Disinfection method and device for inspection instruments of clinical laboratory

Abstract

The invention provides a disinfection method and device for an inspection instrument of a clinical laboratory, and the method comprises the steps: carrying out the feature enhancement processing of a preprocessed video image, and obtaining a feature-enhanced video image; calibrating the video image after feature enhancement to form a training data set; training the deep learning model by using the training data set to obtain an inspection appliance type identification model; inputting the video image of the target inspection appliance into an inspection appliance type identification model, and judging the type of the inspection appliance; the disinfection time and temperature of the steam sterilization cabinet are controlled according to the type of the inspection appliance. According to the invention, the deep learning model is trained by using the feature-enhanced video image to obtain the inspection appliance type identification model, and the disinfection time and temperature of the steam sterilization cabinet can be controlled based on the type of the inspection appliance, so that automatic disinfection management of the inspection appliance can be realized, manual intervention is reduced, and the disinfection efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of disinfection of testing instruments, and more specifically, relates to a disinfection method and device for testing instruments in a laboratory. Background Art

[0002] In the field of medical health, the laboratory is an important part of disease diagnosis. The test instruments used by the laboratory, such as blood collection needles, culture dishes, test tubes, microscope slides, etc., are in direct contact with biological samples such as blood and body fluids. Therefore, ensuring the sterility of these instruments is the key to preventing cross infection and ensuring the accuracy of test results. Traditional disinfection methods include high temperature and high pressure sterilization, chemical disinfectant immersion, ultraviolet irradiation, etc., but these methods require real-time monitoring by medical staff. Therefore, there is an urgent need for an efficient, safe, and automated disinfection method to improve the hygiene level of medical facilities and ensure the health and safety of medical staff. Summary of the invention

[0003] In order to solve the above problems, the object of the present invention is to provide a method and device for disinfecting laboratory testing instruments.

[0004] A method for disinfecting laboratory testing instruments, comprising:

[0005] Step 1: Acquire a video image of the inspection device;

[0006] Step 2: preprocessing the video image to obtain a preprocessed video image;

[0007] Step 3: performing feature enhancement processing on the preprocessed video image to obtain a feature enhanced video image;

[0008] Step 4: Calibrate the feature-enhanced video image to form a training data set;

[0009] Step 5: Use the training data set to train the deep learning model to obtain a test tool type recognition model;

[0010] Step 6: Input the video image of the target inspection tool into the inspection tool type recognition model to determine the type of the inspection tool;

[0011] Step 7: Control the sterilization time and temperature of the steam sterilizer according to the type of test equipment.

[0012] Preferably, the step 2: preprocessing the video image to obtain a preprocessed video image includes:

[0013] Step 2.1: Use a square window to divide the video image into multiple blocks;

[0014] Step 2.2: Calculate the gradient value of the pixel point at the center of the block in the corresponding block;

[0015] Step 2.3: construct a smooth detection model according to the gradient value;

[0016] Step 2.4: Preprocess the video image using the smooth detection model to obtain a preprocessed video image.

[0017] Preferably, the step 2.3: constructing a smooth detection model according to the gradient value comprises:

[0018] Using the formula:

[0019]

[0020] Construct a smooth denoising model; where Represents the gradient value of the pixel point x at the center of the block in the corresponding block, is the pixel mean of the block, u(x) represents the pixel value of pixel x, u median (x) represents the median value of the pixel point x at the center of the block in the corresponding block.

[0021] Preferably, the step 2.4: preprocessing the video image using the smooth detection model to obtain a preprocessed video image includes:

[0022] Step 2.4.1: Calculate the smoothing value of the central pixel of the block using the smoothing detection model;

[0023] Step 2.4.2: Filter out all pixels whose smoothing values ​​are greater than the denoising threshold as abnormal pixels;

[0024] Step 2.4.3: Obtain adjacent frames of the video image where the abnormal pixel is located;

[0025] Step 2.4.4: Preprocess the block where the abnormal pixel point is located based on the pixel values ​​at adjacent frames to obtain a preprocessed video image.

[0026] Preferably, in step 2.4.3, the formula is used:

[0027]

[0028] The block where the abnormal pixel point is located is preprocessed to obtain the preprocessed video image; where I(x j ,y j ) indicates that the video image of the jth adjacent frame is in (x j ,y j ) position, σ represents the standard deviation, N represents the number of adjacent frames, and p x,y(t) represents the pixel value of the block where the abnormal pixel is located.

[0029] Preferably, the step 3: performing feature enhancement processing on the preprocessed video image to obtain a feature-enhanced video image comprises:

[0030] Step 3.1: Convert the preprocessed video image into the HSI color space to form an HSI image;

[0031] Step 3.2: Based on the values ​​of the preprocessed video image in the RGB color channel, the HSI image is enhanced to obtain an enhanced HSI image; wherein the enhancement processing formula is:

[0032]

[0033] Where S″ represents the saturation of the enhanced HSI image, S represents the saturation of the original HSI image, I represents the brightness of the original HSI image, min(R,G,B) represents the minimum value of the preprocessed video image in the red, green and blue color channels, and max(R,G,B) represents the maximum value of the preprocessed video image in the red, green and blue color channels.

[0034] Preferably, the step 5: using the training data set to train the deep learning model to obtain a test tool type recognition model comprises:

[0035] Step 5.1: Input the training data set into the convolutional neural network, and train the convolutional neural network through a loss function; wherein the loss function is:

[0036]

[0037] Among them, i represents the number of images in the training data set, y i represents the true label of the image, Represents the result predicted by the convolutional neural network;

[0038] Step 5.2: Optimize the network parameters in the convolutional neural network using a step-down model to minimize the loss function and obtain a test tool type recognition model; wherein the step-down model is:

[0039]

[0040] Among them, Δ i represents the convolution kernel weight after the i-th iteration, represents the bias value of the l-th layer of the neural network, η represents the learning rate, Represents the partial derivative of the loss function with respect to the convolution kernel weights.

[0041] The present invention also provides a disinfection device for laboratory testing instruments, comprising:

[0042] A video image acquisition module, used to acquire video images of the inspection device;

[0043] A preprocessing module, used for preprocessing the video image to obtain a preprocessed video image;

[0044] A feature enhancement module, used for performing feature enhancement processing on the preprocessed video image to obtain a feature enhanced video image;

[0045] An image calibration module, used to calibrate the feature-enhanced video image to form a training data set; a training module, used to train the deep learning model using the training data set to obtain a test tool type recognition model;

[0046] A judgment module, used for inputting the video image of the target inspection tool into the inspection tool type recognition model to judge the type of the inspection tool;

[0047] The UV sterilization module is used to control the sterilization time and temperature of the steam sterilizer according to the type of test equipment.

[0048] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory and the processor are connected via the bus, and wherein the computer program, when executed by the processor, implements the steps in the above-mentioned method for disinfecting testing instruments in a laboratory.

[0049] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method for disinfecting testing instruments in a laboratory are implemented.

[0050] The beneficial effect of the disinfection method and device for laboratory testing instruments provided by the present invention is that: compared with the prior art, the present invention obtains a testing instrument type recognition model by training a deep learning model using feature-enhanced video images, and can control the disinfection time and temperature of the steam sterilizer based on the type of testing instrument, thereby realizing automated disinfection management of testing instruments, reducing manual intervention, and improving disinfection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 A schematic diagram of a method for disinfecting laboratory instruments provided by an embodiment of the present invention;

[0053] Figure 2 A schematic diagram of a disinfection device for laboratory testing instruments provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0055] To achieve the above object, the technical solution adopted by the present invention is:

[0056] See also Figure 1 , a method for disinfecting laboratory testing instruments, comprising:

[0057] Step 1: Acquire a video image of the inspection device;

[0058] Step 2: preprocessing the video image to obtain a preprocessed video image;

[0059] Furthermore, the step 2 comprises:

[0060] Step 2.1: Use a square window to divide the video image into multiple blocks;

[0061] Step 2.2: Calculate the gradient value of the pixel point at the center of the block in the corresponding block;

[0062] Step 2.3: construct a smooth detection model according to the gradient value;

[0063] In step 2.3, the present invention may adopt the formula:

[0064] f(x)=1-exp(▽u(x)-▽u mean (x) 2 -u(x)-u median (x)

[0065] Construct a smooth denoising model; where ▽u(x) represents the gradient value of the pixel point x at the center of the block in the corresponding block, ▽u mean (x) is the pixel mean of the block, u(x) represents the pixel value of pixel point x, and u median (x) represents the median value of the pixel point x at the center of the block in the corresponding block.

[0066] Step 2.4: Preprocess the video image using the smooth detection model to obtain a preprocessed video image.

[0067] Among them, step 2.4 includes:

[0068] Step 2.4.1: Calculate the smoothing value of the central pixel of the block using the smoothing detection model;

[0069] Step 2.4.2: Filter out all pixels whose smoothing values ​​are greater than the denoising threshold as abnormal pixels;

[0070] Step 2.4.3: Obtain adjacent frames of the video image where the abnormal pixel is located;

[0071] In step 2.4.3, the formula can be used:

[0072]

[0073] The block where the abnormal pixel point is located is preprocessed to obtain the preprocessed video image; where I(x j ,y j ) indicates that the video image of the jth adjacent frame is in (x j ,y j ) position, σ represents the standard deviation, N represents the number of adjacent frames, and p x,y (t) represents the pixel value of the block where the abnormal pixel is located.

[0074] Step 2.4.4: Preprocess the block where the abnormal pixel point is located based on the pixel values ​​at adjacent frames to obtain a preprocessed video image.

[0075] The present invention can reduce abnormal pixels caused by noise by comparing and weighting pixel values ​​of adjacent frames, thereby improving the overall quality of the image.

[0076] Step 3: performing feature enhancement processing on the preprocessed video image to obtain a feature enhanced video image;

[0077] Furthermore, step 3 includes:

[0078] Step 3.1: Convert the preprocessed video image into the HSI color space to form an HSI image;

[0079] Step 3.2: Based on the values ​​of the preprocessed video image in the RGB color channel, the HSI image is enhanced to obtain an enhanced HSI image; wherein the enhancement processing formula is:

[0080]

[0081] Where S″ represents the saturation of the enhanced HSI image, S represents the saturation of the original HSI image, I represents the brightness of the original HSI image, min(R,G,B) represents the minimum value of the preprocessed video image in the red, green and blue color channels, and max(R,G,B) represents the maximum value of the preprocessed video image in the red, green and blue color channels.

[0082] The image enhancement method of the present invention enhances the saturation of pixels while keeping the hue of each pixel unchanged, which not only avoids the occurrence of color cast but also enhances the image interpretation effect.

[0083] Step 4: Calibrate the feature-enhanced video image to form a training data set;

[0084] Step 5: Use the training data set to train the deep learning model to obtain a test tool type recognition model;

[0085] Furthermore, step 5 includes:

[0086] Step 5.1: Input the training data set into the convolutional neural network, and train the convolutional neural network through a loss function; wherein the loss function is:

[0087]

[0088] Among them, i represents the number of images in the training data set, y i represents the true label of the image, Represents the result predicted by the convolutional neural network;

[0089] Step 5.2: Optimize the network parameters in the convolutional neural network using a step-down model to minimize the loss function and obtain a test tool type recognition model; wherein the step-down model is:

[0090]

[0091] Among them, Δ i represents the convolution kernel weight after the i-th iteration, represents the bias value of the l-th layer of the neural network, η represents the learning rate, Represents the partial derivative of the loss function with respect to the convolution kernel weights.

[0092] The present invention optimizes network parameters by utilizing a step-descent model, thereby reducing the time required for convolutional neural network convergence, accelerating training speed, and saving computing resources and time.

[0093] Step 6: Input the video image of the target inspection tool into the inspection tool type recognition model to determine the type of the inspection tool;

[0094] Step 7: Control the sterilization time and temperature of the steam sterilizer according to the type of test equipment.

[0095] The present invention obtains a test instrument type recognition model by using a deep learning model trained with feature-enhanced video images, and can control the disinfection time and temperature of the steam sterilizer based on the type of the test instrument. This can achieve automated disinfection management of the test instruments, reduce manual intervention, and improve disinfection efficiency.

[0096] See also Figure 2 The present invention also provides a disinfection device for laboratory testing instruments, comprising:

[0097] A video image acquisition module, used to acquire a video image of the inspection device;

[0098] A preprocessing module, used for preprocessing the video image to obtain a preprocessed video image;

[0099] A feature enhancement module, used for performing feature enhancement processing on the preprocessed video image to obtain a feature enhanced video image;

[0100] An image calibration module, used to calibrate the feature-enhanced video image to form a training data set; a training module, used to train the deep learning model using the training data set to obtain a test tool type recognition model;

[0101] A judgment module, used for inputting the video image of the target inspection tool into the inspection tool type recognition model to judge the type of the inspection tool;

[0102] The sterilization module is used to control the sterilization time and temperature of the steam sterilizer according to the type of test equipment.

[0103] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory and the processor are connected via the bus, and wherein the computer program, when executed by the processor, implements the steps in the above-mentioned method for disinfecting testing instruments in a laboratory.

[0104] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps in the above-mentioned method for disinfecting laboratory testing instruments are implemented. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present invention are the same as the beneficial effects of the method for disinfecting laboratory testing instruments described in the above-mentioned technical solution, which will not be repeated here.

[0105] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for disinfecting laboratory instruments, characterized in that: include: Step 1: Acquire a video image of the inspection device; Step 2: preprocessing the video image to obtain a preprocessed video image; Step 3: performing feature enhancement processing on the preprocessed video image to obtain a feature enhanced video image; Step 4: Calibrate the feature-enhanced video image to form a training data set; Step 5: Use the training data set to train the deep learning model to obtain a test tool type recognition model; Step 6: Input the video image of the target inspection tool into the inspection tool type recognition model to determine the type of the inspection tool; Step 7: Control the sterilization time and temperature of the steam sterilizer according to the type of test equipment.

2. A method for disinfecting laboratory instruments as claimed in claim 1, characterized in that: The step 2: preprocessing the video image to obtain a preprocessed video image, comprises: Step 2.1: Use a square window to divide the video image into multiple blocks; Step 2.2: Calculate the gradient value of the pixel point at the center of the block in the corresponding block; Step 2.3: construct a smooth detection model according to the gradient value; Step 2.4: Preprocess the video image using the smooth detection model to obtain a preprocessed video image.

3. A method for disinfecting laboratory instruments as claimed in claim 2, characterized in that: The step 2.3: constructing a smooth detection model according to the gradient value, comprising: Using the formula: f(x)=1-exp(|▽u(x)-▽u mean (x)| 2 -|u(x)-u median (x)|) Construct a smooth denoising model; where ▽u(x) represents the gradient value of the pixel point x at the center of the block in the corresponding block, ▽u mean (x) is the pixel mean of the block, u(x) represents the pixel value of pixel point x, and u median (x) represents the median value of the pixel point x at the center of the block in the corresponding block.

4. A method for disinfecting laboratory instruments as claimed in claim 3, characterized in that: The step 2.4: preprocessing the video image using the smooth detection model to obtain a preprocessed video image includes: Step 2.4.1: Calculate the smoothing value of the central pixel of the block using the smoothing detection model; Step 2.4.2: Filter out all pixels whose smoothing values ​​are greater than the denoising threshold as abnormal pixels; Step 2.4.3: Obtain adjacent frames of the video image where the abnormal pixel is located; Step 2.4.4: Preprocess the block where the abnormal pixel point is located based on the pixel values ​​at adjacent frames to obtain a preprocessed video image.

5. A method for disinfecting laboratory instruments according to claim 4, characterized in that: In step 2.4.3, the formula is used: The block where the abnormal pixel point is located is preprocessed to obtain the preprocessed video image; where I(x j ,y j ) indicates that the video image of the jth adjacent frame is in (x j ,y j ) position, σ represents the standard deviation, N represents the number of adjacent frames, and p x,y (t) represents the pixel value of the block where the abnormal pixel is located.

6. A method for disinfecting laboratory instruments according to claim 5, characterized in that: The step 3: performing feature enhancement processing on the preprocessed video image to obtain a feature-enhanced video image, comprises: Step 3.1: Convert the preprocessed video image into the HSI color space to form an HSI image; Step 3.2: Based on the values ​​of the preprocessed video image in the RGB color channel, the HSI image is enhanced to obtain an enhanced HSI image; wherein the enhancement processing formula is: Where S″ represents the saturation of the enhanced HSI image, S represents the saturation of the original HSI image, I represents the brightness of the original HSI image, min(R,G,B) represents the minimum value of the preprocessed video image in the red, green and blue color channels, and max(R,G,B) represents the maximum value of the preprocessed video image in the red, green and blue color channels.

7. A method for disinfecting laboratory instruments as claimed in claim 6, characterized in that: The step 5: using the training data set to train the deep learning model to obtain a test tool type recognition model, includes: Step 5.1: Input the training data set into the convolutional neural network, and train the convolutional neural network through a loss function; wherein the loss function is: Among them, i represents the number of images in the training data set, y i represents the true label of the image, Represents the result predicted by the convolutional neural network; Step 5.2: Optimize the network parameters in the convolutional neural network using a step-down model to minimize the loss function and obtain a test tool type recognition model; wherein the step-down model is: Among them, Δ i represents the convolution kernel weight after the i-th iteration, represents the bias value of the l-th layer of the neural network, η represents the learning rate, Represents the partial derivative of the loss function with respect to the convolution kernel weights.

8. A disinfection device for laboratory testing instruments, characterized in that: include: A video image acquisition module, used to acquire a video image of the inspection device; A preprocessing module, used for preprocessing the video image to obtain a preprocessed video image; A feature enhancement module, used for performing feature enhancement processing on the preprocessed video image to obtain a feature enhanced video image; An image calibration module is used to calibrate the feature-enhanced video image to form a training data set; A training module, used to train the deep learning model using the training data set to obtain a test tool type recognition model; A judgment module, used for inputting the video image of the target inspection tool into the inspection tool type recognition model to judge the type of the inspection tool; The sterilization module is used to control the sterilization time and temperature of the steam sterilizer according to the type of test equipment.

9. An electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, wherein: When the computer program is executed by the processor, the steps of a method for disinfecting laboratory instruments as described in claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for disinfecting laboratory instruments as described in claims 1 to 7 are implemented.

Images