Rpr shaking table detection method and system combined with digital imaging and ai

Abstract

The application relates to the technical field of medical detection, and particularly discloses an RPR shaking table detection method and system combined with digital imaging and AI, which comprises the following steps: uniformly shaking and mixing a serum sample by using an RPR shaking table, adjusting the shaking frequency and amplitude through a real-time monitoring and feedback system; collecting a serum sample image after RPR shaking table treatment in real time through a high-resolution CMOS / CCD digital camera or a camera; pre-processing the collected serum sample image, including denoising, enhancement, segmentation and edge detection, and extracting key features in the serum sample; and the application provides an intelligent RPR shaking table detection method and system combined with digital imaging, AI recognition and remote management, which can significantly improve the detection efficiency, reduce the misjudgment rate, support data storage and remote review, and has a wide market application prospect.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of medical detection, and particularly relates to an RPR shaking table detection method and system based on digital imaging and AI fusion. BACKGROUND

[0002] RPR (Rapid Plasma Reagin) detection is a serological detection method commonly used for the diagnosis of diseases such as syphilis, which evaluates the infection situation by judging the degree of antigen-antibody reaction in serum. Traditional RPR detection methods mostly rely on manual visual observation of the changes of serum samples after reaction on a shaking table, and the reaction intensity is judged by the naked eye. However, this approach has several shortcomings, which seriously affect the accuracy and efficiency of detection.

[0003] Firstly, manual visual judgment is easily affected by subjective factors, resulting in a high misjudgment rate. Due to differences in observation standards and experience of different operators, misjudgment and inconsistency are likely to occur during batch detection. Secondly, the detection efficiency of traditional methods is low, especially when performing batch detection, which requires manual observation and recording of results one by one, consuming time and labor, and is not suitable for modern medical needs. In addition, traditional RPR detection methods usually cannot be digitized for archiving, making it difficult to achieve data storage, management, and remote review, resulting in the inability to effectively trace historical data in the later stage and increasing the difficulty of data management. Finally, existing RPR detection methods lack intelligent analysis functions and cannot automatically distinguish between negative, weak positive, and positive detection results, relying on manual judgment and being easily disturbed by human errors, which cannot meet the needs of high precision and high efficiency.

[0004] Although there are some digital detection devices on the market, the main problems of these devices have not been effectively solved. First of all, most existing devices only have image magnification function and cannot perform intelligent recognition and automatic analysis, still requiring manual intervention and failing to achieve true automation and intelligentization. Secondly, the imaging quality of the device is affected by light and sample position, which may result in unclear imaging or large errors, thereby affecting the accuracy of the interpretation results. More critically, most existing devices cannot be seamlessly connected with hospital information systems (HIS / LIS), cannot support batch management and real-time sharing of data, and limit their popularization and promotion in clinical applications.

[0005] Therefore, it is necessary to propose an RPR shaking table detection method and system based on digital imaging and AI fusion to solve the problems in the prior art that existing devices only have image magnification function, cannot perform intelligent recognition, and the imaging quality is limited, and light and sample position may affect the interpretation results.

[0006] The above information disclosed in this BACKGROUND section is only for increasing the understanding of the background of the application, therefore, it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0007] The present application aims to provide a digital imaging and AI integrated RPR shaker detection method and system to solve the problems raised in the background.

[0008] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0009] The digital imaging and AI integrated RPR shaker detection method comprises:

[0010] The serum sample is uniformly shaken and mixed using the RPR shaker, and the shaking frequency and amplitude are adjusted through the real-time monitoring and feedback system;

[0011] The serum sample image after RPR shaker treatment is collected in real time by a high-resolution CMOS / CCD digital camera or camera;

[0012] The collected serum sample image is preprocessed, including denoising, enhancement, segmentation and edge detection, and key features in the serum sample are extracted;

[0013] The preprocessed serum sample image is analyzed based on a deep learning model to infer whether the serum sample contains target substances, and an analysis result is generated;

[0014] The analysis result is compared with historical data, and a detection report containing the concentration of target substances in the serum sample, the reaction mode and the abnormality conclusion is output;

[0015] The detection report is fed back to the operation interface in real time, and if an abnormal reaction is found, the relevant measures are automatically prompted, and the RPR shaker running state is periodically or according to the preset conditions for self-diagnosis and performance check.

[0016] Preferably, the serum sample is uniformly shaken and mixed using the RPR shaker, and the shaking frequency and amplitude are adjusted through the real-time monitoring and feedback system, comprising:

[0017] Start the RPR shaker, set the initial shaking frequency to 100 rpm and the amplitude to 10 degrees;

[0018] During the shaking process, the sensor collects the serum sample state data in real time, including shaking frequency, amplitude and sample temperature;

[0019] During the shaking process, the particle precipitation in the serum sample is monitored in real time, and once the particle precipitation is found, the shaking frequency and amplitude are immediately adjusted;

[0020] During the shaking process, the shaking frequency and amplitude are remotely adjusted through a wireless connection protocol according to the collected sample state data.

[0021] Preferably, the high-resolution CMOS / CCD digital camera or camera is used to collect the image of the serum sample after RPR shaking table processing in real time, including:

[0022] The high-resolution CMOS / CCD digital camera or camera is initialized, and the collection frequency is set to 30 frames per second;

[0023] The adaptive focusing function of the camera is configured to automatically adjust the focal length according to the distance of the serum sample and the light conditions;

[0024] The multi-spectral LED lighting is enabled, and the optimal lighting spectrum is selected according to the characteristics of the serum sample;

[0025] The serum sample image is collected from multiple angles and synthesized into a panoramic image;

[0026] During the collection process, the position change of the serum sample is monitored in real time, and the camera focus is automatically adjusted to obtain the serum sample image at different positions.

[0027] Preferably, the collected serum sample image is preprocessed, including denoising, enhancement, segmentation, and edge detection, to extract key features in the serum sample, including:

[0028] The deep convolutional denoising autoencoder algorithm is used to denoise the collected serum sample image to eliminate noise in the image;

[0029] The adaptive histogram equalization algorithm is applied to enhance the denoised image to optimize the contrast of the image;

[0030] Adaptive histogram equalization algorithm formula:

[0031]

[0032] In the formula, N(x, y) is the neighborhood of the target pixel point (x, y), h(i, j) is the histogram value of the pixel in the neighborhood, and N is the neighborhood size;

[0033] The segmentation and edge detection of the enhanced image are combined with the adaptive threshold-based segmentation algorithm and the semantic segmentation algorithm to extract the boundary between the serum components and the background.

[0034] Preferably, the preprocessed serum sample image is analyzed based on a deep learning model to determine whether the serum sample contains a target substance, and an analysis result is generated, including:

[0035] A deep learning model is constructed, which adopts a CNN+YOLO target detection architecture and combines an LSTM network to process time sequence information to infer the dynamic changes of particles in serum sample images.

[0036] CNN output features:

[0037] F CNN =CNN(I)

[0038] YOLO target detection:

[0039] D=YOLO(F CNN )

[0040] LSTM network:

[0041]

[0042] In the formula, I is an input image, h t is the hidden state of the LSTM network.

[0043] The model training adopts a clinically labeled RPR shaking table reaction image dataset, and synthesizes sample image data of rare types through data enhancement technology.

[0044] The trained deep learning model is used to analyze the preprocessed serum sample images, and an enhanced learning algorithm is used to dynamically adjust and optimize the errors of the analysis results.

[0045] Preferably, the analysis results are compared with historical data, and a detection report containing the concentration of target substances in the serum sample, the reaction mode, and the abnormality conclusion is output, including:

[0046] The report generation is based on an AI model trained from historical detection data, which automatically analyzes the concentration change trend of various substances in the serum, and intuitively displays through a combination of charts and images.

[0047] The detection report is uploaded to the cloud storage result, and a remote diagnosis function is provided, which uploads data to the hospital LIS / HIS system in real time, and integrates into the hospital laboratory automation system.

[0048] The AI model adopts a federated learning technology to train and optimize the model among multiple medical institutions.

[0049] Local model update and global aggregation formula:

[0050]

[0051] In the formula, is the model parameter of the kth client, w t+1 is the global model parameter.

[0052] Preferably, the RPR shaker performs self-diagnosis and performance check on the running state periodically or according to preset conditions, including:

[0053] When potential faults or performance degradation are found, the operator is immediately warned for timely maintenance or replacement of parts;

[0054] An abnormality detection algorithm based on deep learning is adopted to monitor and warn in real time the performance of key components of the RPR shaker;

[0055] Abnormality detection algorithm formula:

[0056]

[0057] In the formula, ξ z is a slack variable, s is a weight vector of the decision boundary, ρ is a bias term, and R is the total number of samples in the data set.

[0058] The RPR shaker detection system based on digital imaging and AI fusion includes:

[0059] The RPR shaker is used to uniformly shake and mix the serum samples and adjust the shaking frequency and amplitude through a real-time monitoring and feedback system;

[0060] A high-resolution CMOS / CCD digital camera or camera is used to collect images of the serum samples processed by the RPR shaker in real time;

[0061] An image processing unit is used to preprocess the collected serum sample images and extract key features in the serum samples;

[0062] A deep learning model is used to analyze the preprocessed serum sample images, infer whether the serum samples contain target substances, and generate analysis results;

[0063] A report generation unit is used to compare the analysis results with historical data and output a detection report containing the concentration of target substances in the serum samples, reaction patterns, and abnormality conclusions;

[0064] A self-diagnosis and performance check unit is used to perform self-diagnosis and performance check on the running state of the RPR shaker periodically or according to preset conditions.

[0065] Preferably, the RPR shaker includes multiple independently controllable shaking units, each of which automatically adjusts the shaking frequency and amplitude according to the characteristics of the serum samples;

[0066] The real-time monitoring and feedback system includes multiple sensors for collecting serum sample state data in real time and introducing an adaptive control algorithm to adjust the shaking frequency and amplitude through real-time feedback data;

[0067] Adaptive control algorithm formula:

[0068]

[0069] In the formula, u(t) is a control signal, e(t) is an error, M(t) is a time-varying gain, alpha is a compensation coefficient, is the error integral from time 0 to the current time t;

[0070] The high-resolution CMOS / CCD digital camera or camera includes an adaptive focusing function, automatically adjusts the focal length according to the distance of the serum sample and the light condition, and enables multi-spectral LED illumination, selects the best illumination spectrum according to the characteristics of the serum sample;

[0071] The image processing unit adopts a deep convolution denoising autoencoder algorithm to denoise the collected images, and applies an adaptive histogram equalization algorithm to enhance the denoised images;

[0072] The generated adversarial network is introduced to enhance the image quality, and the loss function formula of the generated adversarial network is as follows:

[0073]

[0074] In the formula, D(a) is the output probability of the discriminator for real data, G(r) is the output of the generator, Pdata(a) is the real data distribution, P r (r) is the distribution of the latent space.

[0075] Preferably, the deep learning model adopts a CNN+YOLO architecture combined with a long short-term memory network of time sequence information to infer the dynamic changes of particles in the serum sample image, and dynamically adjusts and optimizes the errors of the analysis results through a reinforcement learning algorithm;

[0076] The graph neural network is introduced to optimize the spatial relationship and dynamic change between different regions in the serum sample image;

[0077] Information propagation formula of graph neural network:

[0078]

[0079] In the formula, is the representation of node d in the l+1 layer, B(d) is the neighbor node set of node d, A (l) and b (l) are the weights and biases of the l layer, and sigma is the activation function;

[0080] The report generation unit automatically analyzes the concentration trend of various substances in the serum based on the AI model trained by historical detection data, and intuitively displays through a combination of charts and images.

[0081] The self-diagnosis and performance check unit adopts an abnormality detection algorithm based on deep learning to monitor the performance of key components of the RPR shaker in real time and provide early warning.

[0082] Compared with the prior art, the present application has the following advantages:

[0083] The present application significantly improves detection accuracy and efficiency, reduces errors caused by manual intervention, automatically identifies target substances and their concentration changes in serum samples, and provides high-quality detection reports. In addition, the present application also has self-diagnosis and performance check functions to ensure the stability and reliability of the equipment during operation, meeting the needs of modern medicine for high precision and high efficiency.

[0084] In summary, the present application proposes an intelligent RPR shaker detection method and system combining digital imaging, AI recognition and remote management, which can significantly improve detection efficiency, reduce misjudgment rate, and support data storage and remote review, and has broad market application prospects. BRIEF DESCRIPTION OF DRAWINGS

[0085] Figure 1 The present application is a digital imaging and AI fusion RPR shaker detection method flowchart;

[0086] Figure 2 The present application is a digital imaging and AI fusion RPR shaker detection system framework diagram. DETAILED DESCRIPTION

[0087] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0088] Embodiment 1:

[0089] Please refer to Figure 1 The digital imaging and AI fusion RPR shaker detection method includes:

[0090] The RPR shaker is used to uniformly shake and mix the serum sample, and the shaking frequency and amplitude are adjusted by the real-time monitoring and feedback system;

[0091] Start the RPR shaker, set the initial shaking frequency to 100 revolutions per minute and the amplitude to 10 degrees;

[0092] During the shaking process, the state data of the serum sample is collected in real time by the sensor, including the shaking frequency, amplitude, and sample temperature;

[0093] During the shaking process, the particle precipitation in the serum sample is monitored in real time, and once the particle precipitation is found, the shaking frequency and amplitude are immediately adjusted;

[0094] During the shaking process, the shaking frequency and amplitude are remotely adjusted through the wireless connection protocol according to the collected sample state data.

[0095] Further, the RPR shaking table device can ensure uniform mixing of the serum sample during the shaking process and effectively avoid particle precipitation by real-time monitoring and feedback adjustment of the shaking frequency and amplitude. At the same time, the device has real-time data acquisition and remote adjustment functions, dynamically optimizes the shaking parameters according to the sample state, improves the detection efficiency and intelligent level, and solves the problems of manual operation and low efficiency in the traditional method.

[0096] Real-time acquisition of the serum sample image after RPR shaking table processing through high-resolution CMOS / CCD digital camera or camera;

[0097] Initialize the high-resolution CMOS / CCD digital camera or camera, resolution ≥ 12 million pixels, set the acquisition frequency to 30 frames per second;

[0098] Configure the adaptive focusing function of the camera, automatically adjust the focal length according to the distance and light conditions of the serum sample;

[0099] Enable multi-spectral LED lighting, select the best lighting spectrum according to the characteristics of the serum sample;

[0100] Collect serum sample images from multiple angles and synthesize panoramic images;

[0101] Real-time monitoring of the position change of the serum sample during the acquisition process, and automatically adjusting the camera focus to obtain serum sample images at different positions.

[0102] Further, the image acquisition method realizes high-precision real-time image acquisition of the serum sample through high-resolution CMOS / CCD digital camera, intelligent focusing, multi-spectral LED lighting, and polarized light reflection elimination technology. It supports multi-angle shooting and panoramic image synthesis, improves the comprehensiveness and accuracy of image acquisition, and helps subsequent intelligent analysis and result interpretation.

[0103] Preprocessing of the collected serum sample images, including denoising, enhancement, segmentation, and edge detection, to extract key features in the serum sample;

[0104] The deep convolutional denoising autoencoder algorithm is used to denoise the collected serum sample images, eliminating the noise in the images.

[0105] The adaptive histogram equalization algorithm is applied to enhance the denoised images, optimizing the contrast of the images.

[0106] The adaptive threshold-based segmentation algorithm and semantic segmentation algorithm are combined to segment and edge detect the enhanced images, extracting the boundaries of serum components and background.

[0107] Further, the image preprocessing process provides high-quality and detailed image data for subsequent analysis through deep convolutional denoising autoencoder, histogram equalization, adaptive threshold segmentation, and semantic segmentation algorithms, thereby improving the accuracy and reliability of serum sample analysis.

[0108] The preprocessed serum sample images are analyzed based on a deep learning model to determine whether the serum sample contains target substances, generating analysis results.

[0109] A deep learning model is constructed using the CNN+YOLO target detection architecture and combining the LSTM network to process time series information to infer the dynamic changes of particles in the serum sample images.

[0110] The model training uses the clinically annotated RPR shaking bed reaction image dataset (including negative, weak positive, positive, and multiple concentration gradient image data), and synthesizes rare sample image data through data augmentation techniques.

[0111] The trained deep learning model is used to analyze the preprocessed serum sample images, and the reinforcement learning algorithm is used to dynamically adjust and optimize the errors in the analysis results.

[0112] Classification criteria:

[0113] The results are automatically classified as negative (-), weak positive (±), positive (+), and strong positive (++) based on the size, shape, and distribution density of the agglutination particles.

[0114] The results are automatically classified as negative (-), weak positive (±), positive (+), and strong positive (++) based on the size, shape, and distribution density of the agglutination particles.

[0115] The results are automatically classified as negative (-), weak positive (±), positive (+), and strong positive (++) based on the size, shape, and distribution density of the agglutination particles.

[0116] Further, through the clinically annotated RPR shaking bed reaction image dataset and data augmentation techniques, the model can accurately process various types of sample data and automatically classify them as negative, weak positive, positive, or strong positive. Combined with the reinforcement learning algorithm, the system dynamically optimizes the analysis results, reduces the error rate, and ensures that the accuracy is over 98%, greatly improving the accuracy and efficiency of serum sample analysis.

[0117] The analysis result is compared with historical data, and a detection report containing the concentration of the target substance in the serum sample, the reaction mode, and whether the conclusion is abnormal is output;

[0118] The report generation is based on an AI model trained on historical detection data, which automatically analyzes the concentration trend of various substances in the serum. The results are displayed intuitively through charts and images.

[0119] The detection report is uploaded to the cloud storage and provides remote diagnosis function. It supports Wi-Fi / Bluetooth / USB interface, real-time data upload to hospital LIS / HIS system, and integration into hospital laboratory automation system through API open interface.

[0120] The AI model uses federated learning technology to train and optimize the model among multiple medical institutions.

[0121] Further, by automatically analyzing the serum sample and combining historical data to generate detailed detection reports, the concentration of substances, reaction patterns, and abnormal conclusions are provided to intuitively display the test results. Using AI models and federated learning techniques, data is shared among multiple medical institutions and the model is optimized to improve diagnostic accuracy. At the same time, cloud storage and remote diagnosis functions are supported, enabling real-time data upload and integration with hospital LIS / HIS systems, promoting efficient circulation and intelligent management of medical data.

[0122] The detection report is fed back to the operation interface in real time. If an abnormal reaction is found, the relevant measures are automatically prompted, and the RPR shaker running state is periodically or according to the preset conditions for self-diagnosis and performance check;

[0123] When potential faults or performance degradation are found, the operator is immediately warned for timely maintenance or replacement of parts.

[0124] An abnormal detection algorithm based on deep learning is used to monitor and warn the performance of key components of the RPR shaker in real time.

[0125] Further, by feeding back the detection report in real time and automatically prompting the relevant measures for abnormal reactions, the reaction speed and accuracy during operation are improved. Combined with periodic self-diagnosis and deep learning-based abnormal detection algorithms, the performance of key components of the RPR shaker is monitored in real time, and warnings are issued in a timely manner when potential faults or performance degradation occur, helping operators to perform timely maintenance or replacement of parts, ensuring the stability and reliability of the equipment.

[0126] Embodiment 2:

[0127] Please refer to Figure 2 As shown in the figure, the RPR shaker detection system based on digital imaging and AI fusion includes:

[0128] RPR shaker, for uniform shaking and mixing of serum samples, and adjusting the shaking frequency and amplitude through a real-time monitoring and feedback system;

[0129] The RPR shaker includes multiple independently controllable shaking units, each of which automatically adjusts the shaking frequency and amplitude according to the characteristics of the serum sample;

[0130] The RPR shaker base is driven by a brushless motor, ensuring stability during long-term operation, and can adjust the oscillation frequency and amplitude (range: 50-150 rpm), optimizing the antigen-antibody mixing effect;

[0131] The real-time monitoring and feedback system includes multiple sensors for real-time acquisition of serum sample state data, and introduces an adaptive control algorithm to adjust the shaking frequency and amplitude through real-time feedback data;

[0132] High-resolution CMOS / CCD digital camera or camera, for real-time acquisition of serum sample images after RPR shaker processing;

[0133] High-resolution CMOS / CCD digital camera or camera includes adaptive focusing function, automatically adjusts focal length according to distance and light conditions of serum sample, and enables multi-spectral LED lighting, selects the best lighting spectrum according to the characteristics of the serum sample;

[0134] Image processing unit for preprocessing of collected serum sample images, extracting key features in serum sample;

[0135] The image processing unit uses a deep convolutional denoising autoencoder algorithm to denoise the collected images, and applies an adaptive histogram equalization algorithm to enhance the denoised images;

[0136] Deep learning model for analyzing preprocessed serum sample images, inferring whether the serum sample contains target substances, and generating analysis results;

[0137] The deep learning model uses a CNN+YOLO architecture combined with a long short-term memory network that incorporates temporal information to infer the dynamic changes of particles in the serum sample image, and uses reinforcement learning algorithms to dynamically adjust and optimize the errors in the analysis results;

[0138] Introducing a graph neural network to optimize the spatial relationships and dynamic changes between different regions in the serum sample image;

[0139] Report generation unit for comparing analysis results with historical data, outputting a detection report containing the concentration of target substances in the serum sample, reaction patterns, and whether the conclusion is abnormal;

[0140] The report generation unit automatically analyzes the concentration trend of various substances in the serum based on the AI model trained by historical detection data, and intuitively displays the results through charts and images.

[0141] The self-diagnosis and performance check unit is used for periodic or preset condition-based self-diagnosis and performance check of the RPR shaker running state.

[0142] The self-diagnosis and performance check unit uses a deep learning-based anomaly detection algorithm to monitor and warn the performance of the RPR shaker's key components in real time.

[0143] The user interface uses a 7-inch touch screen to display the detection progress and analysis results, and supports remote operation (PC / mobile terminal synchronous monitoring).

[0144]

[0145] Application example: medical staff use RPR shaker to detect patient serum samples

[0146] In a hospital's clinical laboratory, medical staff need to use the RPR shaker to detect a patient's serum sample to determine whether it contains specific target substances such as certain pathogens or biomarkers.

[0147] I. Application process

[0148] The medical staff first prepare the patient's serum sample and place it on the RPR shaker.

[0149] Start the RPR shaker through the operation interface, set the initial shaking frequency to 100 rpm and the amplitude to 10 degrees. The shaker starts to uniformly shake the serum sample to ensure that the components in the sample are fully mixed.

[0150] During the shaking process, the real-time monitoring and feedback system collects real-time state data of the serum sample through sensors, including shaking frequency, amplitude, and sample temperature.

[0151] The medical staff can observe these data through the operation interface, and once they find abnormal conditions such as particle precipitation, the system will automatically adjust the shaking frequency and amplitude to ensure uniform mixing of the sample.

[0152] After the serum sample has been shaken and mixed for a period of time, the medical staff start the high-resolution CMOS / CCD digital camera or camera to collect real-time images of the serum sample processed by the RPR shaker. The camera or camera is equipped with adaptive focusing function, multi-spectral LED lighting and polarization light reflection elimination function to ensure the quality of the collected images is clear and accurate.

[0153] The collected serum sample images are transmitted to the image processing unit for preprocessing. Preprocessing includes steps such as denoising, enhancement, segmentation, and edge detection to extract key features in the serum sample.

[0154] After preprocessing, the preprocessed serum sample images are analyzed based on a deep learning model (CNN+YOLO architecture combined with LSTM network). The model infers whether the serum sample contains target substances and generates analysis results.

[0155] The analysis results are compared with historical data to generate a detection report containing the concentration of target substances in the serum sample, reaction patterns, and abnormality conclusions.

[0156] The detection report is fed back to the operation interface in real time, and medical personnel can intuitively view and analyze the report. If an abnormal reaction is found, the system will also automatically prompt relevant measures so that medical personnel can take further action in a timely manner.

[0157] During the entire detection process, the RPR shaker also performs regular or preset condition-based self-diagnosis and performance checks. An abnormality detection algorithm based on deep learning is used to monitor and warn the performance of key components in real time. Once potential failures or performance degradation are detected, the system immediately alerts medical personnel for timely maintenance or replacement of components.

[0158] II. Advantages and Experience

[0159] Improve detection accuracy and efficiency: Through high-resolution imaging, image preprocessing, deep learning model analysis, and real-time feedback system, the RPR shaker significantly improves detection accuracy and efficiency, allowing medical personnel to obtain accurate detection results more quickly.

[0160] Reduce human intervention errors: Automatically identify target substances and their concentration changes in serum samples, reducing human intervention errors and improving detection accuracy and reliability.

[0161] Support data storage and remote review: Provide high-quality detection reports and support data storage, historical data comparison, and remote diagnosis, enhancing data management and sharing capabilities. Medical personnel can easily view and analyze historical data for remote review and consultation.

[0162] Self-diagnosis and performance check function: Ensure the stability and reliability of the device during operation, reducing maintenance costs. Medical personnel can use the RPR shaker for detection work with greater confidence.

[0163] Example 3:

[0164] The embodiment of the present application also provides a computer readable storage medium, and the computer readable storage medium stores the program of the RPR shaking table detection system of digital imaging and AI fusion as any one of the above, the program is executed by the processor to realize the processes of the RPR shaking table detection system embodiments, and the same technical effects can be achieved, to avoid repetition, which will not be repeated here. The computer readable storage medium is, for example, a read-only memory (Read-Only Memory, ROM for short), a random access memory (Random Access Memory, RAM for short), a magnetic disk or an optical disk, etc.

[0165] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example" or "some examples" means that the specific features, structures, materials or characteristics described in combination with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, the different embodiments or examples described in the present specification and the features of the different embodiments or examples can be combined and combined by those skilled in the art without contradiction.

[0166] In the drawings of the embodiments of the present application, only the structures related to the embodiments of the present application are involved, and other structures can be referred to the general design. In the case of no conflict, the same embodiment and different embodiments of the present application can be combined with each other.

[0167] The flowchart shown in the drawings is only an example description, and does not necessarily include all the contents and operations / steps, and does not necessarily be executed in the order described. For example, some operations / steps can also be decomposed, combined or partially combined, so that the actual execution order may be changed according to the actual situation.

[0168] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various changes, modifications, replacements and deformations can be made to the embodiments without departing from the principles and spirits of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A RPR shaker detection method combining digital imaging and AI, characterized in that, The application comprises the following steps: Using an RPR shaker to uniformly shake and mix the serum sample, adjusting the shaking frequency and amplitude through a real-time monitoring and feedback system; Using a high-resolution CMOS / CCD digital camera or camera to real-time capture the image of the serum sample processed by the RPR shaker; Pretreating the captured serum sample image, including denoising, enhancing, segmentation, and edge detection, to extract key features in the serum sample; Analyzing the pretreated serum sample image based on a deep learning model to infer whether the serum sample contains target substances and generate an analysis result; Comparing the analysis result with historical data to output a detection report containing the concentration of target substances in the serum sample, the reaction mode, and whether the conclusion is abnormal; Real-time feeding the detection report to the operation interface and automatically prompting relevant measures if an abnormal reaction is found, and periodically or according to preset conditions, performing self-diagnosis and performance check on the RPR shaker's running state; The use of an RPR shaker to uniformly shake and mix the serum sample, adjusting the shaking frequency and amplitude through a real-time monitoring and feedback system, includes: Starting the RPR shaker, setting the initial shaking frequency to 100 revolutions per minute and the amplitude to 10 degrees; During the shaking process, real-time capturing the serum sample state data, including shaking frequency, amplitude, and sample temperature, through sensors; During the shaking process, real-time monitoring the particle sedimentation in the serum sample and adjusting the shaking frequency and amplitude immediately once particle sedimentation is found; During the shaking process, remotely adjusting the shaking frequency and amplitude through a wireless connection protocol based on the collected sample state data.

2. The digital imaging and AI-fused RPR shaker detection method of claim 1, wherein: The use of a high-resolution CMOS / CCD digital camera or camera to real-time capture the image of the serum sample processed by the RPR shaker, includes: Initializing the high-resolution CMOS / CCD digital camera or camera, setting the capture frequency to 30 frames per second; Configuring the camera's adaptive focusing function to automatically adjust the focal length based on the distance and light conditions of the serum sample; Enabling multi-spectral LED lighting to select the best lighting spectrum based on the characteristics of the serum sample; Capturing the serum sample image from multiple angles and synthesizing a panoramic image; Real-time monitoring the position changes of the serum sample during the capturing process and automatically adjusting the camera focus to obtain serum sample images at different positions.

3. The digital imaging and AI-fused RPR shaker detection method of claim 2, wherein: The pretreatment of the captured serum sample image, including denoising, enhancing, segmentation, and edge detection, to extract key features in the serum sample, includes: Using a deep convolutional denoising autoencoder algorithm to denoise the captured serum sample image to eliminate noise in the image; Applying an adaptive histogram equalization algorithm to enhance the denoised image to optimize the contrast of the image; Combining an adaptive threshold-based segmentation algorithm with a semantic segmentation algorithm to segment and edge detect the enhanced image to extract the boundaries of serum components and background.

4. The digital imaging and AI-fused RPR shaker detection method of claim 3, wherein: The analysis of the pretreated serum sample image based on a deep learning model to infer whether the serum sample contains target substances and generate an analysis result, includes: A deep learning model is constructed, which adopts a CNN+YOLO target detection architecture and combines an LSTM network to process time series information to infer the dynamic changes of particles in serum sample images. The model training uses a clinically annotated RPR centrifuge reaction image dataset, and synthesizes rare sample image data through data augmentation techniques. The trained deep learning model is used to analyze preprocessed serum sample images, and an enhanced learning algorithm is used to dynamically adjust and optimize the analysis results.

5. The digital imaging and AI-fused RPR shaker detection method of claim 4, wherein: The analysis results are compared with historical data to output a detection report containing the concentration of target substances in the serum sample, the reaction pattern, and the abnormality conclusion, including: The AI model trained based on historical detection data automatically analyzes the concentration trend of various substances in the serum, and visually displays it through charts and images. The detection report is uploaded to the cloud storage result, and the remote diagnosis function is provided, which uploads data to the hospital LIS / HIS system in real time and integrates it into the hospital laboratory automation system. The AI model uses federated learning technology to train and optimize the model among multiple medical institutions.

6. The digital imaging and AI-fused RPR shaker detection method of claim 5, wherein: The RPR centrifuge performs self-diagnosis and performance check regularly or according to preset conditions, including: When potential faults or performance degradation are found, the operator is immediately warned for timely maintenance or replacement of parts. An abnormality detection algorithm based on deep learning is used to monitor and warn the performance of key components of the RPR centrifuge in real time.

7. The RPR shaking table detection system combined with digital imaging and AI, characterized in that, Including: The RPR centrifuge is used to uniformly shake and mix the serum sample, and the real-time monitoring and feedback system adjusts the shaking frequency and amplitude. A high-resolution CMOS / CCD digital camera or camera is used to capture real-time images of the serum sample processed by the RPR centrifuge. An image processing unit is used to preprocess the collected serum sample images and extract key features from the serum sample. A deep learning model is used to analyze the preprocessed serum sample images, infer whether the serum sample contains target substances, and generate analysis results. A report generation unit is used to compare the analysis results with historical data to output a detection report containing the concentration of target substances in the serum sample, the reaction pattern, and the abnormality conclusion. A self-diagnosis and performance check unit is used to periodically or according to preset conditions, self-diagnose and check the performance of the RPR centrifuge. The RPR centrifuge is used to uniformly shake and mix the serum sample, and the real-time monitoring and feedback system adjusts the shaking frequency and amplitude, including: Start the RPR centrifuge, set the initial shaking frequency to 100 revolutions per minute, and the amplitude to 10 degrees. During the shaking process, the sensor collects real-time sample state data, including shaking frequency, amplitude, and sample temperature. During the shaking process, the particle sedimentation in the serum sample is monitored in real time, and the shaking frequency and amplitude are adjusted immediately once the particle sedimentation is found. During the shaking process, the shaking frequency and amplitude are remotely adjusted through a wireless connection protocol based on the collected sample state data.

8. The digital imaging and AI-fused RPR shaker detection system of claim 7, wherein: The RPR shaker includes multiple independently controllable shaking units, each of which automatically adjusts the shaking frequency and amplitude according to the characteristics of the serum sample; The real-time monitoring and feedback system includes multiple sensors for real-time acquisition of serum sample state data, and introduces an adaptive control algorithm to adjust the shaking frequency and amplitude through real-time feedback data; The high-resolution CMOS / CCD digital camera or camera includes an adaptive focusing function that automatically adjusts the focal length according to the distance and light conditions of the serum sample, and enables multi-spectral LED lighting to select the best lighting spectrum according to the characteristics of the serum sample; The image processing unit uses a deep convolutional denoising autoencoder algorithm to denoise the collected images, and applies an adaptive histogram equalization algorithm to enhance the denoised images; Introducing a generative adversarial network to enhance image quality.

9. The digital imaging and AI-fused RPR shaker detection system of claim 8, wherein: The deep learning model uses a CNN+YOLO architecture combined with a long short-term memory network that incorporates temporal information to infer the dynamic changes of particles in the serum sample image, and uses reinforcement learning algorithms to dynamically adjust and optimize the errors in the analysis results; Introducing a graph neural network to optimize the spatial relationships and dynamic changes between different regions in the serum sample image; The report generation unit automatically analyzes the concentration trend of various substances in the serum based on the AI model trained from historical detection data, and visually displays the results through a combination of charts and images; The self-diagnosis and performance check unit uses a deep learning-based anomaly detection algorithm to monitor and alert the performance of key components of the RPR shaker in real time.

Images