A method, system, readable storage medium and device for plasma waste liquid color analysis

Abstract

The application relates to a plasma waste liquid color analysis method, a system, a readable storage medium and equipment, the method comprising the following steps: collecting an image of a transparent container containing plasma waste liquid to obtain an initial image; wherein the plasma waste liquid is obtained after plasma is filtered by adopting a double filtration plasma replacement method; the initial image is preprocessed and color space conversion is performed to obtain an image to be analyzed; a color feature vector is extracted from the image to be analyzed; and the color feature vector is input into a pre-trained classification model to obtain a classification result; through computer vision technology, non-invasive and rapid identification of the nature of DFPP waste liquid is realized, objective and convenient reference indexes are provided for clinical doctors to adjust treatment parameters, evaluate the metabolic state of patients and the risk of cardiovascular diseases, detection cost is effectively reduced, and diagnosis and treatment efficiency is improved.

Description

Technical Field

[0001] This invention relates to the fields of medical testing and biomedical engineering technology, specifically to a method, system, readable storage medium, and device for color analysis of plasma waste liquid. Background Technology

[0002] Double Filtration Plasmapheresis (DFPP) is a widely used plasma separation technique in clinical blood purification. It separates plasma through a primary plasma separator, then removes large molecular pathological products (such as immune complexes, lipoproteins, inflammatory factors, etc.) through a secondary plasma component separator (with smaller membrane pores), and finally reinfuses beneficial components such as albumin back to the patient.

[0003] During DFPP clinical procedures, the waste plasma discharged from the secondary separator contains high concentrations of pathogenic factors, and its appearance, color, and properties often contain important pathological information. For example: Milky white or cloudy (chylous): This usually indicates that the patient's plasma contains extremely high concentrations of triglycerides or low-density lipoprotein (LDL), which is closely related to severe metabolic syndrome and cardiovascular diseases (such as coronary heart disease and hyperlipidemia); Red or pink: This may indicate tubing compression or excessive transmembrane pressure leading to red blood cell destruction (hemolysis); Dark yellow or brown: may indicate hyperbilirubinemia or specific liver metabolic abnormalities.

[0004] Currently, clinical monitoring of waste liquid properties mainly relies on visual judgment by medical personnel. This method has significant limitations: Highly subjective: The judgment results lack consistency due to the influence of ambient light (such as the color temperature of fluorescent lights), observation angle, and individual differences in visual perception among medical staff. Unquantifiable: Subtle color gradations are not visible to the naked eye (such as a slight increase in the degree of hemolysis), making it difficult to establish a standardized graded early warning mechanism.

[0005] While existing biochemical tests or omics analyses can accurately reflect molecular changes in plasma, their testing cycles are long (requiring laboratory submission), costly, and usually require additional blood sampling, making them unsuitable for rapid bedside assessment or dynamic monitoring during treatment. Therefore, utilizing computer technology to transform readily available waste fluid color information from DFPP into quantifiable, objective indicators reflecting the patient's metabolic status and cardiovascular risk represents a new direction with significant clinical application value. Summary of the Invention

[0006] To address the technical problems of long detection cycles, high costs, and complex operations in existing technologies for DFPP waste liquid, which make it unsuitable for rapid assessment or dynamic monitoring, this invention provides a method, system, readable storage medium, and device for color analysis of plasma waste liquid.

[0007] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: A method for color analysis of blood plasma waste liquid includes the following steps: An initial image is obtained by acquiring an image of a transparent container filled with plasma waste fluid; wherein, the plasma waste fluid is plasma waste fluid obtained after filtering plasma using a double filtration plasma replacement method; The initial image is preprocessed and its color space is converted to obtain the image to be analyzed; Extract color feature vectors from the image to be analyzed; The color feature vector is input into a pre-trained classification model to obtain the classification result.

[0008] The beneficial effects of this invention are: objective quantification and high sensitivity. This invention, through color space conversion (such as HSV or Lab) rather than simple grayscale processing, can effectively preserve the hue and saturation information of the waste liquid, thereby accurately distinguishing trace hemolysis (changes in red tones) or chyle (changes in white / transparency) that are difficult for the human eye to discern, eliminating errors caused by ambient light and subjective human experience. This invention achieves non-invasive and rapid identification of the properties of DFPP waste liquid through computer vision technology, providing clinicians with objective and convenient reference indicators for adjusting treatment parameters, assessing patients' metabolic status and cardiovascular risk, effectively reducing detection costs and improving diagnostic efficiency.

[0009] Based on the above technical solution, the present invention can be further improved as follows.

[0010] Further, images of the transparent container holding the blood plasma waste were acquired to obtain an initial image, including the following steps: Place the transparent container containing the plasma waste liquid under a preset standard light source environment; An initial image is obtained by taking a picture of the plasma waste liquid area of ​​the transparent container placed under a preset standard light source environment using an image acquisition device; wherein, the image acquisition device is any one of an industrial camera, a mobile terminal camera, or a medical imaging probe.

[0011] Furthermore, the initial image is preprocessed and its color space is converted to obtain the image to be analyzed, including the following steps: The waste liquid area in the initial image is identified, and the background area and the high-reflection area are removed to obtain the region of interest image. The region of interest image is converted from the RGB color space to a target color space; wherein the target color space includes at least one of the HSV color space, Lab color space, and YCbCr color space.

[0012] Further, extracting color feature vectors from the image to be analyzed includes the following steps: Calculate the statistical values ​​of the region of interest image in each channel of the target color space; wherein the statistical values ​​include at least the channel pixel mean, standard deviation, or color histogram features; The multiple statistical values ​​are combined into a dataset to obtain the color feature vector.

[0013] Furthermore, the classification model is trained through the following steps: Acquire images of plasma waste fluid samples from multiple groups with known pathological types, and extract their corresponding sample color feature vectors; The sample color feature vector is used as input, and the corresponding known pathological type is used as output label to construct a training dataset; The machine learning model is trained under supervision using the training dataset, and the model parameters are optimized to obtain the classification model.

[0014] Furthermore, the machine learning model includes any one of Support Vector Machine (SVM), Random Forest, K-Nearest Neighbors (KNN), Naive Bayes, or artificial neural networks.

[0015] Furthermore, the known pathological types include at least: normal waste fluid, hemolytic waste fluid, high bilirubin waste fluid, and chylous waste fluid.

[0016] To address the aforementioned technical problems, this invention also provides a system for analyzing the color of plasma waste liquid, the technical details of which are as follows: A system for color analysis of blood plasma waste liquid, comprising: An image acquisition module is used to acquire images of a transparent container filled with plasma waste liquid to obtain an initial image; wherein, the plasma waste liquid is plasma waste liquid obtained after filtering plasma using a double filtration plasma replacement method; The image processing module is used to preprocess and convert the color space of the initial image to obtain the image to be analyzed; The feature extraction module is used to extract color feature vectors from the image to be analyzed; The classification calculation module is used to input the color feature vector into a pre-trained classification model to obtain the classification result.

[0017] To address the aforementioned technical problems, the present invention also provides a computer-readable storage medium, the technical content of which is as follows: A computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, it implements the above-mentioned plasma waste color analysis method.

[0018] To address the aforementioned technical problems, the present invention also provides a computer device, the technical content of which is as follows: A computer device includes a memory and a processor, characterized in that the memory stores a computer program, and when the processor executes the computer program, it implements the above-mentioned plasma waste color analysis method. Attached Figure Description

[0019] Figure 1 This is a flowchart of a plasma waste liquid color analysis method according to an embodiment of the present invention; Figure 2 This is a structural block diagram of a plasma waste liquid color analysis system according to an embodiment of the present invention. Detailed Implementation

[0020] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0021] Example 1: Blood plasma waste liquid color analysis method: like Figure 1 As shown, this embodiment provides a method for color analysis of plasma waste liquid, including the following steps: S1. Collect images of plasma waste to obtain initial images.

[0022] This step aims to obtain macroscopic image data containing rich color information, rather than microscopic cell images.

[0023] The specific steps are as follows: Set up the collection environment: Place the transparent waste liquid bag or waste liquid tube containing DFPP waste liquid in front of a background board equipped with a standard light source (such as high color rendering LED white light, color temperature 5000K-6500K) to eliminate the interference of ambient stray light on color interpretation.

[0024] Image capture: Using image acquisition equipment (such as industrial cameras, high-pixel mobile terminal cameras, or dedicated probes integrated into dialysis equipment) to capture images of the target area of ​​the transparent container and obtain initial images in RGB format.

[0025] Compared to the traditional method of preparing slides and observing with a microscope, this embodiment uses a macroscopic imaging method, which does not require damaging the airtightness of the tubing, is simple to operate, and meets the needs of rapid bedside testing.

[0026] S2. Preprocess and convert the color space of the initial image to obtain the image to be analyzed.

[0027] Raw images often contain background noise and their RGB color space is not suitable for human visual perception, so they need to be processed: Region of Interest (ROI) Extraction: Image segmentation algorithms (such as Otsu thresholding or edge detection operators) are used to identify the liquid surface region in the image, and pipe walls, labels, background plates, and areas of high-brightness reflection on the liquid surface are removed to obtain a clean ROI image.

[0028] Color space conversion: Convert the ROI image from the RGB color space to the target color space. In this embodiment, the HSV (Hue, Saturation, Value) color space or the Lab color space is preferred.

[0029] Principle: In the RGB color space, changes in brightness will cause the R, G, and B components to change simultaneously, making it difficult to distinguish the essence of the color; while in the HSV color space, the H (hue) component can independently express color categories such as "red, yellow, and white", S (saturation) can express the intensity of the color, and V (brightness) can reflect the light transmittance (turbidity) of the liquid.

[0030] S3. Extract color feature vectors from the image to be analyzed.

[0031] Instead of extracting a single grayscale value, a multi-dimensional feature vector is constructed to fully describe the properties of the waste liquid: Statistical feature calculation: Calculate the statistical values ​​of the ROI image in each channel of the target color space (such as HSV).

[0032] Calculate the mean (Hmean) and variance (Hvar) of the H channel: used to distinguish the dominant color tone (e.g., red vs. yellow).

[0033] Calculate the mean of the S channel (Smean): used to determine the vibrancy of colors (e.g., bright red vs. light pink).

[0034] Calculate the mean of the V channel (Vmean): used to determine the turbidity of the liquid (e.g., clear vs. milky).

[0035] Vector construction: Combine the above statistical values ​​to form the color feature vector X = [Hmean, Smean, Vmean, Hvar,…] to be analyzed.

[0036] S4. Input the color feature vector into the pre-trained classification model to obtain the classification result.

[0037] This step utilizes machine learning algorithms to automate the identification process.

[0038] Model training process (preliminary): Data collection: Collect a large number of clinically confirmed historical images of DFPP waste liquid and label them with pathological labels (e.g., label 0 = normal pale yellow, label 1 = hemolytic red, label 2 = chylous white, label 3 = high bilirubin deep yellow).

[0039] Model selection: Choose a machine learning algorithm suitable for multi-classification tasks, such as support vector machine (SVM), random forest, or artificial neural network (ANN).

[0040] Training and optimization: The model is trained under supervision by taking the color feature vector of historical data as input and the pathological label as output. The model parameters are optimized by cross-validation, and finally a well-trained classification model is obtained.

[0041] Online inference: Input the color feature vector of the current waste liquid obtained in step S3 into the above-trained classification model. The model outputs the probability value of the current waste liquid belonging to each pathological type. The type with the highest probability is taken as the final classification result.

[0042] The pathological type and clinical indications of plasma waste fluid are determined based on the classification results.

[0043] Based on the identified pathology type and a pre-defined medical rule base, the corresponding clinical assessment information is output: Emulsified waste liquid (white / turbid): This indicates that the waste liquid contains a high concentration of triglycerides or lipoproteins.

[0044] Clinical significance: It assists doctors in assessing the severity of metabolic syndrome or severe hyperlipidemia in patients, dynamically monitors the effectiveness of lipid-lowering treatment by changes in the "whiteness" of waste fluid, and indicates the risk of cardiovascular disease.

[0045] Hemolytic waste liquid (red): This indicates the presence of free hemoglobin in the waste liquid.

[0046] Clinical significance: This indicates possible tubing compression, excessive transmembrane pressure, or increased fragility of the patient's own red blood cells. Equipment should be checked immediately to prevent medical accidents.

[0047] High bilirubin waste liquid (dark yellow / tea color): indicates abnormal bilirubin metabolism.

[0048] Clinical significance: It helps in assessing liver function status.

[0049] Normal waste fluid (pale yellow and clear): indicates a stable treatment process; continue monitoring.

[0050] Example 2: Plasma Waste Color Analysis System Based on Dual-Filtration Plasma Replacement Example 2: Plasma Waste Color Analysis System Based on Dual-Filtration Plasma Replacement like Figure 2As shown, in some other embodiments, a plasma waste color analysis system is also provided, comprising: Image acquisition module: Used to control the image acquisition device to acquire images of a transparent container filled with blood plasma waste under standard light source conditions, and obtain an initial image; Image processing module: used to preprocess the initial image (such as de-glare and background removal) and perform color space conversion (such as RGB to HSV) to obtain the image to be analyzed; Feature extraction module (original grayscale value extraction module): used to extract multidimensional color feature vectors (including statistical features such as hue, saturation, and brightness) from the image to be analyzed. The classification calculation module (formerly the classification value calculation module) is used to input the color feature vector into a pre-trained machine learning classification model, perform inference calculations, and output the classification result. Based on the classification results, output the pathological type of plasma waste fluid (such as chylous, hemolytic, hyperbilirubinic, etc.) and the corresponding clinical prompts.

[0051] Among them, the image acquisition module, image processing module, feature extraction module, and classification calculation module are program modules, or computer devices (such as medical workstations or tablet computers) that integrate corresponding functional software.

[0052] Summary of the technical effects of the embodiments of the present invention: The present invention can infer the pathological state of a patient or the operating status of equipment by directly analyzing the color features of plasma waste fluid images without the need for additional invasive blood sampling.

[0053] Highly efficient and convenient: The testing process can be completed in seconds, fully meeting the needs of rapid bedside screening or real-time dynamic monitoring during DFPP treatment.

[0054] Objective quantification: Through HSV / Lab color space analysis and machine learning algorithms, the digital correspondence between waste liquid properties and pathological indicators was realized, effectively avoiding subjective errors in manual visual observation (such as ambient light interference and visual fatigue).

[0055] Significant clinical value: This system can not only monitor equipment safety (such as hemolysis warning), but also serve as an auxiliary assessment tool for the treatment effect of cardiovascular diseases such as metabolic syndrome, coronary heart disease, and hyperlipidemia (by monitoring the clearance of chylous waste fluid), as well as an auxiliary prompting tool for abnormal liver and kidney function, and has strong clinical applicability.

[0056] Example 3: Computer-readable storage medium.

[0057] In other embodiments, a computer-readable storage medium is also provided, the storage medium storing a computer program or computer instructions, which, when executed by a computer processor, implement any step of the plasma waste color analysis method described in Embodiment 1 above.

[0058] The storage medium can be an internal storage unit of any data processing device described in any of the foregoing embodiments, such as a hard disk or memory. The storage medium can also be an external storage device of any data processing device, such as a plug-in hard disk, smart memory card, SD card, flash memory card, etc., mounted on the device. Furthermore, the storage medium can include both internal storage units and external storage devices of any data processing device. The computer-readable storage medium is used to store the computer program and other programs and data required by the data processing device, and can also be used to temporarily store data that has been output or will be output.

[0059] Example 4: Computer equipment.

[0060] In other embodiments, a computer device is also provided, including a memory and one or more processors, wherein executable code is stored in the memory, and when the one or more processors execute the executable code, any step of the plasma waste color analysis method described in Embodiment 1 is implemented.

[0061] The memory can be an internal storage unit of any data processing device described in any of the foregoing embodiments, such as a hard disk or RAM. The memory can also be an external storage device of any data processing device. The computer device includes, but is not limited to, desktop computers, laptops, tablets, smartphones, or medical device terminals with integrated computing units.

[0062] It should be noted that the analytical results provided by this invention are only an objective classification of the physical properties of waste liquid, intended to provide intermediate data reference for subsequent medical decisions, and are not directly used as disease diagnosis results or treatment basis.

[0063] 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, improvements, etc., made within the concept and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for color analysis of blood plasma waste, characterized in that, Includes the following steps: An initial image is obtained by acquiring an image of a transparent container filled with plasma waste fluid; wherein, the plasma waste fluid is plasma waste fluid obtained after filtering plasma using a double filtration plasma replacement method; The initial image is preprocessed and its color space is converted to obtain the image to be analyzed; Extract color feature vectors from the image to be analyzed; The color feature vector is input into a pre-trained classification model to obtain the classification result.

2. The method for color analysis of plasma waste liquid according to claim 1, characterized in that, To obtain an initial image, images of a transparent container holding waste plasma are acquired, including the following steps: Place the transparent container containing the plasma waste liquid under a preset standard light source environment; An initial image is obtained by taking a picture of the plasma waste liquid area of ​​the transparent container placed under a preset standard light source environment using an image acquisition device; wherein, the image acquisition device is any one of an industrial camera, a mobile terminal camera, or a medical imaging probe.

3. The method for color analysis of plasma waste liquid according to claim 2, characterized in that, The initial image is preprocessed and its color space is converted to obtain the image to be analyzed, including the following steps: The waste liquid area in the initial image is identified, and the background area and the high-reflection area are removed to obtain the region of interest image. The region of interest image is converted from the RGB color space to a target color space; wherein the target color space includes at least one of the HSV color space, Lab color space, and YCbCr color space.

4. The method for color analysis of plasma waste liquid according to claim 3, characterized in that, Extracting color feature vectors from the image to be analyzed includes the following steps: Calculate the statistical values ​​of the region of interest image in each channel of the target color space; wherein the statistical values ​​include at least the channel pixel mean, standard deviation, or color histogram features; The multiple statistical values ​​are combined into a dataset to obtain the color feature vector.

5. The method for color analysis of plasma waste liquid according to claim 1, characterized in that, The classification model is trained through the following steps: Acquire images of plasma waste fluid samples from multiple groups with known pathological types, and extract their corresponding sample color feature vectors; The sample color feature vector is used as input, and the corresponding known pathological type is used as output label to construct a training dataset; The machine learning model is trained under supervision using the training dataset, and the model parameters are optimized to obtain the classification model.

6. The method for color analysis of plasma waste liquid according to claim 5, characterized in that, The machine learning model includes any one of support vector machines, random forests, K-nearest neighbors, Naive Bayes, or artificial neural networks.

7. The method for color analysis of plasma waste liquid according to claim 5, characterized in that, The known pathological types include at least: normal waste fluid, hemolytic waste fluid, high bilirubin waste fluid, and chylous waste fluid.

8. A system employing the plasma waste color analysis method as described in any one of claims 1 to 7, characterized in that, include: An image acquisition module is used to acquire images of a transparent container filled with plasma waste liquid to obtain an initial image; wherein, the plasma waste liquid is plasma waste liquid obtained after filtering plasma using a double filtration plasma replacement method; The image processing module is used to preprocess and convert the color space of the initial image to obtain the image to be analyzed; The feature extraction module is used to extract color feature vectors from the image to be analyzed; The classification calculation module is used to input the color feature vector into a pre-trained classification model to obtain the classification result.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the plasma waste color analysis method as described in any one of claims 1 to 7.

10. A computer device, comprising a memory and a processor, characterized in that, The memory stores a computer program, and when the processor executes the computer program, it implements the plasma waste color analysis method as described in any one of claims 1 to 7.

Images