WO2020135226A1

WO2020135226A1 - Drainage monitoring system and method based on image recognition

Info

Publication number: WO2020135226A1
Application number: PCT/CN2019/126662
Authority: WO
Inventors: 刘铁楠; 吴春涛; 艾梦池
Original assignee: 刘铁楠
Priority date: 2018-12-29
Filing date: 2019-12-19
Publication date: 2020-07-02
Also published as: JP7211668B2; JP2022516157A; CN109498857B; CN109498857A

Abstract

Disclosed are a drainage monitoring system and method based on image recognition. According to an exemplary embodiment, the drainage monitoring system comprises: at least one video acquisition device for obtaining video stream data of a drainage liquid; an image processing module, which is connected to the at least one video acquisition device and used for receiving the video stream data and processing the video stream data to obtain drainage data of the drainage liquid; and a monitoring module, which is connected to the image processing module and used for receiving and monitoring the drainage data. Moreover, the monitoring system can further be provided with a central monitoring module for monitoring monitored data of multiple monitoring modules in real time. The drainage monitoring system can centrally monitor multiple drainage data such as the drainage volume, color, turbidity of the drainage liquid, and monitor relevant data and start alarm, so that the work burden of medical staff is effectively reduced, and the medical risks are reduced.

Description

Drainage monitoring system and method based on image recognition

Technical field

The present invention generally relates to the field of medical devices, and more particularly, to a drainage monitoring system and method based on image recognition.

Background technique

The body fluid drainage device is one of the most common medical devices in the medical field and is widely used in major clinical departments. The most commonly used drainage devices for existing medical devices are mainly drainage bottles or drainage bags. In the clinical application of diagnosis and treatment, drainage devices rely on gravity or negative pressure to drain, and are mostly placed at the bottom of the bed.

The indicators that characterize the changes in the drainage process include the drainage volume, color, and turbidity of the drainage fluid. The amount of drainage volume reflects the accumulation and symptoms of water and blood pus in the body cavity of the patient. The color and turbidity of the drainage fluid must be To a certain extent, it reflects the status of the drainage site tissue. Because the current medical drainage devices generally still rely on manual viewing for monitoring, automatic monitoring of the liquid in the drainage device cannot be achieved, and patients generally require long drainage treatment times, sometimes up to days or even weeks, so it will consume a lot of manpower It is especially inconvenient to observe and monitor the drainage volume, color and turbidity of the drainage fluid at night or when there is a shortage of personnel. Moreover, there are often cases where the patient's condition deteriorates and the drainage fluid suddenly increases or the traits change for a short time. For example, a blood vessel rupture leads to massive bleeding, and the drainage fluid is doped with an abnormal substance, which changes from clarified and thin to turbid, which requires medical care. Personnel find and perform corresponding treatments in time, and the conventional manual observation and monitoring mode is difficult to detect such changes in time, and it is easy to delay the disease, cause serious consequences and even endanger the lives of patients.

In view of this, it is desirable to provide a drainage monitoring device that can simultaneously monitor drainage data including drainage flow rate, flow rate, drainage fluid color, etc., to minimize medical risks, and find, alarm, and record And treatment, thereby reducing the workload of medical staff, reducing the medical risks that may be caused by artificial monitoring of loopholes.

Summary of the invention

In order to solve the above technical problems, the present invention provides a medical drainage liquid monitoring system and method, which can realize the unified monitoring of important information data in the drainage process such as the drainage flow rate, flow rate, color, turbidity, etc. Collect and record the relevant data, and report to the police when an abnormality occurs.

According to an aspect of the present invention, there is provided a drainage monitoring system based on image recognition, comprising: at least one video acquisition device for acquiring video stream data of drainage fluid; an image processing module, which is connected to the at least one video acquisition device Connected for receiving the video stream data and processing the video stream data to obtain the drainage data of the drainage liquid; and a monitoring module connected to the image processing module for receiving and monitoring Describe the streaming data.

In an embodiment, the video capture device includes a first video capture device aligned with the drainage container, and optionally, a second video capture device aligned with the drainage tube.

In one embodiment, the drainage data includes the color, turbidity and drainage volume of the drainage fluid.

In an embodiment, the monitoring module is further configured to calculate the flow rate of the drainage liquid flowing through the drainage tube based on the drainage flow data of the drainage liquid.

In an embodiment, the monitoring module is further configured to determine a change in the properties of the drainage fluid based on the change in color and turbidity.

In an embodiment, the image processing module includes: an image sampling module for extracting a key frame image from the video stream data; a color extraction module for performing image recognition on the key frame image and extracting the Color information of key frame image; turbidity detection module, used to process the key frame image to determine the turbidity information of the drainage fluid; drainage flow reading module, used to perform image recognition on the key frame image And enhanced processing, read the scale value information in the key frame image.

In an embodiment, the image sampling module is configured to adjust the sampling rate based on changes in the color information.

In one embodiment, the turbidity detection module includes: a brightness extraction module to extract the brightness value of the image from the key frame image; a turbidity determination module to characterize the turbidity of the drainage fluid based on the brightness value Degrees information.

In an embodiment, the image processing module includes: an image sampling module for extracting key frame images from the video stream data; and an image feature extraction module for extracting color features and turbidity from the key frame images Degree features, wherein the image feature extraction module is a model obtained based on neural network training.

In an embodiment, the image sampling module is configured to adjust the sampling rate based on changes in the color characteristics.

In an embodiment, the monitoring module includes: a first data receiving unit for receiving color, turbidity and drainage flow data of the drainage liquid; a data processing unit for calculating flow through based on the drainage flow data The flow rate of the drainage liquid of the drainage tube, and determining the property change of the drainage liquid based on the change of the color data and the turbidity data; the first display unit is used to change the color, turbidity, drainage flow rate, flow rate and properties At least one of them is displayed in real time; and a first alarm unit is used to issue an alarm signal when at least one of the drainage volume, color, turbidity, drainage volume, flow rate, and property changes exceeds a predetermined threshold.

In an embodiment, the drainage monitoring system further includes a central monitoring module for monitoring the monitoring data of a plurality of the monitoring modules in real time.

In an embodiment, the central monitoring module includes: a second data receiving unit configured to receive monitoring data sent by a plurality of the monitoring modules in a wired or wireless manner, the monitoring data including the color, turbidity, At least one of the drainage volume, the flow rate, and the change in the properties; the storage unit for storing the monitoring data; and the second display unit for simultaneously displaying the monitoring data of a plurality of the monitoring modules.

In an embodiment, the central monitoring module further includes: a second alarm unit, configured to issue an alarm signal when the monitoring data of any one of the plurality of monitoring modules exceeds a predetermined threshold.

In an embodiment, the central monitoring module further includes: a threshold setting unit for setting a predetermined threshold of the monitoring data of each monitoring module.

Another aspect of the present invention provides a drainage monitoring method based on image recognition, including: acquiring video stream data of drainage fluid; processing the video stream data to obtain drainage data of the drainage fluid; and receiving and Monitoring the drainage data, wherein the drainage data includes the color, turbidity and drainage volume of the drainage liquid.

Compared with the prior art, the drainage monitoring system according to the embodiment of the present application can accurately monitor the color and change of the drainage liquid, and realizes a variety of drainage data such as the flow rate, drainage volume, and turbidity of the drainage liquid based on image processing Monitoring, to achieve the processing, recording and alarming of the above drainage data, reducing the workload of medical staff and greatly reducing medical risks; the monitoring system does not require the liquid nature of the drainage fluid, and all body fluids are It has better response and higher accuracy; and the monitoring system has a simple structure, is easy to install and deploy, and can be applied to various clinical occasions.

BRIEF DESCRIPTION

By describing the embodiments of the present application in more detail with reference to the accompanying drawings, the above and other objects, features, and advantages of the present application will become more apparent. The drawings are used to provide a further understanding of the embodiments of the present application, and form a part of the specification, and are used to explain the present application together with the embodiments of the present application, and do not constitute a limitation on the present application. In the drawings, the same reference numerals generally represent the same components.

1 is a block schematic diagram of a drainage monitoring system according to an embodiment of the invention;

2 is a block diagram of an image processing module according to an embodiment of the invention;

3 is a block diagram of an image processing module according to another embodiment of the present invention;

4 is a structural block diagram of a monitoring module according to an embodiment of the present invention;

5 is a structural block diagram of a central monitoring module according to an embodiment of the present invention;

6 is a flowchart of a method for monitoring drainage according to an embodiment of the present invention.

detailed description

The present application will be further described in detail below with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the related invention, but not to limit the invention.

The monitoring system of the present invention is suitable for various medical procedures involving drainage. In the drainage process, body fluids such as abdominal fluid and pleural fluid generally flow out of the drainage chamber through the medical drainage catheter, and are finally contained by the drainage container. The drainage container can It is all the drainage containers in the medical process, such as drainage bottles, drainage bags, negative pressure drainage devices or intraoperative suction devices, which are generally placed under the chamber to form a hydraulic pressure difference to achieve the drainage of body fluids.

As described above, the existing drainage devices generally manually monitor the drainage flow, color and other data of the drainage liquid, which is easy to cause risks due to the lack of monitoring. Therefore, the present invention proposes to be able to detect and perform the drainage data during the drainage process Monitoring device. The term "drainage device" is used in the present invention, and its meaning includes both a drainage container for containing drainage fluid, a drainage tube connected to the drainage site of the patient and the drainage container, and a negative pressure device connected to the drainage container. In order to monitor the color of the drainage liquid, the drainage tube and the drainage container are preferably made of sampling transparent materials, such as glass or transparent silica gel.

FIG. 1 is a schematic diagram of a drainage monitoring system according to a first embodiment of the present invention. As shown in FIG. 1, the drainage monitoring device includes: a video collection device 10, which can shoot the drainage liquid to obtain the video stream data of the drainage liquid; and an image processing module 20, which is connected to the video collection device 10 and can receive The video stream data, and process the video stream data to obtain the drainage data of the drainage liquid; and a monitoring module 30, which is connected to the image processing module 20, is used to receive and monitor the drainage data.

The video capture device 10 may use optical sensor devices such as a camera, and the camera continuously collects image data sequences to form video stream data. Compared with pure image data, the present invention uses video data to achieve dynamic real-time monitoring of the drainage process.

As shown in FIG. 1, the video capture device 10 is aimed at the drainage device 40 to shoot the drainage process. At least one video acquisition device is provided. In one embodiment, only one first camera aligned with the drainage container may be provided, so as to implement monitoring of the drainage liquid in the drainage container. In the present invention, the side of the drainage container is provided with Volume scale. For easy identification, volume scale lines can be printed on each side of the drainage container. The volume scale lines are printed in black and are wear-resistant and corrosion-resistant. The video data collected from this contains information such as flow rate and color. In another embodiment, a second camera aligned with the drainage tube may be additionally provided for acquiring color information of the drainage liquid flowing through the drainage tube. If there is a dangerous situation such as blood vessel rupture during the postoperative drainage process, the color change of the drainage liquid in the drainage container is not as obvious as the color change of the drainage liquid in the drainage tube, so the collected data through the second camera can monitor such danger more sensitively.

The video acquisition device 10 is detachably mounted on the drainage container or negative pressure device, and can also be installed on the IV rack together with the drainage container, which ensures that the drainage container is located below the drainage chamber of the patient, and the video acquisition device and the drainage container are positioned opposite Fixed, to ensure the consistency of the format of the video acquisition data. In some ways, a light source can be set to ensure the clarity of the shooting at night during the drainage process, and the light source may not be used according to the specific situation.

The image processing module 20 receives the video data captured by the camera, and obtains the drainage data such as the color, turbidity, and drainage volume of the drainage liquid through the image processing method. In one embodiment, as shown in FIG. 2, the image processing module includes an image sampling module 21, a color extraction module 22, a turbidity detection module 23 and a drainage flow reading module 24.

The video stream data can be regarded as a sequence of images discharged in a certain order. The image sampling module 21 is used to extract key frame images from the video stream data collected by the camera. This article uses "key frame images" to denote images from which drainage data can be obtained. The image acquisition module 21 can extract the key frame image in the video stream data at a fixed sampling frequency, for example, extract one frame image per second, or adjust the sampling rate of the key frame image according to the change of the drainage data as described below, so as to achieve The dynamic and accurate monitoring of a certain period of time during the drainage process.

The color extraction module 22 is used to perform image recognition on the key frame image and extract color information of the key frame image. In one embodiment, the key frame images of the first camera or the second camera may be used for image recognition to extract color information, or the key frame images of the first camera and the second camera may be used for image recognition to extract color information. It can be compared and analyzed to obtain the patient's drainage situation. Since the color information of the image of the second camera aligned with the drainage tube can better characterize the real-time color of the drainage liquid, the color extraction module 22 preferably extracts the color in the key frame image of the second camera information.

In one embodiment, the color extraction module 22 may collect the RGB values of multiple sampling points of the key frame image according to a preset sampling rule, and then estimate the color information of the key frame image according to the color values of the multiple sampling points. Generally speaking, the color of the drainage area of the same key frame image is basically the same, so the color value of a few pixels in this area can represent most of the pixels in the image, and the collection bottle or drainage tube is used as the sampling area. Select multiple sampling points in the sampling area and collect their color values, so that the color information of the image can be extracted.

Due to the influence of factors such as lighting or the transmittance of the drainage container, it will cause interference to the color of the key frame image. To this end, in one embodiment, the acquired color information can be corrected, for example, the camera takes an image of the standard color card and extracts its color, and compares it with the standard color to obtain the deviation between the two as a correction factor, and uses The correction factor corrects the color of the drainage fluid extracted from the key frame image.

In one embodiment, the image sampling module 21 is configured to adjust the sampling rate based on the change of the color information. The change of the color information can be based on the RGB values of the three channels to calculate the change value of the color of the drainage liquid over time, which can be either the amount of change in the overall color of the drainage liquid or the amount of change in the color value of a single channel. In the case of drastic changes in the color of the drainage fluid, for example, when the R value of the image exceeds a threshold, it may indicate that the patient's drainage is abnormal. In this case, the sampling rate of the key frame image extracted from the video stream data should be increased, such as by the second One frame image is adjusted to extract 5 frames per second, which makes the monitoring of the drainage process more accurate.

The turbidity detection module 23 is used to process the key frame image to determine the turbidity information of the drainage fluid. In the present invention, the brightness information of the image is used to characterize the turbidity of the drainage fluid. In one embodiment, the turbidity detection module includes: a brightness extraction module that extracts the brightness value of the image from the key frame image; a turbidity determination module that characterizes the turbidity information of the drainage fluid based on the brightness value.

The brightness value of the key frame image can be extracted in various ways. For example, the key frame image can be converted from RGB space to a YUV format image, and the Y value is used as the turbidity information of the drainage fluid. The specific conversion method is Y=0.299*R+ 0.587*G+0.114*B. In one embodiment, the key frame image may be converted from RGB space to another color mode image, for example, the L component of LAB and the I component of HSI may be used as the brightness value of the pixel.

After extracting, for example, the brightness values of all pixels in the area of the drainage tube in the image, the turbidity information of the drainage fluid can be characterized by calculating its average or mean square value. If the value is greater than a threshold, it indicates that the drainage fluid is normal and clear Status, and if the value is lower than a threshold, it indicates that there may be blood clots in the drainage fluid, causing it to appear cloudy.

The drainage flow reading module 24 is used to perform image recognition and enhancement processing on the key frame image, and read the scale value information in the key frame image. The scale value information represents the volume of the drainage liquid contained in the drainage container. In one embodiment, the scale value information may be determined based on the level of the drainage fluid. Specifically, first convert the key frame image to grayscale mode, and then perform median filtering on the image to enhance the liquid level information in the image, and the processed image is subjected to edge enhancement processing to extract the liquid level in the drainage container Line, based on the pixel coordinates of the liquid level line, the volume of drainage fluid can be calculated. In an example, the first camera is set in a direction that can be aligned with the volume scale lines on the two sides of the drainage container, so that there are two liquid level lines in the key frame image, which can be based on the average of the two liquid level lines Pixel coordinates calculate the volume of the drainage fluid to balance the possible angle difference between the camera and the liquid level line.

In another embodiment, the present invention can also use machine learning to extract color features and turbidity features in the drainage data. As shown in FIG. 3, the image processing module includes: an image sampling module 21', an image feature extraction module 22' and a drainage flow reading module 24', in which the image acquisition module 21' and the drainage flow reading module 24' function The configuration is the same as that of the image acquisition module 21 and the drainage reading module 24 described above, which will not be repeated here.

The image feature extraction module 22' is a model obtained based on neural network training and is used to extract color features and turbidity features from key frame images. In one embodiment, the neural network may use a CNN model of a convolutional neural network, while using the YUV color space as the input feature of the model, which independently reflects the brightness and chrominance information of the image. By training a neural network to obtain a detection model, and using the input of a certain layer in the model as a descriptor of color features and turbidity features, any actual captured image can be input into the model to obtain the representation of the image Drainage data description. Specifically, in the area to be recognized of the drainage tube, a certain number of pixels are extracted and their color values are converted into the color values of the YUV color space, and then the data is input into the trained CNN model for color and turbidity recognition.

In one embodiment, the image sampling module 21' may be configured to adjust the sampling rate based on the change in the color characteristics. Its function and principle are the same as the previous image acquisition module 21, and will not be repeated here.

Returning to FIG. 1, the image processing module 20 is connected to the monitoring module 30 through a data line, which receives drainage data such as the color, turbidity, and drainage volume of the drainage fluid. Meanwhile, as described below, the monitoring module 30 may compare the received drainage data with a set value, thereby monitoring the drainage data.

In an embodiment, the monitoring module 30 may calculate the flow rate of the drainage liquid flowing through the drainage tube based on the drainage flow data of the drainage liquid. Specifically, for example, the image sampling module extracts one frame of image per second, and the drainage reading module reads the scale value information L1 of the first frame image and the scale value information L2 of the 30th frame image respectively, then the drainage volume is L2 within 30 seconds -L1, from which the flow velocity v through the drainage tube of diameter D=4*(L2-L1)/(30*πD ² ) can be calculated.

In addition, the monitoring module 30 may be further configured to determine a change in the properties of the drainage fluid based on the changes in the color and turbidity. The change in the properties may be that the drainage fluid is doped with an abnormal substance, for example, a blood clot or gas appears in the drainage fluid. All of them will cause the color or turbidity of the drainage fluid to change. As described above, the color change can be based on the three-channel RGB value to calculate the color change value of the drainage fluid over time. The change value can be either the amount of change in the overall color of the drainage fluid or the amount of change in the color value of a channel . The change of the turbidity can be monitored based on the brightness value of the image. When the change of the properties of the drainage fluid exceeds the preset threshold, the monitoring module 30 can start an alarm to promptly remind the patient's family or medical staff to view or process it.

In an example, as shown in FIG. 4, the monitoring module 30 includes a first data receiving unit 31, a data processing unit 32 and a first display unit 33, the first data receiving unit 31 and the image processing module 20 are connected by a data line , Used to receive the color, turbidity, and drainage data of the drainage fluid, and transmit the received data to the data processing unit 32. The data processing unit 32 can perform calculations on the relevant data, such as calculating the flow through the drainage tube based on the drainage flow data The flow rate of the drainage fluid and the change in the properties of the drainage fluid based on the change in color and turbidity. In one example, the data processing unit 32 can also calculate the estimated full time based on the flow velocity and the volume of the current drainage bottle. At least one of the drainage data is displayed in real time by the first display unit 33. Because the drainage bottle or drainage bag is mostly placed at the bottom of the bed or is covered, the real-time display of the drainage data can make it easier to observe the drainage flow, drainage color, or turbidity.

The content displayed by the first display unit 33 can be selected according to specific needs. The number can also be adjusted according to the needs, one display unit can be set, or multiple display units can be set, for example, they are set beside the bed and the nurse station respectively. The first display unit 33 may be a display unit of various forms, such as a liquid crystal display, a light column, a digital display adjustable instrument, etc., and may be an integrated type or a split type. Among them, the color display method of the drainage liquid can be visually displayed through the color information, and can also be displayed numerically through the RGB component. In an embodiment, the first display unit 33 may also be an operation device such as a handheld PDA, and the drainage data is wirelessly transmitted to the device for medical personnel to view at any time. The first display unit 33 and the data processing unit 32 are connected in a wired or wireless manner, for example, the drainage data can be transmitted to the first display unit 33 via Bluetooth to achieve the purpose of remote real-time monitoring.

As an example, the data processing unit 32 may use a single chip microcomputer, a processor, and the like with data calculation and processing capabilities to perform calculation processing on the drainage data, and it may further include an input system, an output system, and a wireless communication module connected to the processor. In an example, the data processing unit 32 is connected to the negative pressure treatment device through a wired or wireless method. For example, when it is judged that the drainage fluid property changes and it is judged that there is gas in the drainage fluid, a control signal is sent to the negative pressure treatment device, for example Increase the negative pressure to draw gas.

In an example, the monitoring module 30 further includes a first alarm unit 34 for issuing an alarm when at least one of the determined drainage volume, flow rate, color, turbidity, and property changes exceeds a predetermined threshold For example, the data processing unit 32 may compare the turbidity with a predetermined threshold to determine whether abnormalities such as blood clots appear in the drainage fluid, and accordingly trigger the first alarm unit 34 to generate an alarm signal. The predetermined threshold may be fixedly set in the data processing unit 32, or may be input and set through an input system of a control panel such as the threshold setting unit 36, so that different thresholds may be set according to different drainage types and patient signs. The alarm signal may be an alarm signal such as sound and light, thereby attracting the attention of the patient and his family or medical staff, and reminding the attention of the drainage situation, thereby reducing the workload of the medical staff. For example, in the clinical situation, a sudden change of the drainage fluid properties may occur in a short time due to rupture of blood vessels. If the optimal treatment period is not easily found in time, such abnormal conditions can be found in time by the monitoring device of the present invention. The medical risks caused by the lack or negligence of medical staff during the monitoring process are avoided.

As shown in FIGS. 4-5, the monitoring system further includes a central monitoring module 50, which can be used as a part of the monitoring function of the monitoring center of the nurse station or the monitoring management center of the hospital to monitor the monitoring data of multiple monitoring modules 30 in real time The data processing unit 32 of each monitoring module can associate the drainage data with the identification information (name, bed number, age, etc.) of the corresponding patient, and send the corresponding data to the central monitoring module 50 in a wired or wireless manner through the communication unit 35 , Which includes: a second data receiving unit 51 for receiving monitoring data sent by a plurality of monitoring modules, a storage unit 52 for recording and storing the monitoring data, and a second display unit 53 for displaying multiple monitoring simultaneously The monitoring data of the module, for example, can display patient identification information, drainage data, the estimated full time of the drainage container, etc., and display the data such as the curve of drainage volume, flow rate, color and/or turbidity with time through the image. Based on the above functions, the central monitoring module 50 can centrally monitor the relevant data of the drainage patients on a certain floor or all beds of the hospital ward, which is convenient for unified dispatch management.

In an example, the central monitoring module 50 may further include: an information processing unit 56 configured to process the received identification information and drainage data, and through the information processing unit 56 and the storage unit 52, the patient's identification information and The drainage data is edited and processed, so that the hospital staff can centrally store and query the historical drainage data of each patient. The central monitoring module 50 may further include a threshold setting unit 52 and a second alarm unit 54 which are respectively used to set a predetermined threshold value of the monitoring data of each monitoring module and that the monitoring data of any one of the multiple monitoring modules exceeds When the threshold is predetermined, an alarm signal is issued. It can be understood that the first alarm unit 34 and the second alarm unit 54 can respectively alert the patient/family and nurse/doctor of the abnormal drainage situation, and remind the medical staff to take necessary measures, so that the medical risk can be minimized.

FIG. 6 is a flowchart of a drainage monitoring method according to an embodiment of the present application, including the following steps:

In step S110, the video stream data of the drainage fluid is acquired.

The drainage monitoring method of the present invention can be applied to various medical procedures, and the drainage fluid includes various drainage fluids such as intraoperative lumen drainage and postoperative urine drainage. In one embodiment, a video collection device such as a 2 million pixel camera can be used to align the drainage device to capture the drainage process and obtain video stream data. And can be stored as AVI, MPG and other general formats. Compared with image shooting, the present invention uses video stream data to dynamically monitor the drainage process, thus achieving more accurate drainage monitoring.

In step S120, the video stream data is processed to obtain drainage data of the drainage fluid.

The processing of video stream data can be implemented on a data processor with image processing capabilities such as a microcontroller chip or a DSP processor. The obtained drainage data includes the drainage flow, color, and turbidity of the drainage liquid. Correspondingly, the processing of the video streaming data includes image sampling, color extraction, turbidity detection, and drainage flow reading. The specific processing can be described above The image processing module is implemented and will not be repeated here. Through the processing of the video stream data, data such as the drainage volume, color, and turbidity of the drainage liquid can be obtained in a centralized manner, thereby obtaining all-round information on the drainage of the patient.

In step S130, the drainage data is received and monitored.

You can use single-chip computers and processors with data calculation and processing capabilities to calculate and monitor the drainage data. Further, based on the drainage volume, color, and turbidity, the drainage speed and the properties of the drainage fluid can be calculated. These data can be displayed on the screen or mobile terminal online. For example, through the display and processing of drainage data, it can be used to monitor the signs of the patient during the operation of the drainage process, and can also be used to monitor the recovery of the patient by the medical staff after the drainage. Through monitoring and alarm functions, it can reduce the workload of patients and medical staff, and minimize medical risks, low cost, and can produce good economic benefits.

The above description has been given for the purpose of illustration and description. Furthermore, this description is not intended to limit the embodiments of the present application to the form disclosed herein. Any person skilled in the art can make various changes and modifications in form and detail without departing from the scope and ideas of the present invention. That is, the scope of the present invention is defined by the appended claims and their equivalents.

Claims

A drainage monitoring system based on image recognition, including:

At least one video acquisition device for acquiring video stream data of drainage fluid;

An image processing module, connected to the at least one video acquisition device, for receiving the video stream data and processing the video stream data to obtain drainage data of the drainage fluid; and

A monitoring module connected to the image processing module is used to receive and monitor the drainage data.
The drainage monitoring system according to claim 1, wherein the video collection device includes a first video collection device aligned with the drainage container, and optionally, a second video collection device aligned with the drainage tube.
The drainage monitoring system according to claim 1, wherein the drainage data includes color, turbidity and drainage volume of the drainage fluid.
The drainage monitoring system according to claim 3, wherein the monitoring module is further configured to calculate the flow rate of the drainage liquid flowing through the drainage tube based on the drainage flow data of the drainage liquid.
The drainage monitoring system according to claim 3, wherein the monitoring module is further configured to determine a property change of the drainage liquid based on the change in the color and turbidity.
The drainage monitoring system of claim 1, wherein the image processing module comprises:

An image sampling module, used for extracting key frame images from the video stream data;

A color extraction module, used to perform image recognition on the key frame image and extract color information of the key frame image;

A turbidity detection module, configured to process the key frame image to determine the turbidity information of the drainage fluid;

The drainage reading module is used to perform image recognition and enhancement processing on the key frame image, and read the scale value information in the key frame image.
The drainage monitoring system of claim 6, wherein the image sampling module is configured to adjust the sampling rate based on the change of the color information.
The drainage monitoring system of claim 6, wherein the turbidity detection module comprises:

A brightness extraction module, extracting the brightness value of the image from the key frame image;

The turbidity determination module characterizes the turbidity information of the drainage liquid based on the brightness value.
The drainage monitoring system of claim 1, wherein the image processing module comprises:

An image sampling module, used for extracting key frame images from the video stream data;

An image feature extraction module is used to extract color features and turbidity features from the key frame image, wherein the image feature extraction module is a model obtained based on neural network training.
The drainage monitoring system of claim 9, wherein the image sampling module is configured to adjust the sampling rate based on changes in the color characteristics.
The drainage monitoring system of claim 3, wherein the monitoring module comprises:

A first data receiving unit, configured to receive color, turbidity and drainage data of the drainage liquid;

The data processing unit is used to calculate the flow rate of the drainage liquid flowing through the drainage tube based on the drainage flow data, and determine the property change of the drainage liquid based on the change of the color data and the turbidity data;

A first display unit for real-time display of at least one of the color, turbidity, drainage volume, flow rate, and property changes; and

The first alarm unit is configured to issue an alarm signal when at least one of the color, turbidity, drainage volume, flow rate, and property change exceeds a predetermined threshold.
The drainage monitoring system according to claim 11, further comprising:

The central monitoring module is used to monitor the monitoring data of multiple monitoring modules in real time.
The drainage monitoring system of claim 12, wherein the central monitoring module includes:

A second data receiving unit, configured to receive monitoring data sent by a plurality of the monitoring modules in a wired or wireless manner, the monitoring data including at least one of the color, turbidity, drainage volume, flow rate, and property changes;

A storage unit for storing the monitoring data; and

The second display unit is used to simultaneously display the monitoring data of multiple monitoring modules.
The drainage monitoring system according to claim 13, wherein the central monitoring module further comprises:

A second alarm unit, configured to issue an alarm signal when the monitoring data of any one of the plurality of monitoring modules exceeds a predetermined threshold;

The threshold setting unit is used to set a predetermined threshold of the monitoring data of each monitoring module.
A drainage monitoring method based on image recognition, including:

Obtain the video stream data of drainage fluid;

Processing the video stream data to obtain drainage data of the drainage liquid; and

Receive and monitor the drainage data,

Wherein, the drainage data includes the color, turbidity and drainage volume of the drainage liquid.