Detailed Description
The technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification.
The terms first, second and the like in the description and in the claims and in the above-described figures are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order. Furthermore, the term "include" and any variations thereof is intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, article, or apparatus.
The execution subject of the medical test instrument remote monitoring image analysis method provided by the embodiments of the present disclosure may be the medical test instrument remote monitoring image analysis device provided by the embodiments of the present disclosure, or a server integrated with the medical test instrument remote monitoring image analysis device, where the medical test instrument remote monitoring image analysis device may be implemented in a hardware or software manner.
The present specification, prior to detailing a medical test instrument remote monitoring image analysis method in connection with one or more embodiments, describes a scenario in which the medical test instrument remote monitoring image analysis method is applied.
Referring to fig. 1a, fig. 1a is a schematic view of a scenario of a remote monitoring image analysis system 100 for a medical test apparatus according to an embodiment of the present invention, where the remote monitoring image analysis system 100 for a medical test apparatus may include an internet of things collector, a medical test apparatus 110, an image capturing device 120, a remote monitoring image analysis device 130 for a medical test apparatus, an apparatus sharing platform 140, and so on. The medical test instrument remote monitoring image analysis device 120 is respectively in communication connection with the internet of things collector, the image shooting device 120 and the user side 130.
In this embodiment, the internet of things collector may collect data from various physical environments corresponding to the medical test apparatus, convert the data into digital signals, and then transmit the digital signals to the remote monitoring image analysis device 120 of the medical test apparatus through a network. The internet of things collector can comprise a sensor, a controller, a communication module, a power management component and the like. The sensor is used for sensing external environment such as temperature, humidity, illumination, sound and the like and converting sensed information into electric signals; the controller processes the signals according to a preset program or instructions from other systems, and performs corresponding operations, such as adjusting a switch, driving a motor, and the like. The communication module is used for sending the collected data or control instructions out through a wired or wireless network. These networks may be local area networks, wide area networks, or even the internet. The power management component is used for providing stable power supply for the whole equipment, and the Internet of things collector can also supply power in a renewable energy mode such as solar energy.
In this embodiment, the medical test apparatus 110 may include a plurality of test apparatuses, an upper portion of the test apparatuses may be a specimen movement area 1110, and a lower portion of the test apparatuses is an apparatus panel area 1112, and the apparatus panel area includes an indicator light, a button, and the like. The medical test instrument 110 may be a device for performing various medical experiments and diagnoses, and the medical test instrument 110 may include a clinical chemistry analysis instrument, a radioactive detection instrument, a bioreactor and small animal imaging system, an immunoassay instrument, and the like.
In this embodiment, the image capturing device 120 may be a capturing device for capturing a scene image of a medical test instrument, where the scene image captured by the image capturing device 120 includes the medical test instrument, and the image capturing device 120 may send the captured scene image to the medical test instrument remote monitoring image analysis device 130.
In this embodiment, referring to fig. 1b, fig. 1b is a schematic structural diagram of an instrument sharing platform 140 according to an embodiment of the present invention, where the instrument sharing platform 140 may be integrated in an electronic device, and the electronic device may be a server, a terminal, or other devices. The server may be a single server or a server cluster composed of a plurality of servers. The terminal may be a mobile phone, a tablet computer, a smart bluetooth device, a notebook computer, or a personal computer (Personal Computer, PC), etc. The instrument sharing platform 140 has a visual display interface 142, and the medical test instrument remote monitoring image analysis device 130 may send remote monitoring image analysis data to the instrument sharing platform 140, and a worker may acquire the remote monitoring image analysis data in real time through the display interface 142 of the instrument sharing platform 140.
In this embodiment, the remote monitoring image analysis device 130 of the medical test apparatus may be integrated in an electronic device, which may be a server or other device. The server may be a single server or a server cluster composed of a plurality of servers. In some embodiments, the medical test instrument remote monitoring image analysis apparatus may also be integrated in a plurality of electronic devices, for example, the medical test instrument remote monitoring image analysis apparatus 130 may be integrated in a plurality of servers, and the medical test instrument remote monitoring image analysis method of the present application is implemented by the plurality of servers.
The medical test apparatus remote monitoring image analysis device 130 may include an image acquisition module, an edge detection module, a feature extraction module, an analysis module, and the like. The medical test instrument remote monitoring image analysis device 130 can acquire various physical environment collection data corresponding to the medical test instrument through the internet of things collector, and can also acquire scene images corresponding to the medical test instrument at a plurality of moments in the current time period from the image shooting device 120; performing edge detection on scene images corresponding to a plurality of moments respectively based on a pre-accessed edge detection model to obtain instrument operation areas and specimen state areas corresponding to medical test instruments at the plurality of moments; based on a pre-accessed feature extraction model, respectively extracting features of an instrument operation area and a specimen state area corresponding to the medical test instrument at a plurality of moments to obtain instrument operation features and specimen state features corresponding to the medical test instrument at a plurality of moments; and determining remote monitoring image analysis data according to instrument operation characteristics and specimen state characteristics corresponding to the medical test instruments at a plurality of moments, wherein the remote monitoring image analysis data comprises abnormal analysis data corresponding to the current time period and the like.
It should be noted that, the schematic view of the scenario of the remote monitoring image analysis system of the medical test apparatus shown in fig. 1b is merely an example, and the remote monitoring image analysis system of the medical test apparatus and the scenario described in the embodiments of the present invention are for more clearly describing the technical solution of the embodiments of the present invention, and do not constitute a limitation on the technical solution provided by the embodiments of the present invention, and as the evolution of the remote monitoring image analysis system of the medical test apparatus and the appearance of a new scenario, those skilled in the art can know that the technical solution provided by the embodiments of the present invention is equally applicable to similar technical problems.
Referring to fig. 2, fig. 2 is a flowchart of a remote monitoring image analysis method for a medical test apparatus according to an embodiment of the present invention, where the remote monitoring image analysis method for a medical test apparatus may be performed by the remote monitoring image analysis device 110 for a medical test apparatus shown in fig. 1 a. The remote monitoring image analysis method of the medical test instrument at least comprises the following steps:
200. and acquiring scene images respectively corresponding to a plurality of moments of the medical test instrument in the current time period.
In this embodiment, the medical test apparatus may include a plurality of test apparatuses, where the test apparatuses include a specimen movement area and an apparatus panel area, the specimen movement area is an area where a specimen is placed, and the specimen movement area may be a container made of a transparent material, so that the image capturing device captures a specimen located in the specimen movement area; the instrument panel area includes a plurality of indicator lights and buttons, etc., and different indicator lights are used to indicate different medical test instrument operating parameter states, for example, the operating parameters may include temperature, humidity, oxygen content, etc. corresponding to the specimen active area. The scene image may be an image of a medical test instrument, the image including the medical test instrument. The image shooting device can shoot a plurality of scene images of the medical test instrument in a time period, wherein the time period corresponds to a plurality of shooting moments, and the shooting moments correspond to a plurality of scene images.
In the embodiment, the image capturing device 120 may send a plurality of scene images captured at a plurality of moments in a current time period to the remote monitoring image analysis device 130 of the medical test instrument; the medical test instrument remote monitoring image analysis device 130 is in communication connection with the image shooting device 120, and can acquire scene images respectively corresponding to a plurality of moments of the medical test instrument in the current time period.
In some embodiments, after obtaining scene images respectively corresponding to a plurality of moments in a current time period of the medical test instrument, the method further comprises: preprocessing the scene images corresponding to the moments respectively, wherein the preprocessing process comprises the following steps: performing image scaling processing on the scene image to obtain a scene image after image scaling, wherein the size of the scene image after image scaling meets the preset size; and performing color coding processing on the scene image after image scaling to obtain a scene image after the color coding processing, wherein the scene image after the color coding processing is a scene image after the preprocessing.
The embodiment can reduce the calculation amount of subsequent processing and improve the processing speed by performing image scaling processing on the scene image; in addition, the embodiment may further perform color coding processing on the scene image after image scaling, where the color coding processing may include graying, binarizing, normalizing, histogram equalizing, and the like.
The gray level converts the color image into a gray image, so that the data processing amount can be reduced; the binarization can convert the image into an image with only black and white colors, so that the data size is further reduced, and the object recognition can be effectively performed; normalization can eliminate the influence of factors such as light by carrying out normalization processing on the colors of the images, so that the images have better robustness in different environments; the histogram equalization can adjust the histogram of the image to improve the contrast of the image, so that the image is clearer and is convenient for the identification of the subsequent medical test instrument. According to the embodiment, the image quality can be improved by preprocessing the scene image, so that useful information in the scene image is more prominent, the computing resources required for processing the image can be reduced, and the system processing speed is increased.
210. And carrying out edge detection on scene images corresponding to the moments respectively based on the pre-accessed edge detection model to obtain instrument operation areas and specimen state areas corresponding to the medical test instruments at the moments.
In this embodiment, the edge detection model may be a model for identifying an instrument operation area and a specimen status area corresponding to a medical test instrument in the scene image.
According to the embodiment, the trained edge detection model can be connected into the medical test instrument remote monitoring image analysis system, and the medical test instrument remote monitoring image analysis system performs edge detection on scene images corresponding to a plurality of moments respectively based on the pre-connected edge detection model, so that an instrument operation area and a specimen state area corresponding to the medical test instrument at the plurality of moments are obtained.
In some embodiments, referring to fig. 3, fig. 3 is a schematic flow chart illustrating edge detection of a scene image corresponding to each of a plurality of moments. As shown in fig. 3, based on a pre-accessed edge detection model, performing edge detection on scene images corresponding to a plurality of moments respectively to obtain an instrument operation area and a specimen state area corresponding to a plurality of moment medical test instruments, including:
2100. Performing instrument edge detection on scene images corresponding to a plurality of moments respectively based on a first edge detection model which is accessed in advance to obtain instrument operation areas corresponding to medical test instruments in the scene images corresponding to the moments;
2110. And carrying out specimen edge detection on the scene images corresponding to the moments respectively based on the second edge detection model which is accessed in advance, and obtaining specimen state areas corresponding to the specimens in the scene images corresponding to the moments.
In this embodiment, the edge detection model includes a first edge detection model and a second edge detection model. The edge detection model is a deep convolutional neural network, a multi-scale convolutional network, a spatial pyramid pooling network, an attention mechanism model, a CNN-based migration learning model or a generation countermeasure network and the like.
According to the embodiment, the trained first edge detection model and the trained second edge detection model can be respectively connected into a medical test instrument remote monitoring image analysis system, and the medical test instrument remote monitoring image analysis system can perform instrument edge detection on scene images corresponding to a plurality of moments respectively based on the first edge detection model which is connected in advance, so that an instrument operation area corresponding to a medical test instrument in the scene images corresponding to each moment is obtained; the medical test instrument remote monitoring image analysis system can also perform specimen edge detection on scene images corresponding to a plurality of moments based on a second edge detection model which is accessed in advance, so as to obtain specimen state areas corresponding to specimens in the scene images corresponding to the moments.
For example, referring to fig. 4, fig. 4 shows a schematic view of the identified instrument operation region and specimen state region in a scene image. As shown in fig. 4, the medical test instrument in the scene image includes 3 test instruments, wherein the 3 test instruments are a test instrument a, a test instrument B and a test instrument C, respectively, a sample a is placed in a sample moving area of the test instrument a, a sample B is placed in a sample moving area of the test instrument B, and a sample C is placed in a sample moving area of the test instrument C. According to the embodiment, instrument edge detection is carried out on the scene image through a first edge detection model which is accessed in advance, so that an instrument operation area corresponding to a medical test instrument in the scene image is obtained, namely, an instrument operation area A1 corresponding to a test instrument A, an instrument operation area B1 corresponding to a test instrument B and an instrument operation area C1 corresponding to a test instrument C are obtained; in this embodiment, the specimen edge detection is performed on the scene image by using the second edge detection model that is accessed in advance, so as to obtain a specimen state region corresponding to the specimen in the scene image, that is, a specimen state region a1 corresponding to the specimen a, a specimen state region b1 corresponding to the specimen b, and a specimen state region c1 corresponding to the specimen c.
In some embodiments, based on a pre-accessed edge detection model, before performing edge detection on scene images corresponding to a plurality of moments respectively to obtain an instrument operation area and a specimen state area corresponding to a plurality of moment medical test instruments, the method further comprises: acquiring a first image data training set with a medical test instrument edge mark and a second image data training set with a specimen edge mark; training the first edge detection model through a first image data training set to obtain a trained first edge detection model; training the second edge detection model through a second image data training set to obtain a trained second edge detection model; and respectively accessing the first edge detection model after training and the second edge detection model after training into a remote monitoring image analysis system of the medical test instrument.
Before the edge detection model is connected to the remote monitoring image analysis system of the medical test instrument, the edge detection model needs to be trained to obtain the trained edge detection model.
In the training process of the first edge detection model, the first image data training set with the medical test instrument edge marks is adopted, the first edge detection model is trained through the first image data training set, so that a trained first edge detection model is obtained, and the trained first edge detection model can well identify specimen state areas corresponding to the medical test instrument in scene images corresponding to all moments. The medical test instrument edge marking may be a block diagram that demarcates the medical test instrument edge.
In the training process of the second edge detection model, the second image data training set with the specimen edge marks is adopted, the second edge detection model is trained through the second image data training set, so that a trained second edge detection model is obtained, and the trained second edge detection model can well identify specimen state areas corresponding to medical test instruments in scene images corresponding to all moments. The specimen edge labels may be block diagrams that divide the specimen edges.
220. And respectively extracting the characteristics of the instrument operation areas and the specimen state areas corresponding to the medical test instruments at a plurality of moments based on the pre-accessed characteristic extraction model to obtain the instrument operation characteristics and the specimen state characteristics corresponding to the medical test instruments at a plurality of moments.
In this embodiment, the feature extraction model is a model for extracting features of an instrument operation area and a specimen status area corresponding to the medical test instrument at a plurality of moments, respectively. The feature extraction model may be a convolutional neural network, etc., which may include a LeNet model, alexNet model, googLeNet model, resNet model, etc.
In some embodiments, referring to fig. 5, fig. 5 shows a schematic flow chart of feature extraction for an instrument operation region and a specimen state region. As shown in fig. 5, based on a feature extraction model accessed in advance, feature extraction is performed on an instrument operation area and a specimen state area corresponding to the medical test instrument at a plurality of moments, so as to obtain instrument operation features and specimen state features corresponding to the medical test instrument at a plurality of moments, including:
2200. Extracting the characteristics of instrument operation areas corresponding to the medical test instruments at a plurality of moments through a first characteristic extraction model which is accessed in advance, so as to obtain instrument operation characteristics corresponding to the medical test instruments at the plurality of moments;
2210. and carrying out feature extraction on specimen state areas corresponding to the medical test instruments at a plurality of moments through a second feature extraction model which is accessed in advance, so as to obtain specimen state features corresponding to the medical test instruments at a plurality of moments.
In this embodiment, the feature extraction model includes a first feature extraction model and a second feature extraction model. When feature extraction is performed on the instrument operation areas corresponding to the medical test instruments at a plurality of moments, the embodiment can input the instrument operation areas corresponding to the medical test instruments at the plurality of moments into a first feature extraction model, and obtain a first model response to the instrument operation areas through the first feature extraction model, wherein the first model response is the instrument operation features corresponding to the medical test instruments at the plurality of moments.
When the feature extraction is performed on the specimen state areas corresponding to the medical test instruments at the plurality of moments, the specimen state areas corresponding to the medical test instruments at the plurality of moments can be input into a second feature extraction model, and second model responses to the specimen state areas are obtained through the second feature extraction model, wherein the second model responses are specimen state features corresponding to the medical test instruments at the plurality of moments.
In some embodiments, based on a feature extraction model accessed in advance, feature extraction is performed on an instrument operation region and a specimen state region corresponding to a plurality of medical test instruments at a plurality of moments, and before obtaining instrument operation features and specimen state features corresponding to the medical test instruments at the plurality of moments, the method includes: acquiring an instrument operation area training set and a specimen state area training set, wherein the instrument operation area training set comprises a plurality of instrument operation area data and feature marks corresponding to the instrument operation area data, and the specimen state area training set comprises a plurality of specimen state area data and feature marks corresponding to the specimen state area data; training the first feature extraction model through an instrument operation area training set to obtain a first feature extraction model after training is completed; training the second feature extraction model through the specimen state region training set to obtain a trained second feature extraction model; and accessing the first feature extraction model after training and the second feature extraction model after training into a medical test instrument remote monitoring image analysis system.
According to the embodiment, before the feature extraction models are respectively used for carrying out feature extraction on instrument operation areas and specimen state areas corresponding to the medical test instruments at a plurality of moments, the feature extraction models are required to be trained, in the training process, a training set of the instrument operation areas and a training set of the specimen state areas are used for respectively training a first feature extraction model and a second feature extraction model, and the trained first feature extraction model and second feature extraction model are obtained.
The instrument operation area training set includes a plurality of instrument operation area data and feature marks corresponding to the instrument operation area data, for example, the plurality of instrument operation area data may include a plurality of instrument operation areas obtained at historical moments, the instrument operation area includes 3 indicator lamps, and the feature marks corresponding to the instrument operation area obtained at a certain moment may be: the first indicator light is normal and is set to be 1; the second indicator light is normal and is set to be 1; and the third indicator light is normal and is set to 0.
The sample state area training set comprises a plurality of sample state area data and feature marks corresponding to the sample state area data, for example, the sample state area data can comprise a plurality of sample state areas obtained at historical moments, samples are arranged in the sample state areas, the feature marks corresponding to the sample state areas obtained at a certain moment can be position information corresponding to the samples, and the position information can comprise edge coordinate information of the samples.
230. And determining remote monitoring image analysis data according to instrument operation characteristics and specimen state characteristics corresponding to the medical test instruments at a plurality of moments, wherein the remote monitoring image analysis data comprises abnormal analysis data corresponding to the current time period.
In this embodiment, the remote monitoring image analysis data may include various physical environment collection data corresponding to the medical test apparatus, and may further include abnormality analysis data corresponding to the current time period, and the like.
According to the embodiment, the remote monitoring image analysis data can be determined according to the instrument operation characteristics and the specimen state characteristics corresponding to the medical test instrument at a plurality of moments, so that the abnormal analysis data can be further and accurately determined, and the staff can acquire the abnormal information of the medical test instrument in time. According to the embodiment of the specification, the accuracy of identifying the abnormal analysis data of the medical test instrument can be effectively improved, and the time cost of the staff is reduced.
In some embodiments, referring to fig. 6, fig. 6 shows a flow chart for determining anomaly analysis data. As shown in fig. 6, determining remote monitoring image analysis data according to the instrument operation characteristics and the specimen state characteristics corresponding to the medical test instrument at a plurality of moments, wherein the remote monitoring image analysis data comprises abnormal analysis data corresponding to a current time period, and the remote monitoring image analysis data comprises:
2300. Determining operation information of the medical test instrument according to instrument operation characteristics corresponding to the medical test instrument at a plurality of moments;
2310. Determining specimen state information according to specimen state characteristics corresponding to the medical test instruments at a plurality of moments;
2320. and obtaining abnormal analysis data corresponding to the current time period according to the operation information of the medical test instrument and the specimen state information.
In some embodiments, determining medical test instrument operational information from instrument operational characteristics corresponding to the medical test instrument at a number of moments comprises: determining the on-off information of each indicator light of the medical test instrument at a plurality of moments according to the instrument operation characteristics corresponding to the medical test instrument at the plurality of moments; determining a parameter state time sequence chart according to the on-off information of each indicator light; and determining the operation information of the medical test instrument according to the parameter state time sequence diagram.
For example, referring to fig. 7, fig. 7 shows a parameter status timing diagram. As shown in fig. 7, the instrument panel area of a certain test instrument in this embodiment has a plurality of indicator lamps, wherein the plurality of indicator lamps include an oxygen content indicator lamp, and the embodiment can provide stable oxygen to the specimen status area through the oxygen device for the specimen to perform basic physiological activities, wherein the oxygen content indicator lamp is used for indicating an abnormal condition of providing oxygen content to the specimen status area. According to the embodiment, the on-off information of each indicator light of the medical test instrument at a plurality of moments can be determined according to the instrument operation characteristics corresponding to the medical test instrument at the plurality of moments, then the parameter state time sequence diagram corresponding to each indicator light is determined according to the on-off information of each indicator light, and then the state condition of the operation parameters corresponding to each indicator light is further determined according to the parameter state time sequence diagram corresponding to each indicator light. Fig. 7 is a parameter status timing chart corresponding to oxygen content in a specimen status area of the test instrument within 24 hours on a certain day, and in this embodiment, the operation characteristics of the medical test instrument corresponding to each moment on the same day can be obtained first, then the on-off information of the oxygen content indicator lamp is determined according to the operation characteristics of the instrument, and then the on-off information of the oxygen content indicator lamp can be mapped into a preset parameter status output range, so as to determine the parameter status timing chart corresponding to the oxygen content operation parameter, and further determine the operation information of the medical test instrument about the oxygen content operation parameter according to the parameter status timing chart.
As shown in fig. 7, from the parameter status timing chart, it can be determined that three anomalies occur on the same day, 6:00 to 6:00 respectively, for 30 minutes; 11:40 to 11:55 for 15 minutes; 22:30 to 23:15 for 45 minutes. According to the medical test instrument remote monitoring image analysis device, remote monitoring image analysis data can be sent to the instrument sharing platform in real time, when the instrument sharing platform monitors that the parameter state is abnormal according to the parameter state time sequence diagram, the instrument sharing platform can display abnormal information through the display interface, so that workers are informed of the abnormality of instruments in time, and the workers need to go to the site to process in time.
In this embodiment, the instrument operation area corresponding to the medical test instrument includes a plurality of indicator lamps, and the on-off states of the different indicator lamps represent the operation states of the medical test instrument with respect to different parameters. According to the embodiment, the on-off information of each indicator light of the medical test instrument at a plurality of moments can be determined according to the instrument operation characteristics corresponding to the medical test instrument at a plurality of moments, the on-off information of each indicator light can be represented by numerical values, then the numerical values corresponding to the on-off information of each indicator light can be mapped to a preset parameter state output range linearly according to the on-off information of each indicator light, so that a parameter state time sequence diagram corresponding to each operation parameter is determined, and then the operation information of the medical test instrument in the current time period is determined based on the parameter state time sequence diagram.
In some embodiments, determining specimen state information from specimen state features corresponding to the medical test instrument at a number of moments comprises: calculating the similarity between the specimen state characteristics corresponding to the medical test instruments at two adjacent moments in a plurality of moments to obtain a plurality of similarities; determining a plurality of sample standard deviations corresponding to the similarity; and when the standard deviation of the samples corresponding to the similarities is not in the preset standard deviation range, judging that the sample state information is in an abnormal state.
In this embodiment, the specimen state features corresponding to the medical test instrument may include position information corresponding to the specimen, and the embodiment calculates the similarity between the specimen state features corresponding to the medical test instrument at two adjacent moments in a plurality of moments to obtain a plurality of similarities; then determining a plurality of sample standard deviations corresponding to the similarity, wherein the sample standard deviations can be as followsWherein s is a sample standard deviation, x is one of a plurality of similarities,The average value corresponding to the plurality of similarities is n, and the number corresponding to the plurality of similarities is n. After calculating the sample standard deviations corresponding to the plurality of similarities, judging whether the sample standard deviations corresponding to the plurality of similarities are within a preset standard deviation range, if so, judging that the sample state information is in a normal state; if not, judging that the sample standard deviation corresponding to the plurality of similarities is in an abnormal state.
According to the method, the similarity between the sample state characteristics corresponding to the medical test instruments at two adjacent moments in the current time period can be calculated, then the sample standard deviation corresponding to the similarities can be determined, and whether the sample is in the abnormal activity state can be determined by judging whether the sample standard deviation is in the preset standard deviation range or not, so that whether the position of the sample is frequently moved or not can be determined.
The scene image of the medical test instrument obtained by the embodiment of the specification can show the state of a specimen in the instrument, the medical test instrument remote monitoring image analysis device can send remote monitoring image analysis data to the instrument sharing platform in real time, and when the instrument sharing platform monitors that the specimen is in an abnormal active state according to the remote monitoring image analysis data, the instrument sharing platform can display abnormal information through a display interface, so that workers are informed of the abnormality of the specimen in time, and the situation that the workers need to go to the site is treated in time.
According to the embodiment of the specification, scene images respectively corresponding to a plurality of moments of a medical test instrument in a current time period can be acquired first; then, based on a pre-accessed edge detection model, carrying out edge detection on scene images corresponding to a plurality of moments respectively to obtain instrument operation areas and specimen state areas corresponding to the medical test instruments at the plurality of moments; based on a pre-accessed feature extraction model, respectively extracting features of an instrument operation area and a specimen state area corresponding to the medical test instrument at a plurality of moments to obtain instrument operation features and specimen state features corresponding to the medical test instrument at a plurality of moments; and then determining remote monitoring image analysis data according to the instrument operation characteristics and the specimen state characteristics corresponding to the medical test instruments at a plurality of moments. According to the embodiment of the specification, the scene image of the medical test instrument can be subjected to data analysis in real time, the running characteristics and the specimen state characteristics of the instrument are determined through the pre-accessed edge detection model and the feature extraction model, and further the abnormal analysis data are accurately determined through the running characteristics and the specimen state characteristics of the instrument, so that a worker can acquire the abnormal information of the medical test instrument in time. According to the embodiment of the specification, the accuracy of identifying the abnormal analysis data of the medical test instrument can be effectively improved, and the time cost of the staff is reduced. According to the embodiment of the specification, the remote monitoring image analysis data can be sent to the instrument sharing platform in real time, so that the monitoring effectiveness of the instrument sharing platform on the medical test instrument is improved.
The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
Referring to fig. 8, fig. 8 is a schematic structural diagram of a remote monitoring image analysis system for a medical test apparatus according to an embodiment of the present disclosure.
As shown in fig. 8, the medical test instrument remote monitoring image analysis system may include at least an image acquisition module 800, an edge detection module 810, a feature extraction module 820, and an analysis module 830, wherein:
the image acquisition module 800 is configured to acquire scene images corresponding to a plurality of moments of the medical test instrument in a current time period respectively;
the edge detection module 810 is configured to perform edge detection on scene images corresponding to a plurality of moments based on a pre-accessed edge detection model, so as to obtain an instrument operation area and a specimen status area corresponding to a plurality of moment medical test instruments;
The feature extraction module 820 is configured to perform feature extraction on an instrument operation area and a sample state area corresponding to the medical test instrument at a plurality of moments based on the feature extraction model accessed in advance, so as to obtain instrument operation features and sample state features corresponding to the medical test instrument at a plurality of moments;
The analysis module 830 is configured to determine remote monitoring image analysis data according to the instrument operation features and the specimen state features corresponding to the medical test instrument at a plurality of moments, where the remote monitoring image analysis data includes abnormal analysis data corresponding to the current time period.
In some embodiments, the medical test instrument remote monitoring image analysis system further comprises a preprocessing module for: performing image scaling processing on the scene image to obtain a scene image after image scaling, wherein the size of the scene image after image scaling meets the preset size; and performing color coding processing on the scene image after image scaling to obtain a scene image after the color coding processing, wherein the scene image after the color coding processing is a scene image after the preprocessing.
In some embodiments, the edge detection model includes a first edge detection model and a second edge detection model, and the edge detection module 810 includes a first detection module and a second detection module, the first detection module configured to: performing instrument edge detection on scene images corresponding to a plurality of moments respectively based on a first edge detection model which is accessed in advance to obtain instrument operation areas corresponding to medical test instruments in the scene images corresponding to the moments; the second detection module is used for: and carrying out specimen edge detection on the scene images corresponding to the moments respectively based on the second edge detection model which is accessed in advance, and obtaining specimen state areas corresponding to the specimens in the scene images corresponding to the moments.
In some embodiments, the medical test instrument remote monitoring image analysis system further comprises a first training module for: acquiring a first image data training set with a medical test instrument edge mark and a second image data training set with a specimen edge mark; training the first edge detection model through a first image data training set to obtain a trained first edge detection model; training the second edge detection model through a second image data training set to obtain a trained second edge detection model; and respectively accessing the first edge detection model after training and the second edge detection model after training into a remote monitoring image analysis system of the medical test instrument.
In some embodiments, the feature extraction model includes a first feature extraction model and a second feature extraction model, and the feature extraction module 820 includes a first extraction module and a second extraction module, the first extraction module to: extracting the characteristics of instrument operation areas corresponding to the medical test instruments at a plurality of moments through a first characteristic extraction model which is accessed in advance, so as to obtain instrument operation characteristics corresponding to the medical test instruments at the plurality of moments; the second extraction module is used for: and carrying out feature extraction on specimen state areas corresponding to the medical test instruments at a plurality of moments through a second feature extraction model which is accessed in advance, so as to obtain specimen state features corresponding to the medical test instruments at a plurality of moments.
In some embodiments, the medical test instrument remote monitoring image analysis system further comprises a second training module for: acquiring an instrument operation area training set and a specimen state area training set, wherein the instrument operation area training set comprises a plurality of instrument operation area data and feature marks corresponding to the instrument operation area data, and the specimen state area training set comprises a plurality of specimen state area data and feature marks corresponding to the specimen state area data; training the first feature extraction model through an instrument operation area training set to obtain a first feature extraction model after training is completed; training the second feature extraction model through the specimen state region training set to obtain a trained second feature extraction model; and accessing the first feature extraction model after training and the second feature extraction model after training into a medical test instrument remote monitoring image analysis system.
In some embodiments, analysis module 830 includes an anomaly determination submodule to: determining operation information of the medical test instrument according to instrument operation characteristics corresponding to the medical test instrument at a plurality of moments; determining specimen state information according to specimen state characteristics corresponding to the medical test instruments at a plurality of moments; and obtaining abnormal analysis data corresponding to the current time period according to the operation information of the medical test instrument and the specimen state information.
In some embodiments, the anomaly determination submodule includes a run information determination module to: determining the on-off information of each indicator light of the medical test instrument at a plurality of moments according to the instrument operation characteristics corresponding to the medical test instrument at the plurality of moments; determining a parameter state time sequence chart according to the on-off information of each indicator light; and determining the operation information of the medical test instrument according to the parameter state time sequence diagram.
In some embodiments, the anomaly determination submodule includes a specimen state determination module to: calculating the similarity between the specimen state characteristics corresponding to the medical test instruments at two adjacent moments in a plurality of moments to obtain a plurality of similarities; determining a plurality of sample standard deviations corresponding to the similarity; and when the standard deviation of the samples corresponding to the similarities is not in the preset standard deviation range, judging that the sample state information is in an abnormal state.
Based on the content of the remote monitoring image analysis system of the medical test instrument in the embodiments of the present specification, it can be known that the embodiments of the present specification can obtain scene images corresponding to a plurality of moments of the medical test instrument in a current time period; then, based on a pre-accessed edge detection model, carrying out edge detection on scene images corresponding to a plurality of moments respectively to obtain instrument operation areas and specimen state areas corresponding to medical test instruments at the plurality of moments; based on a pre-accessed feature extraction model, respectively extracting features of an instrument operation area and a specimen state area corresponding to the medical test instrument at a plurality of moments to obtain instrument operation features and specimen state features corresponding to the medical test instrument at a plurality of moments; and then determining remote monitoring image analysis data according to the instrument operation characteristics and the specimen state characteristics corresponding to the medical test instruments at a plurality of moments. According to the embodiment of the specification, the scene image of the medical test instrument can be subjected to data analysis in real time, the running characteristics and the specimen state characteristics of the instrument are determined through the pre-accessed edge detection model and the feature extraction model, and further the abnormal analysis data are accurately determined through the running characteristics and the specimen state characteristics of the instrument, so that a worker can acquire the abnormal information of the medical test instrument in time. According to the embodiment of the specification, the accuracy of identifying the abnormal analysis data of the medical test instrument can be effectively improved, and the time cost of the staff is reduced. In addition, the embodiment of the specification can send the remote monitoring image analysis data to the instrument sharing platform in real time, so that the monitoring effectiveness of the instrument sharing platform on the medical test instrument is improved.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are mutually referred to, and each embodiment mainly describes differences from other embodiments. In particular, for the medical test instrument remote monitoring image analysis system embodiment, since it is substantially similar to the medical test instrument remote monitoring image analysis method embodiment, the description is relatively simple, and the relevant points are referred to in the description of the method embodiment.
Please refer to fig. 9, which is a schematic diagram of an electronic device according to an embodiment of the present disclosure.
As shown in fig. 9, the electronic device 900 may include: at least one processor 910, at least one network interface 940, a user interface 930, a memory 950, and at least one communication bus 920.
Wherein the communication bus 920 may be used to implement the connectivity communications of the various components described above.
The user interface 930 may include keys, and the optional user interface may also include a standard wired interface, a wireless interface, among others.
The network interface 940 may include, but is not limited to, a bluetooth module, an NFC module, a Wi-Fi module, and the like.
Wherein the processor 910 may include one or more processing cores. The processor 910 utilizes various interfaces and lines to connect various portions of the overall electronic device 900, perform various functions of the electronic device 900, and process data by executing or executing instructions, programs, code sets, or instruction sets stored in the memory 950, and invoking data stored in the memory 950. Alternatively, the processor 910 may be implemented in at least one hardware form of DSP, FPGA, PLA. The processor 910 may integrate one or a combination of several of a CPU, GPU, modem, and the like. The CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for rendering and drawing the content required to be displayed by the display screen; the modem is used to handle wireless communications. It will be appreciated that the modem may not be integrated into the processor 910 and may be implemented by a single chip.
The memory 950 may include a RAM or a ROM. Optionally, the memory 950 includes a non-transitory computer readable medium. Memory 950 may be used to store instructions, programs, code, sets of codes, or sets of instructions. The memory 950 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the above-described various method embodiments, etc.; the storage data area may store data or the like referred to in the above respective method embodiments. Memory 950 may also optionally be at least one storage device located remotely from the processor 910. The memory 950, which is a computer storage medium, may include an operating system, a network communication module, a user interface module, and a medical test instrument remote monitoring image analysis application. Processor 910 may be used to invoke the medical test instrument remote monitoring image analysis application program stored in memory 950 and perform the steps of medical test instrument remote monitoring image analysis and formulation set forth in the foregoing embodiments.
Embodiments of the present disclosure also provide a computer-readable storage medium having instructions stored therein, which when executed on a computer or processor, cause the computer or processor to perform the steps of one or more of the embodiments shown in fig. 2-7 described above. The above-described constituent modules of the electronic apparatus may be stored in a computer-readable storage medium if implemented in the form of software functional units and sold or used as independent products.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions in accordance with embodiments of the present description are produced in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in or transmitted across a computer-readable storage medium. The computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital subscriber line (Digital Subscriber Line, DSL)), or wireless (e.g., infrared, wireless, microwave, etc.). Computer readable storage media can be any available media that can be accessed by a computer or data storage devices, such as servers, data centers, etc., that contain an integration of one or more available media. The usable medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a digital versatile disk (DIGITAL VERSATILE DISC, DVD)), or a semiconductor medium (e.g., a Solid state disk (Solid STATE DISK, SSD)), or the like.
Those skilled in the art will appreciate that implementing all or part of the above-described embodiment methods may be accomplished by way of a computer program, which may be stored in a computer-readable storage medium, instructing relevant hardware, and which, when executed, may comprise the embodiment methods as described above. And the aforementioned storage medium includes: various media capable of storing program code, such as ROM, RAM, magnetic or optical disks. The technical features in the present examples and embodiments may be arbitrarily combined without conflict.
The above-described embodiments are merely preferred embodiments of the present disclosure, and do not limit the scope of the disclosure, and various modifications and improvements made by those skilled in the art to the technical solutions of the disclosure should fall within the protection scope defined by the claims of the disclosure without departing from the design spirit of the disclosure.