High-precision self-adaptive temperature regulation and control system for medical instrument

Abstract

The invention relates to a high-precision self-adaptive temperature regulation and control system for a medical instrument. The system comprises a feature analysis module which collects a multi-dimensional temperature information set with a timestamp and deduces a temperature space-time change trend; and the temperature prediction module generates an optimized temperature distribution diagram by using a Kalman filtering algorithm based on the trend. The temperature deviation value of each area is compared with a preset threshold value, if the deviation exceeds the standard, a dynamic parameter configuration table is generated, and an adaptive power regulation scheme is formulated. And the regulation and control execution module regulates and controls the equipment power according to the regulation scheme to obtain an updated temperature monitoring data set. By analyzing the relevance between the data set and the multi-dimensional temperature information set, a stability optimization parameter is generated, so that a control rule of a fuzzy logic algorithm is adjusted, and a self-adaptive temperature regulation and control scheme is formed. According to the system, accurate prediction, regulation and control of the temperature are realized. The adaptive capacity of the system is improved, and the stability and reliability of temperature regulation and control can be improved by dynamically optimizing the control rule.

Description

Technical Field

[0001] The present invention belongs to the field of medical devices, and particularly relates to a high-precision adaptive temperature control system for medical instruments. Background Art

[0002] With the development of medical device technology, high-precision adaptive temperature control technology for medical instruments has emerged. In the modern medical system, the performance of medical instruments is closely related to temperature control technology. High-precision temperature control is crucial for the stable operation of instruments, the improvement of treatment effects, and even the safety of patients. From medical imaging devices to surgical instruments, the temperature control accuracy directly affects the imaging quality and operation accuracy. In the treatment process, the temperature control level of some devices is directly related to the treatment effect. Moreover, the precise regulation of the patient's body temperature is also indispensable. At the same time, the industry's requirements for temperature control technology continue to rise, pursuing intelligence, refinement, integration, and adaptability. Summary of the Invention

[0003] Based on this, it is necessary to provide a high-precision adaptive temperature control system for medical instruments that can improve the adaptive ability of the system and continuously enhance the stability and reliability of temperature control in response to the above technical problems.

[0004] In a first aspect, the present application provides a high-precision adaptive temperature control system for medical instruments, including:

[0005] A feature analysis module, configured to obtain a multi-dimensional temperature information set with timestamps, and further configured to analyze the spatio-temporal features of the multi-dimensional temperature information set using a convolutional neural network to obtain the temperature spatio-temporal change trend.

[0006] A temperature prediction module, configured to generate an optimized temperature distribution map using the Kalman filter algorithm according to the temperature spatio-temporal change trend; further configured to compare the temperature deviation values of each region in the optimized temperature distribution map based on a preset threshold, and if the deviation value is greater than the preset threshold, generate a dynamic parameter configuration table; further configured to calculate the power adjustment amount using the fuzzy logic algorithm according to the dynamic parameter configuration table to generate an adaptive power adjustment scheme.

[0007] A regulation execution module, configured to adjust the device power according to the adaptive power adjustment scheme to obtain an updated temperature monitoring data set; further configured to analyze the correlation between the temperature monitoring data set and the multi-dimensional temperature information set to generate stability optimization parameters; further configured to adjust the control rules of the fuzzy logic algorithm according to the stability optimization parameters to obtain an adaptive temperature control scheme.

[0008] In one embodiment, analyzing the spatio-temporal features of the multi-dimensional temperature information set using a convolutional neural network to obtain the temperature spatio-temporal change trend includes:

[0009] Perform spatio-temporal fusion processing on the multi-dimensional temperature information set to obtain a spatio-temporal feature matrix; the multi-dimensional temperature information set includes temperature data in the time dimension and the space dimension.

[0010] Input the spatio-temporal feature matrix into a convolutional neural network and perform cross-channel feature fusion in combination with a time sliding window to obtain a first feature map.

[0011] Perform multi-scale pooling operations on the first feature map to generate a pooled feature vector, and calculate the spatio-temporal weight coefficient using a formula.

[0012] Perform weighted aggregation on the pooled feature vector according to the spatio-temporal weight coefficient to obtain a temperature spatio-temporal change trend vector.

[0013] Perform a difference calculation between the temperature spatio-temporal change trend vector and a preset reference trend to obtain a trend offset.

[0014] Update the spatio-temporal weight coefficient according to the trend offset and perform secondary weighting with the pooled feature vector to obtain an optimized temperature spatio-temporal change trend.

[0015] In one embodiment, the spatio-temporal weight coefficient is calculated by the following formula:

[0016]

[0017] α i =Softmax(ReLU(S ωi +W ti ))

[0018] where, W t represents the weight coefficient in the time dimension, n represents the feature dimension, m represents the time step, v i represents the i-th feature vector, h i represents the hidden state vector, S ω represents the spatial weight coefficient, H and W respectively represent the height and width of the feature map, M ij represents the attention mask value at the position (i,j), F ij represents the feature value at the position (i,j), α i represents the spatio-temporal weight coefficient, Softmax() represents the normalization function, and ReLU() represents the activation function.

[0019] In one embodiment, use the Kalman filter algorithm according to the temperature spatio-temporal change trend to generate an optimized temperature distribution map, including:

[0020] Input the parameters in the temperature spatio-temporal change trend into a trained spatio-temporal trend model to obtain a sequence of temperature prediction values.

[0021] Calculate the deviation amount based on the temperature prediction value sequence and the multi-dimensional temperature information set to obtain the dynamic weight parameter.

[0022] Use the dynamic weight parameter to control the fusion algorithm to process the temperature prediction value sequence and the multi-dimensional temperature information set to obtain the residual sequence.

[0023] Use the Kalman filter algorithm to correct the residual sequence to obtain the corrected covariance matrix.

[0024] Update the parameters of the spatio-temporal trend model according to the covariance matrix to obtain the optimized temperature distribution map.

[0025] In one embodiment, calculate the power adjustment amount using the fuzzy logic algorithm according to the dynamic parameter configuration table to generate an adaptive power adjustment scheme, including:

[0026] Generate a membership function set according to the real-time parameter type and parameter threshold in the dynamic parameter configuration table; the membership function set includes temperature membership and load rate membership.

[0027] Perform fuzzy inference on the membership function set using the fuzzy logic algorithm to generate the power adjustment amount.

[0028] Adjust the corresponding device according to the power adjustment amount and update the power adjustment rule using the feedback data to obtain the corrected membership function set; the device feedback data includes the actual power fluctuation value and the adjustment error rate.

[0029] Generate an adaptive power adjustment scheme based on the corrected membership function set; the scheme includes a dynamic power curve and adjustment period parameters.

[0030] In one embodiment, analyze the correlation based on the temperature monitoring data set and the multi-dimensional temperature information set to generate stability optimization parameters, including:

[0031] Construct the time series features in the temperature monitoring data set and the spatial distribution pattern extracted from the multi-dimensional temperature information set using the clustering algorithm to obtain a multi-dimensional correlation matrix; the time series features include the temperature fluctuation coefficient and the acquisition timestamp; the spatial distribution pattern includes the device node location and the temperature gradient difference.

[0032] Calculate the device response delay data according to the multi-dimensional correlation matrix matching the historical trend baseline; the historical trend baseline includes the device operating state and the temperature stability threshold.

[0033] Input the device response delay data into the parameter optimization model to obtain stability optimization parameters including the temperature compensation coefficient and the acquisition frequency adjustment value.

[0034] In a second aspect, the present application also provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the foregoing method is implemented.

[0035] In a third aspect, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the foregoing method is implemented.

[0036] For the above-mentioned high-precision adaptive temperature control system for medical instruments, the feature analysis module collects a multi-dimensional temperature information set with timestamps, analyzes the spatio-temporal features in the data by means of a convolutional neural network, and deduces the spatio-temporal change trend of temperature. The temperature prediction module generates an optimized temperature distribution map based on the above trend by using the Kalman filter algorithm. By comparing the temperature deviation values of each region with a preset threshold, if the deviation exceeds the standard, a dynamic parameter configuration table is generated, and then the power adjustment amount is calculated by using the fuzzy logic algorithm to formulate an adaptive power adjustment scheme. The control execution module adjusts the device power according to the adaptive power adjustment scheme to obtain an updated temperature monitoring data set. By analyzing the correlation between this data set and the multi-dimensional temperature information set, stability optimization parameters are generated, and accordingly, the control rules of the fuzzy logic algorithm are adjusted to form an adaptive temperature control scheme. This system realizes the full-process closed-loop management from data collection and analysis, temperature prediction, to control execution and optimization. It improves the accuracy and timeliness of temperature data processing, and realizes the accurate prediction and control of temperature. The coordinated operation among the modules not only improves the adaptive ability of the system to cope with complex and changeable medical scenarios, but also can continuously improve the stability and reliability of temperature control by dynamically optimizing the control rules, providing a strong guarantee for the stable operation of medical instruments. Description of the Drawings

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0038] Figure 1 It is a structural block diagram of a high-precision adaptive temperature control system for medical instruments provided by an embodiment of the present invention;

[0039] Figure 2 It is a flowchart of generating an optimized temperature distribution map by using the Kalman filter algorithm according to the spatio-temporal change trend of temperature provided by an embodiment of the present invention;

[0040] Figure 3 It is a flowchart of generating stability optimization parameters based on the analysis of the correlation between the temperature monitoring data set and the multi-dimensional temperature information set provided by an embodiment of the present invention. Detailed implementation manners

[0041] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0042] In one of the embodiments, as Figure 1 shown, the present application provides a high-precision adaptive temperature control system for medical instruments, which may include:

[0043] A feature analysis module 101, configured to obtain a multi-dimensional temperature information set with timestamps, and further configured to analyze the spatio-temporal features of the multi-dimensional temperature information set by using a convolutional neural network to obtain a temperature spatio-temporal change trend.

[0044] First, temperature sensors are deployed at key parts of the medical instrument to collect temperature data in all directions, and each group of data is marked with an accurate timestamp to construct a multi-dimensional temperature information set covering the time dimension and the space dimension. Subsequently, this information is input into a carefully constructed convolutional neural network model. The network deeply mines the multi-dimensional temperature information set by designing specific convolutional kernels and time sliding windows, automatically extracts the spatio-temporal features contained in the data, including the temperature change rules at different time nodes and the temperature distribution patterns at different spatial positions of the instrument, and finally obtains a temperature spatio-temporal change trend, providing a key basis for subsequent temperature prediction.

[0045] A temperature prediction module 102, configured to generate an optimized temperature distribution map by using a Kalman filtering algorithm according to the temperature spatio-temporal change trend; further configured to compare the temperature deviation values of each region in the optimized temperature distribution map based on a preset threshold, and if the deviation value is greater than the preset threshold, generate a dynamic parameter configuration table; further configured to calculate a power adjustment amount by using a fuzzy logic algorithm according to the dynamic parameter configuration table to generate an adaptive power adjustment scheme.

[0046] Specifically, based on the temperature spatio-temporal change trend output by the feature analysis module, the temperature prediction module inputs the key parameters in the temperature spatio-temporal change trend into the trained spatio-temporal trend model, and uses the model to calculate the temperature prediction value sequence. To improve the prediction accuracy, the module calculates the deviation between the temperature prediction value sequence and the multi-dimensional temperature information set, thereby determining the dynamic weight parameter. Using the dynamic weight parameter, the module controls the fusion algorithm to fuse the temperature prediction value sequence and the multi-dimensional temperature information set, generates a residual sequence, and uses the Kalman filter algorithm to correct the residual sequence to obtain a corrected covariance matrix, and accordingly updates the spatio-temporal trend model parameters to generate an optimized temperature distribution map. In addition, the module compares the temperature deviation values in each region of the optimized temperature distribution map with a preset threshold, and immediately generates a dynamic parameter configuration table once the deviation value exceeds the threshold. Finally, based on the dynamic parameter configuration table, the module uses the fuzzy logic algorithm, taking into account key factors such as temperature and load rate, calculates the power adjustment amount, and formulates an adaptive power adjustment plan.

[0047] The regulation execution module 103 is used to adjust the device power according to the adaptive power adjustment plan to obtain an updated temperature monitoring data set; it is also used to analyze the correlation between the temperature monitoring data set and the multi-dimensional temperature information set to generate stability optimization parameters; it is also used to adjust the control rules of the fuzzy logic algorithm according to the stability optimization parameters to obtain an adaptive temperature regulation plan.

[0048] According to the adaptive power adjustment plan, the module sends instructions to the power adjustment unit of the medical instrument to adjust the device power, collects and integrates new temperature data to form an updated temperature monitoring data set. To further improve the system regulation stability, the module uses a clustering algorithm to analyze the time series features in the temperature monitoring data set, such as the temperature fluctuation coefficient and the acquisition timestamp, and the spatial distribution patterns extracted from the multi-dimensional temperature information set, such as the device node position and the temperature gradient difference, to construct a multi-dimensional correlation matrix. By matching the multi-dimensional correlation matrix with the historical trend baseline, the device response delay data is obtained and input into the parameter optimization model to generate stability optimization parameters including the temperature compensation coefficient and the acquisition frequency adjustment value. Finally, based on the stability optimization parameters, the module adjusts the control rules of the fuzzy logic algorithm to form an adaptive temperature regulation plan, realizing precise and stable regulation of the temperature of the medical instrument.

[0049] The above-mentioned high-precision adaptive temperature control system for medical instruments, the feature analysis module collects a multi-dimensional temperature information set with timestamps, and analyzes the spatio-temporal features in the data by means of a convolutional neural network to deduce the spatio-temporal change trend of temperature. The temperature prediction module generates an optimized temperature distribution map based on the above trend by using the Kalman filter algorithm. By comparing the temperature deviation values of each region with a preset threshold, if the deviation exceeds the standard, a dynamic parameter configuration table is generated, and then the power adjustment amount is calculated by using the fuzzy logic algorithm to formulate an adaptive power adjustment scheme. The control execution module adjusts the device power according to the adaptive power adjustment scheme to obtain an updated temperature monitoring data set. By analyzing the correlation between this data set and the multi-dimensional temperature information set, stability optimization parameters are generated, and accordingly, the control rules of the fuzzy logic algorithm are adjusted to form an adaptive temperature control scheme. This system realizes the full-process closed-loop management from data collection and analysis, temperature prediction to control execution and optimization. It improves the accuracy and timeliness of temperature data processing, and realizes the accurate prediction and control of temperature. The coordinated operation among modules not only improves the adaptive ability of the system to cope with complex and changeable medical scenarios, but also continuously improves the stability and reliability of temperature control by dynamically optimizing the control rules, providing a strong guarantee for the stable operation of medical instruments.

[0050] In one of the embodiments, analyzing the spatio-temporal features of the multi-dimensional temperature information set by using a convolutional neural network to obtain the spatio-temporal change trend of temperature may include the following steps:

[0051] Step S201, perform spatio-temporal fusion processing on the multi-dimensional temperature information set to obtain a spatio-temporal feature matrix; the multi-dimensional temperature information set includes temperature data in the time dimension and the space dimension.

[0052] Step S202, input the spatio-temporal feature matrix into the convolutional neural network and perform cross-channel feature fusion in combination with a time sliding window to obtain a first feature map.

[0053] Step S203, perform multi-scale pooling operations on the first feature map to generate a pooled feature vector, and calculate the spatio-temporal weight coefficient by using a formula.

[0054] Step S204, perform weighted aggregation on the pooled feature vector according to the spatio-temporal weight coefficient to obtain a spatio-temporal change trend vector of temperature.

[0055] Step S205, perform difference calculation on the spatio-temporal change trend vector of temperature and a preset reference trend to obtain a trend offset.

[0056] Step S206, update the spatio-temporal weight coefficient according to the trend offset and perform secondary weighting with the pooled feature vector to obtain an optimized spatio-temporal change trend of temperature.

[0057] Specifically, first, perform spatio-temporal fusion processing on the multi-dimensional temperature information set containing temperature data in time and space dimensions to form a spatio-temporal feature matrix. Then, input this matrix into a convolutional neural network, and perform cross-channel feature fusion using a time sliding window to obtain a first feature map. Subsequently, perform multi-scale pooling operations on the first feature map to generate a pooled feature vector, and calculate the spatio-temporal weight coefficient using a specific formula. After that, perform weighted aggregation on the pooled feature vector according to the spatio-temporal weight coefficient to obtain a temperature spatio-temporal change trend vector. Then, calculate the difference between this vector and a preset reference trend to obtain a trend offset. Finally, update the spatio-temporal weight coefficient according to the trend offset and perform weighted calculation with the pooled feature vector again to obtain an optimized temperature spatio-temporal change trend.

[0058] This embodiment can deeply mine the spatio-temporal features in the multi-dimensional temperature information set and accurately capture the variation laws of temperature in time and space. By using a convolutional neural network and multi-scale pooling operations, the accuracy and comprehensiveness of feature extraction are improved. The calculation and dynamic update of the spatio-temporal weight coefficient, combined with quadratic weighted processing, effectively enhance the representation ability and prediction accuracy of the temperature spatio-temporal change trend, providing a solid and reliable foundation for subsequent applications such as temperature regulation based on the temperature change trend, and helping to improve the accuracy and stability of relevant systems in temperature control.

[0059] In one of the embodiments, the spatio-temporal weight coefficient can be calculated through the following formula:

[0060]

[0061] α i = Softmax(ReLU(S ωi + W ti ))

[0062] Where, W t represents the weight coefficient in the time dimension, n represents the feature dimension, m represents the time step, v i represents the i-th feature vector, h i represents the hidden state vector, S ω represents the spatial weight coefficient, H and W respectively represent the height and width of the feature map, M ij represents the attention mask value at position (i, j), F ij represents the feature value at position (i, j), α i represents the spatio-temporal weight coefficient, Softmax() represents the normalization function, and ReLU() represents the activation function.

[0063] By assigning different weights to features at different positions through the attention mask value, it is possible to adaptively focus on key spatio-temporal information and avoid noise interference. The ReLU and Softmax functions are introduced in the calculation process. The former enhances the model's fitting ability for complex data, and the latter normalizes the weights, enabling the model to allocate weights more reasonably. In addition, by fusing the information in the time and space dimensions, this formula comprehensively captures the spatio-temporal features of the data, provides strong support for the analysis and processing of multi-dimensional data, and effectively improves the accuracy and reliability of the model based on spatio-temporal data in feature extraction and trend prediction.

[0064] In one embodiment, as Figure 2 shown, using the Kalman filter algorithm according to the spatio-temporal change trend of temperature to generate an optimized temperature distribution map may include the following steps:

[0065] Step S301: Input the parameters in the spatio-temporal change trend of temperature into the trained spatio-temporal trend model to obtain a sequence of temperature prediction values.

[0066] Step S302: Calculate the deviation amount based on the sequence of temperature prediction values and the multi-dimensional temperature information set to obtain dynamic weight parameters.

[0067] Step S303: Use the dynamic weight parameters to control the fusion algorithm to process the sequence of temperature prediction values and the multi-dimensional temperature information set to obtain a residual sequence.

[0068] Step S304: Use the Kalman filter algorithm to correct the residual sequence to obtain a corrected covariance matrix.

[0069] Step S305: Update the parameters of the spatio-temporal trend model according to the covariance matrix to obtain an optimized temperature distribution map.

[0070] Specifically, first input the relevant parameters in the spatio-temporal change trend of temperature into the trained spatio-temporal trend model, and obtain a sequence of temperature prediction values through the operation of the model. Then, compare and analyze this sequence of temperature prediction values with the multi-dimensional temperature information set, calculate the deviation amount between the two, and further determine the dynamic weight parameters based on this deviation amount. Subsequently, use these dynamic weight parameters to control the fusion algorithm to process the sequence of temperature prediction values and the multi-dimensional temperature information set, thereby obtaining a residual sequence. Then use the Kalman filter algorithm to carefully correct the residual sequence to obtain a corrected covariance matrix. Finally, update the parameters of the spatio-temporal trend model according to the covariance matrix, and thus obtain an optimized temperature distribution map.

[0071] By combining the spatio-temporal variation trend of temperature with the multi-dimensional temperature information set and using dynamic weight parameters and fusion algorithms, the information value of different data sources is fully exploited, improving the comprehensiveness and accuracy of data processing. The correction of the residual sequence by the Kalman filter algorithm effectively reduces the noise and error interference in the data and improves the accuracy of temperature prediction. The update of the spatio-temporal trend model parameters according to the covariance matrix enables the model to continuously adapt to the changes in the data, continuously optimize the temperature distribution map, provides a more reliable and accurate basis for subsequent decisions and regulations based on the temperature distribution, and helps to improve the performance and stability of the entire temperature regulation system.

[0072] In one embodiment, calculating a power adjustment amount by using a fuzzy logic algorithm according to a dynamic parameter configuration table and generating an adaptive power adjustment scheme may include the following steps:

[0073] Step S401, generating a membership function set according to the real-time parameter type and parameter threshold in the dynamic parameter configuration table; the membership function set includes a temperature membership degree and a load rate membership degree.

[0074] Step S402, performing fuzzy inference on the membership function set by using the fuzzy logic algorithm to generate a power adjustment amount.

[0075] Step S403, adjusting the corresponding device according to the power adjustment amount and updating the power adjustment rule by using the feedback data to obtain a corrected membership function set; the device feedback data includes an actual power fluctuation value and an adjustment error rate.

[0076] Step S404, generating an adaptive power adjustment scheme based on the corrected membership function set; the scheme includes a dynamic power curve and adjustment period parameters.

[0077] Specifically, first, according to the real-time parameter type and parameter threshold in the dynamic parameter configuration table, a membership function set is constructed, which covers the temperature membership degree and the load rate membership degree to quantify the key state information of the device operation. Then, the membership function set is imported into the fuzzy logic algorithm for fuzzy inference, and combined with the device operation environment and the current state, an appropriate power adjustment amount is generated. Subsequently, the system adjusts the corresponding device according to the power adjustment amount and collects the device feedback data, including the actual power fluctuation value and the adjustment error rate, and accordingly updates the power adjustment rule and optimizes the membership function set. Finally, the system formulates an adaptive power adjustment scheme including a dynamic power curve and adjustment period parameters based on the corrected membership function set.

[0078] By constructing a membership function set, the complex operating state of the device is transformed into quantifiable parameters, providing a basis for fuzzy logic reasoning and effectively enhancing the pertinence of power adjustment. The fuzzy logic algorithm fully considers the uncertainties in device operation and can make reasonable power control decisions based on real-time data. Meanwhile, the system dynamically updates the power adjustment rules and the membership function set with the device feedback data, enabling the power adjustment scheme to be continuously optimized according to the actual operating conditions of the device. The finally generated adaptive power adjustment scheme can not only effectively improve the operating stability and energy utilization efficiency of the device, reduce the device failure rate, and extend the service life of the device, but also provide comprehensive and refined power management support for device operation.

[0079] In one of the embodiments, as Figure 3 shown, analyzing the correlation based on the temperature monitoring data set and the multi-dimensional temperature information set to generate stability optimization parameters may include the following steps:

[0080] Step S501, constructing the time series features in the temperature monitoring data set and the spatial distribution patterns extracted from the multi-dimensional temperature information set by using a clustering algorithm to obtain a multi-dimensional correlation matrix; the time series features include the temperature fluctuation coefficient and the acquisition timestamp; the spatial distribution patterns include the device node position and the temperature gradient difference.

[0081] Step S502, calculating the device response delay data according to the multi-dimensional correlation matrix by matching the historical trend baseline; the historical trend baseline includes the device operating state and the temperature stability threshold.

[0082] Step S503, inputting the device response delay data into a parameter optimization model to obtain stability optimization parameters including the temperature compensation coefficient and the acquisition frequency adjustment value.

[0083] First, extract time series features such as the temperature fluctuation coefficient and the acquisition timestamp from the temperature monitoring data set, and at the same time refine spatial distribution patterns such as the device node position and the temperature gradient difference from the multi-dimensional temperature information set. With the help of a clustering algorithm, integrate these features in the time and space dimensions to construct a multi-dimensional correlation matrix. Subsequently, match and analyze this matrix with the historical trend baseline including the device operating state and the temperature stability threshold to calculate the device response delay data. Finally, input the device response delay data into a carefully designed parameter optimization model, and through model operation, obtain stability optimization parameters including the temperature compensation coefficient and the acquisition frequency adjustment value.

[0084] Integrate time - series features and spatial distribution patterns through a clustering algorithm to comprehensively explore the internal connections of temperature data in the spatio - temporal dimension and improve the integrity of data feature extraction. Match the multi - dimensional correlation matrix with the historical trend baseline, making full use of the historical experience of equipment operation, so that the calculated equipment response delay data is more in line with the actual operation situation. Input the response delay data into the parameter optimization model, which can generate stability optimization parameters in a targeted manner. These optimization parameters not only help improve the temperature regulation accuracy of the system, reduce the impact of temperature fluctuations on equipment operation, but also reduce system resource consumption while ensuring data quality by reasonably adjusting the acquisition frequency, further enhancing the stability and reliability of the entire temperature regulation system and providing strong support for the stable operation of the equipment.

[0085] It should be understood that although the steps in the flowcharts involved in the above - described embodiments are shown in sequence according to the arrows, these steps do not necessarily need to be executed in the order indicated by the arrows. Unless there is a clear description in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above - described embodiments may include multiple steps or multiple stages. These steps or stages do not necessarily need to be executed at the same moment, but can be executed at different moments, and the execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0086] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the high - precision adaptive temperature regulation system for medical instruments as described above are implemented.

[0087] In one embodiment, a computer - readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above - described method embodiments are implemented.

[0088] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can refer to the partial descriptions of the method embodiments. The device embodiments described above are only illustrative. The components described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the present disclosure solution. Those of ordinary skill in the art can understand and implement it without creative work.

[0089] The above-described embodiments merely represent several implementation manners of the embodiments of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent for the embodiments of the application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the embodiments of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the embodiments of the present application.

Claims

1. A high-precision adaptive temperature control system for medical instruments, characterized in that, The system includes: A feature analysis module, which is used to obtain a multi-dimensional temperature information set with timestamps, and is also used to analyze the spatio-temporal features of the multi-dimensional temperature information set by using a convolutional neural network to obtain the spatio-temporal temperature change trend; A temperature prediction module, which is used to generate an optimized temperature distribution map by using the Kalman filtering algorithm according to the spatio-temporal temperature change trend; is also used to compare the temperature deviation values of each region in the optimized temperature distribution map based on a preset threshold. If the deviation value is greater than the preset threshold, a dynamic parameter configuration table is generated; is also used to calculate the power adjustment amount by using the fuzzy logic algorithm according to the dynamic parameter configuration table to generate an adaptive power adjustment scheme; A regulation execution module, which is used to adjust the device power according to the adaptive power adjustment scheme to obtain an updated temperature monitoring data set; is also used to analyze the correlation based on the temperature monitoring data set and the multi-dimensional temperature information set to generate stability optimization parameters; is also used to adjust the control rules of the fuzzy logic algorithm according to the stability optimization parameters to obtain an adaptive temperature regulation scheme.

2. The system according to claim 1, wherein The analysis of the spatio-temporal features of the multi-dimensional temperature information set by using a convolutional neural network to obtain the spatio-temporal temperature change trend includes: Performing spatio-temporal fusion processing on the multi-dimensional temperature information set to obtain a spatio-temporal feature matrix; the multi-dimensional temperature information set includes temperature data in the time dimension and the space dimension; Inputting the spatio-temporal feature matrix into a convolutional neural network and performing cross-channel feature fusion in combination with a time sliding window to obtain a first feature map; Performing a multi-scale pooling operation on the first feature map to generate a pooled feature vector, and calculating the spatio-temporal weight coefficient by using a formula; Performing weighted aggregation on the pooled feature vector according to the spatio-temporal weight coefficient to obtain a spatio-temporal temperature change trend vector; Calculating the difference between the spatio-temporal temperature change trend vector and a preset reference trend to obtain a trend offset; Updating the spatio-temporal weight coefficient according to the trend offset and performing secondary weighting with the pooled feature vector to obtain an optimized spatio-temporal temperature change trend.

3. The system according to claim 2, wherein The spatio-temporal weight coefficient is calculated by the following formula: α i = Softmax(ReLU(S ωi + W ti )) Among them, W t represents the weight coefficient in the time dimension, n represents the feature dimension, m represents the time step, and v i represents the i-th feature vector, h i represents the hidden state vector, S ω represents the spatial weight coefficient, H and W respectively represent the height and width of the feature map, M ij represents the attention mask value at position (i, j), F ij represents the feature value at position (i, j), α i represents the spatio-temporal weight coefficient, Softmax( ) represents the normalization function, and ReLU( ) represents the activation function.

4. The system according to claim 1, characterized in that, The generation of an optimized temperature distribution map by using the Kalman filtering algorithm according to the spatio-temporal temperature change trend includes: Inputting the parameters in the spatio-temporal temperature change trend into a trained spatio-temporal trend model to obtain a temperature prediction value sequence; Calculating the deviation amount according to the temperature prediction value sequence and the multi-dimensional temperature information set to obtain dynamic weight parameters; Using the dynamic weight parameters to control a fusion algorithm to process the temperature prediction value sequence and the multi-dimensional temperature information set to obtain a residual sequence; Using the Kalman filtering algorithm to correct the residual sequence to obtain a corrected covariance matrix; Updating the parameters of the spatio-temporal trend model according to the covariance matrix to obtain an optimized temperature distribution map.

5. The system according to claim 1, characterized in that, The calculation of the power adjustment amount by using the fuzzy logic algorithm according to the dynamic parameter configuration table to generate an adaptive power adjustment scheme includes: Generating a membership function set according to the real-time parameter type and parameter threshold in the dynamic parameter configuration table; the membership function set includes temperature membership and load rate membership; Fuzzily infer the set of membership functions using a fuzzy logic algorithm to generate a power adjustment amount; Adjust the corresponding device according to the power adjustment amount and update the power adjustment rule using feedback data to obtain a corrected set of membership functions; the device feedback data includes the actual power fluctuation value and the adjustment error rate; Generate an adaptive power adjustment scheme based on the corrected set of membership functions; the scheme includes a dynamic power curve and adjustment period parameters.

6. The system according to claim 1, wherein Analyze the correlation based on the temperature monitoring data set and the multi-dimensional temperature information set to generate stability optimization parameters, including: Construct the time series features in the temperature monitoring data set and the spatial distribution pattern extracted from the multi-dimensional temperature information set using a clustering algorithm to obtain a multi-dimensional correlation matrix; the time series features include the temperature fluctuation coefficient and the acquisition timestamp; the spatial distribution pattern includes the device node position and the temperature gradient difference; Calculate the device response delay data according to the multi-dimensional correlation matrix by matching the historical trend baseline; the historical trend baseline includes the device operating state and the temperature stability threshold; Input the device response delay data into a parameter optimization model to obtain stability optimization parameters including a temperature compensation coefficient and an acquisition frequency adjustment value.

7. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the system according to any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the system according to any one of claims 1 to 6.

Images