A universal fully automatic calibration device and method for blood culture instrument
By designing a universal fully automatic calibration device for blood culture instruments, using intelligent algorithms and sensor data, the time-consuming and labor-intensive and model adaptation problems of the traditional temperature calibration process are solved, and high-precision and automated temperature calibration are achieved, suitable for different types of blood. Cultivation instrument.
Patent Information
- Application Number
- CN202410145390.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-02-01
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-02-01
AI Technical Summary
The temperature calibration process of traditional blood culture instruments requires manual operation, which is time-consuming and labor-intensive and error-prone. The existing calibration devices cannot adapt to the size differences of different types of blood culture instruments, making it difficult to achieve full automation and versatility.
A universal fully automatic calibration device for blood culture instruments is designed, including a data acquisition system, a data reception processing system and an auxiliary support system. The integrated temperature and illuminance sensors are used to collect data, and the temperature implicit change distribution information is extracted through intelligent algorithms and Spatial topological characteristics, generate temperature correction measurement estimates, and realize automated calibration.
The temperature measurement accuracy of the blood culture instrument is improved, errors are reduced, and the automatic temperature calibration of different types of blood culture instruments is achieved, which significantly improves the calibration efficiency and accuracy.
Smart Images

Figure CN118006429B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of blood culture instruments, and more specifically, to a universal fully automatic calibration device and method for blood culture instruments. Background Art
[0002] The blood bacterial culture analyzer (referred to as blood culture instrument) is an instrument used in clinical laboratories for in vitro continuous culture and automatic detection of microorganisms such as bacteria and yeast-like fungi in human blood or other sterile body fluids. It can effectively determine whether the sample being tested contains microorganisms, and can be effectively used for rapid and sensitive screening of sepsis and bacteremia. The test results of the blood culture instrument can provide a diagnostic basis for clinical treatment and are directly related to the success or failure of treatment. It plays an important role in the hematology department, infection department and other departments of the hospital.
[0003] The blood culture instrument is mainly composed of two parts: a constant temperature incubation system and a detection system. The constant temperature incubation system mainly provides a suitable constant temperature environment for microbial culture, and the detection system is used to determine whether there is bacterial culture growth. The temperature and illumination of the blood culture instrument jointly determine the reliability of the equipment's detection results. Among them, temperature determines the accuracy of microbial culture. Therefore, temperature calibration of the blood culture instrument is necessary. However, the traditional temperature calibration process requires manual operation, which is time-consuming, labor-intensive, and prone to errors. In addition, there are many blood culture instrument manufacturers at home and abroad, and there are differences in the specifications and models of mainstream blood culture instrument products on the market. A set of calibration devices cannot cover the cabin sizes of various models, and it is difficult to achieve universal application for all models. Therefore, an optimized universal fully automatic calibration device and method for blood culture instruments is expected. Summary of the invention
[0004] In order to solve the above technical problems, the present application is proposed. The present application provides a universal fully automatic calibration device and method for a blood culture instrument, which can improve the measurement accuracy of the blood culture instrument and reduce errors through accurate calibration.
[0005] According to one aspect of the present application, there is provided a universal fully automatic calibration device for a blood culture instrument, which includes: a data acquisition system, a data receiving and processing system, and an auxiliary support system;
[0006] Wherein, the data acquisition system includes a plurality of calibration units, each of which includes a housing assembly, a charging interface mounted on the housing assembly, an integrated temperature and illumination sensor, and a calibration unit circuit module;
[0007] The data receiving and processing system comprises a computer host and a display module electrically connected to the computer host;
[0008] The auxiliary support system includes a containing box and a box circuit module installed on the containing box.
[0009] According to another aspect of the present application, a universal fully automatic calibration method for a blood culture instrument is provided, comprising:
[0010] Get temperature values collected by multiple integrated temperature and illuminance sensors;
[0011] Acquire a spatial topology matrix between the plurality of integrated temperature and illumination sensors, wherein the value of each non-diagonal position in the spatial topology matrix is used to represent the spatial distance between the corresponding two integrated temperature and illumination sensors;
[0012] Extracting sample correlation features of temperature values collected by the integrated temperature and illumination sensors to obtain a temperature correlation pattern feature vector;
[0013] Extracting topological features of the spatial topological matrix to obtain a spatial topological feature matrix; and
[0014] A temperature-corrected measurement estimate is generated based on the mapping fusion interaction features of the temperature correlation pattern feature vector and the spatial topology feature matrix.
[0015] Compared with the prior art, the universal fully automatic calibration device and method for blood culture instruments provided in the present application include: a data acquisition system, a data receiving and processing system, and an auxiliary support system; wherein the data acquisition system includes a plurality of calibration units, each of which includes a housing assembly, a charging interface installed on the housing assembly, an integrated temperature and illumination sensor, and a calibration unit circuit module; the data receiving and processing system includes a computer host and a display module electrically connected to the computer host; the auxiliary support system includes a containing box and a box circuit module installed on the containing box. In this way, the measurement accuracy of the blood culture instrument can be improved and errors can be reduced through precise calibration. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. The following drawings are not intentionally scaled to the actual size, and the focus is on illustrating the main purpose of the present application.
[0017] Figure 1 It is a block diagram of a universal fully automatic calibration device for a blood culture instrument according to an embodiment of the present application.
[0018] Figure 2 Flow chart of a universal fully automatic calibration method for a blood culture instrument according to an embodiment of the present application.
[0019] Figure 3 Schematic diagram of the system architecture of a universal fully automatic calibration method for a blood culture instrument according to an embodiment of the present application.
[0020] Figure 4 This is an application scenario diagram of a universal fully automatic calibration device for a blood culture instrument according to an embodiment of the present application.
[0021] Figure 5 It is a cross-sectional schematic diagram of a calibration unit of a universal fully automatic calibration device for a blood culture instrument according to an embodiment of the present application.
[0022] Figure 6 It is a schematic diagram of the structure of the calibration unit of the universal fully automatic calibration device for blood culture instrument according to the embodiment of the present application.
[0023] Figure 7 It is a schematic diagram of the structure of a universal fully automatic calibration device for a blood culture instrument according to an embodiment of the present application.
[0024] Among them, Figure 5 and Figure 6 In the embodiment, 1-housing, 2-telescopic second housing, 3-semi-embedded third housing, 4-charging interface, 5-integrated temperature and illumination sensor, 6-calibration unit circuit module; Figure 7 Among them, 3-semi-embedded third shell, 7-computer host, 8-display module, 9-containing box, 10-box circuit module, 11-calibration unit charging male head. DETAILED DESCRIPTION
[0025] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.
[0026] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.
[0027] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.
[0028] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.
[0029] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.
[0030] The present application provides a universal fully automatic calibration device for a blood culture instrument. Figure 1 FIG. 1 is a block diagram of a universal fully automatic calibration device for a blood culture instrument according to an embodiment of the present application. Figure 1 As shown, a universal fully automatic calibration device 100 for a blood culture instrument according to an embodiment of the present application includes: a data acquisition system 110, a data receiving and processing system 120 and an auxiliary support system 130; wherein, the data acquisition system 110 includes a plurality of calibration units, each of which includes a shell assembly, a charging interface installed on the shell assembly, an integrated temperature and illumination sensor and a calibration unit circuit module; the data receiving and processing system 120 includes a computer host and a display module electrically connected to the computer host; the auxiliary support system 130 includes a containing box and a box circuit module installed on the containing box.
[0031] Among them, in order to solve the problem of low automation level of mainstream blood culture instrument products on the market, the technical concept of the present application is to use an intelligent algorithm to extract the implicit temperature change distribution information in the temperature values collected by multiple integrated temperature and illumination sensors, and integrate the spatial topological feature information between each of the integrated temperature and illumination sensors into the temperature feature expression, so as to more comprehensively reflect the temperature distribution law inside the blood culture instrument, thereby intelligently generating temperature correction measurement estimates and realizing automatic calibration of the temperature of the blood culture instrument.
[0032] Correspondingly, the calibration unit circuit module includes: a data acquisition unit, used to acquire the temperature values collected by multiple integrated temperature and illuminance sensors; a spatial topology matrix acquisition unit, used to acquire the spatial topology matrix between the multiple integrated temperature and illuminance sensors, wherein the values of each non-diagonal position in the spatial topology matrix are used to represent the spatial distance between the corresponding two integrated temperature and illuminance sensors; a sample association feature extraction unit, used to extract the sample association features of the temperature values collected by the multiple integrated temperature and illuminance sensors to obtain the temperature correlation pattern feature vector; a topology feature extraction unit, used to extract the topological features of the spatial topology matrix to obtain the spatial topology feature matrix; and a temperature correction measurement estimate generation unit, used to generate a temperature correction measurement estimate based on the mapping fusion interaction features of the temperature correlation pattern feature vector and the spatial topology feature matrix.
[0033] Specifically, in the technical solution of the present application, the encoding process of the calibration unit circuit module includes: first, obtaining the temperature values collected by the multiple integrated temperature and illumination sensors; and obtaining the spatial topological matrix between the multiple integrated temperature and illumination sensors, wherein the values of each position on the non-diagonal position in the spatial topological matrix are used to represent the spatial distance between the corresponding two integrated temperature and illumination sensors. Here, obtaining the temperature values collected by the multiple integrated temperature and illumination sensors can reflect the temperature information at different spatial positions of the blood culture instrument, thereby constituting the temperature distribution inside the blood culture instrument, and providing an important basis for temperature calibration. In addition, considering that the temperature at each position in the blood culture instrument may differ due to various factors, that is, the temperature may vary or be unevenly distributed in space, by obtaining the spatial distance between each integrated temperature and illumination sensor, the implicit spatial topological association relationship contained therein can be considered to analyze and explore the causes of temperature changes and the law of temperature transfer, guide the model to pay attention to the influence of the positional relationship of the sensor on temperature calibration, and improve the accuracy and reliability of calibration.
[0034] Then, after arranging the temperature values collected by the multiple integrated temperature and illumination sensors into a temperature input vector according to the sample dimension, the temperature input vector is passed through a temperature implicit correlation pattern feature extractor based on a one-dimensional convolutional layer to obtain a temperature correlation pattern feature vector. That is, a one-dimensional convolutional layer is used to construct a temperature implicit correlation pattern feature extraction to capture the implicit correlation and correlation pattern between the temperature values collected by each of the integrated temperature and illumination sensors. Specifically, these correlation patterns may include temperature trends, periodic changes, instantaneous changes, etc. By extracting such implicit correlations and correlation patterns, the global temperature change information and distribution pattern of the blood culture instrument can be characterized and portrayed. At the same time, the spatial topological matrix is passed through a sensor spatial topological feature extractor based on a convolutional neural network model to obtain a spatial topological feature matrix. That is, a sensor-to-sensor spatial topology feature extractor based on a convolutional neural network model is used to capture the complex spatial relationship between the integrated temperature and illumination sensors expressed by the spatial topology matrix, so as to understand the relative positions and spatial layouts of the integrated temperature and illumination sensors, and further reveal the temperature transfer laws and the implicit influence relationships of the temperatures at the locations of the integrated temperature and illumination sensors, which helps to understand a more comprehensive temperature distribution pattern.
[0035] Correspondingly, the sample association feature extraction unit includes: a vectorization subunit, used to arrange the temperature values collected by the multiple integrated temperature and illuminance sensors into a temperature input vector according to the sample dimension; and a temperature implicit association pattern feature extraction subunit, used to pass the temperature input vector through a temperature implicit association pattern feature extractor based on a one-dimensional convolutional layer to obtain the temperature association pattern feature vector.
[0036] The topological feature extraction unit is used to: pass the spatial topological matrix through an inter-sensor spatial topological feature extractor based on a convolutional neural network model to obtain the spatial topological feature matrix.
[0037] Next, a spatial mapping fusion device is used to process the spatial topological feature matrix and the temperature correlation pattern feature vector to obtain a topological information-doped temperature correlation pattern feature vector. It should be understood that in a blood culture instrument, the spatial relationship between the integrated temperature and illumination sensors is closely related to the temperature correlation. In the technical solution of the present application, by fusing the spatial topological feature matrix and the temperature correlation pattern feature vector, the spatial information can be combined with the temperature information, thereby more comprehensively describing the temperature distribution pattern inside the blood culture instrument and the global temperature implicit correlation change trend. Specifically, the spatial mapping fusion device enables the one-dimensional feature vector, i.e., the temperature correlation pattern feature vector, to interact with the high-dimensional feature map, i.e., the spatial topological feature matrix, to directly control the relevant characteristics of each feature channel, and help the network focus on the specific part of each feature channel to consider the correlation and importance between different feature channels in the feature fusion process. This helps to better fuse the feature information of spatial topology and temperature distribution, making the fused features more representative and discriminative.
[0038] Further, the topology information doped temperature correlation pattern feature vector is passed through a decoder-based temperature-corrected measurement estimate generator to obtain a temperature-corrected measurement estimate. The decoder-based temperature-corrected measurement estimate generator can map the topology information doped temperature correlation pattern feature vector back to a numerical space representing the temperature-corrected measurement estimate.
[0039] Correspondingly, the temperature-corrected measurement estimate generating unit comprises: a spatial mapping fusion subunit, used for using a spatial mapping fuser to process the spatial topological feature matrix and the temperature correlation pattern feature vector to obtain a topological information-doped temperature correlation pattern feature vector; and a decoding subunit, used for passing the topological information-doped temperature correlation pattern feature vector through a decoder-based temperature-corrected measurement estimate generator to obtain a temperature-corrected measurement estimate.
[0040] Among them, in a specific example, the spatial mapping fusion subunit includes: a first convolution secondary subunit, used to pass the temperature correlation pattern feature vector through a point convolution layer to obtain a first convolution feature vector; a first correction secondary subunit, used to pass the first convolution feature vector through a corrected linear unit based on a ReLU function to obtain a first corrected convolution feature vector; a second convolution secondary subunit, used to pass the first corrected convolution feature vector through a point convolution layer to obtain a second convolution feature vector; a second correction secondary subunit, used to pass the second convolution feature vector through a corrected linear unit based on a Sigmoid function to obtain a second corrected convolution feature vector; a convolution encoding secondary subunit, used to pass the spatial topological feature matrix through a CNN model to obtain a high-dimensional implicit feature map; and a fusion secondary subunit, used to fuse the second corrected convolution feature vector with the high-dimensional implicit feature map to obtain the topological information doped temperature correlation pattern feature vector.
[0041] In a specific example, the decoding subunit is used to: use the multiple fully connected layers of the decoder-based temperature-corrected measurement estimate generator to decode and regress the topology information-doped temperature correlation pattern feature vector to obtain the temperature-corrected measurement estimate using the following decoding formula, wherein the decoding formula is: Wherein, V1 is the characteristic vector of the correlation pattern between the topological information doping temperature, Y is the temperature correction measurement estimate, m1 is the weight matrix, B is the bias vector, represents matrix multiplication and b(·) is the activation function.
[0042] Furthermore, in the technical solution of the present application, the universal fully automatic calibration device for the blood culture instrument also includes a training module for training the implicit correlation pattern feature extractor between temperatures based on the one-dimensional convolution layer, the spatial topology feature extractor between sensors based on the convolutional neural network model, the spatial mapping fusion device and the decoder-based temperature correction measurement estimate generator.
[0043] In one example, the training module includes: a training data acquisition unit, used to acquire training data, wherein the training data includes training temperature values collected by multiple integrated temperature and illumination sensors, a training space topology matrix between multiple integrated temperature and illumination sensors, and a true value of a temperature-corrected measurement estimate; a training temperature implicit correlation pattern feature extraction unit, used to arrange the training temperature values collected by multiple integrated temperature and illumination sensors into a training temperature input vector according to a sample dimension, and then pass the training temperature input vector through the temperature implicit correlation pattern feature extractor based on a one-dimensional convolutional layer to obtain a training temperature correlation pattern feature vector; a training sensor space topology feature extraction unit, used to pass the training space topology matrix through the sensor space topology feature extractor based on a convolutional neural network model to obtain a training space topology feature matrix; A training space mapping fusion unit, used to use the space mapping fuser to process the training space topology feature matrix and the training temperature correlation pattern feature vector to obtain a training topology information-doped temperature correlation pattern feature vector; a decoding loss function value generation unit, used to pass the training topology information-doped temperature correlation pattern feature vector through the decoder-based temperature correction measurement estimate value generator to obtain a decoding loss function value; and a loss training unit, used to train the one-dimensional convolutional layer-based temperature implicit correlation pattern feature extractor, the convolutional neural network model-based sensor space topology feature extractor, the space mapping fuser and the decoder-based temperature correction measurement estimate value generator with the decoding loss function value, wherein in each round of iteration of the training, the training topology information-doped temperature correlation pattern feature vector is corrected.
[0044] In the above technical scheme, the training temperature-to-temperature correlation pattern feature vector expresses the one-dimensional local correlation characteristics of the temperature values collected by the multiple integrated temperature and illumination sensors, and the training spatial topology feature matrix expresses the spatial topology distribution characteristics of the multiple integrated temperature and illumination sensors. Therefore, after using a spatial mapping fusion device to process the training spatial topology feature matrix and the training temperature-to-temperature correlation pattern feature vector, the training topology information-doped temperature-to-temperature correlation pattern feature vector may include the temperature value correlation characteristics of the multiple integrated temperature and illumination sensors based on their spatial topology distribution, thereby improving the expression effect of the training topology information-doped temperature-to-temperature correlation pattern feature vector for the sample value correlation characteristics of multiple samples with specific spatial topology distribution.
[0045] However, when the correlation features of multiple sample space topological distribution features and their corresponding sample values are integrated, the significance of the feature distribution information based on their specific feature distributions of the training space topological feature matrix and the training temperature correlation pattern feature vector will be affected, making it difficult to stably focus on the significant local distribution of features when the training topological information-doped temperature correlation pattern feature vector is decoded by a decoder, thereby affecting the training speed.
[0046] Based on this, the applicant of the present application corrects the training topology information doped temperature correlation pattern feature vector V each time the training topology information doped temperature correlation pattern feature vector is iterated through decoding regression by the decoder.
[0047] Accordingly, in one example, in each iteration of the training, the training topology information doping temperature correlation pattern feature vector is corrected by the following correction formula to obtain a corrected training topology information doping temperature correlation pattern feature vector; wherein the correction formula is:
[0048]
[0049] Where V is the characteristic vector of the correlation pattern between the training topological information doping temperature, v i is the i-th eigenvalue of the characteristic vector of the correlation pattern between the training topological information doping temperature, and are the squares of the 1-norm and 2-norm of the training topological information doping temperature correlation pattern feature vector V, L is the length of the training topological information doping temperature correlation pattern feature vector V, and ω is a weight hyperparameter, v i is the eigenvalue of the characteristic vector V of the correlation pattern between the training topological information and the doping temperature, log[·] represents the logarithmic function with base 2, v i ′ is the i-th eigenvalue of the characteristic vector of the correlation pattern between the topological information doping temperatures after correction.
[0050] Specifically, by performing geometric registration of the high-dimensional feature manifold shape based on the scale and structural parameters of the training topological information-doped temperature correlation pattern feature vector V, we can focus on features with rich feature semantic information in the feature set composed of the eigenvalues of the training topological information-doped temperature correlation pattern feature vector V, that is, distinguishable and stable interest features based on the dissimilarity of local context information representation when the decoder is decoding, thereby realizing the saliency annotation of feature information in the decoding process and improving the training speed of the decoder.
[0051] In summary, the universal fully automatic calibration device 100 for a blood culture instrument according to the embodiment of the present application is explained, which can intelligently generate a temperature correction measurement estimate to achieve automatic calibration of the temperature of the blood culture instrument.
[0052] As described above, according to the embodiment of the present application, the universal fully-automatic calibration device 100 for blood culture instruments based on the embodiment of the present application can be implemented in various terminal devices, such as a server having a universal fully-automatic calibration algorithm for blood culture instruments based on the embodiment of the present application. In one example, the universal fully-automatic calibration device 100 for blood culture instruments based on the embodiment of the present application can be integrated into the terminal device as a software module and / or a hardware module. For example, the universal fully-automatic calibration device 100 for blood culture instruments based on the embodiment of the present application can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the universal fully-automatic calibration device 100 for blood culture instruments based on the embodiment of the present application can also be one of the many hardware modules of the terminal device.
[0053] Alternatively, in another example, the universal fully-automatic calibration device 100 for a blood culture instrument based on an embodiment of the present application and the terminal device may also be separate devices, and the universal fully-automatic calibration device 100 for a blood culture instrument may be connected to the terminal device via a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.
[0054] Figure 2 Flow chart of a universal fully automatic calibration method for a blood culture instrument according to an embodiment of the present application. Figure 3 Schematic diagram of the system architecture of the universal fully automatic calibration method for blood culture instrument according to the embodiment of the present application. Figure 2 and Figure 3 As shown, according to the universal fully automatic calibration method of the blood culture instrument of the embodiment of the present application, it includes: S110, obtaining temperature values collected by multiple integrated temperature and illumination sensors; S120, obtaining the spatial topological matrix between the multiple integrated temperature and illumination sensors, wherein the values of each non-diagonal position in the spatial topological matrix are used to represent the spatial distance between the corresponding two integrated temperature and illumination sensors; S130, extracting sample correlation features of the temperature values collected by the multiple integrated temperature and illumination sensors to obtain a temperature correlation pattern feature vector; S140, extracting the topological features of the spatial topological matrix to obtain a spatial topological feature matrix; and, S150, generating a temperature correction measurement estimate based on the mapping fusion interaction features of the temperature correlation pattern feature vector and the spatial topological feature matrix.
[0055] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned universal fully automatic calibration method for blood culture instrument have been referred to above. Figure 1 It has been described in detail in the description of the universal fully-automatic calibration device 100 for a blood culture instrument, and therefore, its repeated description will be omitted.
[0056] Figure 4 FIG. 1 is an application scenario diagram of a universal fully automatic calibration device for a blood culture instrument according to an embodiment of the present application. Figure 4 As shown, in this application scenario, first, the temperature values collected by multiple integrated temperature and illumination sensors are obtained (for example, Figure 4 D1 as shown in FIG) and the spatial distance between multiple integrated temperature and illuminance sensors (e.g., Figure 4 Then, the temperature values collected by the multiple integrated temperature and illumination sensors and the spatial distances between the multiple integrated temperature and illumination sensors are input to a server (for example, Figure 4 In S) shown in , the server is able to use the universal fully automatic calibration algorithm of the blood culture instrument to process the temperature values collected by multiple integrated temperature and illumination sensors and the spatial distances between multiple integrated temperature and illumination sensors to obtain temperature-corrected measurement estimates.
[0057] It should be understood that temperature and optical parameters are the main calibration parameters of the automated blood culture system. This embodiment describes a universal fully automatic calibration device for a blood culture instrument, including a data acquisition system, a data receiving and processing system, and an auxiliary support system. Further, in one embodiment of the present application, the data acquisition system includes 9 calibration units.
[0058] like Figure 5 and Figure 6As shown, the calibration unit comprises a shell (1), a telescopic second shell (2), a semi-embedded third shell (3), a charging interface (4), an integrated temperature and illumination sensor (5) and a calibration unit circuit module (6). The shell (1) is used to protect the internal structure of the calibration unit; the telescopic second shell (2) is located at the upper end of the calibration unit and is used to adjust the length of the calibration unit to match the inner hole length of blood culture instruments of different specifications; the semi-embedded third shell (3) can be embedded to the outside of the calibration unit and is used to adjust the length of the calibration unit, and can be configured with different sizes to match the inner hole diameter of blood culture instruments of different specifications; the charging interface (4) is located at the bottom of the calibration unit and is used to charge the calibration unit; the integrated temperature and illumination sensor (5) is located at the bottom of the calibration unit and is used to synchronously monitor temperature and illumination data; the calibration unit circuit module (6) comprises a calibration unit battery, a calibration unit circuit board and a signal transmitter. The calibration unit battery is used to provide energy required for the normal endurance of the calibration unit; the calibration unit circuit board is used to receive temperature and illumination data collected by temperature and illumination sensors; and the signal transmitter is used to wirelessly transmit the temperature and illumination data to a wireless signal receiving device of a computer host (7).
[0059] like Figure 7 As shown, the data receiving and processing system comprises a computer host (7) and a display module (8). The computer host (7) is located inside the upper flip cover of the containing box (9) and is equipped with a wireless signal receiving device and a blood culture instrument calibration device software. The wireless signal receiving device is used to collect temperature and illumination data; the blood culture instrument calibration device software is used for calibration unit control, original record table output and calibration result calculation. The display module (8) is embedded in the upper flip cover surface of the containing box (9) and is used to view the output information of the computer host (7).
[0060] like Figure 7 As shown, the auxiliary support system includes a containing box (9) and a box circuit module (10). The containing box (9) is used to contain the calibration unit, and to embed the computer host (7), the display module (8) and the box circuit module (10), so as to protect the universal fully automatic calibration device of the blood culture instrument from mechanical shock, pressure or humidity. The box circuit module (10) includes a box battery, a calibration unit charging male connector (11) and a box wire. The box battery is located at the lower part of the containing box (9) and is used to provide the energy required for the operation of the universal fully automatic calibration device of the blood culture instrument; the calibration unit charging male connector (11) is located and embedded in the lower surface of the containing box (9), and is provided with a protrusion matching the calibration unit charging interface (4) for synchronous charging of the calibration unit; the box wire is used to connect the upper flip cover and the lower part of the containing box (9) and is used to power the computer host (7) and the display module (8).
[0061] Furthermore, a test method is provided. Specifically, after the blood culture instrument to be tested works normally for 2 hours, the calibration units of the calibration device described in this application and the existing calibration devices on the market are placed adjacent to each other in the two middle chambers of the blood culture instrument, and temperature readings are recorded every 30 seconds. The judgment standard for "temperature stability" is: 10 consecutive temperature readings are all within the range of ±0.1°C of the average value of the 10 readings. The time required for the calibration device described in this application and the existing calibration device on the market to judge the "temperature stability" of the blood culture instrument is recorded respectively.
[0062] Accordingly, the results show that the time required for the existing calibration device on the market to determine that the blood culture instrument is "temperature stable" is 33 minutes, while the time required for the calibration device described in this application to determine that the blood culture instrument is "temperature stable" is 8 minutes. Therefore, the temperature measurement efficiency of the calibration device described in this application is significantly better than that of the existing calibration devices on the market.
[0063] The present application uses specific words to describe the embodiments of the present application. For example, "first / second embodiment", "one embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or multiple times in different positions in this specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present application can be appropriately combined.
[0064] In addition, it will be appreciated by those skilled in the art that various aspects of the present application may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may all be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of the present application may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.
[0065] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.
[0066] The above is an explanation of the present application and should not be considered as a limitation thereof. Although several exemplary embodiments of the present application are described, it will be readily understood by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application. Therefore, all of these modifications are intended to be included within the scope of the present application as defined by the claims. It should be understood that the above is an explanation of the present application and should not be considered to be limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present application is defined by the claims and their equivalents.
Claims
1. A universal fully automatic calibration device for a blood culture instrument, characterized in that: include: Data acquisition system, data receiving and processing system and auxiliary support system; Wherein, the data acquisition system includes a plurality of calibration units, each of which includes a housing assembly, a charging interface mounted on the housing assembly, an integrated temperature and illumination sensor, and a calibration unit circuit module; The data receiving and processing system comprises a computer host and a display module electrically connected to the computer host; The auxiliary support system includes a containing box and a box circuit module installed on the containing box; The calibration unit circuit module comprises: A data acquisition unit, used to acquire temperature values collected by a plurality of the integrated temperature and illumination sensors; A spatial topology matrix acquisition unit, used to acquire a spatial topology matrix between the plurality of integrated temperature and illumination sensors, wherein the value of each non-diagonal position in the spatial topology matrix is used to represent the spatial distance between the corresponding two integrated temperature and illumination sensors; A sample correlation feature extraction unit, used to extract sample correlation features of the temperature values collected by the integrated temperature and illumination sensors to obtain a temperature correlation pattern feature vector; A topological feature extraction unit, used to extract the topological features of the spatial topological matrix to obtain a spatial topological feature matrix; and A temperature correction measurement estimate generating unit, configured to generate a temperature correction measurement estimate based on the mapping fusion interaction features of the temperature correlation pattern feature vector and the spatial topology feature matrix; The sample-related feature extraction unit comprises: a vectorization subunit, configured to arrange the temperature values collected by the plurality of integrated temperature and illumination sensors into a temperature input vector according to a sample dimension; and The temperature implicit correlation pattern feature extraction subunit is used to obtain the temperature correlation pattern feature vector by passing the temperature input vector through the temperature implicit correlation pattern feature extractor based on the one-dimensional convolution layer; The topological feature extraction unit is used to: The spatial topology matrix is passed through an inter-sensor spatial topology feature extractor based on a convolutional neural network model to obtain the spatial topology feature matrix; The temperature-corrected measurement estimate generating unit comprises: A space mapping fusion subunit, configured to use a space mapping fusion device to process the space topology feature matrix and the temperature correlation pattern feature vector to obtain a topology information-doped temperature correlation pattern feature vector; and A decoding subunit, configured to mix the topological information with the temperature correlation pattern feature vector and pass it through a decoder-based temperature correction measurement estimate generator to obtain a temperature correction measurement estimate; The spatial mapping fusion subunit comprises: A first convolution secondary subunit, used for passing the temperature correlation pattern feature vector through a point convolution layer to obtain a first convolution feature vector; A first corrected secondary subunit, configured to obtain a first corrected convolutional feature vector by passing the first convolutional feature vector through a corrected linear unit based on a ReLU function; A second convolution secondary subunit, configured to pass the first modified convolution feature vector through a point convolution layer to obtain a second convolution feature vector; A second corrected secondary subunit, used for passing the second convolution feature vector through a corrected linear unit based on a Sigmoid function to obtain a second corrected convolution feature vector; A convolutional coding secondary subunit, used for passing the spatial topological feature matrix through a CNN model to obtain a high-dimensional implicit feature map; and A fusion secondary subunit, used for fusing the second modified convolution feature vector with the high-dimensional implicit feature map to obtain a feature vector of a correlation pattern between the topological information doping temperature; The decoding subunit is used for: The temperature-corrected measurement estimate value is obtained by decoding and regressing the topological information-doped temperature correlation pattern feature vector using the multiple fully connected layers of the decoder-based temperature-corrected measurement estimate generator according to the following decoding formula, wherein the decoding formula is: Wherein, V1 is the characteristic vector of the correlation pattern between the topological information doping temperature, Y is the temperature correction measurement estimate, M1 is the weight matrix, B is the bias vector, represents matrix multiplication, h(·) is the activation function; The universal fully automatic calibration device for blood culture instrument also includes a training module for training the temperature implicit correlation pattern feature extractor based on the one-dimensional convolution layer, the sensor space topology feature extractor based on the convolutional neural network model, the space mapping fusion device and the decoder-based temperature correction measurement estimation value generator; The training module comprises: A training data acquisition unit, configured to acquire training data, wherein the training data includes training temperature values collected by the plurality of integrated temperature and illumination sensors, a training space topology matrix between the plurality of integrated temperature and illumination sensors, and a true value of a temperature-corrected measurement estimate; A training temperature implicit correlation pattern feature extraction unit, configured to arrange the training temperature values collected by the plurality of integrated temperature and illumination sensors into a training temperature input vector according to the sample dimension, and then pass the training temperature input vector through the temperature implicit correlation pattern feature extractor based on the one-dimensional convolution layer to obtain a training temperature correlation pattern feature vector; A training inter-sensor spatial topology feature extraction unit, used for passing the training spatial topology matrix through the inter-sensor spatial topology feature extractor based on the convolutional neural network model to obtain a training spatial topology feature matrix; A training space mapping fusion unit, used to use the space mapping fusion device to process the training space topology feature matrix and the training temperature correlation pattern feature vector to obtain a training topology information doped temperature correlation pattern feature vector; A decoding loss function value generating unit, configured to pass the training topology information mixed with the temperature correlation pattern feature vector through the decoder-based temperature correction measurement estimate value generator to obtain a decoding loss function value; and A loss training unit, used to train the temperature implicit correlation pattern feature extractor based on the one-dimensional convolution layer, the sensor space topology feature extractor based on the convolutional neural network model, the spatial mapping fusion device and the decoder-based temperature correction measurement estimation value generator with the decoding loss function value, wherein in each round of iteration of the training, the training topology information doped temperature correlation pattern feature vector is corrected; In each round of iteration of the training, the training topology information doping temperature correlation pattern feature vector is corrected by the following correction formula to obtain the corrected training topology information doping temperature correlation pattern feature vector; wherein the correction formula is: Where V is the characteristic vector of the correlation pattern between the training topological information doping temperature, v i is the i-th eigenvalue of the characteristic vector of the correlation pattern between the training topological information doping temperature, and are the squares of the 1-norm and 2-norm of the training topological information doping temperature correlation pattern feature vector V, L is the length of the training topological information doping temperature correlation pattern feature vector V, and ω is a weight hyperparameter, v i is the eigenvalue of the characteristic vector V of the correlation pattern between the training topological information and the doping temperature, log[·] represents the logarithmic function with base 2, v i ′ is the i-th eigenvalue of the characteristic vector of the correlation pattern between the corrected training topological information doping temperatures.
2. A method for universal automatic calibration of a blood culture instrument for use in the universal automatic calibration device of a blood culture instrument as claimed in claim 1, characterized in that: include: Get temperature values collected by multiple integrated temperature and illuminance sensors; Acquire a spatial topology matrix between the plurality of integrated temperature and illumination sensors, wherein the value of each non-diagonal position in the spatial topology matrix is used to represent the spatial distance between the corresponding two integrated temperature and illumination sensors; Extracting sample correlation features of temperature values collected by the integrated temperature and illumination sensors to obtain a temperature correlation pattern feature vector; Extracting topological features of the spatial topological matrix to obtain a spatial topological feature matrix; and A temperature-corrected measurement estimate is generated based on the mapping fusion interaction features of the temperature correlation pattern feature vector and the spatial topology feature matrix.
