CN116211279A - Information collection and analysis system for extracorporeal membrane lung care - Google Patents

Abstract

The invention belongs to the technical field of information collection and analysis for extracorporeal membrane lung care, and discloses an information collection and analysis system for extracorporeal membrane lung care, which comprises the following components: heart rate acquisition module, blood pressure acquisition module, blood flow acquisition module, central control module, thrombus identification module, nursing quality control module, analysis module, early warning module, display module. According to the invention, intelligent recognition is realized when thrombus occurs through the thrombus recognition module, so that a doctor can observe and judge the thrombus conveniently, and corresponding measures are taken in time; meanwhile, the automatic monitoring operation of the extracorporeal membrane oxygenation device can be realized through the nursing quality control module, the operation quality monitoring of medical staff can be realized, and the accuracy of the extracorporeal membrane oxygenation nursing quality control is improved.

Description

Information collection and analysis system for extracorporeal membrane lung care

Technical Field

The invention belongs to the technical field of information collection and analysis for extracorporeal membrane lung care, and particularly relates to an information collection and analysis system for extracorporeal membrane lung care.

Background

The extracorporeal membrane oxygenation is mainly used for providing continuous extracorporeal respiration and circulation for critical cardiopulmonary failure patients, and winning precious time for rescuing critical diseases. The nature of the extracorporeal membrane oxygenation is an improved artificial heart-lung machine, and the most central parts are membrane lungs and blood pumps, which respectively act as artificial lungs and artificial hearts. In the extracorporeal membrane oxygenation operation, blood is led out from veins, oxygen is absorbed through membrane lungs, carbon dioxide is discharged, and the blood subjected to gas exchange can return to veins or arteries under the pushing of a pump. The former is mainly used for external respiratory support, and the latter can be used for external respiratory support and cardiac support because the blood pump can replace the blood pumping function of the heart. When the lung function of a patient is seriously damaged, the oxygenation of the outer membrane of the body can bear the gas exchange task when the conventional treatment is ineffective, so that the lung is in a rest state, precious time is obtained for the rehabilitation of the patient, and the blood pump can replace the function of the heart blood supply to maintain the blood circulation when the heart function of the patient is seriously damaged. The epicardial pulmonary oxygenation is mainly applicable to patients suffering from cardiac arrest, patients suffering from acute severe heart failure, patients suffering from acute severe respiratory failure, other diseases severely threatening respiratory circulatory function, and the like. However, the existing information collection and analysis system for external membrane lung care cannot identify thrombus in time; meanwhile, the quality of the external membrane pulmonary oxygenation care cannot be accurately controlled.

Through the above analysis, the problems and defects existing in the prior art are as follows:

(1) The existing information collection and analysis system for external membrane lung nursing can not identify thrombus in time.

(2) The quality of the external membrane oxygenation care cannot be accurately controlled.

Disclosure of Invention

Aiming at the problems in the prior art, the invention provides an information collection and analysis system for extracorporeal membrane lung care.

The invention is realized in such a way that an information collection and analysis system for extracorporeal membrane lung care comprises:

the device comprises a heart rate acquisition module, a blood pressure acquisition module, a blood flow acquisition module, a central control module, a thrombus recognition module, a nursing quality control module, an analysis module, an early warning module and a display module;

the heart rate acquisition module is connected with the central control module and is used for acquiring heart rate data of a patient;

the blood pressure acquisition module is connected with the central control module and is used for acquiring blood pressure data of a patient;

the blood flow acquisition module is connected with the central control module and used for acquiring heart and lung blood flow data;

the central control module is connected with the heart rate acquisition module, the blood pressure acquisition module, the blood flow acquisition module, the thrombus recognition module, the nursing quality control module, the analysis module, the early warning module and the display module and used for controlling the normal work of each module;

the thrombus identification module is connected with the central control module and used for identifying thrombus;

the nursing quality control module is connected with the central control module and used for controlling the quality of the external membrane pulmonary oxygenation nursing

The analysis module is connected with the central control module and used for analyzing the heart and lung functions of the patient;

the early warning module is connected with the central control module and used for early warning the heart-lung abnormality of the patient;

the display module is connected with the central control module and used for displaying heart rate, blood pressure, blood flow, thrombus identification results, analysis results and early warning information through the display.

Further, the thrombus recognition module recognition method comprises the following steps:

(1) The method comprises the steps that cameras are arranged around a membrane lung and a blood transfusion pipeline to collect images, and images shot by the cameras are obtained at regular time; carrying out enhancement treatment on the acquired images; affine transformation is carried out on the image shot by the camera, and a light field with a complex geometric shape in the image is converted into a light field with uniform space distance interval;

(2) Calculating the texture roughness of each pixel point in the image; judging whether the corresponding pixel point is a foreign object boundary according to the texture roughness; if the foreign object boundary is the foreign object boundary, the foreign object type is further identified, and if the foreign object boundary is the irregular boundary, the thrombus is judged to occur.

Further, the calculating the roughness of each pixel texture in the image includes:

for each pixel point p (x, y) in the image, taking three adjacent window areas with different sizes taking the pixel point p as a center, respectively calculating the average value of three colors of red, green and blue, wherein the three window sizes taking the p point as the center are as follows: 2k×2k, (k=0, 1, 2);

respectively calculating red maximum difference dR, green maximum difference dG and blue maximum difference dB of adjacent window areas of the pixel point p;

calculating the roughness cp=α×dr+β×dg+γ×db of p-point adjacent images; wherein α, β, γ are referred to as weighting factors, α+β+γ=1; α=0.3, β=0.6, γ=0.1.

Further, at least 3 cameras are arranged when thrombus is judged for the membrane lung.

Further, the control method of the nursing quality control module comprises the following steps:

1) Testing whether the external membrane oxygenation device is normal by a testing device; the working parameters of the external membrane oxygenation device are configured, and the operation parameters and the operation site video of the external membrane oxygenation device during operation are obtained in real time through monitoring equipment;

2) Classifying the operation parameters to form index parameters and setting parameters, and detecting the position of the operation site video in real time according to a preset position category detection model;

3) Inputting the classified index parameters into a calibration table for calibration and obtaining calibrated setting parameters, and inputting a real-time part detection video into a video quality classification model for classification, wherein the classification result output by the video quality classification model comprises accurate operation, operation flaws and serious operation errors;

4) Inputting the calibrated setting parameters into the extracorporeal membrane lung oxygenation device, and outputting guidance and corrective measures according to the classification result.

Further, the step of obtaining in real time the in-situ video of the extracorporeal membrane oxygenation device comprises:

identifying an installation site of the extracorporeal membrane lung oxygenation device;

at least one group of shooting cameras is erected according to the identified installation site so that the shooting range of the shooting cameras at least comprises an epicardial lung oxygenation device and a patient.

Further, the training step of the preset part category detection model includes:

dividing the operation site of the extracorporeal membrane oxygenation device into a plurality of part categories according to preset key monitoring parts;

taking standard part videos shot at operation sites of different external membrane oxygenation devices as a first training set, and dividing a set proportion in the first training set to be used as a first verification set;

training the first training set through a convolutional neural network to form a part category detection model;

and verifying the part category detection model through a first verification set, if the verification is passed, ending training, otherwise, increasing the sample number of the first training set to retrain.

Further, the calibration table includes a plurality of groups of standard index parameters and corresponding standard setting parameters, and the specific steps of inputting the classified index parameters into a calibration model to calibrate and obtaining calibrated setting parameters are as follows:

searching the classified index parameters in a standard table, and if the standard index parameters with the difference smaller than a set threshold value are found, calling standard setting parameters corresponding to the standard index parameters;

comparing the standard setting parameters with the classified post-setting parameters to obtain the parameter difference value;

the step of inputting the calibrated setting parameters into the extracorporeal membrane lung oxygenation device comprises the following steps:

inputting the parameter difference value into the extracorporeal membrane lung oxygenation device;

and the extracorporeal membrane oxygenation device is adjusted to operate according to the parameter difference value and the set parameter.

Further, the training step of the video quality classification model includes:

a plurality of part videos detected by a part category detection model are used as a second training set, and a set proportion is divided in the second training set to be used as a second verification set;

training the second training set through a convolutional neural network to form a video quality classification model;

and verifying the video quality classification model through a second verification set, if the verification is passed, ending training, otherwise, increasing the sample number of the second training set to retrain.

Further, the step of inputting the real-time part detection into the video quality classification model for classification specifically includes:

inputting real-time part detection video into a video quality classification model;

if the part detection video is matched with the video quality classification model, outputting a result with accurate operation;

if the part detection video is not matched with the video quality classification model, continuing to judge the similarity between the part detection video and the video quality classification model, outputting a result of the operation defect if the similarity is greater than or equal to a preset threshold value, and outputting a result of serious operation error if the similarity is less than the preset threshold value.

By combining the technical scheme and the technical problems, the invention has the following advantages and positive effects:

according to the invention, the illuminating device and the camera device are respectively arranged for the blood pipeline and the membranous lung through the thrombus recognition module, and the thrombus formation condition of the blood pipeline and the membranous lung is judged through the artificial intelligent algorithm, so that the intelligent recognition is realized when the thrombus appears, a doctor can observe and judge the thrombus conveniently, and corresponding measures are taken in time; meanwhile, classifying the operation parameters through a nursing quality control module to form index parameters and setting parameters, calibrating the index parameters and obtaining calibrated setting parameters, adjusting the operation of the extracorporeal membrane lung oxygenation device through the calibrated setting parameters, classifying the video quality of the actual operation process of medical staff, and outputting guidance and correction measures according to classification results; the invention can realize the automatic monitoring operation of the external membrane pulmonary oxygenation device, can also realize the operation quality monitoring of medical staff, and improves the accuracy of the external membrane pulmonary oxygenation nursing quality control.

Drawings

Fig. 1 is a block diagram of an information collection and analysis system for external membrane lung care according to an embodiment of the present invention.

Fig. 2 is a flowchart of a thrombus recognition module recognition method according to an embodiment of the present invention.

Fig. 3 is a flowchart of a control method of a nursing quality control module according to an embodiment of the present invention.

In fig. 1: 1. a heart rate acquisition module; 2. a blood pressure acquisition module; 3. a blood flow collection module; 4. a central control module; 5. a thrombus recognition module; 6. a care quality control module; 7. an analysis module; 8. an early warning module; 9. and a display module.

Detailed Description

The present invention will be described in further detail with reference to the following examples in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

As shown in fig. 1, an information collection and analysis system for external membrane lung care according to an embodiment of the present invention includes: heart rate collection module 1, blood pressure collection module 2, blood flow collection module 3, central control module 4, thrombus identification module 5, nursing quality control module 6, analysis module 7, early warning module 8, display module 9.

The heart rate acquisition module 1 is connected with the central control module 4 and is used for acquiring heart rate data of a patient;

the blood pressure acquisition module 2 is connected with the central control module 4 and is used for acquiring blood pressure data of a patient;

the blood flow acquisition module 3 is connected with the central control module 4 and is used for acquiring cardiopulmonary blood flow data;

the central control module 4 is connected with the heart rate acquisition module 1, the blood pressure acquisition module 2, the blood flow acquisition module 3, the thrombus recognition module 5, the nursing quality control module 6, the analysis module 7, the early warning module 8 and the display module 9 and is used for controlling the normal work of each module;

a thrombus recognition module 5 connected to the central control module 4 for recognizing thrombus;

a nursing quality control module 6 connected with the central control module 4 for controlling the quality of the external membrane pulmonary oxygenation nursing

The analysis module 7 is connected with the central control module 4 and is used for analyzing the heart and lung functions of the patient;

the early warning module 8 is connected with the central control module 4 and is used for early warning the heart-lung abnormality of the patient;

the display module 9 is connected with the central control module 4 and is used for displaying heart rate, blood pressure, blood flow, thrombus identification result, analysis result and early warning information through a display.

As shown in fig. 2, the method for identifying the thrombus identification module 5 provided by the invention is as follows:

s101, acquiring images by arranging cameras around a membrane lung and a blood transfusion pipeline, and acquiring the images shot by the cameras at regular time; carrying out enhancement treatment on the acquired images; affine transformation is carried out on the image shot by the camera, and a light field with a complex geometric shape in the image is converted into a light field with uniform space distance interval;

s102, calculating the texture roughness of each pixel point in the image; judging whether the corresponding pixel point is a foreign object boundary according to the texture roughness; if the foreign object boundary is the foreign object boundary, the foreign object type is further identified, and if the foreign object boundary is the irregular boundary, the thrombus is judged to occur.

The method for calculating the texture roughness of each pixel point in the image comprises the following steps:

for each pixel point p (x, y) in the image, taking three adjacent window areas with different sizes taking the pixel point p as a center, respectively calculating the average value of three colors of red, green and blue, wherein the three window sizes taking the p point as the center are as follows: 2k×2k, (k=0, 1, 2);

respectively calculating red maximum difference dR, green maximum difference dG and blue maximum difference dB of adjacent window areas of the pixel point p;

calculating the roughness cp=α×dr+β×dg+γ×db of p-point adjacent images; wherein α, β, γ are referred to as weighting factors, α+β+γ=1; α=0.3, β=0.6, γ=0.1.

When judging thrombus for the membranous lung, at least 3 cameras are arranged.

As shown in fig. 3, the control method of the nursing quality control module 6 provided by the invention is as follows:

s201, testing whether an external membrane lung oxygenation device is normal or not through test equipment; the working parameters of the external membrane oxygenation device are configured, and the operation parameters and the operation site video of the external membrane oxygenation device during operation are obtained in real time through monitoring equipment;

s202, classifying the operation parameters to form index parameters and setting parameters, and detecting the parts of the operation site video in real time according to a preset part type detection model;

s203, inputting the classified index parameters into a calibration table for calibration and obtaining calibrated setting parameters, and inputting the real-time part detection video into a video quality classification model for classification, wherein the classification result output by the video quality classification model comprises accurate operation, operation flaws and serious operation errors;

s204, inputting the calibrated setting parameters into the extracorporeal membrane lung oxygenation device, and outputting guidance and correction measures according to the classification result.

The method for acquiring the field video of the extracorporeal membrane oxygenation device in real time comprises the following steps of:

identifying an installation site of the extracorporeal membrane lung oxygenation device;

at least one group of shooting cameras is erected according to the identified installation site so that the shooting range of the shooting cameras at least comprises an epicardial lung oxygenation device and a patient.

The training steps of the preset part category detection model provided by the invention comprise the following steps:

dividing the operation site of the extracorporeal membrane oxygenation device into a plurality of part categories according to preset key monitoring parts;

taking standard part videos shot at operation sites of different external membrane oxygenation devices as a first training set, and dividing a set proportion in the first training set to be used as a first verification set;

training the first training set through a convolutional neural network to form a part category detection model;

and verifying the part category detection model through a first verification set, if the verification is passed, ending training, otherwise, increasing the sample number of the first training set to retrain.

The invention provides a calibration table comprising a plurality of groups of standard index parameters and corresponding standard setting parameters, wherein the specific steps of inputting the classified index parameters into a calibration model for calibration and obtaining the calibrated setting parameters are as follows:

searching the classified index parameters in a standard table, and if the standard index parameters with the difference smaller than a set threshold value are found, calling standard setting parameters corresponding to the standard index parameters;

comparing the standard setting parameters with the classified post-setting parameters to obtain the parameter difference value;

the step of inputting the calibrated setting parameters into the extracorporeal membrane lung oxygenation device comprises the following steps:

inputting the parameter difference value into the extracorporeal membrane lung oxygenation device;

and the extracorporeal membrane oxygenation device is adjusted to operate according to the parameter difference value and the set parameter.

The training steps of the video quality classification model provided by the invention comprise:

a plurality of part videos detected by a part category detection model are used as a second training set, and a set proportion is divided in the second training set to be used as a second verification set;

training the second training set through a convolutional neural network to form a video quality classification model;

and verifying the video quality classification model through a second verification set, if the verification is passed, ending training, otherwise, increasing the sample number of the second training set to retrain.

The method for inputting the real-time part detection into the video quality classification model for classification specifically comprises the following steps:

inputting real-time part detection video into a video quality classification model;

if the part detection video is matched with the video quality classification model, outputting a result with accurate operation;

if the part detection video is not matched with the video quality classification model, continuing to judge the similarity between the part detection video and the video quality classification model, outputting a result of the operation defect if the similarity is greater than or equal to a preset threshold value, and outputting a result of serious operation error if the similarity is less than the preset threshold value.

The working principle of the invention is as follows:

when the heart rate acquisition system works, firstly, heart rate data of a patient are acquired through a heart rate acquisition module 1; collecting blood pressure data of a patient through a blood pressure collecting module 2; collecting cardiopulmonary blood supply flow data through a blood flow collection module 3; secondly, the central control module 4 identifies thrombus through the thrombus identification module 5; quality control of extracorporeal membrane oxygenation care by a care quality control module 6 the cardiopulmonary function of a patient is analyzed by an analysis module 7; then, early warning is carried out on the heart and lung abnormality of the patient through an early warning module 8; finally, heart rate, blood pressure, blood flow, thrombus identification result, analysis result and early warning information are displayed by the display module 9 through a display.

It should be noted that the embodiments of the present invention can be realized in hardware, software, or a combination of software and hardware. The hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or special purpose design hardware. Those of ordinary skill in the art will appreciate that the apparatus and methods described above may be implemented using computer executable instructions and/or embodied in processor control code, such as provided on a carrier medium such as a magnetic disk, CD or DVD-ROM, a programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The device of the present invention and its modules may be implemented by hardware circuitry, such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., as well as software executed by various types of processors, or by a combination of the above hardware circuitry and software, such as firmware.

The invention has the positive effects that:

according to the invention, the illuminating device and the camera device are respectively arranged for the blood pipeline and the membranous lung through the thrombus recognition module, and the thrombus formation condition of the blood pipeline and the membranous lung is judged through the artificial intelligent algorithm, so that the intelligent recognition is realized when the thrombus appears, a doctor can observe and judge the thrombus conveniently, and corresponding measures are taken in time; meanwhile, classifying the operation parameters through a nursing quality control module to form index parameters and setting parameters, calibrating the index parameters and obtaining calibrated setting parameters, adjusting the operation of the extracorporeal membrane lung oxygenation device through the calibrated setting parameters, classifying the video quality of the actual operation process of medical staff, and outputting guidance and correction measures according to classification results; the invention can realize the automatic monitoring operation of the external membrane pulmonary oxygenation device, can also realize the operation quality monitoring of medical staff, and improves the accuracy of the external membrane pulmonary oxygenation nursing quality control.

The foregoing is merely illustrative of specific embodiments of the present invention, and the scope of the invention is not limited thereto, but any modifications, equivalents, improvements and alternatives falling within the spirit and principles of the present invention will be apparent to those skilled in the art within the scope of the present invention.

Claims (10)

1. An information collection and analysis system for extracorporeal membrane lung care, characterized in that the system comprises:

the device comprises a heart rate acquisition module, a blood pressure acquisition module, a blood flow acquisition module, a central control module, a thrombus recognition module, a nursing quality control module, an analysis module, an early warning module and a display module;

the heart rate acquisition module is connected with the central control module and is used for acquiring heart rate data of a patient;

the blood pressure acquisition module is connected with the central control module and is used for acquiring blood pressure data of a patient;

the blood flow acquisition module is connected with the central control module and used for acquiring heart and lung blood flow data;

the central control module is connected with the heart rate acquisition module, the blood pressure acquisition module, the blood flow acquisition module, the thrombus recognition module, the nursing quality control module, the analysis module, the early warning module and the display module and used for controlling the normal work of each module;

the thrombus identification module is connected with the central control module and used for identifying thrombus;

the nursing quality control module is connected with the central control module and used for controlling the quality of the external membrane pulmonary oxygenation nursing

The analysis module is connected with the central control module and used for analyzing the heart and lung functions of the patient;

the early warning module is connected with the central control module and used for early warning the heart-lung abnormality of the patient;

the display module is connected with the central control module and used for displaying heart rate, blood pressure, blood flow, thrombus identification results, analysis results and early warning information through the display.

2. The information collection and analysis system for extracorporeal membrane lung care according to claim 1, wherein the thrombus recognition module recognition method is as follows:

(1) The method comprises the steps that cameras are arranged around a membrane lung and a blood transfusion pipeline to collect images, and images shot by the cameras are obtained at regular time; carrying out enhancement treatment on the acquired images; affine transformation is carried out on the image shot by the camera, and a light field with a complex geometric shape in the image is converted into a light field with uniform space distance interval;

(2) Calculating the texture roughness of each pixel point in the image; judging whether the corresponding pixel point is a foreign object boundary according to the texture roughness; if the foreign object boundary is the foreign object boundary, the foreign object type is further identified, and if the foreign object boundary is the irregular boundary, the thrombus is judged to occur.

3. The system for collecting and analyzing information for extracorporeal membrane lung care of claim 2, wherein the calculating the roughness of each pixel texture in the image comprises:

for each pixel point p (x, y) in the image, taking three adjacent window areas with different sizes taking the pixel point p as a center, respectively calculating the average value of three colors of red, green and blue, wherein the three window sizes taking the p point as the center are as follows: 2k×2k, (k=0, 1, 2);

respectively calculating red maximum difference dR, green maximum difference dG and blue maximum difference dB of adjacent window areas of the pixel point p;

calculating the roughness cp=α×dr+β×dg+γ×db of p-point adjacent images; wherein α, β, γ are referred to as weighting factors, α+β+γ=1; α=0.3, β=0.6, γ=0.1.

4. The information collection and analysis system for extracorporeal membrane lung care according to claim 2, wherein at least 3 cameras are provided for determining thrombus for the membrane lung.

5. The information collection and analysis system for extracorporeal membrane lung care according to claim 1, wherein the care quality control module control method comprises the following steps:

1) Testing whether the external membrane oxygenation device is normal by a testing device; the working parameters of the external membrane oxygenation device are configured, and the operation parameters and the operation site video of the external membrane oxygenation device during operation are obtained in real time through monitoring equipment;

2) Classifying the operation parameters to form index parameters and setting parameters, and detecting the position of the operation site video in real time according to a preset position category detection model;

3) Inputting the classified index parameters into a calibration table for calibration and obtaining calibrated setting parameters, and inputting a real-time part detection video into a video quality classification model for classification, wherein the classification result output by the video quality classification model comprises accurate operation, operation flaws and serious operation errors;

4) Inputting the calibrated setting parameters into the extracorporeal membrane lung oxygenation device, and outputting guidance and corrective measures according to the classification result.

6. The information collecting and analyzing system for extracorporeal membrane lung care of claim 5, wherein the step of acquiring in real time an in situ video of an extracorporeal membrane lung oxygenation device comprises:

identifying an installation site of the extracorporeal membrane lung oxygenation device;

at least one group of shooting cameras is erected according to the identified installation site so that the shooting range of the shooting cameras at least comprises an epicardial lung oxygenation device and a patient.

7. The information collection and analysis system for extracorporeal membrane lung care of claim 5, wherein the training step of the preset site category detection model comprises:

dividing the operation site of the extracorporeal membrane oxygenation device into a plurality of part categories according to preset key monitoring parts;

taking standard part videos shot at operation sites of different external membrane oxygenation devices as a first training set, and dividing a set proportion in the first training set to be used as a first verification set;

training the first training set through a convolutional neural network to form a part category detection model;

and verifying the part category detection model through a first verification set, if the verification is passed, ending training, otherwise, increasing the sample number of the first training set to retrain.

8. The information collecting and analyzing system for lung care of an external membrane according to claim 5, wherein the calibration table comprises a plurality of groups of standard index parameters and corresponding standard setting parameters, and the specific steps of inputting the classified index parameters into a calibration model to calibrate and obtaining the calibrated setting parameters are as follows:

searching the classified index parameters in a standard table, and if the standard index parameters with the difference smaller than a set threshold value are found, calling standard setting parameters corresponding to the standard index parameters;

comparing the standard setting parameters with the classified post-setting parameters to obtain the parameter difference value;

the step of inputting the calibrated setting parameters into the extracorporeal membrane lung oxygenation device comprises the following steps:

inputting the parameter difference value into the extracorporeal membrane lung oxygenation device;

and the extracorporeal membrane oxygenation device is adjusted to operate according to the parameter difference value and the set parameter.

9. The in vitro membrane lung care information collection analysis system of claim 5 wherein the training step of the video quality classification model comprises:

a plurality of part videos detected by a part category detection model are used as a second training set, and a set proportion is divided in the second training set to be used as a second verification set;

training the second training set through a convolutional neural network to form a video quality classification model;

and verifying the video quality classification model through a second verification set, if the verification is passed, ending training, otherwise, increasing the sample number of the second training set to retrain.

10. The information collecting and analyzing system for extracorporeal membrane lung care of claim 9, wherein the step of inputting the real-time part detection into the video quality classification model for classification specifically comprises:

inputting real-time part detection video into a video quality classification model;

if the part detection video is matched with the video quality classification model, outputting a result with accurate operation;

if the part detection video is not matched with the video quality classification model, continuing to judge the similarity between the part detection video and the video quality classification model, outputting a result of the operation defect if the similarity is greater than or equal to a preset threshold value, and outputting a result of serious operation error if the similarity is less than the preset threshold value.

Images