Intelligent integrated instrument management and counting system for operating room

Abstract

The invention relates to the technical field of operating room instrument management, and discloses an operating room intelligent integrated instrument management and counting system which comprises an intelligent instrument identification and AI information input module, a system database, a counting platform, a real-time tracking and positioning module, an AI path optimization module, an intelligent data analysis and management module, an AI prediction module and a man-machine interaction module. And an early warning and AI decision auxiliary module. The AI intelligent recognition technology runs through the whole surgical instrument management process, the intelligent level and accuracy of instrument counting are greatly improved, the surgical risk caused by manual counting errors is effectively avoided, the surgical preparation efficiency is remarkably improved, the fluency and efficiency of the working process of an operating room are improved, meanwhile, the instruments are prevented from being lost or stolen, and the working efficiency of the operating room is improved. Hospital asset safety is guaranteed; the counting platform realizes multi-dimensional accurate identification and analysis by means of an AI counting algorithm, comprehensively monitors the state of the instrument, timely discovers potential problems, and provides more reliable guarantee for operation safety.

Description

Technical Field

[0001] The present invention relates to the technical field of operating room instrument management, and particularly to an intelligent integrated instrument management and inventory system for operating rooms. Background Art

[0002] Surgical instruments refer to various tools used during surgical operations, generally divided into cutting instruments, clamping instruments, suture instruments, and traction instruments. On the operating table, the instruments need to be reasonably positioned. The commonly used instruments are placed in easily accessible positions, generally arranged in the order of the surgical steps. For example, in an ordinary surgical operation, when starting to incise the skin, the scalpel and tissue scissors are placed near the surgical incision; when performing hemostasis, the hemostatic forceps are placed in a prominent and easily accessible place, so that the doctor can quickly pick them up and use them when needed. For the instruments that are not temporarily used, they should also be placed in a position that does not interfere with the surgical operation to avoid instrument chaos or loss. They can be placed at the edge of the instrument table or in a dedicated placement area. After the operation, the instruments are inventoried to ensure that the quantity and status are correct.

[0003] The precise management and inventory of surgical instruments in the operating room are key links to ensure the smooth progress of the operation and avoid medical accidents. The traditional method relying on manual inventory and simple recording is inefficient and error-prone, and it is difficult to meet the requirements of the high-efficiency operation and refined management of modern operating rooms. Moreover, there are no effective intelligent means in aspects such as instrument status monitoring and usage process optimization, and potential problems of the instruments cannot be detected in time and intelligent assistance for the surgical process cannot be provided. For this reason, we have proposed an intelligent integrated instrument management and inventory system for operating rooms. Summary of the Invention

[0004] In view of the deficiencies of the prior art, the present invention provides an intelligent integrated instrument management and inventory system for operating rooms, which realizes an intelligent operating room system integrating instrument tracking, intelligent inventory, status monitoring, and data management by means of AI intelligent recognition technology, and solves the above technical problems.

[0005] To achieve the above object, the present invention provides the following technical solution: An intelligent integrated instrument management and inventory system for operating rooms, comprising:

[0006] An intelligent instrument identification and AI information input module, surgical instruments are equipped with intelligent tags, the intelligent tags integrate ultra-high frequency RFID and NFC technologies, when inputting, the characteristics of the surgical instruments equipped with intelligent tags are identified, and at the same time, the module is built-in with AI image recognition technology, and the surgical instruments are scanned comprehensively by using the AI image recognition technology, and the identified information and the scanned image data are stored in the system database;

[0007] The system database stores the electronic tags of surgical instruments and their corresponding characteristics;

[0008] Inventory platform: An inventory platform is set up in the operating room. Using the built-in AI inventory algorithm, the instruments are inventoried.

[0009] Real-time tracking and positioning module: Positioning base stations are installed in the operating room and work in coordination with the intelligent tags of surgical instruments to achieve real-time three-dimensional positioning and tracking.

[0010] AI path optimization module: According to the surgical arrangement, the usage frequency of surgical instruments, and the location information, it intelligently plans the optimal movement path of surgical instruments from the preparation area to the operating table and dynamically adjusts according to the actual situation during the operation.

[0011] Intelligent data analysis and management module: Collects and processes the data of other modules and makes analyses.

[0012] AI prediction module: Through in-depth analysis of the surgical instrument data in a large number of surgeries, it predicts the types, quantities, and usage sequences of surgical instruments required for different surgical types.

[0013] Human-computer interaction module: Conducts human-computer interaction through multiple terminals at different locations, including querying surgical instrument information, querying inventory results, and abnormal situation warnings.

[0014] Warning and AI decision-making assistance module: When the system detects abnormal situations in the surgical instrument data, it gives warnings in a timely manner. At the same time, it generates decision-making suggestions to assist in handling abnormal situations.

[0015] Preferably, the characteristics of surgical instruments include name, specification, model, manufacturer, weight, service life, and appearance information.

[0016] Preferably, the AI image recognition technology continuously learns the image changes of different instruments in various usage states for subsequent intelligent judgment of the instrument states.

[0017] Preferably, the inventory platform is built-in with several groups of weight sensors, image recognition cameras, and electromagnetic induction coils. The weight sensors measure the total weight, and in combination with the weight information of individual instruments in the database, the quantity range is initially estimated. The image recognition cameras take pictures of the instrument placement, and the electromagnetic induction coils detect the electromagnetic characteristics of metal instruments to assist in judging integrity.

[0018] Preferably, the AI inventory algorithm is the Faster R-CNN or YOLO (You Only Look Once) algorithm based on convolutional neural networks, and continuous training is carried out using the inventory results each time to continuously self-optimize, improve the inventory accuracy and speed, and adapt to the inventory requirements of different types of surgical instruments.

[0019] Preferably, the AI path optimization algorithm is the Dijkstra algorithm, A* algorithm, or ant colony algorithm.

[0020] Preferably, when there is an urgent surgical need, the transportation route of commonly used surgical instruments is preferentially scheduled to reduce the surgical instrument preparation time; meanwhile, the movement trajectory of the surgical instruments is recorded, and if the surgical instruments deviate from the preset path or enter a restricted area, the system will automatically alarm.

[0021] Preferably, the AI prediction algorithm built in the AI prediction module is a linear regression algorithm, a decision tree algorithm, a random forest algorithm or an LSTM algorithm.

[0022] Preferably, the abnormal conditions of the surgical instruments include: the number of surgical instruments does not match, the weight does not match, damage or approaching expiration, and being offline.

[0023] Preferably, the generated decision suggestions include: when it is found that there is a shortage of urgently needed surgical instruments, recommend alternative instruments or rapid deployment solutions; if the instrument is slightly damaged, evaluate whether it can continue to be used and give suggestions on the safe use range to assist medical staff in making correct decisions quickly and ensure the smooth progress of the surgery.

[0024] Compared with the prior art, the present invention provides an intelligent integrated instrument management and inventory system for the operating room, which has the following beneficial effects:

[0025] 1. The present invention penetrates the entire surgical instrument management process through AI intelligent recognition technology, greatly improving the intelligent level and accuracy of instrument inventory, effectively avoiding surgical risks caused by manual inventory errors, and significantly enhancing the surgical preparation efficiency.

[0026] 2. Combining the instrument tracking and path optimization functions of AI reduces the instrument transfer time, improves the fluency and efficiency of the operating room work process, and at the same time prevents the loss or theft of instruments, ensuring the safety of hospital assets.

[0027] 3. The inventory platform realizes multi-dimensional accurate recognition and analysis by means of the AI inventory algorithm, comprehensively monitors the instrument status, discovers potential problems in a timely manner, and provides a more reliable guarantee for surgical safety.

[0028] 4. The AI prediction module provides a forward-looking decision-making basis for the operating room instrument management, optimizes the procurement, inventory management and instrument use plan, reduces the operating cost, and improves the overall management efficiency and medical service quality of the operating room. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a system architecture diagram of an intelligent integrated instrument management and inventory system for the operating room of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] The following further describes in detail the embodiments of the present invention in conjunction with the drawings. The following embodiments are used to illustrate the present invention, but cannot be used to limit the scope of the present invention.

[0031] In the description of the present invention, unless otherwise specified, "a plurality of" means two or more; the orientation or positional relationship indicated by terms such as "upper", "lower", "left", "right", "inner", "outer", "front end", "rear end", "head", "tail", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be construed as a limitation on the present invention. In addition, terms such as "first", "second", "third", etc. are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0032] In the description of the present invention, it should be noted that, unless otherwise clearly specified and defined, the terms "connected" and "coupled" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0033] Please refer to Figure 1 , an intelligent integrated instrument management and inventory system for the operating room, comprising:

[0034] An intelligent instrument identification and AI information input module, which configures intelligent tags for surgical instruments. The intelligent tags integrate ultra-high frequency RFID and NFC technologies. The intelligent tags include RFID tags and NFC tags. The RFID tags are used for long-distance batch identification and tracking, and the NFC tags are convenient for reading detailed information at close range; when inputting, the characteristics of the surgical instruments equipped with intelligent tags are identified. At the same time, the module incorporates AI image recognition technology, uses the AI image recognition technology to perform an all-round scan of the surgical instruments, and stores the identified information and scanned image data in the system database;

[0035] The system database stores the electronic tags of surgical instruments and their corresponding characteristics;

[0036] An inventory platform is set up in the operating room. Using the built-in AI inventory algorithm, the instruments are inventoried. The AI inventory algorithm uses deep learning image recognition technology to not only identify the types of instruments, but also identify the individual quantities of instruments in complex placement states such as stacking and crossing, and cross-verify with the weight data; the AI inventory algorithm continuously self-optimizes using continuously increasing data samples to improve the inventory accuracy and speed, and adapt to the inventory requirements of different types of surgical instruments;

[0037] A real-time tracking and positioning module installs a positioning base station on the roof of the operating room or at the center of the operating room, and works in coordination with the intelligent tags of the surgical instruments to achieve real-time three-dimensional positioning and tracking;

[0038] AI Path Optimization Module: According to the surgical arrangement, the usage frequency and location information of surgical instruments, it intelligently plans the optimal moving path of surgical instruments from the preparation area to the operating table and dynamically adjusts according to the actual situation during the operation.

[0039] Intelligent Data Analysis and Management Module: It collects and processes the data of other modules, makes analysis, and issues instructions to other modules for feedback. For example, it intelligently analyzes the inventory turnover rate, provides accurate procurement suggestions for the procurement department, and reduces inventory costs.

[0040] AI Prediction Module: Through in-depth analysis of a large amount of surgical instrument data during surgeries, it predicts the types, quantities, and usage sequences of surgical instruments required for different surgical types. It can also analyze historical data to obtain the probabilities of specific instruments required at each stage of a certain type of cardiac surgery, prepare in advance and optimize the instrument placement sequence, predict the instrument wear situation, estimate the remaining life of the instrument based on multiple factors such as the number of uses, disinfection frequency, and stress analysis, and reasonably arrange maintenance, renewal, or scrapping plans.

[0041] Human-Machine Interaction Module: It conducts human-machine interaction through multiple terminals at different locations, including querying surgical instrument information, querying inventory results, and abnormal situation warnings.

[0042] Warning and AI Decision-Assisting Module: When the system detects abnormal situations in surgical instrument data, it gives timely warnings. At the same time, it generates decision suggestions to assist in handling abnormal situations.

[0043] Among them, the characteristics of surgical instruments include name, specification, model, manufacturer, weight, service life, and appearance information. The AI image recognition technology continuously learns the image changes of different instruments in various usage states for subsequent intelligent judgment of the instrument status.

[0044] The inventory platform is built-in with several groups of weight sensors, image recognition cameras, and electromagnetic induction coils. The weight sensors measure the total weight, and combined with the single-instrument weight information in the database, it preliminarily estimates the quantity range. The image recognition cameras take pictures of the instrument placement, and the electromagnetic induction coils detect the electromagnetic characteristics of metal instruments to assist in judging integrity. The AI inventory algorithm is the FasterR-CNN or YOLO (You Only Look Once) algorithm based on convolutional neural networks, and it uses the inventory results of each time for continuous training, continuously self-optimizes, improves the inventory accuracy and speed, and adapts to the inventory requirements of different types of surgical instruments.

[0045] As one of the embodiments, the AI inventory counting algorithm is the YOLO (You Only Look Once) algorithm. The YOLO algorithm regards the object detection task as a regression problem, divides the image into multiple grid cells, and predicts multiple bounding boxes for each grid cell, as well as the confidence and class probabilities of the objects contained in these bounding boxes. In the surgical instrument inventory counting scenario: First, a large amount of image data with labeled surgical instrument types and positions is used to train the YOLO model. During the training process, the model learns the visual features of different instruments in the image, such as shape, color, texture, etc., as well as their relative positions and scale information in the image; during actual inventory counting, the image of the instrument placement captured by the image recognition camera of the high-precision inventory platform is input into the trained YOLO model. The model will analyze each grid cell in the image, predict the possible bounding boxes of surgical instruments, and give the class (such as scalpel, forceps, etc.) and confidence score of the instrument corresponding to each bounding box, indicating the possibility that the instrument of this class actually exists within the bounding box. Then, by integrating and post-processing the prediction results of all grid cells, overlapping or low-confidence bounding boxes are removed, and finally the quantity and position information of different types of surgical instruments in the image are determined. Moreover, as the system continuously runs and more actual instrument image data is collected, these new data can be used to further fine-tune the YOLO model, enabling it to better adapt to changes in different surgical scenarios, instrument placement methods, and lighting conditions, etc., and continuously improving the accuracy and robustness of the inventory counting.

[0046] Among them, the AI path optimization algorithm is the Dijkstra algorithm, the A* algorithm, or the ant colony algorithm; when an emergency surgical need occurs, the transportation route of commonly used surgical instruments is preferentially scheduled to reduce the surgical instrument preparation time. At the same time, the movement trajectory of the surgical instrument is recorded. If the surgical instrument deviates from the preset path or enters a restricted area, the system will automatically alarm.

[0047] As one of the embodiments, the AI path optimization algorithm is the Dijkstra algorithm, which is an algorithm for calculating the single-source shortest path in a weighted graph. In the operating room instrument transportation path planning, each area of the operating room (such as the instrument preparation area, the operating table, the disinfection area, etc.) is regarded as a node in the graph, the passageways between the areas are regarded as edges, and the weight of the edge can be the length of the passageway, the time required for transportation, etc. The Dijkstra algorithm starts from the starting node (such as the instrument storage location), gradually expands the search scope, each time finds the unvisited node closest to the starting node, updates the shortest distance from this node to the starting node, and marks it as visited. Repeat this process until reaching the target node (such as the operating table), so as to obtain the shortest path from the starting node to the target node. It is applicable to the situation of a static operating room layout and fixed transportation costs (such as fixed distance, constant transportation speed). For example, in the instrument preparation stage of a general operation, when the location of the operating table and the instrument storage location are determined, the Dijkstra algorithm can efficiently calculate the shortest transportation path to reduce the instrument transportation time.

[0048] As one of the embodiments, the AI path optimization algorithm is the A* algorithm, which is a heuristic search algorithm that combines the optimality guarantee of the Dijkstra algorithm and heuristic information to improve the search efficiency. In path planning, in addition to considering the actual cost from the starting node to the current node (such as distance, time, etc.), it also considers the cost estimated by a heuristic function from the current node to the target node. This heuristic function is usually an estimated value that underestimates the actual cost. For example, in the operating room path planning, the straight-line distance between two points can be used as the heuristic function. The A* algorithm comprehensively considers these two costs, preferentially searches for those paths that seem more likely to lead to the target node, and thus finds the optimal path faster. When the operating room layout is relatively complex, with multiple alternative paths and obstacles, the A* algorithm can more effectively find the optimal path. For example, in an emergency operation, it is necessary to quickly transport the instruments from one area to the operating table, and there may be obstacles such as personnel and equipment on the way. The A* algorithm can quickly plan the shortest path to avoid obstacles according to the current situation and the target location.

[0049] Among them, the AI prediction algorithm built in the AI prediction module is the linear regression algorithm, the decision tree algorithm, the random forest algorithm or the LSTM algorithm.

[0050] As one of the embodiments, the AI prediction algorithm is a linear regression algorithm, and its principle is as follows: Assume that there is a linear relationship between the dependent variable (such as the wear degree of surgical instruments) and the independent variables (such as the number of uses, the number of disinfections, etc.). The general form of the linear regression model is \(y = \beta_0+\beta_1x_1+\beta_2x_2+\cdots+\beta_nx_n+\epsilon\), where \(y\) is the predicted dependent variable, \(x_i\) is the independent variable, \(\beta_i\) is the regression coefficient, and \(\epsilon\) is the error term. - The regression coefficient \(\beta_i\) is estimated by the least squares method to minimize the sum of the squared errors between the predicted value \(y\) and the actual observed value. In surgical instrument management, for example, if we want to predict the remaining service life of surgical instruments, we can use the usage time, the number of disinfections, etc. as independent variables and the remaining life as the dependent variable to construct a linear regression model. It can be used to predict the remaining service life of surgical instruments. According to the relationship between factors such as the usage time and the number of disinfections of the instruments in historical data and the final scrapping time of the instruments, a linear regression model is established to predict the remaining life of new instruments. It can also be used to estimate the required quantity of a certain type of instrument for different surgical types. Taking the surgical type, the surgical duration, the severity of the patient's condition, etc. as independent variables and the required quantity of instruments as the dependent variable to construct a model for instrument estimation in the surgical preparation stage.

[0051] As one of the embodiments, the AI prediction algorithm is a decision tree algorithm, which is an algorithm based on a tree structure for making decisions. Starting from the root node, it is divided according to different features (such as the type of surgical instruments, the usage frequency, etc.). Each non-leaf node represents a test on a feature, each branch represents an output of the test, and the leaf node represents a category (such as whether the instrument needs maintenance) or a value (such as the estimated next maintenance time of the instrument). For example, when evaluating whether a surgical instrument needs to be replaced, it may first be divided according to the service life of the instrument. If the service life exceeds a certain threshold, it is further divided according to the number of failures of the instrument until the final decision is obtained. When used to classify the maintenance decisions of surgical instruments, according to features such as the usage situation and performance indicators of the instruments, a decision tree is constructed to determine whether the instrument should continue to be used, repaired, or scrapped. When used for surgical risk classification, a decision tree is constructed with features such as the patient's age, underlying diseases, surgical type, etc. to predict the surgical risk level (high, medium, low) so as to make corresponding preparations before the surgery.

[0052] Among them, the abnormal situations of surgical instruments include: the number of surgical instruments does not match, the weight does not match, damage or approaching expiration, and being offline; the generated decision-making suggestions include: when it is found that there is a lack of urgently needed surgical instruments, recommending alternative instruments or rapid deployment plans; if the instrument is slightly damaged, evaluating whether it can continue to be used and giving suggestions on the safe use range to assist medical staff in making correct decisions quickly and ensuring the smooth progress of the operation.

[0053] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent integrated instrument management and inventory system for operating rooms, characterized by: include: Intelligent equipment identification and AI information entry module. Surgical equipment is equipped with intelligent labels. The intelligent labels integrate ultra-high frequency RFID and NFC technologies. The characteristics of surgical equipment equipped with intelligent labels are identified during entry. At the same time, the module has built-in AI image recognition technology, which is used to perform a full-scale scan of surgical equipment, and store the identified information and scanned image data in the system database. A system database that stores the electronic tags of surgical instruments and their corresponding features; The counting platform is set up in the operating room, and the instruments are counted using the built-in AI counting algorithm; Real-time tracking and positioning module, with positioning base stations installed in the operating room, working in conjunction with the smart tags of surgical instruments to achieve real-time three-dimensional positioning tracking; The AI ​​path optimization module intelligently plans the optimal movement path of surgical instruments from the preparation area to the operating table based on the surgical schedule, the frequency of use of surgical instruments, and the location information, and dynamically adjusts the path according to the actual situation during the operation; Intelligent data analysis and management module collects and processes data from other modules and makes analysis; The AI ​​prediction module predicts the type, quantity and order of use of surgical instruments required for different types of surgeries through in-depth analysis of massive surgical instrument data; Human-computer interaction module, which conducts human-computer interaction through multiple terminals in different locations, including querying surgical instrument information, querying inventory results and abnormal situation warning; The early warning and AI decision-making assistance module will issue timely warnings when the system detects abnormalities in surgical instrument data. At the same time, it will generate decision-making suggestions to assist in handling abnormal situations.

2. The intelligent integrated operating room equipment management and inventory system according to claim 1 is characterized by: The characteristics of surgical instruments include name, specification, model, manufacturer, weight, expiration date and appearance information.

3. The intelligent integrated operating room equipment management and inventory system according to claim 1 is characterized by: AI image recognition technology continuously learns the image changes of different devices under various usage conditions, so as to make intelligent judgments on the status of the devices later.

4. The intelligent integrated operating room equipment management and inventory system according to claim 1 is characterized by: The inventory platform has several sets of weight sensors, image recognition cameras and electromagnetic induction coils built in. The weight sensor measures the total weight and combines the weight information of each device in the database to preliminarily estimate the quantity range. The image recognition camera takes pictures of the device placement, and the electromagnetic induction coil detects the electromagnetic properties of metal devices to assist in judging their integrity.

5. The intelligent integrated operating room equipment management and inventory system according to claim 4 is characterized by: The AI ​​counting algorithm is the FasterR-CNN or YOLO algorithm based on convolutional neural networks. It uses the results of each counting to conduct continuous training and continuous self-optimization to improve the counting accuracy and speed and adapt to the counting needs of different types of surgical instruments.

6. The intelligent integrated operating room equipment management and inventory system according to claim 1 is characterized by: The AI ​​path optimization algorithm is Dijkstra algorithm, A* algorithm or ant colony algorithm.

7. The intelligent integrated operating room equipment management and inventory system according to claim 1 is characterized by: When an emergency surgery is needed, the transportation routes of commonly used surgical instruments are prioritized to reduce the preparation time of surgical instruments. At the same time, the movement trajectory of surgical instruments is recorded. If the surgical instruments deviate from the preset path or enter the restricted area, the system will automatically alarm.

8. The intelligent integrated operating room equipment management and inventory system according to claim 1 is characterized by: The AI ​​prediction algorithm built into the AI ​​prediction module is the linear regression algorithm, decision tree algorithm, random forest algorithm, or LSTM algorithm.

9. The intelligent integrated operating room equipment management and inventory system according to claim 1 is characterized by: Abnormal situations of surgical instruments include: inconsistent quantity, inconsistent weight, damaged or about to expire, and not online.

10. The intelligent integrated operating room equipment management and inventory system according to claim 9, characterized in that: The generated decision recommendations include: when it is found that an instrument urgently needed for surgery is missing, alternative instruments or quick deployment plans are recommended; if the instrument is slightly damaged, an assessment is made as to whether it can continue to be used and recommendations on the safe range of use are given, to assist medical staff in making correct decisions quickly and ensure the smooth progress of the operation.

Images