Portable detection device and method for rapidly identifying peritonitis

Abstract

The invention provides a portable detection device and method for rapidly identifying peritonitis, and relates to the technical field of medical instruments, the device comprises: an illumination module for irradiating a peritoneal dialysis liquid sample; the sensor module is used for capturing a sensing data set of the state change of the peritoneal dialysis liquid sample in the irradiation process; the detection chip downloads the trained peritonitis risk detection model, performs data detection on the sensing data set and outputs a risk detection result; and the display module is used for displaying on a display screen. The peritonitis risk detection device can solve the technical problem that a user is difficult to independently complete rapid and accurate detection in a family or community environment due to poor portability and complex operation of a traditional peritonitis detection device, and achieves the purposes of rapidly analyzing a peritonitis sample through optical characteristics and a machine learning algorithm, outputting a peritonitis risk detection result in real time and improving the detection efficiency. And the accuracy and the efficiency of early peritonitis recognition of the user or the medical personnel are improved.

Description

Technical Field

[0001] This application relates to the technical field of medical devices, and particularly to a portable detection device and method for quickly identifying peritonitis. Background Art

[0002] Peritonitis is one of the common and serious complications for peritoneal dialysis users, and its early identification is crucial for the prognosis of users. With the popularization of peritoneal dialysis technology, more and more users need to conduct daily monitoring in the home or community environment to promptly detect the early symptoms of peritonitis. However, the existing peritonitis detection devices are usually bulky, complex to operate, and require professional personnel to operate, making it difficult to meet the needs of users for daily self-monitoring. In addition, these devices often have low detection efficiency and cannot quickly output results, resulting in users being unable to take treatment measures in a timely manner. At the same time, the existing devices lack standardized detection means during the detection process, and the detection results are easily affected by subjective factors, thereby increasing the detection time and reducing the detection efficiency. Summary of the Invention

[0003] The purpose of this application is to provide a portable detection device and method for quickly identifying peritonitis, so as to solve the technical problem that traditional peritonitis detection devices are difficult for users to independently complete rapid and accurate detection in the home or community environment due to poor portability and complex operation.

[0004] In view of the above problems, in the first aspect of this application, a portable detection device for quickly identifying peritonitis is provided. The device includes: a lighting module, which takes a peritoneal dialysis fluid sample and activates the lighting module to irradiate the peritoneal dialysis fluid sample; a sensor module, which captures a sensing data set of the state change of the peritoneal dialysis fluid sample during the irradiation process through the sensor module; a detection chip, which has a trained peritonitis risk detection model downloaded, and performs data detection on the sensing data set according to the peritonitis risk detection model, and outputs a risk detection result; a display module, which is used to display the risk detection result on the display screen of the display module.

[0005] In the second aspect of this application, a portable detection method for quickly identifying peritonitis is provided. The method is implemented by the portable detection device for quickly identifying peritonitis described in the first aspect. Among them, the method includes: taking a peritoneal dialysis fluid sample and irradiating the peritoneal dialysis fluid sample; capturing a sensing data set of the state change of the peritoneal dialysis fluid sample during the irradiation process; performing data detection on the sensing data set according to the peritonitis risk detection model, and outputting a risk detection result; and displaying the risk detection result on the display screen.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The device provided by the embodiment of the present application is composed of a light module, a sensor module, a detection chip, and a display module; among them, the light module takes a peritoneal dialysis fluid sample and starts the light module to irradiate the peritoneal dialysis fluid sample; the sensor module captures a sensing data set of the state change of the peritoneal dialysis fluid sample during the irradiation process through the sensor module; the detection chip has a trained peritonitis risk detection model downloaded, and performs data detection on the sensing data set according to the peritonitis risk detection model, and outputs a risk detection result; the display module is used to display the risk detection result on the display screen of the display module. It effectively solves the technical problem that the traditional peritonitis detection device is difficult for users to independently complete rapid and accurate detection in a home or community environment due to poor portability and complex operation, and achieves the technical effect of rapidly analyzing the peritoneal dialysis fluid sample through optical characteristics and machine learning algorithms, and real-time outputting the peritonitis risk detection result, improving the accuracy and efficiency of early identification of peritonitis by users or medical staff.

[0008] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specific embodiments of the present application are specifically given. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. Brief Description of the Drawings

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings required to be used in the description of the embodiments or the prior art. Obviously, the drawings described below are only exemplary, and for those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0010] Figure 1 It is a schematic structural diagram of a portable detection device for rapidly identifying peritonitis in the present application.

[0011] Figure 2 It is a schematic flowchart of a portable detection method for rapidly identifying peritonitis in the present application.

[0012] Description of the reference numerals: light module 11, sensor module 12, detection chip 13, display module 14. Detailed Description of the Embodiments

[0013] By providing a portable detection device and method for quickly identifying peritonitis, this application solves the technical problem that traditional peritonitis detection devices are difficult for users to independently complete rapid and accurate detection in home or community environments due to poor portability and complex operations, achieving the technical effect of quickly analyzing peritoneal dialysis fluid samples through optical characteristics and machine learning algorithms, and real-time outputting the detection results of peritonitis risk, improving the accuracy and efficiency of early identification of peritonitis by users or medical staff.

[0014] Next, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited by the example embodiments described here. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of this application. Additionally, it should be noted that for the sake of description, only the parts related to this application are shown in the drawings rather than all of them.

[0015] Embodiment 1, please refer to the attached Figure 1 drawings. This application provides a portable detection device for quickly identifying peritonitis, and the device specifically includes the following components:

[0016] The light irradiation module 11 takes the peritoneal dialysis fluid sample and starts the light irradiation module 11 to irradiate the peritoneal dialysis fluid sample.

[0017] Specifically, in the light irradiation module 11, first, it is necessary to take the peritoneal dialysis fluid sample of the user and place it in the detection device. Subsequently, start the light irradiation module 11 in the device, and this module will emit light of a specific wavelength to irradiate the peritoneal dialysis fluid sample. Through the irradiation of light, the components in the peritoneal dialysis fluid sample will interact with the light, for example, optical phenomena such as absorption, scattering, or reflection. These optical characteristics will change due to the presence or absence of inflammatory markers (such as white blood cells, bacteria, or other biomolecules) in the peritoneal dialysis fluid. Subsequently, the subsequent sensor module 12 will capture these optical changes for analysis by the machine learning model. The whole process is fast, non-invasive, and does not require complex operations, providing a convenient detection means for users.

[0018] Furthermore, the light irradiation module 11 includes a light irradiation adjustment unit, and the light irradiation adjustment unit includes:

[0019] A parameter definition unit for defining a light irradiation adjustment parameter library; a sensitivity association unit for establishing a sensitivity index between the preset difference features and each light irradiation adjustment parameter in the defined light irradiation adjustment parameter library; a parameter configuration unit for extracting N light irradiation adjustment parameters greater than the preset sensitivity according to the magnitude of the sensitivity index to configure the adjustment parameters of the light irradiation adjustment unit, where N is a positive integer greater than or equal to 1.

[0020] In a feasible implementation, the light module 11 includes a light adjustment unit. This light adjustment unit is used to dynamically adjust the light parameters according to the characteristics of the peritoneal dialysis fluid sample and the detection requirements. Its internal includes a parameter definition unit, a sensitivity correlation unit, and a parameter configuration unit. The parameter definition unit is responsible for defining the light adjustment parameter library. This parameter library contains a series of light parameters that may affect the optical characteristics of the peritoneal dialysis fluid sample. For example, the wavelength, intensity, irradiation time, incident angle, etc. of light. These parameters are preset according to the physical and chemical characteristics of the peritoneal dialysis fluid sample and the optical response characteristics of peritonitis-related markers. The sensitivity correlation unit, through data analysis, establishes the sensitivity indexes of each parameter in the preset difference characteristics and the light adjustment parameter library. The preset difference characteristics refer to the optical characteristics differences of the peritoneal dialysis fluid in the healthy state and the peritonitis state, which are obtained by comparing the distribution differences between the healthy sensing data set and the non-healthy sensing data set, including changes in light absorption rate, scattering intensity, or reflectivity, etc. In the data analysis process, since the units and value ranges of the light adjustment parameters (such as wavelength, intensity, irradiation time, etc.) and the preset difference characteristics (such as white blood cell concentration, bacteria count, protein content, etc.) may vary greatly, directly performing sensitivity analysis may cause some parameters or characteristics to occupy too large a weight in the analysis, thus affecting the accuracy of the results. Therefore, min-max normalization or Z-score standardization is used to process the light adjustment parameters and the preset difference characteristics, so that all parameters and characteristics are converted to the same scale to ensure the fairness and accuracy of the analysis. Subsequently, the Pearson correlation coefficient calculation formula is used to calculate the correlation coefficients of each light adjustment parameter and each preset difference characteristic for the processed light adjustment parameters and preset difference characteristics. In this process, the mean, covariance, and standard deviation will be calculated. After obtaining the Pearson correlation coefficients between each light adjustment parameter and each preset difference characteristic, the absolute values of these Pearson correlation coefficients will be used as the sensitivity indexes of the corresponding light adjustment parameters and preset difference characteristics. This is because the value range of the Pearson correlation coefficient is [-1, 1], and the larger its absolute value, the stronger the linear correlation between the light adjustment parameter and the difference characteristic. That is to say, the preset difference characteristic is more sensitive to this light adjustment parameter. After that, the sensitivity indexes of all light adjustment parameters are sorted, and the parameters with sensitivity indexes higher than the preset sensitivity threshold (such as 0.7) are selected. Among them, the preset sensitivity threshold is determined based on expert decision and historical experience. The selected light adjustment parameters will be considered to have a significant impact on the difference characteristics and will be used for subsequent light adjustment optimization. Finally, the sensitivity indexes and the selected key parameters are output for the parameter configuration unit to use to optimize the adjustment parameters of the light module 11. For example, if ultraviolet light shows the highest sensitivity index when detecting proteins, the light source switcher will preferentially select ultraviolet light for irradiation.Through the above process, the sensitive correlation unit can systematically analyze the influence of light parameters on the optical properties of peritoneal dialysis fluid samples, providing a scientific basis for the accurate detection of peritonitis.

[0021] Furthermore, the light intensity adjustment unit further includes a light source switcher for switching different light source types to irradiate the peritoneal dialysis fluid sample. The types of the light source switcher include ultraviolet light, visible light, and near-infrared light.

[0022] In a feasible implementation manner, the light intensity adjustment unit further includes a light source switcher, which is a key component that can switch different types of light sources to irradiate the peritoneal dialysis fluid sample according to needs. These light source types include ultraviolet light, visible light, and near-infrared light. Each light source has a unique wavelength range and optical properties, and is suitable for detecting different components or markers in the peritoneal dialysis fluid sample. Ultraviolet light has a short wavelength and high energy, and is suitable for exciting fluorescent substances in the sample or detecting certain specific biomolecules. In the detection of peritonitis, ultraviolet light may be used to identify inflammation-related proteins or bacterial metabolites; visible light has a wide wavelength range and is suitable for observing changes in the color, turbidity, or suspended particles of the sample. Peritoneal dialysis fluid may exhibit different colors or turbidities in the healthy state and the peritonitis state, and visible light can help capture these intuitive optical differences; near-infrared light has a long wavelength and strong penetration ability, and is suitable for detecting deep structures or changes in water content in the sample. In the detection of peritonitis, near-infrared light may be used to analyze the cell concentration or inflammation-related water distribution in the peritoneal dialysis fluid. The light source switcher can automatically or manually switch the light source type according to the detection requirements to ensure the use of the most suitable light source at different detection stages. For example, visible light is used to observe the macroscopic changes of the sample during preliminary screening, while ultraviolet light or near-infrared light is switched to obtain more detailed optical information during further analysis. By flexibly switching the light source type, the light intensity adjustment unit can comprehensively and accurately analyze the peritoneal dialysis fluid sample, thereby improving the detection accuracy of the peritonitis detection device while ensuring the detection efficiency of the peritonitis detection device.

[0023] The sensor module 12 captures a sensing data set of the state change of the peritoneal dialysis fluid sample during the irradiation process through the sensor module 12.

[0024] Specifically, in the sensor module 12, the sensor module 12 captures the state changes of the sample in real time through the collaborative work of the light source and the sensor. The light source irradiates the sample according to the configured light parameters (such as wavelength, intensity, irradiation time, etc.), and the sensor records the responses of the sample under different lighting conditions, including color, light absorption rate, light scattering rate, etc. These response data are recorded in the form of a time series, forming a multi-dimensional sensing data set. To ensure the accuracy of the data, the collected data will undergo preprocessing steps such as noise filtering, alignment, and normalization. Subsequently, this sensing data set will be transmitted to the detection chip 13 through the data transmission channel for model analysis, thereby providing a scientific basis for the accurate detection of peritonitis.

[0025] Furthermore, the sensor module 12 includes a plurality of sensors, and the sensor module 12 further includes:

[0026] A transmission connection unit for establishing a data transmission channel between the plurality of sensors and the detection chip 13; a data receiving unit for the detection chip 13 to receive the plurality of sensing data sets corresponding to the plurality of sensors.

[0027] In a preferred embodiment, the sensor module 12 not only includes a plurality of sensors (such as optoelectronic sensors, spectral sensors, etc.), but also integrates a transmission connection unit and a data receiving unit, which cooperate together to achieve efficient data acquisition and processing. The transmission connection unit is responsible for establishing a stable data transmission channel between the plurality of sensors and the detection chip 13, ensuring that the optical response data collected by the sensors can be transmitted to the detection chip 13 in real time and accurately. This process is usually achieved through wired or wireless communication technologies, and the specific methods can include USB, Bluetooth, Wi-Fi, or a dedicated data bus protocol, etc.; the data receiving unit is located inside the detection chip 13 and is responsible for receiving the sensing data sets from the plurality of sensors and integrating them into a unified format for subsequent processing and analysis. Through the collaborative work of the transmission connection unit and the data receiving unit, the sensor module 12 can efficiently capture the state changes of the peritoneal dialysis fluid sample during irradiation and transmit this data to the detection chip 13, providing a data basis for the rapid detection of peritonitis.

[0028] A detection chip 13, the detection chip 13 has downloaded a trained peritonitis risk detection model, and performs data detection on the sensing data set according to the peritonitis risk detection model, and outputs a risk detection result.

[0029] Specifically, the detection chip 13 is the core processing unit of the entire device. It has downloaded a trained peritonitis risk detection model internally. This model can intelligently analyze the data collected by the sensor module 12 and output the peritonitis risk detection results. The detection chip 13 first receives the raw data from the sensor module 12 through the transmission connection unit. This data includes multi-dimensional information such as color, transparency, and scattering. The received data has been pre-processed, such as denoising, normalization, or feature extraction, to ensure the accuracy and consistency of the data. Subsequently, the detection chip 13 uses the pre-trained peritonitis risk detection model to perform inference analysis on the processed data. This model is trained based on a large number of peritoneal dialysis fluid sample data, can identify the characteristic patterns related to peritonitis, and generate the peritonitis risk detection results. This risk detection result is usually output in the form of a risk level (such as low risk, medium risk, high risk) or a probability value for users or medical staff to reference. The entire process is efficient and real-time, capable of quickly providing accurate detection results, providing strong support for the early detection and intervention of peritonitis.

[0030] Furthermore, the detection chip 13 includes:

[0031] A sampling unit for obtaining healthy peritoneal dialysis fluid samples and non-healthy peritoneal dialysis fluid samples, extracting the healthy sensing data set corresponding to the healthy peritoneal dialysis fluid samples, and the non-healthy sensing data set corresponding to the non-healthy peritoneal dialysis fluid samples; a difference comparison unit for performing distribution difference comparison based on the healthy sensing data set and the non-healthy sensing data set, and extracting preset difference features; a model training unit for training according to the preset difference features and the labels indicating the degree of inflammation risk to obtain a trained peritonitis risk detection model.

[0032] In a preferred embodiment, the sampling unit respectively collects peritoneal dialysis fluid data from the historical medical data of healthy individuals and users with peritonitis to form a healthy peritoneal dialysis fluid sample and an unhealthy peritoneal dialysis fluid sample, and then extracts sensing data from the healthy peritoneal dialysis fluid sample and the unhealthy peritoneal dialysis fluid sample respectively to obtain a healthy sensing data set and an unhealthy sensing data set. Both of these sensing data sets include multi-dimensional data such as color, transparency, scattering degree, and spectral response, providing a basis for subsequent training and analysis. Subsequently, the multi-dimensional data such as color, transparency, scattering degree, and spectral response in the healthy sensing data set are respectively subjected to mean calculation to obtain a standard healthy sensing data set, which is used to compare the distribution differences of the unhealthy sensing data set. After that, the difference between each data in the unhealthy sensing data set and the corresponding data in the standard healthy sensing data set is calculated, and these differences are summarized to obtain a preset difference feature, which contains the distribution differences of the unhealthy sensing data set in features such as color, transparency, scattering degree, and spectral response. Then, the unhealthy sensing data identifiers of each group of difference features are extracted from the preset difference feature, and the corresponding inflammation risk degree labels are extracted from the unhealthy sensing data set using this unhealthy sensing data. This inflammation risk degree label is determined by doctors based on historical experience. By labeling the inflammation risk degree label to the corresponding difference feature, sample data for training the peritonitis risk detection model is obtained, and based on this sample data, training data and validation data are divided. Finally, using the training data to train the peritonitis risk detection model constructed based on machine learning algorithms (such as support vector machine, random forest, neural network), taking the multi-layer perceptron (MLP) in the neural network as an example, the training data is input into the peritonitis risk detection model, and the peritonitis risk detection model is iteratively optimized through steps such as forward propagation, loss calculation, backward propagation, and parameter optimization. When the preset number of iterations is reached or the loss tends to be stable, the validation data is used to evaluate the current peritonitis risk detection model. If the accuracy meets the expected accuracy, the current peritonitis risk detection model is output; otherwise, hyperparameters such as the learning rate and the number of training batches are adjusted to further improve the detection effect of the peritonitis risk detection model, providing reliable technical support for the rapid detection of peritonitis.

[0033] Further, the detection chip 13 further includes:

[0034] A model analysis unit, configured to load the peritonitis risk detection model and detect the real-time difference features corresponding to the preset difference feature items of the sensing data set, where the sensing data set includes color data, transparency data, scattering degree data, and spectral response data; a detection output unit, configured to match the real-time difference features with the preset difference features to obtain a matching label indicating the inflammation risk degree and output a risk detection result.

[0035] In a feasible implementation, when using the detection chip 13 for data detection, a pre-trained peritonitis risk detection model is loaded, and then differential analysis is performed on the collected sensing data set. During this process, the feature data in the sensing data set is compared with the preset differential feature items, and this preset differential feature item is a feature type determined according to the preset differential features. By calculating the difference between the feature data in the sensing data set and the standard healthy sensing data corresponding to each preset differential feature item in the standard healthy sensing data set, the real-time differential features of this sensing data set are obtained. Among them, the sensing data set includes color data, transparency data, scattering data, and spectral response data. The color data reflects the color characteristics of the sample (such as RGB values); the transparency data reflects the transparency of the sample, which is represented by the light transmittance or turbidity; the scattering data reflects the light scattering characteristics of the sample, which is represented by the scattering intensity or scattering angle; the spectral response data reflects the spectral characteristics of the sample at different wavelengths, which is represented by the peak wavelength, peak intensity, integral area, etc. Subsequently, the real-time differential features are input into the peritonitis risk detection model, and the peritonitis risk detection model will match an inflammation risk degree label for the real-time differential features according to the mapping relationship learned from the preset differential features, and use this label to identify the real-time differential features, form a risk detection result and output it. In summary, through the above process, automated peritonitis risk detection can be achieved, ensuring the accuracy and timeliness of the detection results.

[0036] Furthermore, the detection chip 13 further includes a data fusion unit, and the data fusion unit includes:

[0037] A feature convolution unit for performing multi-scale feature convolution on the multiple sensing data sets and outputting multiple sensing feature groups; a feature alignment unit for aligning the multiple sensing feature groups and outputting the aligned multiple sensing feature groups; a feature fusion unit for fusing the aligned multiple sensing feature groups and outputting a fused sensing data set, and the detection chip 13 performs data detection on the fused sensing data set according to the peritonitis risk detection model and outputs a risk detection result.

[0038] In a feasible implementation, to ensure the accuracy of the risk detection results, multiple data acquisitions are performed to form multiple sensing data sets. Then, the data fusion unit in the detection chip 13 is used to fuse these sensing data sets. Among them, the main task of the data fusion unit is to perform multi-scale feature extraction, feature alignment, and feature fusion on multiple sensing data sets (such as color data, transparency data, scattering data, spectral response data, etc.), and finally generate a fused sensing data set for the peritonitis risk detection model to analyze. In this process, the feature convolution unit is used to perform multi-scale feature convolution on multiple sensing data sets to extract features of different scales, such as local and global features of color distribution, local and global features of transparency change, etc., so as to form multiple sensing feature groups. Each feature group contains feature information of different scales. This multi-scale feature convolution is based on a convolutional neural network (CNN). The construction process of the convolutional neural network is similar to the foregoing, and is all carried out through steps such as forward propagation, loss calculation, backpropagation, and parameter optimization, which will not be elaborated here again. Subsequently, the feature alignment unit is used to perform feature alignment on multiple sensing feature groups, that is, the feature groups are adjusted according to the timestamp of each feature to make the multiple sensing feature groups consistent in time. In addition, if there are differences in spatial resolution, they will also be adjusted to the same spatial dimension through interpolation or downsampling to facilitate subsequent feature fusion. After that, the feature fusion unit is used to fuse the aligned multiple sensing feature groups into a fused sensing data set, that is, the mean value of each sensing feature group is calculated by using the average method, and all the calculated mean values are used to form a fused sensing data set. Finally, the fused sensing data set is input into the peritonitis risk detection model, and the risk detection results are output through the same analysis process as described above. In summary, the data fusion unit integrates multiple sensing data sets into a unified fused data set through feature convolution, feature alignment, and feature fusion, providing a more comprehensive and accurate data basis for peritonitis risk detection, thereby improving the timeliness and robustness of detection.

[0039] The display module 14 is used to display the risk detection results on the display screen of the display module 14.

[0040] Specifically, the display module 14 is the core part of the detection chip 13 for presenting the risk detection results. It is responsible for converting complex detection data into easily understandable information and clearly presenting it on the display screen. When the peritonitis risk detection model completes the analysis, the display module 14 will receive the detection results and display them in the form of text, graphics, or color markings. For example, the risk level is directly displayed in text such as low risk, medium risk, or high risk, and at the same time, it is accompanied by color coding of green, yellow, or red, enabling users to clearly understand the current situation at a glance. In addition, the display module 14 will also display specific detection values, such as transparency, color value, and scattering degree, as well as the change trends of these data, to help users more comprehensively grasp the detection information. To further improve the user experience, the display module 14 will also provide corresponding suggestions according to the risk level, such as no treatment required, recommended reexamination, or immediate medical treatment, and will update the status information in real time during the operation of the device, such as detecting or detection completed, to ensure that users can timely understand the operation status of the device. Through this intuitive and comprehensive display method, the display module 14 provides users with a convenient and efficient experience for viewing risk detection results.

[0041] Further, the device further includes a placement container, and the placement container includes:

[0042] A housing, the housing being made of a translucent material; a light-transmitting window, one side of the placement container is provided with a light-transmitting window, so that the light illumination module 11 irradiates the peritoneal dialysis fluid sample inside the container through the light-transmitting window.

[0043] In a preferred embodiment, the detection device further includes a placement container, which is mainly used to carry and fix the peritoneal dialysis fluid sample for detection. The design of the placement container includes two key parts, namely the housing and the light-transmitting window. The housing is made of a translucent material, which can not only protect the sample inside the container but also allow light to pass through moderately, facilitating observation and detection. On one side of the placement container, there is a light-transmitting window, whose function is to enable the light of the light illumination module 11 to directly irradiate the peritoneal dialysis fluid sample inside the container. Through the light-transmitting window, the light illumination module 11 can efficiently perform light analysis on the sample, ensuring the accuracy and reliability of the detection process. This design not only improves the convenience of detection but also ensures the safety and stability of the sample during the detection process.

[0044] Further, the placement container further includes:

[0045] An adjustable notch, which is arranged inside the placement container; a liquid level indicator, used to sense the liquid height of the peritoneal dialysis fluid sample in the placement container. When the liquid height reaches a preset height, the height of the adjustable notch is adjusted to make the heights of the peritoneal dialysis fluid samples placed multiple times consistent.

[0046] In a feasible implementation, the placement container is also equipped with two important functions, namely an adjustable notch and a liquid level indicator. The adjustable notch is set inside the container, and its function is to flexibly adjust the space inside the container to accommodate different amounts of peritoneal dialysis fluid samples; the liquid level indicator is used to monitor the liquid height of the sample in the container in real time. When the liquid reaches the preset height, the liquid level indicator will trigger the adjustment function of the adjustable notch, causing it to automatically adjust the height to ensure that the height of each placed peritoneal dialysis fluid sample remains consistent. This design not only improves the accuracy of detection, but also avoids detection errors caused by inconsistent sample amounts. At the same time, it simplifies the operation process and enhances the user experience.

[0047] In summary, the portable detection device for quickly identifying peritonitis provided by the embodiments of the present application has the following technical effects:

[0048] The light module 11 takes a peritoneal dialysis fluid sample and activates the light module 11 to irradiate the peritoneal dialysis fluid sample; the sensor module 12 captures a sensing data set of the state change of the peritoneal dialysis fluid sample during the irradiation process through the sensor module 12; the detection chip 13 has a trained peritonitis risk detection model downloaded, and performs data detection on the sensing data set according to the peritonitis risk detection model, and outputs a risk detection result; the display module 14 is used to display the risk detection result on the display screen of the display module 14. Through the above steps, the technical problem that traditional peritonitis detection devices are difficult for users to independently complete rapid and accurate detection in a home or community environment due to poor portability and complex operation is solved, and the technical effect of quickly analyzing peritoneal dialysis fluid samples through optical characteristics and machine learning algorithms, and real-time outputting peritonitis risk detection results, improving the accuracy and efficiency of early identification of peritonitis by users or medical staff is achieved.

[0049] Embodiment 2, based on the same inventive concept as the portable detection device for quickly identifying peritonitis in the foregoing embodiment, as Figure 2 shown, the embodiments of the present application provide a portable detection method for quickly identifying peritonitis, and the method includes:

[0050] S100: Take a peritoneal dialysis fluid sample and irradiate the peritoneal dialysis fluid sample; S200: Capture a sensing data set of the state change of the peritoneal dialysis fluid sample during the irradiation process; S300: Perform data detection on the sensing data set according to the peritonitis risk detection model and output a risk detection result; S400: Display the risk detection result on the display screen.

[0051] Further, step S100 further includes:

[0052] Step S110: Define a library of light adjustment parameters; Step S120: Establish a sensitivity index between the preset difference features and each light adjustment parameter in the defined library of light adjustment parameters; Step S130: Extract N light adjustment parameter configurations for adjusting parameters that are greater than a preset sensitivity according to the magnitudes of the sensitivity indices, where N is a positive integer greater than or equal to 1.

[0053] Further, step S300 further includes:

[0054] Step S310: Obtain a healthy peritoneal dialysis fluid sample and an unhealthy peritoneal dialysis fluid sample, extract the healthy sensing data set corresponding to the healthy peritoneal dialysis fluid sample, and the unhealthy sensing data set corresponding to the unhealthy peritoneal dialysis fluid sample; Step S320: Perform a distribution difference comparison based on the healthy sensing data set and the unhealthy sensing data set, and extract preset difference features; Step S330: Train according to the preset difference features and the label indicating the degree of inflammation risk to obtain a trained peritonitis risk detection model.

[0055] Further, step S300 further includes:

[0056] Step S340: A model analysis unit, configured to load the peritonitis risk detection model and detect the real-time difference features corresponding to the preset difference feature items of the sensing data set, where the sensing data set includes color data, transparency data, scattering data, and spectral response data; Step S350: A detection output unit, configured to match the real-time difference features with the preset difference features to obtain a matching label indicating the degree of inflammation risk and output a risk detection result.

[0057] Further, step S300 further includes:

[0058] Step S360: Perform multi-scale feature convolution on the multiple sensing data sets and output multiple groups of sensing features; Step S370: Align the multiple groups of sensing features and output the aligned multiple groups of sensing features; Step S380: Fuse the aligned multiple groups of sensing features, output a fused sensing data set, and perform data detection on the fused sensing data set according to the peritonitis risk detection model to output a risk detection result.

[0059] Any step of the method described above can be stored as computer instructions or programs in an unrestricted computer memory and can be called and recognized by an unrestricted computer processor to implement any one of the methods in the embodiments of the present application, and no redundant limitations are made here.

[0060] Furthermore, the first or second as described above may not only represent an order relationship, but may also represent a certain specific concept, and / or refer to the selection of multiple elements either individually or in whole. Obviously, those skilled in the art can make various changes and modifications to this application without departing from the scope of this application. Thus, if these modifications and variations of this application fall within the scope of this application and its equivalent technologies, this application is intended to include these changes and modifications.

Claims

1. A portable detection device for quickly identifying peritonitis, characterized in that, The device includes: A light module that takes a peritoneal dialysis fluid sample and activates the light module to irradiate the peritoneal dialysis fluid sample; A sensor module that captures a sensing data set of the state change of the peritoneal dialysis fluid sample during the irradiation process through the sensor module; A detection chip with a trained peritonitis risk detection model downloaded, which performs data detection on the sensing data set according to the peritonitis risk detection model and outputs a risk detection result; A display module for displaying the risk detection result on the display screen of the display module.

2. The portable detection device for rapid identification of peritonitis according to claim 1, wherein The detection chip includes: A sampling unit for obtaining healthy peritoneal dialysis fluid samples and non-healthy peritoneal dialysis fluid samples, extracting the healthy sensing data set corresponding to the healthy peritoneal dialysis fluid samples, and the non-healthy sensing data set corresponding to the non-healthy peritoneal dialysis fluid samples; A difference comparison unit for performing distribution difference comparison based on the healthy sensing data set and the non-healthy sensing data set and extracting preset difference features; A model training unit for training according to the preset difference features and labels indicating the degree of inflammation risk to obtain a trained peritonitis risk detection model.

3. The portable detection device for quickly identifying peritonitis according to claim 2, wherein The detection chip further includes: A model analysis unit for loading the peritonitis risk detection model and detecting the real-time difference features of the sensing data set corresponding to the preset difference feature items, where the sensing data set includes color data, transparency data, scattering data, and spectral response data; A detection output unit for matching the real-time difference features with the preset difference features to obtain a matching label indicating the degree of inflammation risk and outputting a risk detection result.

4. The portable detection device for quickly identifying peritonitis according to claim 2, wherein, The light module includes a light adjustment unit, and the light adjustment unit includes: A parameter definition unit for defining a light adjustment parameter library; A sensitivity association unit for establishing a sensitivity index between the preset difference features and each light adjustment parameter in the defined light adjustment parameter library; A parameter configuration unit for extracting N light adjustment parameters greater than a preset sensitivity according to the magnitude of the sensitivity index and configuring the adjustment parameters of the light adjustment unit, where N is a positive integer greater than or equal to 1.

5. The portable detection device for rapid identification of peritonitis according to claim 1, characterized in that, The sensor module includes multiple sensors, and the sensor module further includes: A transmission connection unit for establishing a data transmission channel between the multiple sensors and the detection chip; A data receiving unit for the detection chip to receive multiple sensing data sets corresponding to the multiple sensors.

6. The portable detection device for rapidly identifying peritonitis according to claim 5, wherein, The detection chip further includes a data fusion unit, and the data fusion unit includes: A feature convolution unit for performing multi-scale feature convolution on the multiple sensing data sets and outputting multiple sensing feature groups; A feature alignment unit for aligning the multiple sensing feature groups and outputting the aligned multiple sensing feature groups; A feature fusion unit for fusing the aligned multiple sensing feature groups and outputting a fused sensing data set, and the detection chip performs data detection on the fused sensing data set according to the peritonitis risk detection model and outputs a risk detection result.

7. The portable detection device for rapid identification of peritonitis according to claim 1, wherein, The device further includes a placement container, and the placement container includes: A housing, and the housing is made of a semi-transparent material; A light-transmitting window is provided on one side of the placement container, so that the light module irradiates the peritoneal dialysis fluid sample inside the container through the light-transmitting window.

8. The portable detection device for rapid identification of peritonitis according to claim 7, characterized in that, The placement container further includes: An adjustable notch, which is arranged inside the placement container; A liquid level indicator, which is used to sense the liquid height of the peritoneal dialysis fluid sample in the placement container. When the liquid height reaches a preset height, the height of the adjustable notch is adjusted to make the heights of the peritoneal dialysis fluid samples placed multiple times consistent.

9. The portable detection device for rapid identification of peritonitis according to claim 4, wherein, The light intensity adjustment unit further includes a light source switcher, which is used to switch different light source types to irradiate the peritoneal dialysis fluid sample. The types of the light source switcher include ultraviolet light, visible light, and near-infrared light.

10. A portable detection method for quickly identifying peritonitis, characterized in that, The portable detection method for quickly identifying peritonitis is executed by the portable detection device for quickly identifying peritonitis according to any one of claims 1 to 9, and includes: Taking a peritoneal dialysis fluid sample and irradiating the peritoneal dialysis fluid sample; Capturing a sensing data set of the state change of the peritoneal dialysis fluid sample during the irradiation process; Performing data detection on the sensing data set according to the peritonitis risk detection model and outputting a risk detection result; Displaying the risk detection result on the display screen.

Images