Wearable heat preservation and temperature measurement monitoring system for anesthesia patient during operation

Abstract

The invention relates to the technical field of medical apparatus and instruments, and discloses a wearable heat preservation and temperature measurement monitoring system in an anesthesia patient operation, which comprises a body temperature sensing module, a temperature field modeling module, a heat preservation regulation and control characteristic optimization module, a heat balance decision scheduling module, a temperature control instruction generation module and a biological safety module. The body temperature sensing module collects body surface temperature distribution data through a multi-point flexible sensor array; the temperature field modeling module performs spatial interpolation reconstruction and heat conduction feature extraction on the data; the heat preservation regulation and control characteristic optimization module performs dynamic correlation characteristic compensation based on a thermodynamic model; the heat balance decision scheduling module reconstructs field intensity features in combination with physiological parameter constraints; the temperature control instruction generation module generates a heating diaphragm control signal according to the decision result; the biological safety module guarantees safety through a medical-grade insulating layer and a temperature fusing device. The system is suitable for heat preservation and temperature measurement monitoring in an anesthesia patient operation, and the operation safety is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular to a wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery. Background Art

[0002] Traditional intraoperative thermal insulation methods mainly include heating blankets and air heaters, but these methods often have problems such as uneven thermal insulation effects and the inability to accurately monitor the patient's body surface temperature distribution in real time. Existing body temperature monitoring equipment can usually only provide body temperature data at a single point or a limited number of points. It is difficult to fully reflect the temperature gradient distribution and heat conduction characteristics of the patient's body surface, and cannot provide a more accurate basis for thermal insulation control. In addition, traditional thermal insulation systems lack dynamic optimization and intelligent decision-making capabilities. They cannot be adaptively adjusted according to the patient's physiological parameters and real-time temperature field changes, which can easily lead to insufficient or excessive thermal insulation, affecting patient safety and surgical results.

[0003] With the development of flexible electronics and intelligent sensing technologies, the application of wearable devices in the medical field is gradually increasing. Wearable flexible sensor arrays can better conform to the patient's body surface, enabling real-time collection of temperature data from multiple points. However, how to effectively process and analyze this massive amount of temperature data, establish an accurate temperature field model, and implement intelligent temperature control based on this data remains a technical challenge that needs to be solved. At the same time, ensuring the biosafety of the system during the temperature preservation process and avoiding safety accidents such as patient burns due to heating device failures are also important considerations in the design of intraoperative temperature preservation systems.

[0004] Currently, existing intraoperative thermal insulation and temperature measurement monitoring systems have deficiencies in comprehensive temperature data collection, accurate temperature field modeling, intelligent thermal control, and biosafety. These deficiencies cannot meet the clinical need for precise thermal insulation and safety monitoring of anesthetized patients during surgery. Therefore, there is an urgent need to develop a wearable thermal insulation and temperature measurement monitoring system that can achieve multi-point real-time temperature collection, precise temperature field modeling and optimization, intelligent thermal balance decision-making, and reliable biosafety protection. Summary of the Invention

[0005] The purpose of the present invention is to provide a wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery to solve the problems raised in the above background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery, the system comprising:

[0007] Body temperature sensing module, used to collect patient surface temperature distribution data in real time through a multi-point flexible sensor array;

[0008] a temperature field modeling module, configured to perform spatial interpolation reconstruction and heat conduction feature extraction on the body surface temperature distribution data to obtain a set of temperature gradient distribution tensors;

[0009] a heat preservation and control feature optimization module, configured to perform dynamic correlation feature compensation based on a thermodynamic model on the set of temperature gradient distribution tensors to obtain a set of temperature field optimization feature tensors;

[0010] a thermal balance decision scheduling module, configured to reconstruct the field intensity characteristics of the set of temperature field optimization feature tensors based on physiological parameter constraints to obtain a global thermal balance decision feature map;

[0011] The temperature control instruction generation module is used to generate a heating diaphragm control signal according to the global thermal balance decision characteristic diagram.

[0012] Preferably, the temperature field modeling module includes: a temperature data partitioning unit, used to spatially divide the body surface temperature distribution data according to preset anatomical area rules to obtain a set of temperature partition data units; a heat conduction feature encoding unit, used to input each unit data in the set of temperature partition data units into a heat conduction encoder based on a wearable flexible sensor array to obtain a set of temperature gradient distribution tensors.

[0013] Preferably, the thermal insulation control feature optimization module includes: a temperature tensor reorganization unit, which performs tensor expansion on the temperature gradient distribution tensor along the spatial dimension to obtain a set of temperature gradient distribution vectors; a thermal correlation calculation unit, which is used to calculate the spatial heat conduction correlation between adjacent distribution vectors in the set of temperature gradient distribution vectors to generate a thermal conduction correlation matrix; a thermal compensation weight correction unit, which is used to perform weight balancing correction on the thermal conduction correlation matrix according to the topological relationship between adjacent distribution vectors in the set of temperature gradient distribution vectors to obtain a thermal constraint correlation matrix; a temperature field optimization unit, which is used to perform three-dimensional convolution fusion on the thermal constraint correlation matrix and the set of temperature gradient distribution vectors to obtain the temperature field optimization feature tensor.

[0014] Preferably, the thermal correlation calculation unit includes: a thermal space mapping subunit, used to map each distribution vector in the set of temperature gradient distribution vectors to the thermodynamic parameter space to obtain a set of mapped temperature gradient distribution vectors; a heat conduction correlation analysis subunit, used to calculate the Spearman correlation coefficient between adjacent mapped distribution vectors in the set of mapped temperature gradient distribution vectors to generate the thermal conduction correlation matrix composed of multiple thermal conduction correlation values.

[0015] Preferably, the temperature field optimization unit is specifically implemented as follows: performing spatial convolution processing on the thermal constraint correlation matrix to obtain a thermal constraint correlation feature matrix; inputting the set of temperature gradient distribution vectors and the thermal constraint correlation feature matrix into a thermodynamic encoding network to obtain a set of temperature field context feature vectors; and performing tensor superposition on the set of temperature field context feature vectors to obtain the temperature field optimization feature tensor.

[0016] Preferably, the thermal balance decision scheduling module includes: an optimization feature aggregation unit, which is used to perform regional dimension maximum pooling processing on each tensor in the set of temperature field optimization feature tensors to obtain a set of temperature field optimization feature vectors; a physiological feature deviation calculation unit, which is used to calculate the physiological deviation value of each feature vector in the set of temperature field optimization feature vectors to generate a temperature physiological deviation set; a control reference center determination unit, which is used to select the temperature field optimization feature vector corresponding to the maximum deviation value in the temperature physiological deviation set as the initial control center vector; a thermal compensation weight allocation unit, which is used to calculate the dynamic compensation weight of each feature vector in the set of temperature field optimization feature vectors and the initial control center vector according to the spatial distance between each feature vector in the set of temperature field optimization feature vectors and the initial control center vector and the physiological deviation value of each feature vector to generate a dynamic compensation weight set; a global field strength reconstruction unit, which is used to use the dynamic compensation weight set to perform weighted fusion on the set of temperature field optimization feature vectors to generate the global thermal balance decision feature map.

[0017] Preferably, the physiological characteristic deviation calculation unit is specifically implemented as follows: calculating the mean vector and variance vector of the temperature field optimized characteristic vector; performing element-by-element difference calculation between the temperature field optimized characteristic vector and the mean vector, and squaring the difference result to obtain a temperature characteristic difference vector; calculating the overall average value of the temperature characteristic difference vector; multiplying the average value by the square value of the variance vector, and inputting the result into a normalization function to obtain the physiological deviation value.

[0018] Preferably, the thermal compensation weight allocation unit is specifically implemented as follows: multiplying the physiological deviation value of the temperature field optimization feature vector and the physiological deviation value of the initial control center vector by a first adjustment coefficient to obtain a first dynamic compensation factor; multiplying the Euclidean distance between the temperature field optimization feature vector and the initial control center vector by a second adjustment coefficient to obtain a second dynamic compensation factor; and performing weighted summation on the first dynamic compensation factor and the second dynamic compensation factor to obtain the dynamic compensation weight.

[0019] Preferably, the temperature control instruction generation module is specifically implemented as follows: inputting the global thermal balance decision characteristic diagram into a temperature control signal generator based on a PID controller to obtain the heating diaphragm control signal, and the heating diaphragm control signal is used to indicate the power adjustment strategy of the heating diaphragm.

[0020] Preferably, the system also includes a biosafety module, which is composed of a medical-grade insulation layer and a temperature fuse device, wherein: the medical-grade insulation layer covers the surface of the heating diaphragm, and is used to convert the heating diaphragm control signal into a physical isolation barrier; the temperature fuse device is integrated into the flexible sensor array circuit, and is used to monitor the surface temperature of the heating diaphragm in real time, and generate a fuse protection instruction after double comparison with a preset safety threshold; the fuse protection instruction is transmitted to the system main control unit through a wireless sensor network to execute an emergency power-off.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] The wearable thermal insulation and temperature measurement monitoring system for anesthetized patients provided by the present invention collects patient surface temperature distribution data in real time through a multi-point flexible sensor array, and can comprehensively and accurately obtain the patient's surface temperature information, overcoming the limitations of traditional single-point or limited-point monitoring, and providing a rich data basis for subsequent temperature field modeling and thermal insulation control.

[0023] The temperature field modeling module performs spatial interpolation reconstruction and heat conduction feature extraction on the body surface temperature distribution data, generating a collection of temperature gradient distribution tensors. This process spatially divides the temperature data according to pre-set anatomical regions and processes the data from each partition using a thermal conductivity encoder based on a wearable flexible sensor array. This effectively reveals the spatial distribution characteristics of the patient's body surface temperature and the laws of heat conduction, providing a more targeted basis for thermal regulation.

[0024] The thermal insulation control feature optimization module performs a series of operations on the temperature gradient distribution tensor, including tensor expansion, thermal correlation calculation, weight balancing correction, and three-dimensional convolution fusion. This module compensates for the dynamic correlation characteristics of the temperature field, resulting in a set of optimized temperature field feature tensors. This module fully considers the spatial heat conduction correlation and topological relationships between adjacent temperature distribution vectors, making thermal insulation control more consistent with thermodynamic principles and improving the accuracy and effectiveness of temperature field optimization.

[0025] The thermal balance decision scheduling module reconstructs the field intensity characteristics of the temperature field optimization feature tensor based on physiological parameter constraints, generating a global thermal balance decision feature map. Through maximum pooling, physiological deviation calculation, control reference center determination, dynamic compensation weight allocation, and weighted fusion, this module comprehensively considers the patient's physiological parameters and temperature field characteristics, achieving intelligent and personalized thermal balance decision-making and ensuring the scientific and rationality of thermal insulation control.

[0026] The temperature control command generation module generates a heating diaphragm control signal based on the global thermal balance decision characteristic map, and precisely adjusts the heating diaphragm power through a temperature control signal generator based on a PID controller. This closed-loop control method can adjust the heating power in real-time based on temperature field changes and thermal balance decision results, achieving dynamic adaptive regulation of the insulation process and improving the stability and accuracy of the insulation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a working principle diagram of the wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery according to the present invention;

[0028] Figure 2 This is the working principle diagram of the thermal insulation control feature optimization module;

[0029] Figure 3 This is the working principle diagram of the heat balance decision scheduling module;

[0030] Figure 4 This is a working principle diagram of the physiological characteristic deviation calculation unit. DETAILED DESCRIPTION

[0031] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0032] See also Figure 1-Figure 4 The present invention relates to a wearable thermal insulation and temperature measurement monitoring system for anesthetized patients during surgery. The system includes a body temperature sensing module, a temperature field modeling module, a thermal insulation control feature optimization module, a thermal balance decision scheduling module, and a temperature control instruction generation module. The specific implementation method is as follows:

[0033] The body temperature sensing module collects the patient's surface temperature distribution data in real time through a multi-point flexible sensor array. The flexible sensor array can fit various parts of the patient's body to ensure the accuracy and comprehensiveness of the collected data.

[0034] The temperature field modeling module performs spatial interpolation reconstruction and heat conduction feature extraction on the body surface temperature distribution data to obtain a set of temperature gradient distribution tensors.

[0035] The thermal insulation control feature optimization module performs dynamic correlation feature compensation on the set of temperature gradient distribution tensors based on the thermodynamic model to obtain a set of temperature field optimization feature tensors.

[0036] The thermal balance decision scheduling module reconstructs the field strength characteristics of the set of temperature field optimization feature tensors based on physiological parameter constraints to obtain a global thermal balance decision feature map.

[0037] The temperature control instruction generation module generates a heating diaphragm control signal based on the global thermal balance decision characteristic diagram, thereby achieving precise control of the patient's temperature.

[0038] The present invention will be further described below in conjunction with Examples 1 to 5:

[0039] Example 1:

[0040] The system's temperature field modeling module includes a temperature data partitioning unit and a heat conduction feature encoding unit. The temperature data partitioning unit spatially divides the body surface temperature distribution data according to preset anatomical region rules to obtain a collection of temperature partitioning data units. The preset anatomical region rules here are determined based on the human anatomical structure and the key areas of body temperature monitoring commonly used during surgery. For example, during abdominal surgery, the body surface is divided into regions such as the head and neck, chest and back, abdomen, upper limbs, and lower limbs. Each region can be further subdivided. For example, the abdomen can be divided into sub-regions such as the left upper abdomen, right upper abdomen, left lower abdomen, and right lower abdomen to more accurately capture temperature changes in different parts. During the division process, the temperature data partitioning unit will assign the temperature data corresponding to each sensor to the corresponding anatomical region based on the location information of the temperature data collected by the multi-point flexible sensor array.

[0041] The thermal conductivity feature encoding unit inputs the data from each temperature partition data unit into a thermal conductivity encoder based on a wearable flexible sensor array, generating a set of temperature gradient distribution tensors. The thermal conductivity encoder is designed based on research into the thermal conductivity characteristics of wearable flexible sensor arrays in practical applications. The encoder first preprocesses each temperature partition data unit, including denoising and smoothing, to eliminate random errors and interference that may occur during the acquisition process. The encoder then analyzes the temporal series and spatial distribution characteristics of the temperature data within each partition to extract key heat conduction parameters, such as the heat conduction rate and temperature gradient.

[0042] During processing, the heat conduction feature encoding unit considers the impact of the flexible sensor array's fit with the human body surface on heat conduction. Due to its excellent flexibility, the flexible sensor array adheres closely to the human body surface, reducing the presence of air layers and enabling more direct heat conduction. Based on this characteristic, the encoder adjusts the heat conduction model to more accurately reflect the actual heat conduction process. For example, when calculating the temperature gradient, the distance and positional relationship between sensors are taken into account. Appropriate interpolation algorithms, such as cubic spline interpolation, are used to spatially reconstruct discrete temperature data, resulting in a continuous temperature distribution field.

[0043] Furthermore, the heat conduction feature encoding unit also considers the impact of the physiological characteristics of different anatomical regions on heat conduction. For example, the extremities have relatively few blood vessels and slower heat conduction, while the chest and abdomen are rich in blood vessels and faster heat conduction. Based on these physiological differences, the encoder optimizes the heat conduction feature extraction algorithm for different regions to ensure that the extracted temperature gradient distribution tensor accurately reflects the actual temperature changes in each region.

[0044] The temperature field modeling module achieves accurate modeling of the patient's surface temperature distribution through the collaborative work of the temperature data partitioning unit and the heat conduction feature encoding unit. This modeling approach not only considers the spatial distribution of temperature data but also deeply analyzes the characteristics of heat conduction, providing high-quality input data for the subsequent thermal insulation control feature optimization module. By converting the surface temperature distribution data into a set of temperature gradient distribution tensors, the system can more intuitively understand the changing trends and distribution characteristics of the patient's surface temperature, laying the foundation for formulating a reasonable thermal insulation control strategy. In actual application, the temperature field modeling module continuously receives real-time temperature data from the temperature sensing module and updates the set of temperature gradient distribution tensors in real time to ensure that the system can promptly respond to changes in the patient's body temperature. For example, if the patient's body temperature begins to drop in a certain area due to surgery, the temperature field modeling module can quickly detect this change and reflect it through changes in the temperature gradient distribution tensor, providing timely information for subsequent thermal insulation control.

[0045] Example 2:

[0046] The system's thermal insulation control feature optimization module includes a temperature tensor reorganization unit, a thermal correlation calculation unit, a thermal compensation weight correction unit, and a temperature field optimization unit. The temperature tensor reorganization unit performs tensor expansion on the temperature gradient distribution tensor along the spatial dimension to obtain a set of temperature gradient distribution vectors. Specifically, the unit will decompose the three-dimensional temperature gradient distribution tensor according to the spatial coordinate axis (such as the x, y, and z axes), and convert the temperature gradient data at each spatial position into the corresponding element in the one-dimensional vector, thereby realizing the conversion from tensor structure to vector set for subsequent calculation and analysis of thermal correlation.

[0047] The thermal space mapping subunit in the thermal correlation calculation unit maps each distribution vector in the set of temperature gradient distribution vectors to the thermodynamic parameter space to obtain a set of mapped temperature gradient distribution vectors. The thermodynamic parameter space here covers a variety of physical quantities related to the heat conduction process, such as thermal conductivity, specific heat capacity, temperature gradient change rate, etc. During the mapping process, the elements in each original temperature gradient distribution vector will establish a mapping relationship with the corresponding thermodynamic parameters according to the preset thermodynamic conversion rules. For example, the temperature change rate parameter at a certain position in the vector will be converted into a heat conduction efficiency index in the thermodynamic parameter space through a specific conversion function, so that the original data can be analyzed in a space that is more in line with the physical laws of heat conduction.

[0048] The heat conduction correlation analysis subunit calculates the Spearman correlation coefficient between adjacent mapped distribution vectors in the set of mapped temperature gradient distribution vectors to generate a heat conduction correlation matrix. The selection of adjacent mapped distribution vectors is based on the spatial layout of the sensor array. For example, in a flexible sensor array, the mapping vectors corresponding to physically adjacent sensors are considered adjacent vectors. The calculation of the Spearman correlation coefficient is used to measure the degree of monotonic correlation between temperature gradient changes at different locations. The value of this coefficient ranges from -1 to 1, and the larger the absolute value, the stronger the correlation between the change trends of the two vectors. By traversing all adjacent vector pairs and calculating their correlation coefficients, an N×N heat conduction correlation matrix is ​​finally formed (N is the number of mapped vectors). Each element in the matrix represents the heat conduction correlation strength between sensors at the corresponding location.

[0049] The thermal compensation weight correction unit performs weight-balancing corrections on the thermal conduction correlation matrix based on the topological relationships between adjacent distribution vectors in the set of temperature gradient distribution vectors, resulting in a thermal constraint correlation matrix. The topological relationships between adjacent distribution vectors primarily refer to the spatial arrangement of sensors on the human body surface, such as linear arrangement, planar distribution, or three-dimensional grid distribution. During the correction process, the unit first analyzes the topological structure of the sensor array to determine the physical distances and spatial adjacencies between sensors at different locations. For sensors that are physically close, even if their calculated Spearman correlation coefficients are low, their correlation weights are appropriately increased based on the topological relationships. For sensors that are physically far apart but have high correlation coefficients, the weights are appropriately adjusted based on the topological structure to avoid deviations in the correlation matrix from the actual physical laws of heat conduction due to a single data metric. For example, if the sensor array on the torso is distributed planarly, the topological relationship weights for adjacent sensors within the same anatomical region will be higher than for sensors across regions, thus ensuring that the correlation matrix is ​​more consistent with the actual thermal conduction laws of the human body surface.

[0050] The temperature field optimization unit performs spatial convolution processing on the thermal constraint correlation matrix to obtain a thermal constraint correlation feature matrix. Spatial convolution processing uses a preset convolution kernel (such as a 3×3 or 5×5 matrix) to perform a sliding window operation on the thermal constraint correlation matrix. The convolution operation extracts local correlation features in the matrix and highlights the changing trend and spatial distribution pattern of the thermal conduction correlation between adjacent regions. For example, when the convolution kernel slides on the matrix, the elements in each window are weighted and summed with the convolution kernel weight to generate new feature matrix elements, thereby achieving feature extraction and dimensionality reduction of the original matrix.

[0051] Subsequently, the temperature field optimization unit inputs the set of temperature gradient distribution vectors and the thermal constraint association feature matrix into the thermodynamic encoding network to obtain a set of temperature field context feature vectors. The thermodynamic encoding network is a multi-layer neural network structure. Its input layer receives the fusion data of the temperature gradient distribution vector and the thermal constraint association feature matrix. The hidden layer performs a nonlinear transformation on the data through an activation function (such as the ReLU function) to extract the deep semantic relationship between the temperature field and the thermal conduction correlation. The design of the network takes into account the coupling characteristics of the temperature distribution and the physical process of heat conduction. For example, special neuron nodes are set in the hidden layer to process the cross-features of the temperature gradient and the thermal conduction correlation, so that the output context feature vector can simultaneously contain temperature distribution information and thermal conduction association constraint information.

[0052] Finally, the temperature field optimization unit performs tensor superposition on the set of temperature field context feature vectors to obtain the temperature field optimization feature tensor. During the tensor superposition process, each context feature vector will be reconverted into a tensor form and stacked according to the spatial dimension to eventually form a new high-order tensor. This tensor not only retains the spatial information of the original temperature gradient distribution, but also incorporates the constraint characteristics of the heat conduction correlation, making the representation of the temperature field more comprehensive and accurate, and providing more physically meaningful input data for the subsequent thermal balance decision scheduling module. Through this series of processing, the thermal insulation control feature optimization module can dynamically compensate for the heat conduction correlation characteristics in the temperature field, improve the system's prediction and control accuracy of changes in the human body surface temperature field, and ensure that accurate thermal insulation intervention can be performed according to the actual heat conduction conditions during surgery.

[0053] Example 3:

[0054] The thermal balance decision scheduling module of the system consists of an optimization feature aggregation unit, a physiological feature deviation calculation unit, a control reference center determination unit, a thermal compensation weight allocation unit, and a global field strength reconstruction unit. The optimization feature aggregation unit performs regional dimension maximum pooling processing on each tensor in the set of temperature field optimization feature tensors to obtain a set of temperature field optimization feature vectors. Specifically, in the regional dimension, the unit will divide each temperature field optimization feature tensor into blocks according to the preset anatomical regions (such as head and neck, chest and abdomen, etc.), and take the maximum value of all elements in each block to highlight the key temperature features in each region, reduce the data dimension while retaining the main information, and thus convert the high-order tensor into a low-dimensional feature vector set.

[0055] The physiological characteristic deviation calculation unit calculates the mean vector and variance vector of the temperature field optimized feature vector. First, for the set of temperature field optimized feature vectors, the mean value on each dimension is calculated to form a mean vector μ, where μ i Represents the mean of the i-th feature dimension; at the same time, the variance of each dimension is calculated to form a variance vector σ 2 ,in represents the variance of the i-th feature dimension.

[0056] Next, the temperature field optimized feature vector and the mean vector are interpolated element by element, and the interpolated result is squared to obtain the temperature feature difference vector. Let the temperature field optimized feature vector be v, then each element of the temperature feature difference vector is (v i -μ i ) 2 , where v i is the i-th element of vector v, μ i is the i-th element of the mean vector μ.

[0057] Then calculate the overall average of the temperature feature difference vector, that is, Where n is the number of dimensions of the feature vector.

[0058] This mean value is multiplied by the square of the variance vector, and the result is input into a normalization function to obtain a physiological deviation value. This normalization function typically uses a sigmoid function or min-max normalization to ensure that the physiological deviation value remains within a reasonable range. This series of operations quantifies the degree of deviation of each temperature field's optimized eigenvector from the overall mean, reflecting the degree of abnormality in its physiological characteristics.

[0059] The control reference center determination unit selects the temperature field optimized eigenvector corresponding to the maximum deviation value in the temperature physiological deviation set as the initial control center vector. The temperature physiological deviation set is composed of the physiological deviation values ​​of all temperature field optimized eigenvectors. The unit compares these values ​​and finds the vector with the largest deviation. This vector is used as the initial control reference because the area represented by this vector is likely to have the most significant body temperature abnormality and requires priority control.

[0060] The thermal compensation weight allocation unit multiplies the physiological deviation value of the temperature field optimization feature vector and the physiological deviation value of the initial control center vector by a first adjustment coefficient to obtain a first dynamic compensation factor; multiplies the Euclidean distance between the temperature field optimization feature vector and the initial control center vector by a second adjustment coefficient to obtain a second dynamic compensation factor; and performs weighted summation of the first dynamic compensation factor and the second dynamic compensation factor to obtain a dynamic compensation weight.

[0061] Assume that the optimized characteristic vector of temperature field is v j , its physiological deviation value is d j , the initial control center vector is v0, its physiological deviation value is d0, the first adjustment coefficient is α, then the first dynamic compensation factor is

[0062] Temperature field optimization characteristic vector v j The Euclidean distance between the initial control center vector v0 is ) where v j,i and v 0,i are vectors v j and the i-th element of v0, the second adjustment coefficient is β, then the second dynamic compensation factor is β×dist(v j ,v0).

[0063] Dynamic compensation weight w j The calculation formula is: Where λ is a weighting coefficient used to balance the influence of the two dynamic compensation factors.

[0064] The global field strength reconstruction unit uses the dynamic compensation weight set to perform weighted fusion on the set of temperature field optimization feature vectors to generate a global thermal balance decision feature map. Specifically, for each temperature field optimization feature vector v j , multiplied by its corresponding dynamic compensation weight w j All weighted vectors are then summed to produce the final global thermal balance decision feature map. This feature map comprehensively considers the physiological deviation and spatial distance of each region, reflecting the thermal balance state of the entire patient's body surface temperature field and providing an accurate basis for generating temperature control instructions.

[0065] In actual applications, the thermal balance decision scheduling module will receive a set of temperature field optimization feature tensors in real time and process them according to the above steps. For example, when the body temperature of a certain part of the patient changes abnormally, the physiological deviation value of the temperature field optimization feature vector corresponding to that part will increase. The control reference center determination unit will promptly use it as the initial control center vector. The thermal compensation weight allocation unit will recalculate the dynamic compensation weights of each vector based on the new center vector. The global field strength reconstruction unit will generate a new global thermal balance decision feature map, so that the system can quickly control the abnormal body temperature area and maintain the patient's body temperature stable. Through the collaborative work of each unit, the thermal balance decision scheduling module realizes the reconstruction of field strength features based on physiological parameter constraints, providing scientific decision support for the intraoperative temperature monitoring and heat preservation of anesthetized patients.

[0066] Example 4:

[0067] The system's temperature control instruction generation module inputs the global thermal balance decision characteristic map into the temperature control signal generator based on the PID controller to obtain the heating diaphragm control signal, which is used to indicate the power adjustment strategy of the heating diaphragm.

[0068] In practical applications, the global thermal balance decision feature map is a comprehensive data matrix containing temperature field information and thermal equilibrium status for each region of the patient's body. For example, when anesthetized patients undergo abdominal surgery, the chest, abdomen, and limbs may experience varying degrees of temperature drops due to surgical exposure and the effects of anesthetics. The global thermal balance decision feature map uses different pixel values ​​or data indicators to reflect the degree of temperature deviation and heat conduction trends in each region.

[0069] The temperature control signal generator based on a PID controller is the core component of the entire temperature control command generation module. The PID controller consists of three control steps: proportional, integral, and derivative. Its operating principle is to calculate the control variable based on the system's error signal to achieve precise regulation of the controlled object. In this system, the controlled object is a heating diaphragm, and the control objective is to maintain the patient's surface temperature within a set safety range (e.g., 36°C-37°C).

[0070] When the global thermal balance decision characteristic map is input into the temperature control signal generator, it is first compared with the preset temperature reference value to generate an error signal. For example, if the characteristic map shows that the temperature of the patient's abdominal area is 35.2°C, and the preset reference value is 36.5°C, the temperature error in this area is -1.3°C. The proportional link will output a control quantity proportional to the error based on the current error. The larger the error, the greater the output of the proportional link, so that the heating diaphragm quickly generates a certain amount of heating power. In the above example, the proportional link will calculate a corresponding proportional control quantity based on the -1.3°C error, prompting the heating diaphragm to heat the abdominal area.

[0071] The integral link will accumulate and integrate the error signal, and its function is to eliminate the static error of the system and ensure that the temperature can eventually stabilize at the reference value. Over time, even if the temperature error gradually decreases, the integral link will continue to output a certain control amount due to the accumulation of previous errors to prevent the temperature from fluctuating or failing to reach the reference value. For example, when the temperature in the abdominal area gradually rises to 36.0°C under proportional control, there is still an error of 0.5°C. The integral link will continue to output a control amount based on the accumulated error value over the previous period of time, so that the heating diaphragm maintains a certain power until the temperature reaches 36.5°C.

[0072] The differential link is used to predict the error's changing trend and output a control variable based on the error's rate of change, thereby reducing system overshoot and settling time. As the temperature approaches the reference value, the error's rate of change gradually decreases. The differential link adjusts the control variable in advance based on this trend to prevent the heating diaphragm from overheating and causing the temperature to exceed the reference value. For example, as the abdominal temperature rises from 36.0°C to 36.5°C, the error's rate of change gradually decreases. The differential link outputs a control variable related to the rate of change, fine-tuning the heating power to ensure the temperature smoothly reaches the reference value without significantly exceeding it.

[0073] The temperature control signal generator takes the weighted sum of the outputs from the proportional, integral, and differential stages to generate the final heater diaphragm control signal. This control signal is transmitted to the heater diaphragm in the form of an electrical signal, instructing it to adjust its power output. The heater diaphragm is typically made of a material with good thermal conductivity and safety, such as carbon fiber heating film. Its power can be adjusted between 0% and 100% depending on the control signal.

[0074] For example, when the control signal instructs the heating diaphragm to operate at 60% power, the current flows through the heating diaphragm, generating heat that is transferred to the patient's body surface through heat conduction. The system collects temperature data in real time through the body temperature sensor module and feeds this data back to the temperature control signal generator, forming a closed-loop control circuit. If the temperature rises too quickly, the temperature control signal generator adjusts the control variable based on the new error signal, reducing the heating power. If the temperature rises slowly, the heating power is increased, thereby achieving dynamic and precise control of the patient's body temperature.

[0075] During surgery, the patient's body temperature may change dynamically due to factors such as the duration of the surgery, the depth of anesthesia, and the exposed area of ​​the body surface. For example, a prolonged abdominal surgery may cause the temperature of the patient's lower extremities to gradually drop. In this case, the global thermal balance decision feature map will reflect an increase in the temperature error in the lower extremities. The temperature control instruction generation module will recalculate the control variable through the PID controller based on the new feature map data, adjust the power output of the heating diaphragm in the lower extremities, and ensure that the temperature of all regions of the body is maintained within a reasonable range.

[0076] The temperature control command generation module also considers the heater's power regulation accuracy and response speed. The heater's power regulation accuracy typically reaches 1%, enabling subtle power adjustments based on control signals to accommodate even small changes in the patient's body temperature. Response speed, however, depends on the heater's material properties and circuit design; it typically responds to control signals within seconds, allowing for rapid adjustment of heating power.

[0077] Through the operation of the temperature control instruction generation module, the system realizes the conversion from temperature field data to heating control signals, providing an effective execution method for maintaining the body temperature of anesthetized patients during surgery. Combining the PID control algorithm and the global thermal balance decision characteristic map, this module can accurately calculate the required heating power, allowing the heating diaphragm to dynamically adjust according to the actual patient's body surface temperature, ensuring that the patient's body temperature remains stable during surgery and reducing the risk of complications caused by hypothermia.

[0078] Example 5:

[0079] The system also includes a biosafety module, comprised of a medical-grade insulation layer and a thermal cutout. The medical-grade insulation layer, covering the surface of the heating diaphragm, converts the control signal from the heating diaphragm into a physical isolation barrier. The thermal cutout, integrated into the flexible sensor array circuit, monitors the surface temperature of the heating diaphragm in real time and compares it against a preset safety threshold. This triggers a fuse protection command, which is transmitted via a wireless sensor network to the system's main control unit for emergency power outage.

[0080] In actual applications, the medical-grade insulation layer is usually made of polymer materials that meet medical standards, such as polytetrafluoroethylene or silicone rubber, and its thickness is controlled between 0.1-0.3 mm, which can not only ensure the insulation performance, but also not affect the heat conduction efficiency of the heating membrane. Taking abdominal surgery as an example, when the heating membrane is attached to the patient's abdominal skin, the medical-grade insulation layer will form a transparent isolation barrier. This barrier can effectively block the direct contact between the internal circuit of the heating membrane and the human body, preventing the risk of electric shock caused by current leakage. At the same time, the surface of the insulation layer has been specially treated to have anti-static and anti-body fluid penetration properties. Even if blood or flushing fluid drips during the operation, it will not penetrate into the interior of the heating membrane, avoiding circuit short circuits. For example, when a surgical nurse accidentally spills saline on the surface of the heating membrane, the hydrophobic coating of the insulation layer will cause the liquid to form water droplets and slide down, and will not penetrate into the internal circuit, ensuring that the system can still operate safely in a humid environment.

[0081] The core of the temperature fuse device is a temperature-sensitive element integrated into the flexible sensor array circuit, usually a resettable temperature fuse or thermistor. Taking thermistors as an example, their resistance changes linearly with temperature. When the surface temperature of the heating diaphragm rises, the resistance of the thermistor decreases accordingly, and the current changes in the circuit are collected in real time and transmitted to the signal processing unit. The device has preset dual safety thresholds, namely the first-level warning threshold and the second-level fuse threshold. The first-level warning threshold is usually set at 40°C, and the second-level fuse threshold is set at 42°C. The setting of these two thresholds is based on medical research on the temperature tolerance of human skin, ensuring that the protection mechanism can be triggered in stages when the temperature is abnormal.

[0082] When the system is operating, the thermal cutoff device continuously monitors the surface temperature of the heating diaphragm at a frequency of 100 milliseconds. For example, during a certain operation, the heating diaphragm's resistance increased due to poor local circuit contact, resulting in an abnormally high heating power, and the surface temperature began to rise rapidly from the normal operating 38°C. When the temperature reached 40°C, the thermal cutoff device first triggered the first-level warning mechanism, sending an early warning signal to the monitor next to the operating table via the wireless sensor network. The monitor screen displayed the prompt "Abnormal local temperature of the heating diaphragm" and a low-frequency buzzer alarm to alert medical staff. At this time, the system's main control unit automatically reduced the output power of the heating diaphragm and attempted to control the temperature rise through software adjustment.

[0083] If the temperature continues to rise to 42°C, the temperature fuse device will activate the secondary fuse protection mechanism. At this time, the mechanical or electronic trigger structure inside the device will be activated to cut off the power supply circuit of the heating membrane. At the same time, a fuse protection instruction is sent to the system main control unit through wireless sensor network protocols such as ZigBee or Bluetooth. After receiving the instruction, the main control unit will immediately execute the emergency power-off procedure, and at the same time display a red alarm on the surgical monitoring interface, record the time and location of the fuse, and lock the control authority of the heating membrane to prevent accidental restart. For example, at the moment the temperature reaches 42°C, the bimetallic strip in the temperature fuse device is deformed due to heat, pushing open the microswitch in the circuit, cutting off the power supply of the heating membrane, and transmitting the fuse signal to the main control unit through the wireless module. The main control unit then shuts down the power supply of the entire heating system to prevent high temperature from causing burns to the patient's skin.

[0084] The wireless sensor network plays a key role in signal transmission within the biosafety module. It utilizes a low-power communication protocol to ensure stable data transmission within the surgical environment. Network nodes include the transmitting module in the thermal fuse device and the receiving module in the system's main control unit. Communication between the two modules occurs via encrypted data packets to prevent signal interference and data tampering. For example, a CRC checksum is added to the fuse protection instruction during transmission. Upon receipt, the main control unit verifies the instruction to ensure its integrity and accuracy, preventing accidental power-off or unintended power-off due to signal interference.

[0085] The biosafety module forms a linkage mechanism with other modules of the system. When the temperature fuse device triggers the fuse protection, the body temperature sensor module will encrypt and collect temperature data near the fuse position and transmit it to the main control unit in real time so that medical staff can assess the skin heating condition; the thermal balance decision scheduling module will suspend the thermal balance calculation of the area to prevent erroneous data from affecting the overall control strategy. For example, after the heating diaphragm triggers the fuse in the abdominal area, the body temperature sensor module will increase the frequency of collecting abdominal skin temperature from the conventional 1 time / second to 5 times / second. The main control unit draws a temperature change curve based on the high-frequency data to provide a basis for subsequent safety assessments.

[0086] The physical isolation of the medical-grade insulation layer and the dual protection of the thermal cutout complement each other. The insulation layer prevents the risk of leakage during daily operation, while the thermal cutout responds to emergency situations caused by abnormal heating. For example, if the insulation layer of a wire inside the heating diaphragm deteriorates due to long-term use, the medical-grade insulation layer prevents direct contact between the wire and the human body, preventing electric shock. If the heating diaphragm is locally short-circuited due to external pressure, causing a sharp rise in temperature, the thermal cutout will immediately cut off the power supply when the temperature exceeds the specified limit, preventing burns.

[0087] After the surgery, the biosafety module also provides safety status feedback. The system's main control unit records the biosafety module's operating status during the surgery, including whether a warning or fuse was triggered, the number of triggers, the trigger location, and other information, and generates a safety report for medical staff to review. For example, in the anesthesia recovery room, medical staff can use the main control unit's historical record interface to check whether the patient experienced any abnormal heating membrane temperature during surgery, allowing them to conduct targeted observations on the patient's skin condition.

[0088] All components of the biosafety module have passed medical device safety certification. The electrical strength of its insulation layer can withstand voltages above 1000V, and the response time of the temperature fuse device is controlled within 50 milliseconds, ensuring rapid action in emergency situations.

[0089] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0090] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery, characterized in that: include: Body temperature sensing module, used to collect patient surface temperature distribution data in real time through a multi-point flexible sensor array; a temperature field modeling module, configured to perform spatial interpolation reconstruction and heat conduction feature extraction on the body surface temperature distribution data to obtain a set of temperature gradient distribution tensors; a heat preservation and control feature optimization module, configured to perform dynamic correlation feature compensation based on a thermodynamic model on the set of temperature gradient distribution tensors to obtain a set of temperature field optimization feature tensors; a thermal balance decision scheduling module, configured to reconstruct the field intensity characteristics of the set of temperature field optimization feature tensors based on physiological parameter constraints to obtain a global thermal balance decision feature map; The temperature control instruction generation module is used to generate a heating diaphragm control signal according to the global thermal balance decision characteristic diagram.

2. A wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery according to claim 1, characterized in that: The temperature field modeling module includes: a temperature data partitioning unit for spatially partitioning the body surface temperature distribution data according to preset anatomical region rules to obtain a set of temperature partition data units; and a heat conduction feature encoding unit for inputting each unit data in the set of temperature partition data units into a heat conduction encoder based on a wearable flexible sensor array to obtain a set of temperature gradient distribution tensors.

3. A wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery according to claim 2, characterized in that: The thermal insulation control feature optimization module includes: a temperature tensor reorganization unit, which performs tensor expansion on the temperature gradient distribution tensor along the spatial dimension to obtain a set of temperature gradient distribution vectors; a thermal correlation calculation unit, which is used to calculate the spatial heat conduction correlation between adjacent distribution vectors in the set of temperature gradient distribution vectors to generate a thermal conduction correlation matrix; a thermal compensation weight correction unit, which is used to perform weight balancing correction on the thermal conduction correlation matrix according to the topological relationship between adjacent distribution vectors in the set of temperature gradient distribution vectors to obtain a thermal constraint correlation matrix; and a temperature field optimization unit, which is used to perform three-dimensional convolution fusion on the thermal constraint correlation matrix and the set of temperature gradient distribution vectors to obtain the temperature field optimization feature tensor.

4. A wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery according to claim 3, characterized in that: The thermal correlation calculation unit includes: a thermal space mapping subunit, which is used to map each distribution vector in the set of temperature gradient distribution vectors to a thermodynamic parameter space to obtain a set of mapped temperature gradient distribution vectors; and a heat conduction correlation analysis subunit, which is used to calculate the Spearman correlation coefficient between adjacent mapped distribution vectors in the set of mapped temperature gradient distribution vectors to generate the thermal conduction correlation matrix composed of multiple thermal conduction correlation values.

5. A wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery according to claim 4, characterized in that: The temperature field optimization unit is specifically implemented as follows: performing spatial convolution processing on the thermal constraint correlation matrix to obtain a thermal constraint correlation feature matrix; inputting the set of temperature gradient distribution vectors and the thermal constraint correlation feature matrix into a thermodynamic encoding network to obtain a set of temperature field context feature vectors; and performing tensor superposition on the set of temperature field context feature vectors to obtain the temperature field optimization feature tensor.

6. A wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery according to claim 5, characterized in that: The thermal balance decision scheduling module includes: an optimization feature aggregation unit, which is used to perform regional dimension maximum pooling processing on each tensor in the set of temperature field optimization feature tensors to obtain a set of temperature field optimization feature vectors; a physiological feature deviation calculation unit, which is used to calculate the physiological deviation value of each feature vector in the set of temperature field optimization feature vectors to generate a temperature physiological deviation set; a control reference center determination unit, which is used to select the temperature field optimization feature vector corresponding to the maximum deviation value in the temperature physiological deviation set as the initial control center vector; a thermal compensation weight allocation unit, which is used to calculate the dynamic compensation weight of each feature vector in the set of temperature field optimization feature vectors and the initial control center vector based on the spatial distance between each feature vector and the initial control center vector and the physiological deviation value of each feature vector to generate a dynamic compensation weight set; a global field strength reconstruction unit, which is used to use the dynamic compensation weight set to perform weighted fusion on the set of temperature field optimization feature vectors to generate the global thermal balance decision feature map.

7. A wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery according to claim 6, characterized in that: The physiological characteristic deviation calculation unit is specifically implemented as follows: calculating the mean vector and variance vector of the temperature field optimized feature vector; performing element-by-element difference calculation between the temperature field optimized feature vector and the mean vector, and performing a square operation on the difference result to obtain a temperature characteristic difference vector; Calculating the overall average value of the temperature characteristic difference vector; performing a product operation on the average value and the square value of the variance vector, and inputting the result into a normalization function to obtain the physiological deviation value.

8. A wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery according to claim 7, characterized in that: The thermal compensation weight allocation unit is specifically implemented as follows: multiplying the physiological deviation value of the temperature field optimization feature vector and the physiological deviation value of the initial control center vector by a first adjustment coefficient to obtain a first dynamic compensation factor; multiplying the Euclidean distance between the temperature field optimization feature vector and the initial control center vector by a second adjustment coefficient to obtain a second dynamic compensation factor; and performing weighted summation on the first dynamic compensation factor and the second dynamic compensation factor to obtain the dynamic compensation weight.

9. A wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery according to claim 8, characterized in that: The temperature control instruction generation module is specifically implemented as follows: inputting the global thermal balance decision characteristic diagram into a temperature control signal generator based on a PID controller to obtain the heating diaphragm control signal, and the heating diaphragm control signal is used to indicate the power adjustment strategy of the heating diaphragm.

10. A wearable heat preservation and temperature measurement monitoring system for anesthetized patients during surgery according to claim 9, characterized in that: It also includes a biosafety module, which is composed of a medical-grade insulation layer and a temperature fuse device, wherein: the medical-grade insulation layer covers the surface of the heating diaphragm, and is used to convert the heating diaphragm control signal into a physical isolation barrier; the temperature fuse device is integrated into the flexible sensor array circuit, and is used to monitor the surface temperature of the heating diaphragm in real time, and generate a fuse protection instruction after double comparison with the preset safety threshold; the fuse protection instruction is transmitted to the system main control unit through a wireless sensor network to execute an emergency power-off.

Images