CN115252962A - Intelligent infusion alarm method and system based on machine vision - Google Patents

Abstract

The invention discloses an intelligent infusion alarm method based on machine vision, which belongs to the technical field of medical treatment and machine vision. The invention collects images of infusion containers and dripping pots by relying on monitoring cameras in existing infusion scenarios, and obtains pre-trained images of infusion containers. The machine learning model can effectively obtain the volume data of different infusion containers and the diameter data of different types of droppers, and then obtain the dripping speed data, which can realize the timely alarm of the infusion status, which is compared with the existing infusion alarm methods. In terms of cost, the present invention is low in cost, can realize one-to-many infusion monitoring, and can display the remaining infusion time in real time; in addition, the present invention does not need to set a fixed reference line or a fixed reference value, but according to different infusion container types and infusion container capacities The relative change setting is performed, so it can perform intelligent infusion monitoring for different infusion container types, infusion container capacity and infusion speed, with strong versatility and can meet different usage scenarios.

Description

A kind of intelligent transfusion alarm method and system based on machine vision

technical field

The invention relates to the technical fields of medical treatment and machine vision, in particular to an intelligent transfusion alarm method and system based on machine vision.

Background technique

Intravenous infusion is a method of intravenously injecting a large amount of sterile liquid, electrolytes, and drugs into the body by using the principles of atmospheric pressure and hydrostatic pressure. It is a routine drug delivery technology in clinical medicine; Nurses effectively monitor the infusion process to prevent blood return caused by the completion of the infusion. If the infusion is not reminded in time, it may cause a big accident; Methods of monitoring infusion conditions such as weight sensors, infrared sensors, or pressure sensors for alarming, for example: Chinese Patent No. CN103463698B discloses a pressure-sensitive sensor infusion alarm system, which requires pre-installed sensors on infusion bottles or infusion bags to Realizing infusion monitoring and alarming, this method undoubtedly increases the production cost of infusion bottles or infusion bags, and further increases the cost of treatment for patients. Even if these sensors are placed externally and reused in the form of reusable equipment, the cost It is also very high, and in the scene with many patients, it is also necessary to install one-to-one reusable equipment for each infusion bottle or infusion bag. The follow-up maintenance cost is higher, and compatibility needs to be considered. And it cannot realize one-to-many infusion monitoring;

Compared with the sensor-based monitoring and alarm system, image recognition can be used to solve the above problems. At present, although some methods for infusion monitoring and alarm through image recognition have appeared, for example: Chinese Patent No. CN111298238B discloses a method based on An image recognition infusion early warning method, but this method needs to preset the early warning line on the infusion container, and its practical applicability is low; in addition, Chinese Patent No. CN105498042B discloses a non-light-proof infusion automatic alarm method and its device based on video, Although the invention can realize real-time alarm according to the liquid level in the drip pot and the liquid speed in the drip pot, the alarm is too fast and there is no buffer time for the nurse to change the dressing or remove the needle. If there are many patients, it will make the nurse In addition, due to the different types of infusion containers (commonly there are infusion bottles or bags), the capacity of infusion containers, and the infusion speed adjusted by nurses, it is difficult to compare by setting fixed preset values in practice, because regardless of the speed preset The set value is still the preset value of the warning line of the infusion container. It is affected by many factors, and it is difficult to achieve a unified one. If the nurse presets and enters one by one, there will undoubtedly be a risk of misoperation;

Therefore, need badly a kind of intelligent transfusion warning method to solve the above problems.

Contents of the invention

The purpose of the present invention is to solve the defects existing in the prior art, and propose a kind of intelligent transfusion alarm method and system based on machine vision.

In order to achieve the above object, the present invention adopts the following technical solutions:

A kind of intelligent infusion warning method based on machine vision, comprises the steps:

1) Based on the monitoring camera in the existing infusion scene, the image of the patient's infusion container at the first infusion time point is collected, and the image of the infusion container is recognized based on the first machine learning model to obtain the volume data of the infusion container;

2) After obtaining the volume data of the infusion container, collect the image of the drip pot based on the monitoring camera, and identify it based on the second machine learning model to obtain the diameter data of the dropper;

3) After obtaining the diameter data of the dropper, use the monitoring camera to collect the video of the dripping pot at the second infusion time point, and perform video processing on it based on the preset time interval to calculate the first dripping speed in the dripping pot data, the preset time interval is 1 minute;

4) Correct the first drop rate data based on the diameter data, and calculate the time difference between the second infusion time point and the first infusion time point, and compare it with the corrected first drop rate data Calculate the speed data to obtain the error data;

5) Subtract the error data from the volume data of the infusion container to obtain the current infusion volume data, and calculate the remaining infusion time according to the corrected first drop rate data;

6) Judging in real time whether the remaining infusion time satisfies the preset early warning time range, and if so, an early warning is given;

7) When the remaining infusion time satisfies the preset warning time range, use the monitoring camera to collect the video of the dripping pot for a second time in real time, calculate the second dripping speed data in the dripping pot, and judge the second dripping rate at the same time. Whether the ratio of the speed data to the first drop speed data exceeds the preset ratio, if it exceeds, a second emergency alarm will be issued.

Further, both the first machine learning model and the second machine learning model are generated based on the labeled basic data in the infusion equipment database as a training set, and the basic data in the infusion equipment database include at least different types of Volume data for infusion containers and diameter data for different types of droppers.

Further, the first machine learning model and the second machine learning model are specifically at least one of logistic regression, random forest, K-nearest neighbor, support vector machine, linear discriminant analysis, naive Bayesian, and neural network .

Further, the video processing based on the preset time interval to calculate the first dripping speed data in the dripping pot includes:

Carrying out the timeline distinction of the collected drip pot video, and extracting the 1-minute video;

Mark the key frame of the 1-minute video, that is, the first key frame when the infusion drop falls into the liquid surface for the first time, and the second key frame when the infusion drop falls into the liquid surface for the second time, and cycle in turn until 1 All keyframes in minutes, according to the keyframes, the first dripping speed data in the dripping pot is obtained.

Further, the error data refers to the infusion volume data that has been input into the human body within the time difference between the second infusion time point and the first infusion time point, that is, the period from when the patient starts inputting the time point to after the data processing is completed The amount of infusion completed.

Further, the first drip rate data is corrected based on the diameter data, and the time difference between the second infusion time point and the first infusion time point is calculated, and compared with the corrected The first drop speed data is calculated to obtain error data, including:

Firstly, according to the diameter data, that is, the first drop speed data is corrected according to the required drop volume of 1 ml of droppers with different diameters, that is, the corrected first drop speed data is obtained;

Then, subtracting the first infusion time point from the second infusion time point to obtain a time difference;

Finally, the error data is obtained by multiplying the corrected first drop speed data and the time difference, and the formula is as follows:

S _W ＝V _X ×t _c

In the formula: S _W represents the error data; V _X represents the corrected first drop speed data; t _c represents the time difference.

Further, the formula for calculating the remaining infusion time is as follows:

In the formula: T _Z represents the remaining infusion time; S _Z represents the volume data of the infusion container; S _W represents the error data; V _X represents the corrected first drip speed data.

An intelligent infusion alarm system based on machine vision, including:

The first collection module is used to collect the image of the patient's infusion container at the first infusion time point;

The first identification module is used to identify the image of the infusion container based on the first machine learning model, so as to obtain volume data of the infusion container;

The second collection module is used to collect the image of the dripping pot after obtaining the volume data of the infusion container;

The second identification module is used to identify the image of the dripping pot based on the second machine learning model, so as to obtain the diameter data of the dropper;

The third collection module is used to collect the video of the dripping pot at the second infusion time point after obtaining the diameter data of the dropper;

The first calculation processing module is used to perform video processing on the dripping pot video based on a preset time interval, so as to calculate the first dripping speed data in the dripping pot;

The second calculation processing module is used to calculate the time difference between the second infusion time point and the first infusion time point, and calculate and obtain error data;

The third calculation processing module is used to subtract the error data from the volume data of the infusion container to obtain the current infusion volume data, and calculate the remaining infusion time;

The infusion early warning module is used to judge in real time according to the remaining infusion time whether it satisfies the preset early warning time range, and if it is satisfied, an early warning will be performed, and the remaining infusion time will be displayed at the same time;

The emergency alarm module is used to collect the video of the dripping pot for the second time in real time when the remaining infusion time meets the preset warning time range, calculate the second dripping speed data in the dripping pot, and judge the second dripping speed data at the same time Whether the ratio with the first drop speed data exceeds the preset ratio, and if it exceeds, a second emergency alarm will be issued.

Further, the second calculation and processing module includes a drop speed correction unit, which is used to correct the first drop according to the diameter data, that is, according to the drop volume required for 1 ml of droppers with different diameters. The speed data is corrected, that is, the corrected first drop speed data is obtained.

Compared with the prior art, the beneficial effects of the present invention are:

(1) This application discloses an intelligent infusion alarm method and system based on machine vision, which relies on the monitoring camera in the existing infusion scene to collect images of infusion containers and drip pots, and relies on pre-trained machine learning models The identification can effectively obtain the volume data of different infusion containers and the diameter data of different types of droppers, and then obtain the drip speed data, which can realize timely alarm for the infusion status, which is compared with the existing sensor-based monitoring and alarm system In terms of low cost, one-to-many infusion monitoring can be realized, and the remaining infusion time can be displayed in real time, so that the assistant nurse can change the dressing and remove the needle in time;

(2) This application discloses an intelligent infusion alarm method and system based on machine vision, which corrects the drip speed data according to the diameter data of different types of droppers, so as to obtain accurate remaining infusion time. The time is compared with the preset early warning time range to realize timely early warning of the infusion situation, thereby helping to reserve enough buffer time for nurses to change dressings, prepare medicines and remove needles for multiple people; in addition, through real-time judgment of the second If the ratio of the drip rate data to the first drip rate data exceeds the preset ratio, a second emergency alarm can be issued to prevent the guard from forgetting to change the dressing or remove the needle for a certain patient because there are many patients;

(3) This application discloses an intelligent infusion alarm method and system based on machine vision. Compared with the existing image recognition method for infusion monitoring and alarm, the present invention does not need to set a fixed reference line or a fixed reference value. Instead, the relative changes are set according to different infusion container types and infusion container capacities. Therefore, the present invention can perform intelligent infusion monitoring on different infusion container types, infusion container capacities, and infusion speeds. It has strong versatility and can meet different usage scenarios. Certain social and economic benefits.

Description of drawings

The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the description, and are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention.

Fig. 1 is the overall flowchart of a kind of intelligent transfusion alarm method based on machine vision that the present invention proposes;

Fig. 2 is a schematic diagram of the overall structure of an intelligent infusion alarm system based on machine vision proposed by the present invention.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention.

In describing the present invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", " The orientation or positional relationship indicated by "outside", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, so as to Specific orientation configurations and operations, therefore, are not to be construed as limitations on the invention.

In one embodiment, with reference to Fig. 1, a kind of intelligent infusion warning method based on machine vision is provided, comprising the steps:

1) Based on the monitoring camera in the existing infusion scene, the image of the patient's infusion container at the first infusion time point is collected, and the image of the infusion container is recognized based on the first machine learning model to obtain the volume data of the infusion container;

2) After obtaining the volume data of the infusion container, collect the image of the drip pot based on the monitoring camera, and identify it based on the second machine learning model to obtain the diameter data of the dropper;

Specifically, both the first machine learning model and the second machine learning model are generated based on the labeled basic data in the infusion equipment database as a training set, and the basic data in the infusion equipment database at least contain different types of Volume data for infusion containers and diameter data for different types of droppers.

Specifically, the first machine learning model and the second machine learning model are specifically at least one of logistic regression, random forest, K nearest neighbor, support vector machine, linear discriminant analysis, naive Bayesian, and neural network ;

What needs to be explained here is: the first machine learning model and the second machine learning model can be the same model, or different models, and any model that can realize this technical point can be applied. The invention is not specifically limited thereto.

3) After obtaining the diameter data of the dropper, use the monitoring camera to collect the video of the dripping pot at the second infusion time point, and perform video processing on it based on the preset time interval to calculate the first dripping speed in the dripping pot data, the preset time interval is 1 minute;

Specifically, the video processing is performed based on the preset time interval to calculate the first dripping speed data in the dripping pot, including:

Carrying out the timeline distinction of the collected drip pot video, and extracting the 1-minute video;

Mark the key frame of the 1-minute video, that is, the first key frame when the infusion drop falls into the liquid surface for the first time, and the second key frame when the infusion drop falls into the liquid surface for the second time, and cycle in turn until 1 All keyframes in minutes, according to the keyframes, the first dripping speed data in the dripping pot is obtained.

4) Correct the first drop rate data based on the diameter data, and calculate the time difference between the second infusion time point and the first infusion time point, and compare it with the corrected first drop rate data Calculate the speed data to obtain the error data;

Specifically, the error data refers to the infusion volume data that has been input into the human body within the time difference between the second infusion time point and the first infusion time point, that is, the time period from when the patient starts to input the time point to when the data processing is completed The amount of infusion completed.

Specifically, the first drip rate data is corrected based on the diameter data, and the time difference between the second infusion time point and the first infusion time point is calculated, and it is compared with the corrected The first drop speed data is calculated to obtain error data, including:

Firstly, according to the diameter data, that is, the first drop speed data is corrected according to the required drop volume of 1 ml of droppers with different diameters, that is, the corrected first drop speed data is obtained;

Then, subtracting the first infusion time point from the second infusion time point to obtain a time difference;

Finally, the error data is obtained by multiplying the corrected first drop speed data and the time difference, and the formula is as follows:

S _W ＝V _X ×t _c

In the formula: S _W represents the error data; V _X represents the corrected first drop speed data; t _c represents the time difference.

5) Subtract the error data from the volume data of the infusion container to obtain the current infusion volume data, and calculate the remaining infusion time according to the corrected first drop rate data;

Specifically, the formula for calculating the remaining infusion time is as follows:

In the formula: T _Z represents the remaining infusion time; S _Z represents the volume data of the infusion container; S _W represents the error data; V _X represents the corrected first drip speed data.

6) Judging in real time whether the remaining infusion time satisfies the preset early warning time range, and if so, an early warning is given;

7) When the remaining infusion time satisfies the preset warning time range, use the monitoring camera to collect the video of the dripping pot for a second time in real time, calculate the second dripping speed data in the dripping pot, and judge the second dripping rate at the same time. Whether the ratio of the speed data to the first drop speed data exceeds the preset ratio, if it exceeds, a second emergency alarm will be issued.

In one embodiment, referring to Fig. 2, a kind of intelligent transfusion alarm system based on machine vision is provided, comprising:

The first collection module is used to collect the image of the patient's infusion container at the first infusion time point;

The first identification module is used to identify the image of the infusion container based on the first machine learning model, so as to obtain volume data of the infusion container;

The second collection module is used to collect the image of the dripping pot after obtaining the volume data of the infusion container;

The second identification module is used to identify the image of the dripping pot based on the second machine learning model, so as to obtain the diameter data of the dropper;

The third collection module is used to collect the video of the dripping pot at the second infusion time point after obtaining the diameter data of the dropper;

The first calculation processing module is used to perform video processing on the dripping pot video based on a preset time interval, so as to calculate the first dripping speed data in the dripping pot;

The second calculation processing module is used to calculate the time difference between the second infusion time point and the first infusion time point, and calculate and obtain error data;

The third calculation processing module is used to subtract the error data from the volume data of the infusion container to obtain the current infusion volume data, and calculate the remaining infusion time;

The infusion early warning module is used to judge in real time according to the remaining infusion time whether it satisfies the preset early warning time range, and if it is satisfied, an early warning will be performed, and the remaining infusion time will be displayed at the same time;

The emergency alarm module is used to collect the video of the dripping pot for the second time in real time when the remaining infusion time meets the preset warning time range, calculate the second dripping speed data in the dripping pot, and judge the second dripping speed data at the same time Whether the ratio with the first drop speed data exceeds the preset ratio, and if it exceeds, a second emergency alarm will be issued.

The second calculation and processing module includes a drop speed correction unit, and the drop speed correction unit is used to perform a calculation on the first drop speed data according to the diameter data, that is, according to the drop volume required for 1 ml of droppers with different diameters. Correction, that is, to obtain the corrected first drop speed data.

In one embodiment, a machine vision-based intelligent infusion alarm method and system proposed in this application will be further explained in detail in combination with specific cases;

In this embodiment, it is assumed that a patient undergoes an infusion at 18:00:15, and at this time, the first infusion time point is collected by the monitoring camera in the infusion scene (assuming that the first infusion time point is 18:01:00 seconds) of the patient's infusion container image and identify the infusion container image based on the first machine learning model to obtain the volume data of the infusion container;

In this embodiment, it is assumed that the infusion container image is identified as an infusion bottle type, and the infusion container volume data of the infusion bottle is 200 ml;

Afterwards, collect the image of the dripping pot based on the monitoring camera, and identify it based on the second machine learning model to obtain the diameter data of the dripper;

In this embodiment, it is assumed that the diameter data of the dropper is 0.5 cm, and each drop of liquid is 0.5 ml, that is to say, every two drops of liquid is 1 ml on average;

What needs to be explained here is: the diameter data of the dropper and the drop volume corresponding to each drop of liquid are not necessarily exactly the same as the existing infusion equipment, and it is only used as an example here;

After obtaining the diameter data of the dropper, collect the video of the dripping pot at the second infusion time point (assuming that the first infusion time point is 18:05:00 seconds), and perform video processing on it based on the preset time interval, To calculate the first dripping speed data in the dripping pot;

In this embodiment, it is assumed that the first drop speed data in the drip pot is 30 drops/minute within 1 minute;

Then, according to the diameter data, that is, the first drop speed data is corrected according to the required drop volume of 1 ml of droppers with different diameters, that is, the first drop speed data after correction is 15 drops/minute; then , subtract the first infusion time point from the second infusion time point to obtain a time difference; finally, use the formula S _W =V _X ×t _c to combine the corrected first infusion rate data with the When the time difference is multiplied, the error data obtained is 60 milliliters;

Then, subtract the error data from the volume data of the infusion container, that is, 200-60=140 milliliters;

计算剩余输液时间为140÷15≈9.3分钟；Then, according to the formula

Calculate the remaining infusion time as 140÷15≈9.3 minutes;

Finally, it is judged in real time whether the remaining infusion time satisfies the preset early warning time range. Assuming that the preset early warning time range is 1 minute, the system will give an early warning when the remaining infusion time is less than 1 minute;

At the same time, when the remaining infusion time satisfies the preset warning time range, the monitoring camera will capture the video of the dripping pot twice in real time, calculate the second dripping speed data in the dripping pot, and judge the second dripping speed at the same time Whether the ratio of the data to the first drop speed data exceeds the preset ratio, if so;

In this embodiment, assuming that the second drip speed data is 25 drops/minute, the ratio of the second drip speed data to the first drip speed data is 25:30, which is 5 to 6;

In this embodiment, assuming that the preset ratio is 1, when the second drip speed data is 25 drops/minute, it is less than the preset ratio, so a second emergency alarm is issued.

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto, any person familiar with the technical field within the technical scope disclosed in the present invention, according to the technical solution of the present invention Any equivalent replacement or change of the inventive concepts thereof shall fall within the protection scope of the present invention.

Claims (9)

1. an intelligent transfusion alarm method based on machine vision, is characterized in that, comprises the steps: 1) Based on the monitoring camera in the existing infusion scene, the image of the patient's infusion container at the first infusion time point is collected, and the image of the infusion container is recognized based on the first machine learning model to obtain the volume data of the infusion container; 2) After obtaining the volume data of the infusion container, collect the image of the drip pot based on the monitoring camera, and identify it based on the second machine learning model to obtain the diameter data of the dropper; 3) After obtaining the diameter data of the dropper, use the monitoring camera to collect the video of the dripping pot at the second infusion time point, and perform video processing on it based on the preset time interval to calculate the first dripping speed in the dripping pot data, the preset time interval is 1 minute; 4) Correct the first drop rate data based on the diameter data, and calculate the time difference between the second infusion time point and the first infusion time point, and compare it with the corrected first drop rate data Calculate the speed data to obtain the error data; 5) Subtract the error data from the volume data of the infusion container to obtain the current infusion volume data, and calculate the remaining infusion time according to the corrected first drop rate data; 6) Judging in real time whether the remaining infusion time satisfies the preset early warning time range, and if so, an early warning is given; 7) When the remaining infusion time satisfies the preset warning time range, use the monitoring camera to collect the video of the dripping pot for a second time in real time, calculate the second dripping speed data in the dripping pot, and judge the second dripping rate at the same time. Whether the ratio of the speed data to the first drop speed data exceeds the preset ratio, if it exceeds, a second emergency alarm will be issued. 2. A kind of intelligent infusion alarm method based on machine vision according to claim 1, is characterized in that, described first machine learning model and described second machine learning model are all based on the infusion equipment database after marking The basic data is generated as a training set, and the basic data in the infusion equipment database at least include volume data of different types of infusion container and diameter data of different types of droppers. 3. A kind of intelligent infusion alarm method based on machine vision according to claim 1, is characterized in that, described first machine learning model and described second machine learning model are specifically logistic regression, random forest, K nearest neighbor , support vector machine, linear discriminant analysis, naive Bayesian, neural network at least one. 4. A kind of intelligent transfusion alarm method based on machine vision according to claim 1, is characterized in that, described based on preset time interval it is carried out video processing, to calculate the first drop speed data in drip pot, comprise : Carrying out the timeline distinction of the collected drip pot video, and extracting the 1-minute video; Mark the key frame of the 1-minute video, that is, the first key frame when the infusion drop falls into the liquid surface for the first time, and the second key frame when the infusion drop falls into the liquid surface for the second time, and cycle in turn until 1 All keyframes in minutes, according to the keyframes, the first dripping speed data in the dripping pot is obtained. 5. A kind of intelligent infusion alarm method based on machine vision according to claim 1, characterized in that, the error data refers to the time difference between the second infusion time point and the first infusion time point that has been input into the human body. Infusion volume data, that is, the infusion volume completed during the period from the time point when the patient starts inputting to the completion of data processing. 6. A kind of intelligent infusion alarm method based on machine vision according to claim 1, characterized in that, the first drip speed data is corrected based on the diameter data, and the second infusion time is calculated Point and the time difference of the first infusion time point, at the same time, it is calculated with the corrected first drop rate data to obtain error data, including: Firstly, according to the diameter data, that is, the first drop speed data is corrected according to the required drop volume of 1 ml of droppers with different diameters, that is, the corrected first drop speed data is obtained; Then, subtracting the first infusion time point from the second infusion time point to obtain a time difference; Finally, the error data is obtained by multiplying the corrected first drop speed data and the time difference, and the formula is as follows: S _W ＝V _X ×t _c In the formula: S _W represents the error data; V _X represents the corrected first drop speed data; t _c represents the time difference. 7. A kind of intelligent infusion alarm method based on machine vision according to claim 1, is characterized in that, the calculation formula of described remaining infusion time is as follows:

In the formula: T _Z represents the remaining infusion time; S _Z represents the volume data of the infusion container; S _W represents the error data; V _X represents the corrected first drip speed data. 8. An intelligent transfusion alarm system based on machine vision, characterized in that it comprises: The first collection module is used to collect the image of the patient's infusion container at the first infusion time point; The first identification module is used to identify the image of the infusion container based on the first machine learning model, so as to obtain volume data of the infusion container; The second collection module is used to collect the image of the dripping pot after obtaining the volume data of the infusion container; The second identification module is used to identify the image of the dripping pot based on the second machine learning model, so as to obtain the diameter data of the dropper; The third collection module is used to collect the video of the dripping pot at the second infusion time point after obtaining the diameter data of the dropper; The first calculation processing module is used to perform video processing on the dripping pot video based on a preset time interval, so as to calculate the first dripping speed data in the dripping pot; The second calculation processing module is used to calculate the time difference between the second infusion time point and the first infusion time point, and calculate and obtain error data; The third calculation processing module is used to subtract the error data from the volume data of the infusion container to obtain the current infusion volume data, and calculate the remaining infusion time; The infusion early warning module is used to judge in real time according to the remaining infusion time whether it satisfies the preset early warning time range, and if it is satisfied, an early warning will be performed, and the remaining infusion time will be displayed at the same time; The emergency alarm module is used to collect the video of the dripping pot for the second time in real time when the remaining infusion time meets the preset warning time range, calculate the second dripping speed data in the dripping pot, and judge the second dripping speed data at the same time Whether the ratio with the first drop speed data exceeds the preset ratio, and if it exceeds, a second emergency alarm will be issued. 9. A kind of intelligent transfusion alarm system based on machine vision according to claim 8, characterized in that, the second calculation processing module includes a drip speed correction unit, and the drip speed correction unit is used to , that is, the first drop speed data is corrected according to the drop volume required for 1 ml of droppers with different diameters, that is, the corrected first drop speed data is obtained.

Images