CN111028933B - Hospital consumable inventory management system and method based on behavior recognition - Google Patents
Hospital consumable inventory management system and method based on behavior recognition Download PDFInfo
- Publication number
- CN111028933B CN111028933B CN201911337107.4A CN201911337107A CN111028933B CN 111028933 B CN111028933 B CN 111028933B CN 201911337107 A CN201911337107 A CN 201911337107A CN 111028933 B CN111028933 B CN 111028933B
- Authority
- CN
- China
- Prior art keywords
- consumable
- elbow
- wrist
- stock
- consumables
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/28—Recognition of hand or arm movements, e.g. recognition of deaf sign language
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W90/00—Enabling technologies or technologies with a potential or indirect contribution to greenhouse gas [GHG] emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Business, Economics & Management (AREA)
- Mathematical Physics (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- General Business, Economics & Management (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biophysics (AREA)
- Human Computer Interaction (AREA)
- Evolutionary Biology (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Epidemiology (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention belongs to the technical field related to hospital inventory management, and discloses a hospital consumable inventory management system and method based on behavior recognition, wherein the method comprises the following steps: (1) collecting skeleton point data of a nurse, further obtaining the position range of the disposable consumables and the wrist, elbow and shoulder included angles, and further identifying the behavior of the nurse in real time based on the skeleton point data detected in real time; (2) after the behavior of the nurse is judged to be that consumables are thrown away, pictures of the consumables thrown away by the nurse are input into the convolutional neural network to identify the consumables; the convolutional neural network takes pictures of different consumable items as training data; (3) performing subtraction 1 operation on the consumable stock of the corresponding type according to the identified type of the consumable, judging whether the current consumable stock is lower than the safe stock, and if so, sending an order to the main warehouse to replenish the corresponding consumable; otherwise, no operation is performed. The invention reduces the cost and has stronger applicability.
Description
Technical Field
The invention belongs to the technical field related to hospital inventory management, and particularly relates to a hospital consumable inventory management system and method based on behavior recognition.
Background
With the continuous development of medical science and technology, medical consumables as an indispensable important component of medical behaviors have been widely applied to the whole process of clinical medical diagnosis, treatment and nursing. Meanwhile, the logistics management level of the medical consumables is relatively lagged, a series of resource waste problems exist in the process that the medical consumables are circulated in the logistics link of the hospital due to a rough management mode, and meanwhile, the work of departments is seriously influenced due to untimely consumable data updating, so that the running state of the hospital is seriously influenced, the medical accident risk is increased, and the medical expense cost is increased. Therefore, attention is paid to medical consumable logistics management, and the information technology and logistics management information system are used for performing fine management on the logistics and the inventory of the hospital, so that the intensive research content for improving the scientific management level of the hospital is provided.
The existing hospital consumable inventory management has the following defects: nurses need to spend a great deal of time and energy to manage the inventory every day, which wastes time and labor, thus causing large workload of the nurses and low working efficiency; the straightforward consumable management mode can cause resource waste by performing inventory replenishment and inventory checking according to experience.
Disclosure of Invention
Aiming at the defects or improvement requirements of the prior art, the invention provides a hospital consumable inventory management system and method based on behavior recognition. According to the method, firstly, skeleton point data of nurses are collected to identify the behavior of the nurses throwing consumables, after the behavior of the nurses is judged to be the behavior of throwing consumables, the consumables thrown by the nurses are identified and classified based on the convolutional neural network, and then corresponding consumable stock minus 1 operation is carried out according to the identified consumable categories, so that the function of automatic counting is achieved, and a fixed interval period and a fixed quantity system are used as ordering strategies. When the stock of consumptive material reduces below the safety stock, the system can send the order to main storehouse automatically, supplements the stock to the biggest stock, avoids out of stock phenomenon to take place, has so improved nurse's work efficiency and enthusiasm, has reduced the wasting of resources of consumptive material stock replenishment and inventory in-process, has reduced the medical expenses cost, provides convenience for the normal operation of hospital, and makes consumptive material management more accurate and convenient.
To achieve the above object, according to one aspect of the present invention, there is provided a hospital consumable inventory management method based on behavior recognition, the method comprising the steps of:
(1) collecting skeleton point data of a nurse, performing least square spherical fitting on the three-dimensional coordinate data of the wrist to obtain the position range of the disposable consumables, and calculating the wrist shoulder included angle of the disposable consumables according to the vector formed by the elbow wrist joint and the vector formed by the elbow shoulder joint; then, detecting skeleton points of nurses in real time, and identifying the behaviors of the nurses in real time based on the real-time detected skeleton point data, the position range of the disposable consumables and the wrist, elbow and shoulder included angles;
(2) after the behavior of the nurse is judged to be disposable consumables, inputting pictures of the disposable consumables thrown by the nurse into a convolutional neural network, and identifying the consumables by the convolutional neural network according to the received pictures; the convolutional neural network takes pictures of different consumable items as training data;
(3) performing subtraction 1 operation on the consumable stock of the corresponding type according to the identified type of the consumable, judging whether the current consumable stock is lower than the safe stock, and if so, sending an order to the main warehouse to replenish the corresponding consumable; otherwise, no operation is performed.
Furthermore, the bone point data comprises three-dimensional coordinate data of the right shoulder, the right elbow and the right wrist, and included angle data between a vector formed by joints of the right shoulder and the right elbow and a vector formed by joints of the right elbow and the right wrist.
Furthermore, the included angle of the elbow and shoulder of the disposable consumables is larger than or equal to 90 degrees.
Further, the spherical equation obtained by the least square method spherical fitting is (x-0.249126) ^2+ (y-0.200749) ^2+ (z-1.294224) ^2 ^ 0.004822, (x, y, z) are wrist joint coordinates.
Furthermore, the included angle of the elbow and the shoulder of the disposable consumables is [145.9920, 176.3612] °.
Further, in the step (3), if the consumable stock quantity is reduced to be below the safe stock quantity, a single order is issued to the large warehouse, the order quantity is the amount of replenishing the existing stock to a preset level, and otherwise, whether an order requirement is issued is considered after one month.
According to another aspect of the invention, a hospital consumable inventory management system based on behavior recognition is provided, the system comprises a behavior recognition subsystem, a consumable type recognition subsystem and a management subsystem, wherein the consumable type recognition subsystem is respectively connected with the behavior recognition subsystem and the management subsystem;
the behavior recognition subsystem is used for collecting skeleton point data of nurses, performing least square spherical fitting on the three-dimensional coordinate data of the wrists to obtain the position range of the disposable consumables, and calculating the elbow shoulder included angle of the disposable consumables according to the vector formed by elbow wrist joints and the vector formed by elbow shoulder joints; then, detecting skeleton points of nurses in real time, and identifying the behaviors of the nurses in real time based on the real-time detected skeleton point data, the position range of the disposable consumables and the wrist, elbow and shoulder included angles;
the consumable type identification subsystem is used for inputting a picture of a consumable thrown by a nurse into a convolutional neural network after the consumable is thrown away by the behavior of the nurse to be judged, and the convolutional neural network identifies the consumable according to the received picture;
the management subsystem is used for subtracting 1 from the corresponding type of consumable stock according to the identified type of the consumable, judging whether the current consumable stock is lower than the safe stock, and if so, sending an order to the main warehouse to replenish the corresponding consumable; otherwise, no operation is performed.
Generally, compared with the prior art, the hospital consumable inventory management system and method based on behavior recognition provided by the invention have the following beneficial effects:
1. according to the method, the skeleton point data of a nurse are collected to identify the behavior of throwing the consumables by the nurse, after the behavior of the nurse is judged to be the behavior of throwing the consumables, the consumables thrown by the nurse are identified and classified based on the convolutional neural network, and then corresponding consumable inventory minus 1 operation is carried out according to the identified consumable categories, so that the function of automatic counting is achieved, the cost is reduced, and the applicability is high.
2. When the stock of consumptive material reduces below the safety stock, the system can send the order to main storehouse automatically, supplements the stock to the biggest stock, avoids out of stock phenomenon to take place, has so improved nurse's work efficiency and enthusiasm, has reduced the wasting of resources of consumptive material stock replenishment and inventory in-process, has reduced the medical expenses cost, provides convenience for the normal operation of hospital, and makes consumptive material management more accurate and convenient.
3. If the consumable stock quantity is reduced to be below the safe stock quantity, an order is sent to the large warehouse, the order quantity is that the existing stock is supplemented to a preset level, otherwise, whether the order requirement is sent or not is considered after one month, and therefore the reliability of stock management is improved under the condition of ensuring efficiency and low cost.
4. The hospital consumable inventory management method is simple, easy to implement and high in applicability.
Drawings
FIG. 1 is a schematic flow chart of a hospital consumable inventory management method based on behavior recognition according to the present invention;
fig. 2 is a schematic diagram illustrating a control flow of the inventory related to the hospital consumable inventory management method based on behavior recognition in fig. 1.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
Referring to fig. 1 and 2, according to the hospital consumable inventory management system and method based on behavior recognition provided by the present invention, firstly, the behavior of the consumables of the nurse is recognized, and whether the action of the nurse is "throw" or not is determined; then, identify the type of the consumable in the hand of the nurse who judges that the action is "throw away" to determine which type of consumable belongs to, and subtract 1 operation to the corresponding type of consumable stock, finally realize the function of automatic counting. On the basis, the inventory can be checked manually at regular intervals, so that the problem of system technology error is avoided, and the accuracy of inventory management is further improved.
The hospital consumable inventory management method based on behavior recognition mainly comprises the following steps:
step one, identifying the behavior of a nurse.
Collecting skeleton point data of a nurse, performing least square spherical fitting on the three-dimensional coordinate data of the wrist to obtain the position range of the disposable consumables, and calculating the wrist shoulder included angle of the disposable consumables according to the vector formed by the elbow wrist joint and the vector formed by the elbow shoulder joint; then, the skeleton points of the nurses are detected in real time, and the behaviors of the nurses are identified in real time based on the skeleton point data detected in real time, the position range of the disposable consumables and the wrist, elbow and shoulder included angles.
Specifically, firstly, a depth camera Kinect is adopted to track the position of a human skeleton, and VS2013 programming is utilized to obtain skeleton data of the right shoulder, the right wrist and the right elbow of the human body; then, carrying out actual simulation on the way that the nurses throw away consumables, and obtaining 50 groups of three-dimensional coordinate data of the right shoulder, the right elbow and the right wrist and included angle data between a vector formed by the right shoulder joint and the right elbow joint and a vector formed by the right elbow joint and the right wrist joint by using a program as a data source for behavior recognition; then, performing least square spherical fitting on the three-dimensional coordinate data of the wrist to obtain the position range of the throwing motion, and calculating the included angle range of the throwing motion according to the three-dimensional coordinate data of the wrist joint, the elbow joint and the shoulder joint; and then, detecting new bone point data in real time through the Kinect sensor, detecting and identifying each data, and simultaneously, considering the action of the wrist coordinate data in the throwing action position range and the wrist shoulder included angle in the throwing action range as throwing.
In the process of behavior identification, a large amount of bone point data is needed to be used as a support, and human body example simulation is used as a data source in the embodiment. Wherein, under the condition that the nurse uses the medical trolley (namely, the height of the small basket for recycling consumables is specific), three skeleton points can be selected as characteristic points to represent the throwing action, and the three skeleton points are respectively a right wrist joint, a right elbow joint and a right shoulder joint. In addition, in order to more accurately identify the behavior of throwing away consumables, an included angle between a vector formed by the right wrist joint and the right elbow joint and a vector formed by the right elbow joint and the right shoulder joint is calculated to be used as a fourth characteristic point for representing the throwing-away action.
In the embodiment, hardware equipment used for collecting the bone point data is a Kinect camera, and software equipment is Opencv, Visual Studio2013 and Kinect for Windows SDK 2.0; the VS2013 programming is utilized to realize a program for acquiring skeleton data of the right shoulder, the right wrist and the right elbow of the human body; the mode of discarding consumables by nurses is actually simulated, and 50 groups of three-dimensional coordinate data of the right shoulder, the right elbow and the right wrist and included angle data between a vector formed by the right shoulder and the right elbow joint and a vector formed by the right elbow joint and the right wrist joint are obtained by a program and serve as data sources of behavior recognition.
100 sets of data are collected for training, wherein 50 sets of data are data of 'throwing' action, and 50 sets of data are not data of 'throwing' action (including actions of operating on a workbench, walking and the like). Each group of data comprises four values of wrist joint coordinates (x, y, z) and elbow included angle, and as the wrist elbow shoulder included angle is larger than 90 degrees when the user throws the wrist, elbow and elbow movement, the obtained bone point data of 50 groups are firstly subjected to data preprocessing, the wrist elbow shoulder included angle is eliminated by less than 90 degrees, and the rest data are used as training samples.
The least squares fitting procedure is as follows: matlab reads sample data and performs least square spherical fitting on the wrist data, wherein the fitted spherical equation is (x-0.249126) ^2+ (y-0.200749) ^2+ (z-1.294224) ^2 ^ 0.004822; the elbow-shoulder angle is calculated based on the vector formed by the elbow-wrist joint and the vector formed by the elbow-shoulder joint, and the range of the angle obtained by the sample is [145.9920, 176.3612] °. Based on the two ranges, the data of 5 groups of 'throw' actions and 5 groups of 'non-throw' actions which are acquired additionally are checked, and the checking accuracy is over 80 percent.
And step two, identifying the type of the consumable.
Taking pictures of different kinds of consumables as training data, extracting image features from the pictures through a convolutional neural network, learning, and continuously modifying the weight of the convolutional neural network until the recognition accuracy of the convolutional neural network reaches a set value, thereby obtaining a convolutional neural network model; and after the behavior of the nurse is judged to be disposable consumables, inputting pictures of the disposable consumables thrown by the nurse into the convolutional neural network, and identifying the consumables by the convolutional neural network according to the received pictures.
In this embodiment, the identification of consumable type is divided into three stages: data acquisition, network training and consumable identification, wherein in the data acquisition stage, pictures of a large number of consumables of different types need to be shot and acquired to serve as training data; in the network training stage, image features are extracted and learned through a convolutional neural network algorithm, and the weight of the network is continuously modified until the recognition accuracy reaches the requirement; and in the consumable identification stage, the picture to be identified is input into the trained neural network, so that the consumables are classified.
When data is obtained, the training data obtaining comprises the steps of respectively shooting 20 pictures at different angles for three consumables, namely an infusion bag, an infusion apparatus and an injector, and ensuring that the shooting environment is sufficient in light and the background is the same; 20 background images of no consumables were taken to simulate the case of no consumables. With these 4 types of consumables, a total of 80 pictures are used as training data for the subsequent training of the convolutional neural network.
When identifying consumables based on a convolutional neural network, firstly, preprocessing pictures, namely adjusting the obtained infusion bag, infusion apparatus, injector and non-consumable pictures into pictures with the size of 200x200, reading Matlab, and performing gray image processing to obtain a two-dimensional pixel point matrix; then convolutional neural network training is carried out, and the convolutional neural network autonomous machine learning training corresponding to 80 pictures is completed through calculation processing of an input layer, a convolutional layer, a pooling layer, a full connection layer and a Softmax layer of the convolutional neural network, so that 1 x 4 characteristic vectors are finally obtained; and then, the consumable identification result and analysis are carried out, 5 groups (20) of pictures are selected for detection and identification, 19 pictures are correctly identified, the accuracy is up to 95%, and the more the training pictures are, the better the effect is.
And step three, implementing the inventory management strategy.
Performing subtraction 1 operation on the consumable stock of the corresponding type according to the identified type of the consumable, judging whether the current consumable stock is lower than the safe stock, and if so, sending an order to the main warehouse to replenish the corresponding consumable; otherwise, no operation is performed.
Specifically, the inventory management comprises an order strategy adopting a fixed interval period and a fixed quantity system, and a replenishment strategy of an inventory automatic counting function and safety inventory early warning is added on the basis.
In this embodiment, the inventory management phase includes an order policy of a fixed interval period and a fixed quantity system, and a policy of safety inventory early warning; in order to ensure accuracy, inventory can be manually checked every month, the number of system inventory is corrected, if the inventory quantity is reduced below an order point RP, an order is sent to a large warehouse, the order quantity is that the existing inventory is supplemented to a level S, otherwise, whether an order request is sent or not is considered after one month.
Behavior recognition and consumable type detection realize carrying out the automatic counting to the consumable type, and application system detects daily stock promptly, and when the fixed quantity of certain consumable dropped to safe stock, send the early warning and generate the order, supplemented level S with the stock. The ward consumable inventory management belongs to the situation that the demand changes and the lead period is a fixed constant, firstly, the change situation of the demand is supposed to accord with the state distribution, and the lead period is a fixed numerical value, so that the mean value and the standard deviation of the demand distribution in the lead period can be directly solved, or the demand situation in the past lead period can be used as the basis through direct expectation prediction, so as to determine the expected mean value and the standard deviation of the demand, and then:
wherein LT is lead period/days;
the daily average consumption is; z is a safety factor for demanding certain customer service level (the service level refers to the satisfaction degree of the demand condition); σ is the standard deviation of the daily average consumption.
The conditions of the number and date of the needles of the disposable precise filtering infusion set taken by the nurses in the ward in a certain period of time from the large storehouse are shown in the following table 1:
TABLE 1 use of a consumable
Assuming the lead time is one day and the satisfaction is 0.95, namely the safety factor Z of the corresponding customer service level requirement is 1.65, and the calculated safety stock is 85.2. When the stock of the disposable precise filtering infusion set with the needle is lower than 85.2 safe stocks, the system sends out early warning and generates an order; otherwise, waiting for the order cycle to manually count the inventory and modify the inventory system data. If the stock is lower than the order point, sending an order to the large warehouse to supplement the stock; if the stock is larger than or equal to the order point, the order is not sent, and the manual checking of the stock in the next order period is waited.
The invention also provides a hospital consumable inventory management system based on behavior recognition, which comprises the following components:
the behavior recognition subsystem is used for collecting skeleton point data of nurses, performing least square spherical fitting on the three-dimensional coordinate data of wrists to obtain the position range of the disposable consumables, and calculating the wrist, elbow and shoulder included angle of the disposable consumables according to the vector formed by the elbow and wrist joints and the vector formed by the elbow and shoulder joints; and then, detecting the skeletal points of the nurses in real time, and identifying the behaviors of the nurses in real time based on the skeletal point data detected in real time, the position range of the discarded consumables and the elbow shoulder included angle. In this embodiment, the behavior recognition subsystem includes a depth camera Kinect and a first microprocessor, the depth camera Kinect is used for collecting bone point data, and the first microprocessor is used for processing data, fitting and behavior recognition.
The consumable type identification subsystem is used for taking pictures of different types of consumables as training data, extracting image features from the pictures through a convolutional neural network, learning, and continuously modifying the weight of the convolutional neural network until the identification accuracy of the convolutional neural network reaches a set value, so that a convolutional neural network model is obtained; and after the disposable consumables of the behavior of the nurse are judged, inputting the pictures of the disposable consumables of the nurse into the convolutional neural network, and identifying the consumables by the convolutional neural network according to the received pictures.
The management subsystem is used for subtracting 1 from the corresponding type of consumable stock according to the identified type of the consumable, judging whether the current consumable stock is lower than the safe stock, and if so, sending an order to the main warehouse to replenish the corresponding consumable; otherwise, no operation is performed.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (7)
1. A hospital consumable inventory management method based on behavior recognition is characterized by comprising the following steps:
(1) collecting skeleton point data of a nurse, performing least square spherical fitting on the three-dimensional coordinate data of the wrist to obtain the position range of the disposable consumables, and calculating the wrist shoulder included angle of the disposable consumables according to the vector formed by the elbow wrist joint and the vector formed by the elbow shoulder joint; then, detecting skeleton points of nurses in real time, and identifying the behaviors of the nurses in real time based on the real-time detected skeleton point data, the position range of the disposable consumables and the wrist, elbow and shoulder included angles;
(2) after the behavior of the nurse is judged to be disposable consumables, inputting pictures of the disposable consumables thrown by the nurse into a convolutional neural network, and identifying the consumables by the convolutional neural network according to the received pictures; the convolutional neural network takes pictures of different consumable items as training data;
(3) performing subtraction 1 operation on the consumable stock of the corresponding type according to the identified type of the consumable, judging whether the current consumable stock is lower than the safe stock, and if so, sending an order to the main warehouse to replenish the corresponding consumable; otherwise, no operation is performed.
2. The hospital consumable inventory management method based on behavioral recognition according to claim 1, characterized in that: the bone point data comprises three-dimensional coordinate data of the right shoulder, the right elbow and the right wrist, and included angle data between a vector formed by joints of the right shoulder and the right elbow and a vector formed by joints of the right elbow and the right wrist.
3. The hospital consumable inventory management method based on behavioral recognition according to claim 1, characterized in that: the included angle of the elbow and shoulder of the wrist which throws away the consumables is more than or equal to 90 degrees.
4. The hospital consumable inventory management method based on behavior recognition according to claim 1, characterized in that: the spherical equation obtained by the least square method spherical fitting is (x-0.249126) ^2+ (y-0.200749) ^2+ (z-1.294224) ^2 ^ 0.004822, (x, y, z) are wrist joint coordinates.
5. The hospital consumable inventory management method based on behavior recognition according to claim 1, characterized in that: the included angle between the wrist, elbow and shoulder of the disposable consumables is [145.9920, 176.3612] °.
6. The hospital consumable inventory management method based on behavioral recognition according to any one of claims 1 to 5, characterized in that: and (3) if the consumable stock quantity is reduced to be below the safe stock quantity, sending a primary order to the large warehouse, wherein the order quantity is the amount of replenishing the existing stock to a preset level, and otherwise, considering whether an order demand is sent after one month.
7. The utility model provides a hospital's consumptive material inventory management system based on action discernment which characterized in that: the hospital consumable inventory management system comprises a behavior identification subsystem, a consumable type identification subsystem and a management subsystem, wherein the consumable type identification subsystem is respectively connected with the behavior identification subsystem and the management subsystem;
the behavior recognition subsystem is used for collecting skeletal point data of nurses, performing least square spherical fitting on the three-dimensional coordinate data of wrists to obtain the position range of the disposable consumables, and calculating the wrist, elbow and shoulder included angle of the disposable consumables according to the vector formed by the elbow and wrist joints and the vector formed by the elbow and shoulder joints; then, detecting skeleton points of nurses in real time, and identifying the behaviors of the nurses in real time based on the real-time detected skeleton point data, the position range of the disposable consumables and the wrist, elbow and shoulder included angles;
the consumable type identification subsystem is used for inputting a picture of a consumable thrown by a nurse into a convolutional neural network after the consumable is thrown away by the behavior of the nurse to be judged, and the convolutional neural network identifies the consumable according to the received picture;
the management subsystem is used for subtracting 1 from the corresponding type of consumable stock according to the identified type of the consumable, judging whether the current consumable stock is lower than the safe stock, and if so, sending an order to the warehouse to supplement the corresponding consumable; otherwise, no operation is performed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911337107.4A CN111028933B (en) | 2019-12-23 | 2019-12-23 | Hospital consumable inventory management system and method based on behavior recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911337107.4A CN111028933B (en) | 2019-12-23 | 2019-12-23 | Hospital consumable inventory management system and method based on behavior recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111028933A CN111028933A (en) | 2020-04-17 |
| CN111028933B true CN111028933B (en) | 2022-07-12 |
Family
ID=70211595
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911337107.4A Active CN111028933B (en) | 2019-12-23 | 2019-12-23 | Hospital consumable inventory management system and method based on behavior recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111028933B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113793103A (en) * | 2021-09-18 | 2021-12-14 | 中广核风电有限公司 | Method and device for determining safety stock |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108038420A (en) * | 2017-11-21 | 2018-05-15 | 华中科技大学 | A kind of Human bodys' response method based on deep video |
| WO2019060767A1 (en) * | 2017-09-21 | 2019-03-28 | Fellow, Inc. | Intelligent inventory management and related systems and methods |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040122709A1 (en) * | 2002-12-18 | 2004-06-24 | Avinash Gopal B. | Medical procedure prioritization system and method utilizing integrated knowledge base |
| US20070118399A1 (en) * | 2005-11-22 | 2007-05-24 | Avinash Gopal B | System and method for integrated learning and understanding of healthcare informatics |
| US20160019479A1 (en) * | 2014-07-18 | 2016-01-21 | Rebecca S. Busch | Interactive and Iterative Behavioral Model, System, and Method for Detecting Fraud, Waste, and Abuse |
| US20180308569A1 (en) * | 2017-04-25 | 2018-10-25 | S Eric Luellen | System or method for engaging patients, coordinating care, pharmacovigilance, analysis or maximizing safety or clinical outcomes |
-
2019
- 2019-12-23 CN CN201911337107.4A patent/CN111028933B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019060767A1 (en) * | 2017-09-21 | 2019-03-28 | Fellow, Inc. | Intelligent inventory management and related systems and methods |
| CN108038420A (en) * | 2017-11-21 | 2018-05-15 | 华中科技大学 | A kind of Human bodys' response method based on deep video |
Non-Patent Citations (3)
| Title |
|---|
| 《Accurency of kinect"s skeleton tracking for upper body rehabilitation application》;Amir Mobini et al.;《Disabil Rehabil Assist Technol》;20140709;344-352 * |
| 《移动医疗耗材管理系统的开发》;高正 等;《中国医疗器械杂志》;20180130;38-40 * |
| 基于多支点骨骼模型的实时行为识别方法;王军等;《华中科技大学学报(自然科学版)》;20140110;155-159 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111028933A (en) | 2020-04-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109035187B (en) | Medical image labeling method and device | |
| CN109692821A (en) | Sorting system | |
| KR20190088375A (en) | Surgical image data learning system | |
| CN109598229A (en) | Monitoring system and its method based on action recognition | |
| DE112017007394B4 (en) | Information processing device, gripping system, distribution system, program and information processing method | |
| CN104156068B (en) | Virtual maintenance interaction operation method based on virtual hand interaction feature layer model | |
| CN109886143A (en) | Multi-tag disaggregated model training method and equipment | |
| He et al. | Development of distributed control system for vision-based myoelectric prosthetic hand | |
| CN112989947A (en) | Method and device for estimating three-dimensional coordinates of human body key points | |
| CN111028933B (en) | Hospital consumable inventory management system and method based on behavior recognition | |
| CN113435236A (en) | Home old man posture detection method, system, storage medium, equipment and application | |
| CN109685765A (en) | A kind of X-ray pneumonia prediction of result device based on convolutional neural networks | |
| CN109949280A (en) | Image processing method, device, equipment storage medium and growth and development assessment system | |
| CN113936335B (en) | Intelligent sitting posture reminding method and device | |
| CN113688825A (en) | AI intelligent garbage recognition and classification system and method | |
| CN110276276A (en) | The determination method and system of examinee's face direction of visual lines in a kind of Driving Test | |
| CN103049747B (en) | The human body image utilizing the colour of skin knows method for distinguishing again | |
| CN114240874A (en) | Bone age assessment method and device based on deep convolutional neural network and feature fusion and computer readable storage medium | |
| CN109997199A (en) | Tuberculosis inspection method based on deep learning | |
| CN112288793A (en) | Livestock individual backfat detection method and device, electronic equipment and storage medium | |
| US20210158029A1 (en) | Action-estimating device | |
| WO2019058963A1 (en) | Medical image processing device, medical image processing method, and processing program used for same | |
| Chu et al. | 3D human body reconstruction for worker ergonomic posture analysis with monocular video camera | |
| CN116246350A (en) | Motion monitoring method, device, equipment and storage medium based on motion capture | |
| CN112801118B (en) | Pork pig marketing benefit evaluation system and method based on artificial intelligence and big data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |