CN118304514A - Infusion monitoring system and method based on computer vision recognition - Google Patents
Infusion monitoring system and method based on computer vision recognition Download PDFInfo
- Publication number
- CN118304514A CN118304514A CN202410454364.0A CN202410454364A CN118304514A CN 118304514 A CN118304514 A CN 118304514A CN 202410454364 A CN202410454364 A CN 202410454364A CN 118304514 A CN118304514 A CN 118304514A
- Authority
- CN
- China
- Prior art keywords
- infusion
- time
- obtaining
- nurse
- target
- 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.)
- Pending
Links
- 238000001802 infusion Methods 0.000 title claims abstract description 276
- 238000000034 method Methods 0.000 title claims abstract description 77
- 238000012544 monitoring process Methods 0.000 title claims abstract description 27
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 73
- 238000012545 processing Methods 0.000 claims abstract description 5
- 239000003814 drug Substances 0.000 claims description 45
- 230000000007 visual effect Effects 0.000 claims description 26
- 238000003780 insertion Methods 0.000 claims description 9
- 230000037431 insertion Effects 0.000 claims description 9
- 239000007788 liquid Substances 0.000 claims description 8
- 230000001172 regenerating effect Effects 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 3
- 238000007689 inspection Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010253 intravenous injection Methods 0.000 description 1
- 210000003462 vein Anatomy 0.000 description 1
Landscapes
- Infusion, Injection, And Reservoir Apparatuses (AREA)
Abstract
The application discloses an infusion monitoring system and method based on computer vision recognition, which relates to the technical field of computer vision recognition, wherein the system comprises a video processing module: the method comprises the steps of obtaining video pictures of drip chambers in an infusion hose to obtain interval time for dripping water in the drip chambers; infusion analysis module: for obtaining an infusion rate; and a time determining module: a working time node for generating a nurse; a nurse distribution module: for obtaining a response time of the nurse, determining a target nurse based on the response time; an operation reminding module: the infusion system is used for acquiring the current infusion process of the patient, obtaining infusion operation required by the patient after the completion of the infusion of the round, generating operation reminding based on the infusion operation, sending the operation reminding to a target nurse, and reminding the nurse according to response time and working time nodes. The application has the effect of reducing the working intensity of nurses in the process of patient transfusion.
Description
Technical Field
The application relates to the technical field of computer vision recognition, in particular to an infusion monitoring system and method based on the computer vision recognition.
Background
Infusion is a common treatment mode for some patients in hospitals, and the medical liquid is directly conveyed into the body of the patient in an intravenous injection mode, so that the patient can be better helped to absorb the medical force.
Patients often use a plurality of bottles of medicaments simultaneously in the process of transfusion, so that the patients need to pay attention to the quantity of the medicaments in the medicament bottles at any time during transfusion, and the patients need to call a nurse in time to replace the medicaments when the doses are insufficient so as to prevent dangerous situations such as blood backflow or air input into veins.
In the prior art, the dosage of the medicine bottle in the transfusion process of a patient is mainly observed by the patient or other people. However, in the actual infusion process, the patient falls asleep or is addicted to other things in the infusion process due to the long time of the infusion process, so that the patient forgets to observe the situation. In addition, some infusion sites can be specially arranged for nurses to carry out inspection, when medicines in the medicine bottles are completely infused, the nurses are required to be arranged for uninterrupted inspection, the attention is required to be relatively concentrated, and once the nurses have distraction or the workload of the nurses is large, accidents of patients in the infusion process can be caused with high probability.
Disclosure of Invention
The invention aims to provide an infusion monitoring system and method based on computer vision recognition, which are used for solving the problems in the background technology.
In a first aspect, the present application provides an infusion monitoring system based on computer vision recognition, the system comprising:
And a video processing module: the method comprises the steps of obtaining a video picture of a drip cup in an infusion hose at the initial stage of infusion, and obtaining the interval time of water drop in the drip cup based on the video picture;
Infusion analysis module: the infusion device comprises a device, a control unit and a control unit, wherein the device is used for acquiring parameter information of an infusion hose, extracting inner wall diameter data of the infusion hose based on the parameter information, acquiring water drop capacity data of water drops based on the inner wall diameter data, and acquiring an infusion rate by combining the water drop capacity data and the interval time;
And a time determining module: the infusion control method comprises the steps of obtaining medicine information when a patient infuses, obtaining the infusion quantity of the patient based on the medicine information, obtaining infusion time according to the infusion quantity and the infusion rate, and generating a working time node of a nurse according to the infusion time;
A nurse distribution module: the system comprises a plurality of nurses, a plurality of infusion time node, a plurality of nurse response time and a target nurse, wherein the plurality of nurses are used for acquiring current working states of the nurses, obtaining busy values of the plurality of nurses when the working time node is combined with the infusion time and the working states, obtaining response time of the nurses based on the busy values, and determining the target nurse based on the response time;
An operation reminding module: the infusion reminding device is used for obtaining the current infusion process of the patient, obtaining infusion operation required by the patient after the completion of the infusion of the round according to the current infusion process, generating operation reminding based on the infusion operation, sending the operation reminding to a target nurse, and reminding the nurse according to the response time and the working time node.
Through adopting above-mentioned technical scheme, when the infusion begins, obtain the video picture of drip chamber in the infusion hose, and carry out analysis and obtain the interval time that the water droplet falls, obtain the water droplet capacity data of every water droplet according to the parameter information of infusion software again, further obtain infusion rate, obtain patient's infusion volume according to the medicine information when patient infuses, obtain infusion time according to infusion volume and infusion rate, confirm nurse's operating time node according to infusion time again, confirm target nurse according to nurse's current operating condition, send target nurse with operating time node, remind the nurse according to target nurse's response time. The working intensity of nurses in the transfusion process of patients is reduced.
Preferably, a step of obtaining a video picture of a drip cup in an infusion hose and obtaining the interval time of dripping in the drip cup based on the video picture, specifically comprises the following steps:
Acquiring a video picture of a drip cup in an infusion hose, analyzing the video picture, and extracting a drip cup frame picture in the video picture;
Extracting drip cup features in the drip cup frame picture based on the drip cup frame picture, and obtaining a picture visual angle according to the drip cup features;
judging whether the visual angle of the picture is correct or not, and if the visual angle of the picture is not correct, correcting the visual angle of the picture to obtain a target kettle-dropping frame picture with the correct visual angle;
and obtaining a target video picture with a correct visual angle based on the target drip cup frame picture, and obtaining the interval time of dripping in the drip cup according to the target video picture.
By adopting the technical scheme, the video picture of the drip cup is analyzed to obtain the drip cup frame picture in the video picture, the drip cup characteristics are obtained according to the drip cup frame picture, the picture visual angle is obtained according to the drip cup characteristics, whether the picture visual angle is correct or not is judged, if not, the picture visual angle is corrected to obtain the target drip cup frame picture, the target video picture with correct visual angle is obtained according to the target drip cup frame picture, and the interval time for dripping water is obtained according to the target video picture. The accuracy of analyzing the dropping of the water drops in the drip cup is improved.
Preferably, parameter information of the infusion hose is obtained, inner wall diameter data of the infusion hose is obtained based on the parameter information, water drop capacity data of water drops is obtained based on the inner wall diameter data, and the infusion rate is obtained by combining the water drop capacity data and the interval time, specifically:
Acquiring parameter information of the infusion hose, and extracting to obtain inner wall diameter data of the infusion hose based on the parameter information;
obtaining an interfacial tension value at a liquid inlet in the drip cup based on the inner wall diameter data, obtaining the size of water drops falling in the drip cup based on the interfacial tension, and obtaining water drop capacity data of the water drops based on the size of the water drops;
Obtaining the drop frequency of drops based on the interval time, and obtaining the number of drops in unit time based on the drop frequency;
And obtaining the infusion rate based on the water drop quantity and the water drop capacity data in unit time.
By adopting the technical scheme, the inner wall diameter data of the infusion hose is obtained according to the parameter information of the infusion hose, the interfacial tension value of the liquid in the drip cup when the liquid enters the throat is obtained according to the inner wall diameter data, the water drop size is obtained according to the interfacial arrangement data, the water drop capacity data is obtained according to the water drop size, the water drop frequency is obtained according to the interval time, the water drop quantity in unit time is further obtained, and the infusion rate is obtained by combining the water drop quantity and the water drop capacity data. The accuracy of monitoring the infusion rate during infusion is improved.
Preferably, the step of acquiring medicine information when a patient infuses, obtaining the infusion amount of the patient based on the medicine information, obtaining infusion time according to the infusion amount and the infusion rate, and generating a working time node of a nurse according to the infusion time, specifically comprises the following steps:
acquiring medicine information when a patient infuses, and acquiring the infusion quantity of the patient in the current infusion medicine bottle based on the medicine information;
obtaining a current time point, obtaining infusion time according to the infusion amount and the infusion rate, and generating a working time node of a nurse according to the current time point and the infusion time.
By adopting the technical scheme, the infusion quantity in the infusion medicine bottle is obtained according to the medicine information when the patient infuses, the infusion time is obtained according to the infusion quantity and the infusion rate, and then the working time node of the nurse is determined according to the current time point and the infusion time. The automaticity of arranging the work of nurses is improved.
Preferably, the step of obtaining current working states of a plurality of nurses, obtaining busy values of the plurality of nurses when the working time node is combined with the infusion time and the working states, obtaining response time of the nurses based on the busy values, and determining a target nurse based on the response time comprises the following steps:
Acquiring current working states of a plurality of nurses, and analyzing the working states to obtain the workload and the working difficulty of the current work of the nurses;
predicting future work busyness of the nurse based on the workload and the work difficulty to obtain a predicted busyness curve;
obtaining a busy value of a nurse at the working time node based on the predicted busy curve and the infusion time;
And obtaining response time of each nurse to the operation request of the patient based on the busy value, sequencing the response time to obtain a response time sequence, taking the shortest response time in the response time sequence as target response time, and determining a target nurse according to the target response time.
By adopting the technical scheme, the current working state of the nurses is analyzed to obtain the current workload and working difficulty of the nurses, a busy curve of the nurses is generated according to the workload and the working difficulty, the busy value of the nurses when the working time nodes are obtained according to the busy curve and the transfusion time, the response time of each nurse to the operation request of the patients is determined according to the busy value, and the nurse with the smallest response time is selected as the target nurse. The rationality in selecting the target nurse is improved.
Preferably, the current infusion process of the patient is obtained, the infusion operation required by the patient after the completion of the infusion of the round is obtained according to the current infusion process, an operation reminder is generated based on the infusion operation, the operation reminder is sent to a target nurse, and the nurses are reminded according to the response time and the working time node, specifically comprising the following steps:
Acquiring a current infusion process of a patient, and analyzing the current infusion process to obtain infusion operation required to be performed after the completion of the infusion of the patient's own round;
Analyzing the infusion operation based on the infusion operation to obtain operation time and operation flow of the infusion operation, and generating operation reminding based on the operation time and the operation flow;
Generating a worksheet according to the workload of the target nurse, sending the operation prompt to the worksheet according to the working time node, and determining an initial prompt time node;
And generating redundant time by combining the operation time and the target response time, adjusting the initial reminding time according to the redundant time to obtain a target reminding time node, and reminding a target nurse after the time reaches the target reminding time node.
By adopting the technical scheme, the infusion operation which needs to be carried out by the patient after the transfusion of the round is completed is obtained according to the current transfusion process of the patient, and comprises the steps of changing the medicine or pulling out the needle. Obtaining operation time and operation flow according to the infusion operation, generating operation prompt according to the operation time and the operation flow, generating a worksheet according to the workload of a target nurse, sending the operation prompt to the worksheet, determining an initial prompt time node, generating redundant time according to the operation time and the target response time, adjusting the initial prompt time node according to the redundant time to obtain a target prompt time node, and prompting the target nurse according to the target prompt time node. The accuracy of reminding time setting is improved.
Preferably, before the step of reminding the target nurse after the time reaches the target reminding time node, the method further comprises:
Monitoring a worksheet of a target nurse, and judging whether emergency insertion work occurs before the target reminding time node;
If the worksheet of the target nurse is judged to have emergency insertion work before the target reminding time node, determining a nurse to be selected according to the response time sequence;
And regenerating a waiting reminding time node according to the waiting response time of the waiting nurse, and transferring the operation reminding to a waiting worksheet of the waiting nurse.
By adopting the technical scheme, whether the target nurse has emergency insertion work before the target reminding time node is judged, after the emergency insertion work is performed, the candidate nurse is redetermined according to the response time sequence, the candidate reminding time node is generated according to the candidate response time of the candidate nurse, and the operation reminding is transferred to the candidate worksheet of the candidate nurse. The probability of patient unmanageable due to other emergency transactions is reduced.
In a second aspect, the present application provides a method for monitoring infusion based on computer vision recognition, the method comprising:
acquiring a video picture of a drip cup in an infusion hose at the initial stage of infusion, and acquiring the interval time of water drop in the drip cup based on the video picture;
acquiring parameter information of an infusion hose, extracting inner wall diameter data of the infusion hose based on the parameter information, acquiring water drop capacity data of water drops based on the inner wall diameter data, and acquiring an infusion rate by combining the water drop capacity data and the interval time;
acquiring medicine information of a patient during transfusion, acquiring transfusion quantity of the patient based on the medicine information, acquiring transfusion time according to the transfusion quantity and the transfusion rate, and generating a working time node of a nurse according to the transfusion time;
acquiring current working states of a plurality of nurses, obtaining busy values of the nurses when the working time nodes are combined with the infusion time and the working states, obtaining response time of the nurses based on the busy values, and determining a target nurse based on the response time;
the method comprises the steps of obtaining the current infusion process of a patient, obtaining infusion operation required by the patient after the completion of the infusion of a round of infusion according to the current infusion process, generating operation reminding based on the infusion operation, sending the operation reminding to a target nurse, and reminding the nurse according to response time and working time nodes.
In summary, the present application includes at least one of the following beneficial technical effects:
1. When the transfusion starts, a video picture of a drip cup in a transfusion hose is acquired, the interval time of dripping is obtained through analysis, then the drip capacity data of each drip is obtained according to the parameter information of transfusion software, the transfusion rate is further obtained, the transfusion quantity of a patient is obtained according to the medicine information when the patient infuses, the transfusion time is obtained according to the transfusion quantity and the transfusion rate, the working time node of a nurse is determined according to the transfusion time, the target nurse is determined according to the current working state of the nurse, the working time node is sent to the target nurse, and the nurse is reminded according to the response time of the target nurse. The working intensity of nurses in the transfusion process of patients is reduced.
2. And obtaining the infusion operation which needs to be carried out by the patient after the current infusion process of the patient is finished, wherein the operation comprises changing the medicine or pulling out the needle. Obtaining operation time and operation flow according to the infusion operation, generating operation prompt according to the operation time and the operation flow, generating a worksheet according to the workload of a target nurse, sending the operation prompt to the worksheet, determining an initial prompt time node, generating redundant time according to the operation time and the target response time, adjusting the initial prompt time node according to the redundant time to obtain a target prompt time node, and prompting the target nurse according to the target prompt time node. The accuracy of reminding time setting is improved.
Drawings
FIG. 1 is a block diagram of an infusion monitoring system based on computer vision recognition provided in an embodiment of the present application;
fig. 2 is a flowchart of steps of an infusion monitoring method based on computer vision recognition according to an embodiment of the present application.
Reference numerals illustrate: 1. a video processing module; 2. an infusion analysis module; 3. a time determination module; 4.a nurse distribution module; 5. and (5) operating a reminding module.
Detailed Description
The present application will be described in further detail with reference to fig. 1 to 2, but embodiments of the present application are not limited thereto.
The embodiment of the application discloses an infusion monitoring system and method based on computer vision identification.
In this embodiment, an infusion monitoring system based on computer vision recognition, the system comprising:
And a video processing module: the method comprises the steps of obtaining a video picture of a drip cup in an infusion hose at the initial stage of infusion, and obtaining the interval time of water drop in the drip cup based on the video picture;
infusion analysis module: the infusion device comprises a device, a device and a control system, wherein the device is used for acquiring parameter information of an infusion hose, extracting inner wall diameter data of the infusion hose based on the parameter information, acquiring water drop capacity data of water drops based on the inner wall diameter data, and acquiring infusion rate by combining the water drop capacity data and interval time;
and a time determining module: the infusion control method comprises the steps of obtaining medicine information when a patient infuses, obtaining the infusion quantity of the patient based on the medicine information, obtaining infusion time according to the infusion quantity and the infusion rate, and generating a working time node of a nurse according to the infusion time;
A nurse distribution module: the method comprises the steps of obtaining current working states of a plurality of nurses, obtaining busy values of the nurses when working is performed at a working time node by combining infusion time and the working states, obtaining response time of the nurses based on the busy values, and determining target nurses based on the response time;
An operation reminding module: the infusion system is used for acquiring the current infusion process of the patient, obtaining infusion operation required by the patient after the completion of the round of infusion according to the current infusion process, generating operation reminding based on the infusion operation, sending the operation reminding to a target nurse, and reminding the nurse according to response time and working time nodes.
It should be noted that the above modules are merely basic modules of the present embodiment, and in the implementation process, some modules may be added, reduced or modified appropriately without affecting the overall implementation effect.
The method comprises the steps of obtaining a video picture of a drip cup in an infusion hose, and obtaining the interval time of water drop in the drip cup based on the video picture, wherein the steps are as follows:
Acquiring a video picture of a drip cup in the infusion hose, analyzing the video picture, and extracting a drip cup frame picture in the video picture;
Extracting drip cup features in the drip cup frame picture based on the drip cup frame picture, and obtaining a picture visual angle according to the drip cup features;
judging whether the visual angle of the picture is correct or not, and if the visual angle of the picture is not correct, correcting the visual angle of the picture to obtain a target kettle-dropping frame picture with the correct visual angle;
and obtaining a target video picture with a correct visual angle based on the target drip cup frame picture, and obtaining the interval time of dripping in the drip cup according to the target video picture.
In the application, taking a process of transfusion of a patient as an example, the patient uses a mobile phone to shoot a video picture of a transfusion hose, analyzes the video picture to obtain a drip cup frame picture, analyzes the drip cup frame picture to find that a visual angle is from top to bottom, corrects the visual angle to obtain a front visual angle of the drip cup frame picture, adjusts the video picture according to the front visual angle to obtain a target video picture, and obtains the time for water in the drip cup to drop according to the target video picture to be 1 second.
Acquiring parameter information of an infusion hose, extracting inner wall diameter data of the infusion hose based on the parameter information, acquiring water drop capacity data of water drops based on the inner wall diameter data, and acquiring an infusion rate by combining the water drop capacity data and the interval time, wherein the method specifically comprises the following steps of:
Acquiring parameter information of the infusion hose, and extracting inner wall diameter data of the infusion hose based on the parameter information;
Obtaining an interfacial tension value at a liquid inlet in the drip cup based on the inner wall diameter data, obtaining the size of water drops falling in the drip cup based on the interfacial tension, and obtaining water drop capacity data of the water drops based on the size of the water drops;
obtaining the drop frequency of drops based on the interval time, and obtaining the number of drops in unit time based on the drop frequency;
Based on the number of water drops per unit time and the water drop capacity data, the infusion rate is obtained.
In the application, taking a process of infusion of a patient as an example, acquiring parameter information of the infusion hose, acquiring inner wall diameter data of the infusion hose to be 2 millimeters according to the parameter information, acquiring an interfacial tension value of a liquid inlet in the drip cup to be 2 according to the inner wall diameter data, acquiring the size of water drops according to the interfacial tension value, and acquiring water drop capacity data to be 1 milliliter according to the size of the water drops. And obtaining the water drop frequency of 60 drops per minute according to the interval time of 1 second, obtaining the water drop quantity in unit time, and obtaining the infusion rate of 60 milliliters per minute according to the water drop quantity and the attribute capacity data.
Obtaining medicine information of a patient during infusion, obtaining the infusion quantity of the patient based on the medicine information, obtaining infusion time according to the infusion quantity and the infusion rate, and generating a working time node of a nurse according to the infusion time, wherein the working time node comprises the following specific steps:
acquiring medicine information when a patient infuses, and acquiring the infusion quantity of the patient in the current infusion medicine bottle based on the medicine information;
the method comprises the steps of obtaining a current time point, obtaining infusion time according to the infusion quantity and the infusion rate, and generating a working time node of a nurse according to the current time point and the infusion time.
In the application, taking the process of infusion of a patient as an example, acquiring medicine information when the patient infuses, acquiring the current infusion amount of 600 milliliters in an infusion bottle of the patient according to the medicine information, acquiring the current time point as eight early hours, acquiring the infusion time as 10 minutes according to the infusion rate, and acquiring the working time node of a nurse as eight ten minutes according to the infusion time.
The method comprises the steps of obtaining current working states of a plurality of nurses, obtaining busy values of the nurses when working time nodes are combined with infusion time and the working states, obtaining response time of the nurses based on the busy values, and determining target nurses based on the response time, wherein the steps specifically comprise:
The method comprises the steps of obtaining current working states of a plurality of nurses, and analyzing the working states to obtain the current working load and the working difficulty of the nurses;
predicting the future work busyness of the nurse based on the workload and the work difficulty to obtain a predicted busyness curve;
obtaining a busy value of a nurse at a working time node based on the predicted busy curve and the infusion time;
And obtaining response time of each nurse to the operation request of the patient based on the busy value, sequencing the response time, obtaining a response time sequence, taking the shortest response time in the response time sequence as target response time, and determining the target nurse according to the target response time.
In the application, taking the process of infusion of a certain patient as an example, acquiring the working states of three nurses, analyzing the working states to obtain the current workload and working difficulty of each nurse, analyzing the workload and working difficulty to predict the working busyness of the nurses to obtain a predicted busyness curve, obtaining busyness values of the three nurses when the eight nurses are eight in a node according to the eight-point working time, respectively being 1, 2 and 3, obtaining response time of each nurse according to the busyness values, namely 2 minutes, 4 minutes and 6 minutes, arranging the nurses, selecting a target response time 2 with the shortest response time, and determining the nurses corresponding to the nurses as target nurses.
The method comprises the steps of obtaining the current infusion process of a patient, obtaining infusion operation required by the patient after the completion of the infusion of the round according to the current infusion process, generating operation reminding based on the infusion operation, sending the operation reminding to a target nurse, and reminding the nurse according to response time and working time nodes, wherein the steps are as follows:
acquiring a current infusion process of a patient, and analyzing the current infusion process to obtain infusion operation required to be performed after the current round of infusion of the patient is completed;
Analyzing the infusion operation based on the infusion operation to obtain the operation time and operation flow of the infusion operation, and generating an operation reminder based on the operation time and the operation flow;
Generating a worksheet according to the workload of the target nurse, sending an operation prompt to the worksheet according to the working time node, and determining an initial prompt time node;
And generating redundant time by combining the operation time and the target response time, adjusting the initial reminding time according to the redundant time to obtain a target reminding time node, and reminding a target nurse after the time reaches the target reminding time node.
In the application, taking the process of infusion of a patient as an example, the current infusion process of the patient is the infusion process of a first bottle of liquid medicine, the infusion operation after the completion of the infusion of the patient's own round is the dressing change operation, the operation time when the dressing change operation is obtained is 1 minute according to the dressing change operation, the operation flow is that an infusion tube is pulled out from one bottle of medicine and is inserted into the other bottle of medicine, and the operation reminder is generated according to the operation flow. The method comprises the steps of obtaining the workload of a target nurse, generating a worksheet, sending an operation prompt to the worksheet, determining the initial prompt time to be 8:10, adjusting the initial prompt time according to the response time of the target nurse for 2 minutes and combining the response time and the operation time to obtain the target prompt time to be 8:07, and prompting the target nurse when the target prompt time is 8:07.
Before the step of reminding the nurse after the time reaches the target reminding time node, the method further comprises the following steps:
Monitoring a worksheet of a target nurse, and judging whether emergency insertion work occurs before a target reminding time node;
If the worksheet of the target nurse is judged to have emergency insertion work before the target reminding time node, determining a nurse to be selected according to the response time sequence;
and regenerating a waiting reminding time node according to the waiting response time of the waiting nurse, and transferring the operation reminding to the waiting worksheet of the waiting nurse.
In the application, taking the process of infusion of a patient as an example, monitoring a worksheet of a target nurse, finding that urgent insertion work occurs in the worksheet of the nurse before 8:07 minutes, and requiring the nurse to go to process, determining a candidate nurse according to a response time sequence, wherein the candidate response time of the candidate nurse is 4 minutes, generating a candidate reminding time node according to the candidate response to be 8:05, transferring an operation reminder to the worksheet of the candidate nurse, and reminding the nurse when 8:05 minutes.
The embodiment of the invention provides an infusion monitoring method based on computer vision recognition, which uses any one of the infusion monitoring systems based on computer vision recognition, and comprises the following steps:
s100: acquiring a video picture of a drip cup in an infusion hose at the initial stage of infusion, and acquiring the interval time of dripping of water in the drip cup based on the video picture;
S200: acquiring parameter information of the infusion hose, extracting inner wall diameter data of the infusion hose based on the parameter information, acquiring water drop capacity data of water drops based on the inner wall diameter data, and acquiring the infusion rate by combining the water drop capacity data and the interval time;
S300: acquiring medicine information of a patient during transfusion, acquiring transfusion quantity of the patient based on the medicine information, acquiring transfusion time according to the transfusion quantity and the transfusion rate, and generating a working time node of a nurse according to the transfusion time;
S400: acquiring current working states of a plurality of nurses, acquiring busy values of the nurses when working is performed at a working time node by combining the infusion time and the working states, acquiring response time of the nurses based on the busy values, and determining a target nurse based on the response time;
S500: the method comprises the steps of obtaining the current infusion process of a patient, obtaining infusion operation required by the patient after the completion of the infusion of the round according to the current infusion process, generating operation reminding based on the infusion operation, sending the operation reminding to a target nurse, and reminding the nurse according to response time and working time nodes.
The above embodiments are not intended to limit the scope of the present application, so: all equivalent changes in structure, shape and principle of the application should be covered in the scope of protection of the application.
Claims (8)
1. An infusion monitoring system based on computer vision recognition, comprising:
And a video processing module: the method comprises the steps of obtaining a video picture of a drip cup in an infusion hose at the initial stage of infusion, and obtaining the interval time of water drop in the drip cup based on the video picture;
Infusion analysis module: the infusion device comprises a device, a control unit and a control unit, wherein the device is used for acquiring parameter information of an infusion hose, extracting inner wall diameter data of the infusion hose based on the parameter information, acquiring water drop capacity data of water drops based on the inner wall diameter data, and acquiring an infusion rate by combining the water drop capacity data and the interval time;
And a time determining module: the infusion control method comprises the steps of obtaining medicine information when a patient infuses, obtaining the infusion quantity of the patient based on the medicine information, obtaining infusion time according to the infusion quantity and the infusion rate, and generating a working time node of a nurse according to the infusion time;
A nurse distribution module: the system comprises a plurality of nurses, a plurality of infusion time node, a plurality of nurse response time and a target nurse, wherein the plurality of nurses are used for acquiring current working states of the nurses, obtaining busy values of the plurality of nurses when the working time node is combined with the infusion time and the working states, obtaining response time of the nurses based on the busy values, and determining the target nurse based on the response time;
An operation reminding module: the infusion reminding device is used for obtaining the current infusion process of the patient, obtaining infusion operation required by the patient after the completion of the infusion of the round according to the current infusion process, generating operation reminding based on the infusion operation, sending the operation reminding to a target nurse, and reminding the nurse according to the response time and the working time node.
2. The infusion monitoring system based on computer vision recognition according to claim 1, wherein the step of obtaining a video picture of a drip cup in an infusion hose and obtaining an interval time of dropping water drops in the drip cup based on the video picture comprises the following steps:
Acquiring a video picture of a drip cup in an infusion hose, analyzing the video picture, and extracting a drip cup frame picture in the video picture;
Extracting drip cup features in the drip cup frame picture based on the drip cup frame picture, and obtaining a picture visual angle according to the drip cup features;
judging whether the visual angle of the picture is correct or not, and if the visual angle of the picture is not correct, correcting the visual angle of the picture to obtain a target kettle-dropping frame picture with the correct visual angle;
and obtaining a target video picture with a correct visual angle based on the target drip cup frame picture, and obtaining the interval time of dripping in the drip cup according to the target video picture.
3. The infusion monitoring system based on computer vision recognition according to claim 2, wherein the steps of obtaining parameter information of an infusion hose, extracting inner wall diameter data of the infusion hose based on the parameter information, obtaining water drop capacity data of water drops based on the inner wall diameter data, and obtaining an infusion rate in combination with the water drop capacity data and the interval time are specifically as follows:
Acquiring parameter information of the infusion hose, and extracting to obtain inner wall diameter data of the infusion hose based on the parameter information;
obtaining an interfacial tension value at a liquid inlet in the drip cup based on the inner wall diameter data, obtaining the size of water drops falling in the drip cup based on the interfacial tension, and obtaining water drop capacity data of the water drops based on the size of the water drops;
Obtaining the drop frequency of drops based on the interval time, and obtaining the number of drops in unit time based on the drop frequency;
And obtaining the infusion rate based on the water drop quantity and the water drop capacity data in unit time.
4. The infusion monitoring system based on computer vision recognition as set forth in claim 3, wherein the steps of obtaining medicine information of a patient when the patient infuses, obtaining an infusion amount of the patient based on the medicine information, obtaining an infusion time according to the infusion amount and the infusion rate, and generating a working time node of a nurse according to the infusion time are specifically as follows:
acquiring medicine information when a patient infuses, and acquiring the infusion quantity of the patient in the current infusion medicine bottle based on the medicine information;
obtaining a current time point, obtaining infusion time according to the infusion amount and the infusion rate, and generating a working time node of a nurse according to the current time point and the infusion time.
5. The infusion monitoring system based on computer vision recognition according to claim 4, wherein the steps of obtaining current working states of a plurality of nurses, obtaining busy values of the plurality of nurses at the working time node in combination with the infusion time and the working states, obtaining response time of the nurses based on the busy values, and determining target nurses based on the response time are specifically as follows:
Acquiring current working states of a plurality of nurses, and analyzing the working states to obtain the workload and the working difficulty of the current work of the nurses;
predicting future work busyness of the nurse based on the workload and the work difficulty to obtain a predicted busyness curve;
obtaining a busy value of a nurse at the working time node based on the predicted busy curve and the infusion time;
And obtaining response time of each nurse to the operation request of the patient based on the busy value, sequencing the response time to obtain a response time sequence, taking the shortest response time in the response time sequence as target response time, and determining a target nurse according to the target response time.
6. The infusion monitoring system based on computer vision recognition according to claim 5, wherein the steps of obtaining a current infusion process of a patient, obtaining an infusion operation required by the patient after the completion of the current round of infusion according to the current infusion process, generating an operation reminder based on the infusion operation, sending the operation reminder to a target nurse, and reminding the nurse according to the response time and the working time node are specifically as follows:
Acquiring a current infusion process of a patient, and analyzing the current infusion process to obtain infusion operation required to be performed after the completion of the infusion of the patient's own round;
Analyzing the infusion operation based on the infusion operation to obtain operation time and operation flow of the infusion operation, and generating operation reminding based on the operation time and the operation flow;
Generating a worksheet according to the workload of the target nurse, sending the operation prompt to the worksheet according to the working time node, and determining an initial prompt time node;
And generating redundant time by combining the operation time and the target response time, adjusting the initial reminding time according to the redundant time to obtain a target reminding time node, and reminding a target nurse after the time reaches the target reminding time node.
7. The computer vision recognition-based infusion monitoring system of claim 6, further comprising, prior to the step of alerting the target nurse after the time reaches the target alert time node:
Monitoring a worksheet of a target nurse, and judging whether emergency insertion work occurs before the target reminding time node;
If the worksheet of the target nurse is judged to have emergency insertion work before the target reminding time node, determining a nurse to be selected according to the response time sequence;
And regenerating a waiting reminding time node according to the waiting response time of the waiting nurse, and transferring the operation reminding to a waiting worksheet of the waiting nurse.
8. A method for monitoring infusion based on computer vision recognition, using the infusion monitoring system based on computer vision recognition as set forth in any one of claims 1 to 7, comprising the steps of:
acquiring a video picture of a drip cup in an infusion hose at the initial stage of infusion, and acquiring the interval time of water drop in the drip cup based on the video picture;
acquiring parameter information of an infusion hose, extracting inner wall diameter data of the infusion hose based on the parameter information, acquiring water drop capacity data of water drops based on the inner wall diameter data, and acquiring an infusion rate by combining the water drop capacity data and the interval time;
acquiring medicine information of a patient during transfusion, acquiring transfusion quantity of the patient based on the medicine information, acquiring transfusion time according to the transfusion quantity and the transfusion rate, and generating a working time node of a nurse according to the transfusion time;
acquiring current working states of a plurality of nurses, obtaining busy values of the nurses when the working time nodes are combined with the infusion time and the working states, obtaining response time of the nurses based on the busy values, and determining a target nurse based on the response time;
the method comprises the steps of obtaining the current infusion process of a patient, obtaining infusion operation required by the patient after the completion of the infusion of a round of infusion according to the current infusion process, generating operation reminding based on the infusion operation, sending the operation reminding to a target nurse, and reminding the nurse according to response time and working time nodes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410454364.0A CN118304514A (en) | 2024-04-16 | 2024-04-16 | Infusion monitoring system and method based on computer vision recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410454364.0A CN118304514A (en) | 2024-04-16 | 2024-04-16 | Infusion monitoring system and method based on computer vision recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118304514A true CN118304514A (en) | 2024-07-09 |
Family
ID=91722071
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410454364.0A Pending CN118304514A (en) | 2024-04-16 | 2024-04-16 | Infusion monitoring system and method based on computer vision recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118304514A (en) |
-
2024
- 2024-04-16 CN CN202410454364.0A patent/CN118304514A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20240108221A1 (en) | Method and system for providing data management in integrated analyte monitoring and infusion system | |
| EP3285827B1 (en) | Infusion system, device, and method having advanced infusion features | |
| JP6014013B2 (en) | Wireless medical data communication system and method | |
| JP2006024225A (en) | System for processing healthcare related event information and user interface | |
| US11383024B2 (en) | System for providing multiple infusions to a patient | |
| US20080200870A1 (en) | System for guiding a user during programming of a medical device | |
| JP2001321345A (en) | Organism signal monitor | |
| CN107545142B (en) | Virtual medicine bottle system and method for tracking medicine taking records of chronic disease patients | |
| US20050049464A1 (en) | Remote pharmaceutical administration device | |
| JP2003271732A (en) | Hospital information system | |
| CN112233768A (en) | Medicine safe use control method, device and system | |
| KR20130121211A (en) | Medication service terminal and monitoring method therof | |
| CN118304514A (en) | Infusion monitoring system and method based on computer vision recognition | |
| US20140207484A1 (en) | Drug information management device and drug information management method | |
| CN115252962A (en) | Intelligent infusion alarm method and system based on machine vision | |
| US20210228092A1 (en) | Treatment recommendation creation system | |
| US20220384005A1 (en) | System and Method for Monitoring Adherence to Drug Regimen | |
| JP2014016846A (en) | Apparatus, method, and program for information processing | |
| JP2006051244A (en) | System for supporting infusion of medicine or the like in home | |
| CN113963773A (en) | Diagnostic report display method and device, electronic equipment and storage medium | |
| CN106901985A (en) | Compounding medicine control method and device | |
| CN112915316A (en) | Infusion control system | |
| CN112420152A (en) | Intelligent pharmaceutical recommendation method and system for renal insufficiency patients | |
| Rayo et al. | Assessing medication safety technology in the intensive care unit | |
| CN115713999B (en) | Informationized data processing method and system suitable for infusion treatment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |