Perioperative hypothermia management method and system

Abstract

The application discloses a perioperative hypothermia management method and system, wherein the management method comprises the following steps: acquiring the surface temperature time series data of a target patient in a perioperative period in real time; converting the surface temperature time series data into core temperature change characteristics based on a pre-constructed body temperature conversion model, wherein the body temperature conversion model is constructed based on the correlation between the surface temperature time series data and the core temperature change characteristics; dynamically evaluating the hypothermia risk of the target patient based on the core temperature change characteristics to obtain a dynamic risk evaluation result; and automatically controlling a heating device based on the dynamic risk evaluation result. The surface temperature time series data is converted into core temperature change characteristics through the body temperature conversion model. The risk caused by invasive measurement is avoided, the accuracy of non-invasive body temperature monitoring is improved, and the patient's body temperature is dynamically adjusted by automatically controlling the heating device to automatically heat or stop heating in real time according to the dynamic risk evaluation result, so that automatic body temperature regulation is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of body temperature management, in particular to a perioperative hypothermia management method and system. BACKGROUND

[0002] Intraoperative hypothermia (IH) is a common complication under general anesthesia, which has a significant impact on target patients and can lead to increased blood loss, impaired drug metabolism, and infection and other adverse outcomes. There are still the following problems in the perioperative temperature management: (1) Accurate temperature monitoring is helpful to find IH in time, but current core temperature monitoring is mostly invasive operation (for example, nasopharyngeal or esophageal temperature monitoring commonly used under general anesthesia can cause mucosal damage and even bleeding). Traditional non-core temperature monitoring is non-invasive operation (for example, axillary temperature monitoring), but the temperature reading obtained by it has a large error compared with the core temperature, so that medical staff cannot use it as a reference for intraoperative temperature management; (2) As the most effective and commonly used method to prevent and treat IH, the inflatable heat blanket is currently usually relied on clinical experience to judge whether the target patient needs to use it, which can lead to non-individualized treatment and increase the cost of patient hospitalization and waste of medical resources; (3) When the intraoperative temperature of the patient fluctuates, the traditional inflatable heat blanket needs to manually adjust the equipment parameters to maintain normal body temperature, which is not conducive to improving the safety and efficiency of surgery; (4) The existing technology does not fully consider the influence of intraoperative dynamic factors (such as body temperature variables, blood loss, irrigation fluid temperature and dosage, etc.) on the risk of intraoperative hypothermia, resulting in incomplete and inaccurate risk assessment. SUMMARY

[0003] Therefore, the present application provides a perioperative hypothermia management method and system to solve the technical problems mentioned in the background.

[0004] In a first aspect, the present application provides a perioperative hypothermia management method, comprising:

[0005] obtaining the body surface temperature time series data of the target patient in the perioperative period in real time;

[0006] converting the body surface temperature time series data into core temperature change characteristics based on a pre-constructed body temperature conversion model, wherein the body temperature conversion model is constructed based on the correlation between the body surface temperature time series data and the core temperature change characteristics;

[0007] dynamically evaluating the hypothermia risk of the target patient based on the core temperature change characteristics to obtain a dynamic risk evaluation result;

[0008] automatically controlling the heating device based on the dynamic risk evaluation result.

[0009] Optionally, the body temperature conversion model comprises a temperature interval-based body temperature correction model.

[0010] The converting the body surface temperature into the core temperature based on the pre-constructed body temperature conversion model comprises:

[0011] real-time analyzing the body surface temperature to determine a current temperature interval to which the body surface temperature belongs;

[0012] compensating the body surface temperature based on a temperature difference value corresponding to the current temperature interval to obtain the core temperature.

[0013] Optionally, before the real-time acquisition of the body surface temperature time series data of the target patient during the perioperative period, the method further comprises:

[0014] acquiring a surgical process parameter of the target patient;

[0015] performing a hypothermia basic risk assessment on the target patient based on the surgical process parameter to obtain a basic risk assessment result;

[0016] determining a basic heating strategy of the heating device during the perioperative period based on the basic risk assessment result.

[0017] Optionally, the performing the hypothermia basic risk assessment on the target patient based on the surgical process parameter to obtain the basic risk assessment result is performed using the following formula:

[0018] R base = 1 / [1 + exp (-log)] * 100%,

[0019] wherein exp ≈ 2.718; log = aX1 + bX2 + cX3 + dX4 + eX5 + fX6 + gX7 + hX8 + i; a, b, c, d, e, f, h, g, i are coefficients; X1 is pre-anesthesia body temperature, X2 is body index, X3 is general anesthesia combined with nerve block anesthesia or epidural anesthesia, X4 is limb surgery, X5 is chest surgery, X6 is abdominal surgery, X7 is whether a flushing fluid is used, and X8 is surgery duration.

[0020] Optionally, the dynamically assessing the hypothermia risk of the target patient based on the core temperature change feature to obtain a dynamic risk assessment result comprises:

[0021] calculating a first change amount in a first time window and a second change amount in a second time window based on the core temperature change feature;

[0022] determining a first risk correction value based on a change interval to which the first change amount belongs;

[0023] determine a second risk correction value based on a change interval to which the second change amount belongs;

[0024] correct the basic risk assessment result based on the first risk correction value and the second risk correction value.

[0025] Optionally, the dynamic risk assessment result is obtained by dynamically assessing the hypothermia risk of the target patient based on the core temperature change feature includes:

[0026] acquire a dynamic risk parameter and a dynamic risk parameter type of the target patient during a perioperative period; the dynamic risk parameter type includes an endogenous risk parameter, an exogenous direct risk parameter and an exogenous indirect risk parameter;

[0027] determine a corresponding dynamic correction factor according to the dynamic risk parameter and the dynamic risk parameter type;

[0028] correct the basic risk assessment result based on the dynamic correction factor to obtain the dynamic risk assessment result.

[0029] Optionally, the endogenous risk parameter includes a bleeding parameter, the exogenous direct risk parameter includes an exogenous parameter directly placed in the body, and the exogenous indirect risk parameter includes an environmental parameter;

[0030] The corresponding dynamic correction factor is determined according to the dynamic risk parameter and the dynamic risk parameter type, including:

[0031] acquire a risk parameter time sequence change feature of the dynamic risk parameter of different types respectively;

[0032] input the risk parameter time sequence change feature into a supplementary risk assessment model respectively to obtain a heat loss assessment result corresponding to the dynamic risk parameter, wherein the supplementary risk assessment model is constructed based on a time sequence cumulative dependence relationship between the core temperature change feature and the risk parameter time sequence change feature;

[0033] fuse the heat loss assessment results corresponding to all the dynamic risk parameters to obtain the dynamic correction factor.

[0034] Optionally, the automatic control of the heating device based on the dynamic risk assessment result includes:

[0035] predict the core temperature change trend based on the dynamic risk assessment result;

[0036] control the heating device to heat according to a power change trend opposite to the core temperature change trend based on the core temperature change trend.

[0037] According to a second aspect, the embodiments of the present application provide a perioperative hypothermia management system, comprising: a body surface temperature acquisition device, a controller, a heating device and a display device, wherein the body surface temperature acquisition device is arranged on the body surface of a target patient, used to acquire the body surface temperature of the target patient and transmit the body surface temperature to the controller, the controller is used to execute the perioperative hypothermia management method according to any one of the first aspect, and the heating device is used to execute the corresponding heating strategy under the control of the controller.

[0038] Optionally, the body surface temperature acquisition device comprises a patch type wireless body temperature sensor, used to be attached to the body surface of the target patient, and used to wirelessly communicate with the controller and transmit the body surface temperature to the controller in real time; the display device is connected with the controller, used to dynamically display the body surface temperature and the core temperature in real time.

[0039] In the present application, by analyzing the difference between the body surface temperature and the core temperature, the body surface temperature time series data is converted into the core temperature change characteristics by using the pre-constructed body temperature conversion model. Not only the risk caused by invasive measurement (such as mucosal damage and even bleeding) is avoided, but also the accuracy of non-invasive body temperature monitoring is improved, so that medical staff can more accurately perform intraoperative body temperature management. By combining the preoperative static risk score and the intraoperative dynamic factors (such as the amount of bleeding, the temperature and amount of flushing fluid), the intraoperative hypothermia risk score is calculated in real time, a more comprehensive and accurate risk assessment is provided, the warming strategy is optimized, and the incidence of intraoperative hypothermia is reduced. At the same time, according to the dynamic risk assessment result, the heating device is automatically controlled to automatically heat or stop heating to dynamically adjust the body temperature of the patient, realizing automatic body temperature regulation. The automatic heating or cooling program can be automatically performed according to the preset body temperature upper and lower limit parameters, the risk of various adverse events caused by body temperature fluctuation is reduced, and the overall medical service quality and efficiency are improved. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0041] Figure 1 is a flowchart of the perioperative hypothermia management method according to the embodiments of the present application;

[0042] Figure 2 is a schematic diagram of the perioperative hypothermia management system according to the embodiments of the present application;

[0043] Figure 3 Fig. 1 is a schematic diagram of a hardware structure of a controller of a perioperative hypothermia management system according to an embodiment of the present application. DETAILED DESCRIPTION

[0044] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0045] According to the embodiments of the present application, a perioperative hypothermia management method is provided, as shown in the following steps. Figure 1

[0046] S10. Real-time acquisition of body surface temperature time series data of a target patient in a perioperative period. In the present embodiment, the body surface temperature can be collected in real time by a body surface temperature collection device. For example, a high-precision NTC thermistor (precision ±0.1℃) can be used, which is packaged in a flexible silicone patch and adhered to the skin surface of the axillary fossa to collect the body surface temperature of the skin surface of the axillary fossa.

[0047] ​S20. converting the body surface temperature time series data into core temperature change features based on a pre-constructed body temperature conversion model, wherein the body temperature conversion model is constructed based on the correlation between the body surface temperature time series data and the core temperature change features. In this embodiment, the body surface temperature conversion model can be constructed after analyzing the body surface temperature and core temperature of multiple test subjects. For example, 190 sets of axillary temperature (non-core temperature) and nasopharyngeal temperature (core temperature) data of 23 patients are analyzed, and the difference (ΔT = nasopharyngeal temperature - axillary temperature) between the axillary temperature and the nasopharyngeal temperature is calculated. It is found that ΔT is distributed in three zones: zone A (axillary temperature < 35.6°C), zone B (35.6°C ≤ axillary temperature ≤ 36.2°C), and zone C (axillary temperature > 36.2°C). For each temperature zone, the average difference (Mean ΔT) and its standard deviation in that zone are calculated, i.e., the average difference between the axillary temperature and the nasopharyngeal temperature in zone A (Mean ΔTA) = 0.6°C; the average difference between the axillary temperature and the nasopharyngeal temperature in zone B (Mean ΔTB) = 0.3°C; and the average difference between the axillary temperature and the nasopharyngeal temperature in zone C (Mean ΔTC) = 0.1°C. For a new axillary temperature measurement value, its corresponding temperature zone is determined, and the average difference in the corresponding zone is used to compensate the axillary temperature to obtain the core temperature, i.e., the core temperature in zone A = the axillary temperature in zone A + Mean ΔTA; the core temperature in zone B = the axillary temperature in zone B + Mean ΔTB; and the core temperature in zone C = the axillary temperature in zone C + Mean ΔTC. Then, after obtaining the body surface temperature time series data, the body surface temperature is analyzed in real time to determine the current temperature zone to which the body surface temperature belongs; and the body surface temperature is compensated based on the temperature difference corresponding to the current temperature zone to obtain the core temperature.

[0048] S30. dynamically evaluating the hypothermia risk of the target patient based on the core temperature change features to obtain a dynamic risk evaluation result. The base risk can be determined based on the preoperative surgical process parameters, and the base risk is dynamically corrected based on the core temperature change features during the surgery to obtain the dynamic risk evaluation result. For example,

[0049] The surgery process parameters include pre-anesthesia body temperature, body mass index, general anesthesia combined with nerve block anesthesia or epidural anesthesia, limb surgery, chest surgery, abdominal surgery, whether to use flushing fluid, and surgery duration. For a target patient, inputting the pre-anesthesia body temperature, body mass index, general anesthesia combined with nerve block anesthesia or epidural anesthesia, limb surgery, chest surgery, abdominal surgery, whether to use flushing fluid, and surgery duration can obtain the intraoperative hypothermia risk score of the target patient. If the intraoperative hypothermia risk score of the target patient is lower than the set threshold of the prediction model, it is judged that the intraoperative hypothermia is low risk; if the intraoperative hypothermia risk score of the target patient is higher than the set threshold of the prediction model, it is judged that the intraoperative hypothermia is high risk. Medical staff can use the present application to preoperatively evaluate the intraoperative hypothermia risk of the target patient, facilitate to preplan the warming measures and body temperature monitoring scheme for high-risk patients, and save medical resources for low-risk groups. The basic risk calculation formula of the intraoperative hypothermia risk of the target patient is:

[0050] R base = 1 / [1 + exp (-log)] * 100%,

[0051] wherein exp is approximately 2.718; log = aX1 + bX2 + cX3 + dX4 + eX5 + fX6 + gX7 + hX8 + i; a = -3.795, b = -0.100, c = 0.870, d = 1.351, e = 2.465, f = 0.888, h = 0.987, g = 0.004, i = 138.171 are all coefficients; X1 is the pre-anesthesia body temperature, X2 is the body mass index, X3 is the general anesthesia combined with nerve block anesthesia or epidural anesthesia, X4 is the limb surgery, X5 is the chest surgery, X6 is the abdominal surgery, X7 is whether to use flushing fluid, and X8 is the surgery duration.

[0052] If the anesthesia mode of the target patient is general anesthesia combined with nerve block anesthesia or epidural anesthesia, X3 = 1; if the anesthesia mode of the target patient is not general anesthesia combined with nerve block anesthesia or epidural anesthesia, X3 = 0.

[0053] If the surgery site of the target patient is limb surgery, X4 = 1; if the surgery site of the target patient is not limb surgery, X4 = 0.

[0054] If the surgery site of the target patient is chest surgery, X5 = 1; if the surgery site of the target patient is not chest surgery, X5 = 0.

[0055] If the surgery site of the target patient is abdominal surgery, X6 = 1; if the surgery site of the target patient is not abdominal surgery, X6 = 0.

[0056] If the flushing fluid is used in the operation of the target patient, X7=1; if the flushing fluid is not used in the operation of the target patient, X7=0.

[0057] In the dynamic correction of the basic risk by using the intraoperative core temperature change feature, a first change amount in a first time window and a second change amount in a second time window can be calculated based on the core temperature change feature; a first risk correction value is determined based on the change interval to which the first change amount belongs; a second risk correction value is determined based on the change interval to which the second change amount belongs; the basic risk assessment result is corrected by using the first risk correction value and the second risk correction value. For example, the first time window of the body surface temperature time series data can be collected every 15 minutes, and the body temperature data T1, T2,…, T15 (recorded every 1 minute) of every 15 consecutive time points can be collected.

[0058] The degree of body temperature change is calculated. Taking the degree of body temperature drop as an example, the overall drop degree in 15 minutes is calculated:

[0059] ΔT global =T1-T 15

[0060] The first time window of the body surface temperature time series data can be collected every 5 minutes, and the body temperature drop degree can be calculated every 5 minutes in the first time window:

[0061] ΔT k =T k -T k-4 (k=5,6,7,..,15)

[0062] The first risk correction value can be determined based on the change interval to which the first change amount belongs, which can include determining the first risk correction value based on the total body temperature drop degree ΔT global in 15 minutes. For example:

[0063] If the body temperature drop degree in 15 minutes is greater than 0.3℃, the first risk correction value is +30%;

[0064] If the body temperature drop degree in 15 minutes is between 0.1℃ and 0.3℃, the first risk correction value is +20%;

[0065] If the body temperature drop degree in 15 minutes is between 0℃ and 0.1℃, the first risk correction value is +5%.

[0066] Therefore, the first risk correction value α is represented as:

[0067]

[0068] Determining a second risk correction value based on the range of change to which the second change belongs can include: a risk increase β based on the degree of decline within a 5-minute window, for example:

[0069] If the body temperature drops by ΔT within any 5-minute window k If the temperature is greater than 0.3℃, an additional 10% risk is added. Therefore, the second risk correction value β is expressed as:

[0070]

[0071] When correcting the basic risk assessment result using the first risk correction value and the second risk correction value, the basic risk assessment result Rind can be obtained by combining the first correction value and the second correction value.

[0072]

[0073] Rbase is the basic risk assessment result, with a value ranging from 0 to 100%.

[0074] S40. The heating device is automatically controlled based on the dynamic risk assessment results. In this embodiment, by conducting real-time dynamic risk assessment of perioperative hypothermia, the heating device is automatically controlled to heat or stop heating in real time according to the dynamic risk assessment results to dynamically adjust the patient's body temperature. This can reduce the risk of various adverse events caused by body temperature fluctuations and improve the overall quality and efficiency of medical services.

[0075] In this application, by analyzing the difference between body surface temperature and core temperature, a pre-constructed temperature conversion model is used to convert the time-series data of body surface temperature into core temperature change characteristics. This not only avoids the risks associated with invasive measurements (such as mucosal damage or even bleeding) but also improves the accuracy of non-invasive temperature monitoring, enabling medical staff to manage intraoperative temperature more precisely. By combining preoperative static risk scores with intraoperative dynamic factors (such as blood loss, irrigation fluid temperature and volume), an intraoperative hypothermia risk score is calculated in real time, providing a more comprehensive and accurate risk assessment, optimizing warming strategies, and reducing the incidence of intraoperative hypothermia. Simultaneously, based on the dynamic risk assessment results, the heating device is automatically controlled in real time to automatically heat or stop heating to dynamically adjust the patient's body temperature, achieving automated temperature regulation. Automatic warming or cooling programs can be implemented according to preset upper and lower temperature limits, reducing the risk of various adverse events caused by temperature fluctuations and improving the overall quality and efficiency of medical services.

[0076] In an embodiment, the low body temperature basic risk assessment of the target patient based on the surgical process parameters can include: selecting a plurality of potential independent variables related to the intraoperative hypothermia of the target patient, including age, gender, height, weight, BMI, lung noise, smoking history, drinking history, history of malignant tumor, history of chronic pain, history of surgery, cardiovascular disease, respiratory disease, nervous system disease, endocrine system disease, blood pressure system disease, digestive system disease, urinary system disease, connective tissue disease, ASA classification, body temperature in operating room, body temperature before anesthesia induction, inhalation anesthesia, intravenous anesthesia, combined intravenous and inhalation anesthesia, combined intravenous and inhalation anesthesia combined with epidural anesthesia and / or nerve block anesthesia, head and neck surgery, extremity surgery, chest surgery, abdominal surgery, spine surgery, surgical classification, whether to catheterize, whether to flush, blood loss, duration of surgery, duration of anesthesia, etc.

[0077] Logistic regression (a simple linear classification model, easy to understand and interpret. Even with a small sample size, it can be used to develop a model and provide probability estimates of samples belonging to a specific category, rather than just a simple classification result), Lasso regression (Lasso regularization can select some relevant variables and ignore other variables to reduce the complexity of the model and prevent overfitting. This feature selection can enhance the interpretability of the model) and random forest (an ensemble learning method composed of hundreds or thousands of decision trees. It trains each tree on a slightly different set of observations using bootstrapping, and makes a final prediction by averaging the predictions of each tree) are used to establish three target patient intraoperative hypothermia prediction models.

[0078] In ROC, the AUC (95% CI) of Logistic regression model, Lasso regression model and random forest model were 0.845 (0.776-0.914), 0.807 (0.732-0.882) and 0.720 (0.634-0.817), respectively. At the Youden index, the specificity and sensitivity of Logistic regression model, Lasso regression model and random forest model were 87.65% and 72.34%, 61.73% and 85.11%, 54.32% and 82.98%, respectively; the PPV and NPV were 77.27% and 84.52%, 56.34% and 87.72%, 51.32% and 84.62%, respectively. In the calibration curve, the Logistic regression model was closer to the standard line, while the random forest model had the largest error. In the clinical decision curve, the net benefit of the Logistic regression model was the highest, while the net benefit of the random forest model was the lowest. In summary, the Logistic regression model in this embodiment has the best prediction performance and can accurately predict the risk of intraoperative hypothermia in the target patient. The prediction parameters included in the Logistic regression model are: body temperature before anesthesia induction, body mass index, general anesthesia combined with nerve block anesthesia or epidural anesthesia, limb surgery, chest surgery, abdominal surgery, use of flushing fluid, and duration of surgery. Each prediction parameter has a coefficient and an odds ratio: the coefficient is used to measure the degree of influence of each independent variable on the dependent variable, and the positive and negative signs represent the directional relationship between the independent variable and the dependent variable, and the size of the coefficient represents the contribution degree of the variable to the result; the odds ratio is the ratio of the probability of occurrence to the probability of non-occurrence of the independent variable, which represents the change multiple of the probability of occurrence of the dependent variable relative to the probability of non-occurrence when the independent variable increases by one unit.

[0079] In external validation, the AUC (95% CI) of the Logistic regression model was 0.796 (0.7267-0.865), indicating that the Logistic regression model had good generalizability in predicting intraoperative hypothermia in the target patient.

[0080] The specific risk calculation formula for intraoperative hypothermia in the target patient is:

[0081] R base = 1 / [1+exp(-log)]*100%,

[0082] Wherein, exp≈2.718; log=aX1+bX2+cX3+dX4+eX5+fX6+gX7+hX8+i; a=-3.795, b=-0.100, c=0.870, d=1.351, e=2.465, f=0.888, h=0.987, g=0.004, i=138.171 are coefficients; X1 is the body temperature before anesthesia induction, X2 is the body index, X3 is general anesthesia combined with nerve block anesthesia or epidural anesthesia, X4 is limb surgery, X5 is chest surgery, X6 is abdominal surgery, X7 is whether to use irrigation fluid, and X8 is the duration of surgery.

[0083] In other embodiments, the target patient hypothermia prediction model can be expressed as:

[0084] R base =1 / {1+2.718[-(-3.795*X1-0.100*X2+0.870*X3+1.351*X4+2.465*X5+0.888*X7+0.987*X7+0.004*X8+138.171)]}*100%. Wherein, X1 is assigned to the true situation of body temperature before anesthesia induction, such as 35.5℃; X2 is assigned to the true situation of body index, such as 22.13 kg / m2; X3 is assigned to 1 or 0 according to whether it is general anesthesia combined with nerve block anesthesia or epidural anesthesia; X4 is assigned to 1 or 0 according to whether it is limb surgery; X5 is assigned to 1 or 0 according to whether it is chest surgery; X6 is assigned to 1 or 0 according to whether it is abdominal surgery; X7 is assigned to 1 or 0 according to whether irrigation fluid is used; and X8 is assigned to the expected duration of surgery, such as 90 minutes. It is worth noting that the threshold of the target patient intraoperative hypothermia risk score needs to be determined according to the actual situation. Different hospitals and medical staff can set the threshold of the target patient intraoperative hypothermia risk score according to their professional knowledge and actual use, such as 30% or 60%, and the present application does not specify a specific threshold.

[0085] By assessing the basic risk, for the low-risk population, unnecessary warming measures can be avoided through accurate risk assessment, saving medical resources. For high-risk patients, precise and effective warming strategies can be prepared in advance to reduce the incidence of complications.

[0086] In one embodiment, since the prior art does not fully consider the influence of intraoperative dynamic factors (such as body temperature variables, amount of bleeding, irrigation fluid temperature and amount, etc.) on intraoperative hypothermia risk, resulting in insufficient and inaccurate risk assessment, therefore, in this embodiment, when the core temperature change characteristics are used to dynamically assess the target patient hypothermia risk, dynamic risk parameters are considered to dynamically assess the target patient hypothermia risk, which can specifically include:

[0087] acquiring dynamic risk parameters and dynamic risk parameter types of a perioperative target patient; the dynamic risk parameter types include endogenous risk parameters, exogenous direct risk parameters and exogenous indirect risk parameters;

[0088] determining corresponding dynamic correction factors according to the dynamic risk parameters and the dynamic risk parameter types;

[0089] correcting the basic risk assessment result based on the dynamic correction factors to obtain the dynamic risk assessment result.

[0090] For example, the dynamic risk parameters can be illustrated by taking intraoperative bleeding and intraoperative irrigation as examples:

[0091] Influence of bleeding amount on intraoperative hypothermia risk:

[0092] Bleeding amount within 500ml (or 10% of total blood volume): for every 100ml of blood, the risk of hypothermia increases by about 1.5%.

[0093] Bleeding amount within 500-1000ml (or 10%-20% of total blood volume): for every 100ml of blood, the risk of hypothermia increases by about 4%.

[0094] Bleeding amount more than 1000ml (or 20% of total blood volume): 1000-1500ml: for every 100ml of blood, the risk of hypothermia increases by about 7%; more than 1500ml: for every 100ml of blood, the risk of hypothermia increases by about 10%.

[0095] Therefore, the bleeding-related intraoperative hypothermia risk increase amount R B can be expressed as:

[0096]

[0097] Influence of irrigation fluid temperature and amount on intraoperative hypothermia risk:

[0098] For every 500ml of non-temperature-maintained irrigation fluid (<37℃) used, the risk of intraoperative hypothermia increases by 10%.

[0099] For every 500ml of temperature-maintained irrigation fluid (37℃-43℃) used, the risk of intraoperative hypothermia increases by 2%.

[0100] Therefore, the irrigation fluid temperature and amount-related intraoperative hypothermia risk increase amount RI can be expressed as:

[0101]

[0102] Fusing all the heat loss assessment results corresponding to the dynamic risk parameters to obtain the dynamic correction factor:

[0103] R = R ind + min(30%, R B )+ min(20%, R I )

[0104] In another embodiment, the supplementary risk assessment model further corrects the dynamic correction factor based on the time-dependent relationship between the body temperature drop rate and the dynamic risk parameter, on the basis of the dynamic risk parameter and the dynamic risk parameter type determining the corresponding dynamic correction factor.

[0105] The endogenous risk parameter includes a bleeding parameter, the exogenous direct type risk parameter includes an exogenous parameter directly placed in the body, and the exogenous indirect type risk parameter includes an environmental parameter.

[0106] For the bleeding parameter, about 37kcal of heat is taken away per 100ml of whole blood loss, wherein when acute bleeding occurs, the blood volume in the systemic circulation drops suddenly, the body contracts the blood vessels, causing the body surface heat dissipation and heat production to decrease synchronously, and too much bleeding will also cause metabolic inhibition and loss of body heat production.

[0107] The exogenous parameter directly placed in the body can include a body cavity flushing fluid parameter, an infusion parameter, etc. Taking the body cavity flushing fluid parameter as an example, the body cavity flushing heat loss formula is:

[0108] Q 冲洗 = k x V x (T 核心 -T);

[0109] Wherein k is the heat conduction coefficient of the body cavity flushing, T 核心 is the core temperature, T is the flushing fluid temperature; V is the cumulative flushing volume.

[0110] The environmental parameter of the exogenous indirect type risk parameter can include a surgical room environmental parameter, a transfer process exposure environmental parameter, and a ward environmental parameter. In this embodiment, the environmental parameter can include temperature, wind speed and evaporation rate (humidity is negatively correlated), generally, the room temperature is reduced by 2℃, the exposed part of the radiation heat dissipation is increased by 15%; the wind speed of the laminar system is positively correlated with the convective heat dissipation rate of the exposed part, and the hourly water loss of the open surgical incision is positively correlated with the equivalent heat loss.

[0111] And, for the dynamic risk parameters, dynamic risk parameter types, and the mechanism of action of dynamic risk parameters and core body temperature, in the embodiment, a multi-modal fusion neural network model is used as a supplementary risk assessment model to evaluate the dynamic correction factor. In the embodiment, the multi-modal fusion neural network model includes an input layer, a time feature extraction layer cascaded after the input layer, and an output layer. The time feature extraction layer also includes a time dimension attention mechanism module for attention weighting of the time features extracted by the time feature extraction layer, for example, attention weighting of time periods that most significantly affect core temperature, such as the bleeding peak period, the flushing peak period, and the transport exposure period.

[0112] The corresponding dynamic correction factor is determined according to the dynamic risk parameters and the dynamic risk parameter types, including:

[0113] The risk parameter time sequence change characteristics of different types of dynamic risk parameters are obtained respectively. The risk parameter time sequence change characteristics, as model input characteristics, can include time sequence change characteristics of endogenous risk parameters (bleeding parameters), exogenous direct risk parameters (flushing and infusion parameters), and exogenous indirect risk parameters (environmental parameters). The above time sequence change characteristics can be integrated into a multi-dimensional time sequence feature vector. The multi-dimensional time sequence feature vector includes real-time values and historical window statistical values of dynamic parameters such as real-time bleeding amount, bleeding speed, flushing fluid temperature, flushing fluid flow rate, room temperature, humidity, and wind speed, for example, 1-5 minute sliding window mean, standard deviation, etc. All time sequence feature vectors are synchronized based on the surgery time axis to ensure the time sequence consistency of different modal data.

[0114] The risk parameter time sequence change characteristics are input into a supplementary risk assessment model to obtain heat loss evaluation results corresponding to the dynamic risk parameters, wherein the supplementary risk assessment model is constructed based on the time sequence cumulative dependence relationship between the core temperature change characteristics and the risk parameter time sequence change characteristics. All heat loss evaluation results corresponding to the dynamic risk parameters are fused to obtain the dynamic correction factor.

[0115] In one embodiment, the dynamic risk parameters can include at least two dynamic risk types, one continuous and stable influence type risk parameter, and one short-term fluctuation influence type parameter. Therefore, the time feature extraction layer includes a first time scale feature extraction layer and a second time scale feature extraction layer, wherein the first time scale and the second time scale are different. In the embodiment, the first time scale feature extraction layer can use a long short-term memory network to capture long-distance time sequence dependence, such as the cumulative effect of intraoperative continuous bleeding, long-term flushing, infusion, or long-term low-temperature environment on core temperature. The second time scale feature extraction layer can use a gated recurrent unit to capture medium and short-term time sequence dependence, focusing on the cumulative effect of rapidly changing risk events, such as sudden massive bleeding, short-term large flushing, and sudden risk events.

[0116] For the risk parameters with persistent and stable impact, such as operating room temperature, operating room wind speed, infusion, body fluid evaporation of open surgical incision, etc., such parameters change slowly, but have long-acting time, and long-term memory is needed for their persistent impact on core temperature. The LSTM layer can selectively retain historical temperature data through the forget gate to avoid short-term fluctuations interfering with the judgment of long-term trends.

[0117] For parameters with short-term fluctuation impact, such as sudden massive bleeding, irrigation start, irrigation mutation, irrigation end, environmental parameter mutation during transportation, etc., such parameters can change dramatically in a short time, and need to quickly respond to their immediate impact on heat loss. The update gate of the GRU layer can integrate the latest parameter change amount in time, and the reset gate can ignore outdated historical information; the GRU layer can calculate the current heat loss evaluation value through the input x t and the historical hidden state h t-1 in the current time point, and quickly calculate the current heat loss evaluation value to meet the real-time demand.

[0118] Among them, the GRU layer processes high-frequency short-term parameters (such as bleeding, irrigation fluid) to capture the immediate heat loss effect: wherein, is the input time sequence feature vector of the GRU layer at time t, is the hidden state of the GRU layer at the previous time point of time t, is the hidden state of the GRU layer at time t.

[0119] The LSTM layer processes low-frequency long-term parameters (such as environmental temperature, continuous small amount of bleeding, etc.) to capture the cumulative heat loss effect: wherein, is the input time sequence feature vector of the LSTM layer at time t, is the hidden state of the LSTM layer at the previous time point of time t, is the hidden state of the LSTM layer at time t, and Ct is the cell state of the LSTM layer at time t, and Ct-1 is the cell state of the LSTM layer at the previous time point of time t.

[0120] After obtaining the output features of the GRU layer and the LSTM layer, the output features are weighted by attention, wherein,

[0121] The output features are mapped to the query (Query), key (Key), and value (Value) space. Among them, Query can correspond to the reason for the current core temperature drop rate, Key is the risk parameter feature at each time point, and Value is the heat loss contribution at the corresponding time point, and is weighted by the attention mechanism through the following formula:

[0122] Query, Key, Value mapping: Q = XW Q , K = XW K , V = XW V ;

[0123] Attention score:

[0124] Attention weight: a = softmax(Score);

[0125] The output features of the LSTM layer and the GRU are summed by the attention weight, highlighting the risk contribution of the key period. The attention mechanism weights the risk contribution of different time points and types, and outputs a dynamic correction factor for adjusting the basic heating power.

[0126] The importance of the risk parameters at each time point to the current core temperature evaluation is represented. For example: when a sudden massive bleeding occurs during the operation, the attention weight of the corresponding time point is significantly increased, and the model pays more attention to the contribution of the bleeding parameter at that moment to the heat loss. The low outdoor temperature during the transportation process may have a higher weight at a specific time point (such as within 5 minutes of leaving the operating room), reflecting the immediate environmental impact. When fusing endogenous (bleeding), exogenous direct type (irrigation fluid), and exogenous indirect type (environmental) risks, the attention weight automatically identifies the dominant factor (such as the irrigation fluid temperature being the main factor for body temperature drop at a certain time, and its corresponding exogenous direct type parameter weight being higher). Combined with the output of the time series model (LSTM / GRU), attention can further refine the "key risk period" (such as continuous hypotension accompanied by bleeding in the second half of the operation), optimizing the dynamic adjustment strategy of the heating parameter. By capturing long-term cumulative effects with LSTM and responding to short-term mutations with GRU, the attention mechanism dynamically allocates time series weights, forming a multi-dimensional time series dependent modeling of heat loss, which can handle multi-scale time series features of intraoperative risk parameters, and achieve precise real-time adjustment of the dynamic correction factor, thus more effectively predicting and preventing hypothermia.

[0127] In one embodiment, a local feature enhancement layer can also be included between the input layer and the time series feature extraction layer (GRU layer and LSTM layer) to extract local features within a short time window, divide the multi-dimensional time series feature vector processed by the input layer into multiple local features, and combine the features required for the GRU layer and the LSTM layer to achieve combined features of different dimensional features, such as combining the features corresponding to the bleeding parameters and the irrigation parameters, or combining the features corresponding to the bleeding parameters, the irrigation parameters, and the environmental parameters to capture the combined risks (such as mutual promotion or mutual offset) between different dimensional features. In this embodiment, a convolution kernel of K kernel size can be used to slide in the time series dimension to calculate the local feature mapping:

[0128]

[0129] wherein, K is the length of the convolution kernel, wk is the convolution weight, xt-k is the input feature of the past K time points, Ct is the output multi-dimensional local feature vector, and Ct is combined according to the time length required by the GRU layer and the LSTM layer to capture the risk of combination between different dimensional features within, for example, 1 minute, 5 minutes, 10 minutes.

[0130] In this embodiment, the supplementary risk assessment model can adopt a GRU (Gated Recurrent Unit) network, and the flushing fluid parameters are taken as an example for illustration:

[0131] The multi-modal fusion neural network model outputs a "flushing heat loss coefficient γ" (range 0-1, the larger the value, the higher the risk of body heat loss), and γ is combined with the first risk correction value and the second risk correction value to correct the dynamic risk:

[0132] R ind = min[50%, 0.3 x R base + a + b + 0.2 x g]

[0133] In this embodiment, by combining the preoperative static risk score and intraoperative dynamic factors (such as the amount of bleeding, flushing fluid temperature and amount), the intraoperative hypothermia risk score is calculated in real time, providing a more comprehensive and accurate risk assessment, optimizing the warming strategy, and reducing the incidence of intraoperative hypothermia.

[0134] This embodiment also provides a perioperative hypothermia management system, as shown in Figure 2 The management system includes a body surface temperature acquisition device, a controller and a heating device, wherein the body surface temperature acquisition device is arranged on the target patient's body surface to acquire the body surface temperature of the target patient and transmit the body surface temperature to the controller, the controller is used to execute the perioperative hypothermia management method of any one of the above embodiments, and the heating device is used to execute the corresponding heating strategy under the control of the controller.

[0135] In this embodiment, the body surface temperature acquisition device includes a patch type wireless body temperature sensor, which is used to adhere to the target patient's body surface and wirelessly communicates with the controller to transmit the body surface temperature to the controller in real time.

[0136] Exemplarily, a high-precision NTC thermistor (accuracy ±0.1°C) is used, packaged in a flexible silicone patch, attached to the skin surface of the armpit, and wireless data transmission is achieved through a low-power Bluetooth 5.0 module. In this embodiment, a display screen can also be configured to display the received body surface temperature, such as armpit temperature, and the core temperature obtained by the body surface temperature, achieving dual body temperature detection and real-time display of dual body temperature. In this embodiment, the body surface temperature and core temperature can be displayed in the form of a temperature curve or in the form of a scattered point labeled value, and the specific display method is not limited in this embodiment. In addition, the body surface temperature and core temperature can be displayed in different colors of curves or scattered points or values to facilitate effective differentiation by medical personnel.

[0137] Exemplarily, the display screen can be a 1.3-inch OLED display screen (resolution 128x64) that displays the armpit temperature, core temperature, and system operating status in real time.

[0138] The body surface temperature collection device is fixed to the armpit by a medical-grade adhesive and is covered with a breathable elastic bandage to ensure close fitting and avoid motion interference. The heating unit in the heating device uses a PTC ceramic heating sheet (rated power 20W, response time <5 seconds) integrated into an inflatable heat preservation device, which uses a matching inflatable heat preservation blanket to cover the main body surface heat dissipation area. A brushless DC fan (wind speed 0.5-3m / s adjustable) is used to improve the heat convection efficiency.

[0139] In the heating mode, when the core temperature < the lower limit of the preset temperature, start PTC heating and gradually increase the fan speed (PID control, adjustment period 10 seconds). In the cooling mode, when the core temperature > the upper limit of the preset temperature, reduce the PTC heating temperature and gradually reduce the fan speed (PID control, adjustment period 10 seconds).

[0140] The perioperative hypothermia management system also has an interactive design. Exemplarily, in the mobile application: based on BLE Mesh networking, it supports simultaneous connection to multiple devices. Independent controller: equipped with physical knobs and buttons, with preset quick temperature control gears (such as "rapid cooling" and "energy-saving heat preservation").

[0141] A fuzzy PID control algorithm is established to adaptively adjust the temperature control strength according to the difference between the core temperature and the preset body temperature:

[0142] When the core temperature < 35.5°C, start the heating mode: the heating device is set to 43°C with high wind volume.

[0143] When 35.5°C ≤ core temperature ≤ 36°C, start the heat preservation mode: the heating device is set to 38°C with high wind volume.

[0144] (3) When core temperature > 36℃, start basic mode: heating device set to 32℃, low air volume.

[0145] Set up a hardware watchdog circuit, which automatically cuts off the temperature control power supply when the system is unresponsive for more than 30 seconds.

[0146] Dual backup storage of critical data (Flash + SD card) to prevent data loss due to sudden power failure.

[0147] The controller provided by the embodiment of the applicationapplicationinclude a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete mutual communication through the communication bus, the memory is configured to store a computer program, and the processor is configured to execute the method in any one of the above embodiments by running the computer program stored on the memory.

[0148] Figure 3 is a structural block diagram of an optional controller according to the embodiment of the application, as shown in Figure 3 the processor 10, the communication interface 20 and the memory 30 complete mutual communication through the communication bus 40, wherein,

[0149] the memory 30 is configured to store a computer program;

[0150] the processor 10 is configured to execute the computer program stored on the memory 30, so as to realize the method in any one of the above embodiments.

[0151] Optionally, in the embodiment, the communication bus can be a PCI (Peripheral Component Interconnect, Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture, Extended Industry Standard Architecture) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, Figure 3 only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0152] The communication interface is configured to communicate between the above controller and other devices.

[0153] The memoryapplicationinclude a RAM and a non-volatile memory, for example, at least one disk memory. Optionally, the memoryapplicationbe at least one storage device located away from the aforementioned processor.

[0154] The processor can be a general processor, which can include but is not limited to a CPU (Central Processing Unit), a NP (Network Processor), etc. The processor can also be a DSP (Digital Signal Processing), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component.

[0155] Optionally, the specific examples in the embodiments can refer to the examples described in the above embodiments, and the embodiments will not be described here.

[0156] Those skilled in the art can understand that the above-mentioned embodiments can be implemented by hardware, software or a combination of hardware and software. Figure 3 The structure shown is only schematic, and the controller implementing any one of the above-mentioned embodiments can be a terminal device, which can be a smart phone (such as an Android phone, an IOS phone, etc.), a tablet computer, a palm computer, a Mobile Internet Device (MID), a PAD, etc. Figure 3 It does not limit the structure of the above-mentioned controller. For example, the terminal device can further include more or less components (such as a network interface, a display device, etc.) than those shown in the above-mentioned embodiments, or have a different configuration from that shown in the above-mentioned embodiments. Figure 3 For example, the terminal device can further include more or less components (such as a network interface, a display device, etc.) than those shown in the above-mentioned embodiments, or have a different configuration from that shown in the above-mentioned embodiments. Figure 3 For example, the terminal device can further include more or less components (such as a network interface, a display device, etc.) than those shown in the above-mentioned embodiments, or have a different configuration from that shown in the above-mentioned embodiments.

[0157] Those skilled in the art can understand that all or part of the steps of the above-mentioned embodiments can be completed by programs instructing the related hardware of the terminal device, and the programs can be stored in a computer readable storage medium, which can include a flash disk, a ROM, a RAM, a magnetic disk or an optical disk, etc.

[0158] As an exemplary embodiment, the present application also provides a computer readable storage medium, which stores a computer program, wherein the computer program is configured to execute the method steps of any one of the above-mentioned embodiments when running.

[0159] Optionally, in the embodiments, the above-mentioned storage medium can be used for the program code for executing the method steps of the embodiments.

[0160] Optionally, in the embodiments, the above-mentioned storage medium can be located on at least one of the network devices in the network shown in the above-mentioned embodiments.

[0161] Optionally, in the embodiment, the storage medium is configured to store the method in the above embodiment.

[0162] Optionally, the specific examples in the embodiment can refer to the examples described in the above embodiment, and the embodiment will not be described here.

[0163] Optionally, in the embodiment, the storage medium described above can include but is not limited to: a U disk, a ROM, a RAM, a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0164] The serial number of the embodiment of the application described above is only for description, and does not represent the advantages and disadvantages of the embodiment.

[0165] The integrated units in the above embodiment, if realized in the form of a software function unit and sold or used as an independent product, can be stored in the above computer-readable storage medium. Based on such understanding, the technical solutions of the application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing one or more computer devices (which can be personal computers, servers or network devices, etc.) to execute all or part of the steps of the method in the above embodiment.

[0166] In several embodiments provided in the application, it should be understood that the disclosed client can be implemented by other ways. Among them, the device embodiment described above is only schematic, for example, the division of units is only a logical function division, and actual implementation can have another division way, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the coupling or direct coupling or communication connection between the displayed or discussed each other can be through some interface, indirect coupling or communication connection between units or modules, which can be electrical or other forms.

[0167] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place or distributed on multiple network units. Part or all of the units can be selected to achieve the purpose of the scheme provided in the embodiment according to actual needs.

[0168] In addition, each functional unit in each embodiment of the application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software function unit.

[0169] In the above-described embodiments of the present application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0170] The above is only the preferred embodiment of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, can make a number of improvements and refinements, these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A perioperative hypothermia management method, characterized by, The method comprises the following steps: real-time acquisition of body surface temperature time series data of a target patient during a perioperative period; conversion of the body surface temperature time series data into core temperature change characteristics based on a pre-constructed body temperature conversion model, wherein the body temperature conversion model is constructed based on the correlation between the body surface temperature time series data and the core temperature change characteristics; dynamic assessment of the target patient's hypothermia risk based on the core temperature change characteristics, resulting in a dynamic risk assessment result, including acquisition of the target patient's surgical process parameters; and basic risk assessment of the target patient based on the surgical process parameters, resulting in a basic risk assessment result; acquisition of dynamic risk parameters and dynamic risk parameter types of the target patient during the perioperative period; determination of corresponding dynamic correction factors according to the dynamic risk parameters and the dynamic risk parameter types; and correction of the basic risk assessment result based on the dynamic correction factors, resulting in the dynamic risk assessment result, wherein determination of corresponding dynamic correction factors according to dynamic risk parameters and dynamic risk parameter types includes: acquisition of risk parameter time series change characteristics of different types of dynamic risk parameters, including continuous stable influence type risk parameters and short-term fluctuation influence type parameters; inputting the risk parameter time series change characteristics into a supplementary risk assessment model to obtain a heat loss assessment result corresponding to the dynamic risk parameters, wherein the supplementary risk assessment model comprises an input layer, a time feature extraction layer cascaded after the input layer, and an output layer, the time feature extraction layer comprising a long short-term memory network and a gated recurrent unit, weighting and summing the output features of the long short-term memory network and the gated recurrent unit according to attention weights to output dynamic correction factors; and automatic control of a heating device based on the dynamic risk assessment result.

2. The method for perioperative hypothermia management according to claim 1, wherein The body temperature conversion model comprises a temperature interval-based body temperature correction model; conversion of the body surface temperature into core temperature based on the pre-constructed body temperature conversion model includes: real-time analysis of the body surface temperature to determine the current temperature interval to which the body surface temperature belongs; compensation of the body surface temperature based on the temperature difference corresponding to the current temperature interval to obtain the core temperature.

3. The perioperative hypothermia management method according to claim 1, wherein, Before real-time acquisition of body surface temperature time series data of a target patient during a perioperative period, the method further comprises the following steps: acquisition of the target patient's surgical process parameters; basic risk assessment of the target patient based on the surgical process parameters, resulting in a basic risk assessment result; determination of a basic heating strategy of the heating device during the perioperative period based on the basic risk assessment result.

4. The perioperative hypothermia management method of claim 3 wherein, The basic risk assessment of the target patient based on the surgical process parameters, resulting in a basic risk assessment result, is evaluated using the following formula: = 1 / [1 + exp(-log)]*100%, Wherein, exp≈2.718; log=aX1+bX2+cX3+dX4+eX5+fX6+gX7+hX8+i; a, b, c, d, e, f, h, g, i are coefficients; X1 is the body temperature before anesthesia induction, X2 is the body index, X3 is general anesthesia combined with nerve block anesthesia or epidural anesthesia, X4 is limb surgery, X5 is chest surgery, X6 is abdominal surgery, X7 is whether to use flushing fluid, and X8 is the duration of surgery.

5. The perioperative hypothermia management method according to claim 3, wherein, Based on the core temperature change characteristics, the low body temperature risk of the target patient is dynamically evaluated, and a dynamic risk assessment result is obtained. Based on the core temperature change characteristics, a first change amount in a first time window and a second change amount in a second time window are calculated. A first risk correction value is determined based on the change interval to which the first change amount belongs. A second risk correction value is determined based on the change interval to which the second change amount belongs. The first risk correction value and the second risk correction value are used to correct the basic risk assessment result.

6. The perioperative hypothermia management method of claim 3 wherein, The core temperature change characteristics are used to dynamically evaluate the low body temperature risk of the target patient, and a dynamic risk assessment result is obtained. The dynamic risk parameters and dynamic risk parameter types of the target patient during the perioperative period are obtained; the dynamic risk parameter types include endogenous risk parameters, exogenous direct risk parameters, and exogenous indirect risk parameters. The corresponding dynamic correction factors are determined according to the dynamic risk parameters and the dynamic risk parameter types. The dynamic risk assessment result is obtained by correcting the basic risk assessment result based on the dynamic correction factors.

7. The perioperative hypothermia management method of claim 6 wherein, The endogenous risk parameters include bleeding parameters, the exogenous direct risk parameters include exogenous parameters directly placed in the body, and the exogenous indirect risk parameters include environmental parameters. The corresponding dynamic correction factors are determined according to the dynamic risk parameters and the dynamic risk parameter types. The risk parameter time sequence change characteristics of different types of dynamic risk parameters are obtained respectively. The risk parameter time sequence change characteristics are input into a supplementary risk assessment model respectively to obtain heat loss evaluation results corresponding to the dynamic risk parameters, wherein the supplementary risk assessment model is constructed based on the time sequence cumulative dependence relationship between the core temperature change characteristics and the risk parameter time sequence change characteristics. The heat loss evaluation results corresponding to all dynamic risk parameters are fused to obtain the dynamic correction factors.

8. The perioperative hypothermia management method of claim 1 wherein, The dynamic risk assessment result is used to automatically control the heating device, which includes: The core temperature change trend is predicted based on the dynamic risk assessment result. The core temperature change trend is predicted based on the dynamic risk assessment result.

9. A perioperative hypothermia management system, characterized by The core temperature change trend is predicted based on the dynamic risk assessment result. The core temperature change trend is predicted based on the dynamic risk assessment result. The core temperature change trend is predicted based on the dynamic risk assessment result. The body surface temperature acquisition device, the controller, the heating device and the display device, wherein the body surface temperature acquisition device is arranged on the body surface of the target patient, used for acquiring the body surface temperature of the target patient and transmitting the body surface temperature to the controller, the controller is used for executing the method for managing the peroperative hypothermia according to any one of claims 1-8, and the heating device is used for executing the corresponding heating strategy under the control of the controller.

10. The peroperative hypothermia management system according to claim 9, characterized in that, The body surface temperature acquisition device comprises a patch type wireless body temperature sensor, used for being attached to the body surface of the target patient, wirelessly communicating with the controller and transmitting the body surface temperature to the controller in real time. The display device is connected with the controller and used for dynamically displaying the body surface temperature and the core temperature in real time.

Images