triggers the alarm device when no breathing exists or breathing is abnormal. By means of the multi-mode sensing and intelligent analysis technology, dynamic capture of respiration parameters and instant response of abnormal events are achieved, the system is suitable for the scenes of digestive endoscopy, sleep apnea syndrome monitoring, postoperative respiration monitoring and the like, and the real-time performance and reliability of medical monitoring are improved.

The invention relates to the technical field of medical surgical instruments, and discloses a needle holding forceps for minimally invasive surgery under digestive endoscopy, which comprises a forceps head, a forceps body and a control handle, the forceps head comprises a connecting column, a rotating seat is rotatably mounted at the end part of the connecting column, and two clamping plates are rotatably mounted at one end, far away from the connecting column, of the rotating seat; one end of the plier body is fixedly connected with the control handle, the transition sleeve is fixedly installed at the other end of the plier body, and the connecting column is fixedly installed in the transition sleeve. According to the needle holding forceps, the clamping plate is arranged on the rotating seat, the rotating seat is rotationally connected with the connecting column, the third transmission steel wire in the forceps body pulls the transmission arm, the transmission arm circularly moves back and forth with the axis of the connecting column as the axis, rotation of the rotating seat is achieved, and the pressing key is pressed or loosened through the control handle. The two ends of the third transmission steel wire are pulled, forward or reverse rotation of the rotating seat can be controlled, the rotating angle of the rotating seat can be kept, operation is more flexible, and needle holding and sewing actions are more convenient.

The invention discloses a digestive endoscopy report automatic generation method based on a multi-modal large language model, and the method comprises the steps: obtaining a plurality of first endoscopy image groups and corresponding first report texts, and training the multi-modal large language model to obtain a target model; in a reasoning stage, inputting a second endoscopic image group into the target model, generating an anatomical part prediction result through an anatomical part classification head, and generating an abnormal score through an abnormal probability scoring head; and determining a representative image according to the anatomical site prediction result, and generating a corresponding site description and diagnosis conclusion based on the representative image to form a second report text. By introducing an anatomical part identification and anomaly scoring mechanism into the multi-modal model, automatic generation from multiple endoscopic images to a structured report text is realized, the report generation efficiency and consistency are improved, and the method is suitable for digestive endoscopy auxiliary diagnosis scenes.

The invention relates to the technical field of medical images, in particular to a digestive endoscopy image intelligent labeling system based on deep learning and a method thereof. The deep learning algorithm module is in communication connection with the preprocessing module and is used for receiving the standardized digestive tract endoscope image data sent by the preprocessing module; based on the standardized digestive tract endoscopic image data, executing a deep learning algorithm to carry out image annotation; generating labeled digestive tract endoscopic image data and unlabeled digestive tract endoscopic image data; the label storage module is in communication connection with the deep learning algorithm module; the micro-service module is in communication connection with the deep learning algorithm module and the label storage module; the universal interface module is in communication connection with the label storage module and the micro-service module; carrying out standardized storage on the marked digestive tract endoscopic image data; and an API interface is provided for an external system to access labeled digestive tract endoscopic image data, so that the generalization ability and labeling accuracy of the model are remarkably improved.

The invention belongs to the technical field of digestive tract endoscopic surgery instruments, and particularly relates to a staining solution preparation instrument which is used for assisting an operator to calibrate a surgical area in a digestive endoscopic surgery so as to show the outline of a diseased part. Comprising a basic liquid containing part which comprises a first main body part provided with a near end and a far end, and a first containing cavity is formed in the direction from the far end to the near end; the first push rod partially extends into the first accommodating chamber, the staining fluid accommodating part is pre-filled with staining fluid, and the staining fluid accommodating part and the first accommodating chamber have a communicating state and a closed state; the pushing assembly comprises a second main body part which is connected with the staining fluid accommodating part; the second push rod is connected with the second main body part and can axially move relative to the second main body part, the second push rod has a first moving direction, and when the second push rod moves in the first moving direction, the staining fluid containing part and the first containing cavity are converted into a communicating state from a closed state.