Hypoxia risk monitoring and early warning device and method for postoperative patient

Abstract

The invention discloses a hypoxia risk monitoring and early warning device and method for a postoperative patient, and relates to the technical field of medical monitoring, and the method comprises the steps: monitoring the real-time state information of a target user; calling a predetermined state support mechanism, and performing support evaluation on the real-time state information to obtain a support evaluation result; correcting the real-time state information according to the support evaluation result to obtain target state information; introducing a hypoxia evaluation function to evaluate and analyze the target state information to obtain a real-time prediction hypoxia coefficient; judging whether the real-time predicted hypoxia coefficient meets a preset hypoxia coefficient threshold value or not, and if the real-time predicted hypoxia coefficient does not meet the preset hypoxia coefficient threshold value, sending out a hypoxia risk early warning instruction; and based on the hypoxia risk early warning instruction, starting a predetermined hypoxia intervention scheme to carry out risk intervention suggestions on the target user. The technical problem that in the prior art, low-oxygen early-stage signals are difficult to capture in time, and intervention delay is caused is solved, and the technical effects of accurate recognition and rapid intervention of low-oxygen risks are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical monitoring, and particularly relates to a hypoxia risk monitoring and early warning device and method for postoperative patients. Background Art

[0002] Postoperative hypoxia is one of the common complications after anesthesia, which has a significant impact on the postoperative recovery and life safety of patients. At present, although the monitoring of postoperative hypoxia can reflect the changes in blood oxygen saturation in real time, its monitoring results have a certain lag. It often detects abnormalities only after the patient has shown obvious hypoxia symptoms, making it difficult to capture the early signals of hypoxia in a timely manner, resulting in delayed intervention and increasing the risk of postoperative hypoxia. Summary of the Invention

[0003] This application provides a hypoxia risk monitoring and early warning device and method for postoperative patients, which are used to solve the technical problem that it is difficult to capture the early signals of hypoxia in the prior art in a timely manner, resulting in delayed intervention.

[0004] In view of the above problems, this application provides a hypoxia risk monitoring and early warning device and method for postoperative patients.

[0005] In the first aspect of this application, a hypoxia risk monitoring and early warning device for postoperative patients is provided. The device includes:

[0006] A real-time information acquisition module, which is used to dynamically monitor the real-time status information of the target user through a sensor component; a support evaluation module, which is used to retrieve a predetermined status support mechanism and perform a support evaluation on the real-time status information according to the predetermined status support mechanism to obtain a support evaluation result; a correction module, which is used to correct the real-time status information according to the support evaluation result to obtain target status information; an evaluation and analysis module, which is used to introduce a hypoxia evaluation function to evaluate and analyze the target status information to obtain a real-time predicted hypoxia coefficient; a threshold judgment module, which is used to judge whether the real-time predicted hypoxia coefficient meets a predetermined hypoxia coefficient threshold; an early warning module, which is used to issue a hypoxia risk early warning instruction if the real-time predicted hypoxia coefficient does not meet the predetermined hypoxia coefficient threshold; a risk intervention module, which is used to start a predetermined hypoxia intervention plan based on the hypoxia risk early warning instruction to give risk intervention suggestions to the target user.

[0007] In the second aspect of this application, a hypoxia risk monitoring and early warning method for postoperative patients is provided. The method includes:

[0008] Dynamically monitor the real-time status information of the target user through the sensor component; retrieve the predetermined status support mechanism, and perform a support evaluation on the real-time status information according to the predetermined status support mechanism to obtain a support evaluation result; correct the real-time status information according to the support evaluation result to obtain the target status information; introduce a hypoxia evaluation function to evaluate and analyze the target status information to obtain a real-time predicted hypoxia coefficient; determine whether the real-time predicted hypoxia coefficient meets the predetermined hypoxia coefficient threshold; if the real-time predicted hypoxia coefficient does not meet the predetermined hypoxia coefficient threshold, issue a hypoxia risk warning instruction; based on the hypoxia risk warning instruction, start a predetermined hypoxia intervention plan to give risk intervention suggestions to the target user.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] In this application, the real-time status information of the target user is dynamically monitored through the sensor component; the predetermined status support mechanism is retrieved, and a support evaluation is performed on the real-time status information according to the predetermined status support mechanism to obtain a support evaluation result; the real-time status information is corrected according to the support evaluation result to obtain the target status information; a hypoxia evaluation function is introduced to evaluate and analyze the target status information to obtain a real-time predicted hypoxia coefficient; it is determined whether the real-time predicted hypoxia coefficient meets the predetermined hypoxia coefficient threshold; if the real-time predicted hypoxia coefficient does not meet the predetermined hypoxia coefficient threshold, a hypoxia risk warning instruction is issued; based on the hypoxia risk warning instruction, a predetermined hypoxia intervention plan is started to give risk intervention suggestions to the target user. The present invention solves the technical problem that it is difficult for the prior art to capture early hypoxia signals in a timely manner, resulting in delayed intervention. By real-time monitoring the user's status through the sensor component, performing data evaluation and correction based on the predetermined status support mechanism, combining with the analysis of the hypoxia evaluation function to obtain the real-time predicted hypoxia coefficient, determining whether it exceeds the preset threshold, if it exceeds, a hypoxia risk warning instruction is issued, and the hypoxia intervention plan is started to provide personalized intervention suggestions, achieving the technical effect of accurately identifying and quickly intervening in hypoxia risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0012] Figure 1 It is a schematic structural diagram of a hypoxia risk monitoring and warning device for postoperative patients provided by an embodiment of this application;

[0013] Figure 2 Schematic flowchart of a hypoxia risk monitoring and early warning method for postoperative patients provided by an embodiment of the present application.

[0014] Explanation of reference numerals: real-time information acquisition module 11, support evaluation module 12, calibration module 13, evaluation and analysis module 14, threshold judgment module 15, early warning module 16, risk intervention module 17. Detailed implementation manners

[0015] By providing a hypoxia risk monitoring and early warning device and method for postoperative patients, the present application aims to solve the technical problem that it is difficult to capture early hypoxia signals in the prior art in a timely manner, resulting in delayed intervention. The user's status is monitored in real time through a sensor component, data evaluation and calibration are performed based on a predetermined status support mechanism, a real-time predicted hypoxia coefficient is obtained by combining with a hypoxia evaluation function analysis, it is judged whether it exceeds a preset threshold, if it exceeds, a hypoxia risk early warning instruction is issued, and a hypoxia intervention plan is started to provide personalized intervention suggestions, achieving the technical effects of accurate identification and rapid intervention of hypoxia risk.

[0016] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0017] It should be noted that any variations of the terms "including" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, device, product or server including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices.

[0018] Embodiment 1, as Figure 1 shown, an embodiment of the present application provides a hypoxia risk monitoring and early warning device for postoperative patients, and the device includes:

[0019] A real-time information acquisition module 11, configured to dynamically monitor the real-time status information of a target user through a sensor component.

[0020] In the embodiment of the present application, the real-time information acquisition module 11 is designed to dynamically monitor the real-time status information of the target user through the sensor component. Specifically, first, the blood oxygen sensor is activated. This sensor is fixed to the fingertip of the target user and is used to dynamically monitor the real-time blood oxygen saturation, reflecting the oxygen content level in the blood. Subsequently, the respiration sensor is activated to monitor the user's respiration rate and heart rate in real time, and to evaluate the status of the respiratory and circulatory systems. Finally, the blood oxygen saturation, respiration rate, and heart rate data are integrated to form complete real-time status information.

[0021] Furthermore, in the device provided by the embodiment of the application, the real-time information acquisition module 11 is further configured to:

[0022] The blood oxygen sensor activation unit is used to activate the blood oxygen sensor in the sensor component, where the blood oxygen sensor is fixed to the target fingertip of the target user; the blood oxygen saturation monitoring unit is used to dynamically monitor the real-time blood oxygen saturation through the blood oxygen sensor; the respiration sensor activation unit is used to activate the respiration sensor in the sensor component and dynamically monitor the real-time respiration rate and real-time heart rate through the respiration sensor; the real-time status information generation unit is used to form the real-time status information based on the real-time blood oxygen saturation, the real-time respiration rate, and the real-time heart rate.

[0023] In the embodiment of the present application, first, the blood oxygen sensor activation unit activates the blood oxygen sensor in the sensor component. The blood oxygen sensor adopts the photoplethysmography technology and is fixed to the target fingertip of the target user, such as the index finger or the middle finger, because the fingertip is rich in capillaries and has a stable blood flow, which is suitable for optical detection. The blood oxygen saturation monitoring unit is used to emit red light and infrared light through the blood oxygen sensor, penetrate the skin and vascular tissues, and calculate the real-time blood oxygen saturation by using the different absorption characteristics of the light by oxyhemoglobin and deoxyhemoglobin in the blood.

[0024] Next, the respiration sensor activation unit activates the respiration sensor, which can be based on different technical principles, such as resistive strain gauges, thermistors, or piezoelectric sensors, etc. The respiration sensor is mainly used to dynamically monitor the real-time respiration rate, that is, the number of breaths per unit time. The normal respiration rate of an adult is about 12 - 20 times per minute. In addition, this sensor can also indirectly monitor the real-time heart rate, that is, the number of heartbeats per minute. Heart rate monitoring is often achieved through electrocardiogram technology or analysis methods based on pulse waves.

[0025] After the monitoring of the blood oxygen saturation, respiration rate, and heart rate is completed, the real-time status information generation unit integrates these three pieces of data to form complete real-time status information.

[0026] A support evaluation module 12 is configured to retrieve a predetermined status support mechanism and perform a support evaluation on the real-time status information according to the predetermined status support mechanism to obtain a support evaluation result.

[0027] In an embodiment of the present application, the support evaluation module 12 scientifically evaluates the real-time status information of a postoperative patient based on a predetermined status support mechanism. Its working process includes randomly extracting any status indicator and corresponding parameters from the real-time status information, retrieving a relevant historical parameter sequence to construct a fitting spline curve, calculating the vertical distance between the real-time parameter and the fitting spline, and then analyzing this distance to obtain a support index. Then, the support index is compared with a preset threshold value and marked as normal (first mark) or abnormal (second mark). Finally, all the marking results are comprehensively combined to form a support evaluation result.

[0028] Further, in the device provided in the embodiment of the application, the support evaluation module 12 is further configured to:

[0029] A status indicator extraction unit is configured to randomly extract any indicator parameter corresponding to any status indicator from the real-time status information; a parameter sequence retrieval and analysis unit is configured to retrieve any parameter sequence of the any status indicator and analyze the any parameter sequence to obtain any fitting spline; a vertical distance calculation unit is configured to calculate an arbitrary vertical distance from the any indicator parameter to the any fitting spline; a support index analysis unit is configured to analyze the any vertical distance based on the predetermined status support mechanism to obtain any support index; a parameter marking unit is configured to, if the any support index is within a predetermined index threshold, perform a first mark on the any indicator parameter, and if the any support index is not within the predetermined index threshold, perform a second mark on the any indicator parameter; a support evaluation result generation unit is configured to form the support evaluation result based on the first mark and the second mark.

[0030] In an embodiment of the present application, the status indicator extraction unit first randomly extracts any status indicator and its corresponding indicator parameter from the real-time status information, such as blood oxygen saturation, respiratory rate, or heart rate.

[0031] Next, the parameter sequence retrieval and analysis unit retrieves the historical parameter sequence of the selected status indicator from the historical database and generates an arbitrary fitting spline curve through a data analysis method. Specifically, after retrieving the parameter sequence of any status indicator, first form a first sample parameter group according to this sequence and obtain a first verification parameter group through comparison. Then generate a first spline curve based on the first sample parameter group to describe the change trend of the status indicator. Subsequently, use the first verification parameter group to perform a validity verification on the spline curve to obtain a first verification result. If this result meets the predetermined verification constraint, determine this spline curve as an arbitrary fitting spline.

[0032] Subsequently, the vertical distance calculation unit calculates the vertical distance between any index parameter and the fitted spline to quantify the deviation between the real-time data and the historical trend. The calculation of the vertical distance is based on the Euclidean distance formula. After obtaining the vertical distance, the support index analysis unit combines a predetermined state support mechanism to analyze the distance and calculate any support index to quantify the normality of the current state. When calculating the support index, it is calculated through the formula where d max is the preset maximum acceptable deviation value, SI is the support index, and d is any vertical distance.

[0033] Next, the parameter marking unit compares the support index with a preset threshold value to determine the normality of the current state. If the support index is greater than or equal to the preset threshold value (e.g., 0.8), the index parameter is marked as the first mark, indicating a normal state; if the support index is less than the preset threshold value, it is marked as the second mark, indicating a potential physiological abnormality.

[0034] Finally, the support evaluation result generation unit synthesizes the results of all the first marks and the second marks to form a support evaluation result.

[0035] Furthermore, in the device provided by the application embodiment, the support evaluation module 12 is further configured to:

[0036] The sample parameter group construction unit is configured to construct a first sample parameter group according to the arbitrary parameter sequence and obtain a first verification parameter group by comparison; the spline curve generation unit is configured to generate a first spline curve based on the first sample parameter group; the spline curve verification unit is configured to verify the first spline curve by using the first verification parameter group to obtain a first verification result; the fitted spline determination unit is configured to use the first spline curve as the arbitrary fitted spline when the first verification result meets the predetermined verification constraint.

[0037] In the embodiment of the present application, the sample parameter group construction unit first constructs a first sample parameter group according to the arbitrary parameter sequence of the postoperative patient, that is, the real-time monitored physiological data (such as blood oxygen saturation, respiratory rate, heart rate, etc.). The construction of this parameter group is based on historical monitoring data and aims to reflect the index change trend of the patient in a normal or near-normal physiological state. During this process, data preprocessing is performed, including outlier removal, missing value filling, data smoothing processing, etc., to ensure the reliability and consistency of the data, and finally a first sample parameter group is obtained.

[0038] Next, based on the first sample parameter group, a first verification parameter group is obtained by comparison. The first verification parameter group includes a data set different from the sample group, which is the latest real-time monitored data or independent clinical reference data.

[0039] After obtaining the first set of sample parameters, the spline curve generation unit generates the first spline curve based on this data. This step uses the cubic spline interpolation method to construct a smooth and continuous curve between data points, accurately describing the changing trend of physiological indicators over time. The specific operations include calculating the spline function coefficients for each data segment to ensure that the curve has continuous derivatives at the connection points of each segment, guaranteeing smoothness, and generating a spline curve that can reflect the changing trend of physiological data, suitable for capturing minute physiological fluctuations. The first spline curve is obtained through this process.

[0040] After generating the first spline curve, the spline curve verification unit uses the previously obtained first set of verification parameters to verify its validity. This process aims to evaluate the applicability and accuracy of the spline curve in different data environments, ensuring that it is not only applicable to historical data but also can accurately reflect the current real-time state. The verification methods mainly include error analysis, using common error evaluation methods such as the mean squared error (MSE) to quantify the deviation between the spline curve and the actual monitoring data. Through error calculation, the fitting accuracy of the spline curve is determined, obtaining the first verification result and quantifying the deviation degree between the spline curve and the actual data.

[0041] Finally, the fitting spline determination unit compares the first verification result with the predetermined verification constraint conditions. The predetermined verification constraint is usually a set error threshold (such as MSE less than 0.05) to ensure that the spline curve has sufficient fitting accuracy. If the error value of the first verification result is lower than the threshold, it is considered that the spline curve meets the requirements in terms of both accuracy and generalization ability, and it is determined as an arbitrary fitting spline. If the error exceeds the threshold, adjust the sample parameter group or refit the spline until the preset requirements are met, obtaining an arbitrary fitting spline.

[0042] Calibration module 13 is used to calibrate the real-time state information according to the support evaluation result to obtain the target state information.

[0043] In the embodiment of the present application, the calibration module 13 receives the support evaluation result, which includes the deviation analysis between the real-time state information and the normal physiological state, such as the support index and the state marker (the first marker indicates normal, and the second marker indicates abnormal).

[0044] Next, identify the abnormal data points that need to be calibrated. Generally, when the support index of a certain real-time state data is lower than a preset threshold value (for example, 0.8) and is marked with the second marker, this data point is regarded as an abnormal data. At this time, compare these abnormal data points with the previously generated fitting spline curve to determine the specific deviation range.

[0045] After identifying the data points that need to be corrected, error compensation and data adjustment are performed. A simple correction method based on the fitted spline is adopted, that is, the abnormal data points are adjusted to be within the range of the difference from the fitted spline curve, or directly replaced with the corresponding values of the fitted spline curve.

[0046] After completing the data correction, consistency verification is carried out to ensure that the corrected data conforms to the physiological logic and no new abnormalities occur. This verification usually includes re-checking whether the corrected data still exceeds the normal range, or whether the deviation from the fitted spline curve has been reduced to an acceptable range. If the corrected data is within the physiological range and the deviation is significantly reduced, it is determined as the final target state information.

[0047] The evaluation and analysis module 14 is used to introduce a hypoxia evaluation function to evaluate and analyze the target state information, and obtain a real-time predicted hypoxia coefficient.

[0048] In the embodiment of the present application, the evaluation and analysis module 14 introduces a hypoxia evaluation function to evaluate and analyze the target state information. Specifically, first, the target individual characteristics of the target user are obtained, such as individual parameters such as age and weight, to improve the accuracy of the evaluation. Subsequently, the hypoxia probability evaluation function is read, and based on this function, the target individual characteristics are analyzed to calculate the real-time predicted hypoxia probability, that is, the possibility of the target user having hypoxia in the current state. Then, the hypoxia degree evaluation function is read, and combined with the target state information (such as blood oxygen saturation, respiratory rate, heart rate, etc.) for evaluation, to obtain the real-time predicted hypoxia degree, which reflects the severity of the hypoxia risk. Finally, based on the hypoxia evaluation function, the real-time predicted hypoxia probability is multiplied by the real-time predicted hypoxia degree to calculate the comprehensive real-time predicted hypoxia coefficient.

[0049] Furthermore, in the device provided by the embodiment of the application, the evaluation and analysis module 14 is further used for:

[0050] The individual characteristic acquisition unit is used to acquire the target individual characteristics of the target user; the hypoxia probability evaluation unit is used to read the hypoxia probability evaluation function and evaluate and analyze the target individual characteristics according to the hypoxia probability evaluation function to obtain the real-time predicted hypoxia probability; the hypoxia degree evaluation unit is used to read the hypoxia degree evaluation function and evaluate and analyze the target state information according to the hypoxia degree evaluation function to obtain the real-time predicted hypoxia degree; the hypoxia coefficient calculation unit is used to perform a product calculation on the real-time predicted hypoxia probability and the real-time predicted hypoxia degree according to the hypoxia evaluation function to obtain the real-time predicted hypoxia coefficient.

[0051] In the embodiment of the present application, the individual characteristic acquisition unit obtains the target individual characteristics of the target user from a preset database, including the target age, target body mass index, and target inflammation index, etc.

[0052] Next, the hypoxia probability evaluation unit reads the preset hypoxia probability evaluation function, inputs the obtained target individual characteristics into the function, performs calculations and analyzes, and finally obtains the real-time predicted hypoxia probability.

[0053] Subsequently, the hypoxia degree evaluation unit reads the hypoxia degree evaluation function to evaluate the severity of the hypoxia risk in the user's current physiological state. Different from the hypoxia probability evaluation, the hypoxia degree evaluation function mainly depends on target state information, such as physiological indicators like blood oxygen saturation, respiratory rate, and heart rate. These data are processed by the calibration module and have high accuracy. Based on these indicators, the real-time predicted hypoxia degree is calculated through the hypoxia degree evaluation function.

[0054] After obtaining the real-time predicted hypoxia probability and the real-time predicted hypoxia degree, the hypoxia coefficient calculation unit performs a product calculation on these two results according to the hypoxia evaluation function to obtain the real-time predicted hypoxia coefficient. Among them, the expression of the hypoxia evaluation function is L = D(x) × κ(x), where L refers to the real-time predicted hypoxia coefficient, κ(x) refers to the real-time predicted hypoxia probability of the target user x, and D(x) refers to the real-time predicted hypoxia degree of the target user x.

[0055] Furthermore, in the device provided by the application embodiment, the expression of the hypoxia probability evaluation function is:

[0056]

[0057] Among them, κ(x) refers to the real-time predicted hypoxia probability of the target user x, μ0, μ1, μ2, …, μ n refers to the model parameters, and g1(x), g2(x), …, g n (x) refers to the target individual characteristics of the target user, and the target individual characteristics include multiple physical characteristics, multiple surgical characteristics, and multiple anesthesia characteristics.

[0058] In the embodiment of the present application, the expression of the hypoxia probability evaluation function is Among them, κ(x) refers to the real-time predicted hypoxia probability of the target user x, which is used to evaluate the possibility of the patient having hypoxia after surgery or in a specific physiological state. μ0, μ1, μ2, …, μ n refers to the model parameters, which are preset by technical experts based on a large amount of clinical data. g1(x), g2(x), …, g n (x) refers to the target individual characteristics of the target user, and the target individual characteristics include multiple physical characteristics, multiple surgical characteristics, and multiple anesthesia characteristics.

[0059] By inputting the obtained target individual characteristics into the hypoxia probability evaluation function for calculation, the real-time predicted hypoxia probability is obtained.

[0060] Further, in the device provided by the application embodiment, the multiple physical characteristics at least include a target age, a target body mass index, and a target inflammation index; the multiple surgical characteristics at least include a target surgical duration, a target surgical type, and a target surgical position; and the multiple anesthesia characteristics at least include a target anesthesia method, a target postoperative infusion volume, and a target complication index.

[0061] In the embodiment of the present application, the multiple physical characteristics at least include a target age, a target body mass index, and a target inflammation index. The target age is directly represented in numerical form, the target body mass index is calculated by dividing the weight of the target user by the square of the height, and the target inflammation index is represented by the numerical value of the C-reactive protein test result. These data are stored in a preset database after being acquired.

[0062] The multiple surgical characteristics at least include a target surgical duration, a target surgical type, and a target surgical position. Among them, the target surgical duration represents the actual duration of the surgery. The target surgical type is represented by numerical coding for different surgical types. For example, 1 represents cardiac surgery, 2 represents laparoscopic surgery, 3 represents neurosurgery, etc. The target surgical position is also represented by numerical coding (such as 0 = supine position, 1 = prone position, 2 = lateral position, etc.). These data are also stored in the preset database.

[0063] The multiple anesthesia characteristics at least include a target anesthesia method, a target postoperative infusion volume, and a target complication index. The target anesthesia method uses numerical classification to represent different anesthesia types, such as 0 = local anesthesia, 1 = general anesthesia, 2 = epidural anesthesia, etc. The target postoperative infusion volume represents the total infusion volume within 24 hours after the surgery. The target complication index is quantitatively scored based on the severity of postoperative complications (such as infection, atelectasis, arrhythmia, etc.). For example, 0 = no complication, 1 = mild complication, 2 = moderate complication, 3 = severe complication, and this index is set by technical experts. Similarly, these data are stored in the preset database.

[0064] Further, in the device provided by the application embodiment, the expression of the hypoxia degree evaluation function is:

[0065]

[0066] Among them, D(x) refers to the real-time predicted hypoxia degree of the target user x, SP(x), RR(x), and HR(x) respectively refer to the real-time blood oxygen saturation, the real-time respiratory rate, and the real-time heart rate in the target status information, and a, b, and c refer to weight coefficients, and a + b + c = 1.

[0067] In the embodiment of the present application, the expression of the hypoxia degree evaluation function is: Among them, D(x) refers to the real-time predicted hypoxia degree of the target user x, SP(x), RR(x), and HR(x) respectively refer to the real-time blood oxygen saturation, real-time respiratory rate, and real-time heart rate in the target state information, a, b, and c refer to weight coefficients, and a + b + c = 1. a, b, and c are preset by technical experts in advance.

[0068] By inputting the obtained target state information into the hypoxia degree evaluation function for calculation, the real-time predicted hypoxia degree is obtained.

[0069] The threshold judgment module 15 is used to judge whether the real-time predicted hypoxia coefficient meets the predetermined hypoxia coefficient threshold.

[0070] In the embodiment of the present application, the threshold judgment module 15 compares the obtained real-time predicted hypoxia coefficient with the predetermined hypoxia coefficient threshold, where the predetermined hypoxia coefficient threshold is preset by technical experts in advance, such as 0.7. This threshold is used to distinguish normal and abnormal hypoxia risk states. When the real-time predicted hypoxia coefficient is greater than the predetermined hypoxia coefficient threshold, it will be judged as not meeting the threshold condition, indicating that the target user may have a hypoxia risk.

[0071] The early warning module 16 is used to issue a hypoxia risk early warning instruction if the real-time predicted hypoxia coefficient does not meet the predetermined hypoxia coefficient threshold.

[0072] In the embodiment of the present application, when the real-time predicted hypoxia coefficient does not meet the predetermined hypoxia coefficient threshold, that is, when the real-time predicted hypoxia coefficient is greater than the predetermined hypoxia coefficient threshold, it is considered that the target user has a hypoxia risk at this time, and the early warning module 16 issues a hypoxia risk early warning instruction.

[0073] The risk intervention module 17 is used to start a predetermined hypoxia intervention plan based on the hypoxia risk early warning instruction to give risk intervention suggestions to the target user.

[0074] Furthermore, in the device provided by the application embodiment, the predetermined hypoxia intervention plan includes an oxygen inhalation concentration adjustment plan and a respiratory support parameter adjustment plan.

[0075] In the embodiment of the present application, the risk intervention module 17 automatically starts the preset hypoxia intervention plan based on the hypoxia risk early warning instruction issued by the early warning module 16, and gives risk intervention suggestions to the target user to reduce the hypoxia risk and ensure the safety of the patient. After receiving the hypoxia risk early warning instruction, according to the severity of the real-time predicted hypoxia coefficient, the corresponding intervention plan is automatically matched. The predetermined hypoxia intervention plan includes an oxygen inhalation concentration adjustment plan and a respiratory support parameter adjustment plan.

[0076] Specifically, in the oxygen inhalation concentration adjustment plan, according to the physiological indicators of the patient's current blood oxygen saturation and respiratory rate, etc., the oxygen inhalation concentration and oxygen flow rate are automatically recommended for adjustment. For example, when the patient's blood oxygen saturation drops to 88%, the oxygen concentration is increased from 28% to 40%. When the effect is not good after the oxygen inhalation adjustment, the respiratory support parameter adjustment plan is further activated to automatically recommend adjusting the parameters of non-invasive or invasive ventilators, such as inspiratory positive pressure, positive end-expiratory pressure, and respiratory rate. For example, for users with a blood oxygen saturation continuously lower than 85%, the PEEP is adjusted to 6 cmH2O and the IPAP is set to 12 cmH2O to increase the alveolar ventilation volume and oxygenation ability.

[0077] Finally, personalized risk intervention suggestions are generated and pushed to the medical staff terminal to ensure that medical staff can take corresponding measures in time. At the same time, the physiological parameters of the target user are continuously monitored, and the intervention strategy is automatically adjusted to form a closed-loop management mechanism.

[0078] In the embodiments of the present application, in summary, the embodiments of the present application at least have the following technical effects:

[0079] In the present application, the real-time status information of the target user is dynamically monitored through the sensor component; the predetermined status support mechanism is retrieved, and the real-time status information is supported and evaluated according to the predetermined status support mechanism to obtain a support evaluation result; the real-time status information is corrected according to the support evaluation result to obtain target status information; a hypoxic evaluation function is introduced to evaluate and analyze the target status information to obtain a real-time predicted hypoxic coefficient; it is judged whether the real-time predicted hypoxic coefficient meets the predetermined hypoxic coefficient threshold; if the real-time predicted hypoxic coefficient does not meet the predetermined hypoxic coefficient threshold, a hypoxic risk warning instruction is issued; based on the hypoxic risk warning instruction, a predetermined hypoxic intervention plan is activated to provide risk intervention suggestions for the target user. The present invention solves the technical problem that it is difficult to capture early hypoxic signals in the prior art in a timely manner, resulting in delayed intervention. By real-time monitoring the user's status through the sensor component, data evaluation and correction are performed based on the predetermined status support mechanism, and the real-time predicted hypoxic coefficient is analyzed in combination with the hypoxic evaluation function. It is judged whether it exceeds the preset threshold. If it exceeds, a hypoxic risk warning instruction is issued, and a hypoxic intervention plan is activated to provide personalized intervention suggestions, achieving the technical effects of accurate identification and rapid intervention of hypoxic risks.

[0080] Embodiment 2, based on the same inventive concept as the hypoxic risk monitoring and warning device for postoperative patients in the foregoing embodiment, as Figure 2 shown, the embodiments of the present application provide a method for monitoring and warning hypoxic risks for postoperative patients, and the method includes:

[0081] Dynamically monitor the real-time status information of the target user through the sensor component; retrieve the predetermined status support mechanism, and perform a support evaluation on the real-time status information according to the predetermined status support mechanism to obtain a support evaluation result; correct the real-time status information according to the support evaluation result to obtain the target status information; introduce a hypoxia evaluation function to evaluate and analyze the target status information to obtain a real-time predicted hypoxia coefficient; determine whether the real-time predicted hypoxia coefficient meets the predetermined hypoxia coefficient threshold; if the real-time predicted hypoxia coefficient does not meet the predetermined hypoxia coefficient threshold, issue a hypoxia risk warning instruction; based on the hypoxia risk warning instruction, start a predetermined hypoxia intervention plan to give risk intervention suggestions to the target user.

[0082] Further, dynamically monitor the real-time status information of the target user through the sensor component, and the method further includes:

[0083] Activate the blood oxygen sensor in the sensor component, wherein the blood oxygen sensor is fixed to the target finger tip of the target user; dynamically monitor the real-time blood oxygen saturation through the blood oxygen sensor; activate the respiratory sensor in the sensor component, and dynamically monitor the real-time respiratory rate and the real-time heart rate through the respiratory sensor; compose the real-time status information based on the real-time blood oxygen saturation, the real-time respiratory rate, and the real-time heart rate.

[0084] Further, retrieve the predetermined status support mechanism, and perform a support evaluation on the real-time status information according to the predetermined status support mechanism to obtain a support evaluation result, and the method further includes:

[0085] Randomly extract any index parameter corresponding to any status index in the real-time status information; retrieve the arbitrary parameter sequence of the arbitrary status index, and analyze the arbitrary parameter sequence to obtain an arbitrary fitting spline; calculate the arbitrary vertical distance from the arbitrary index parameter to the arbitrary fitting spline; analyze the arbitrary vertical distance based on the predetermined status support mechanism to obtain an arbitrary support index; if the arbitrary support index is within the predetermined index threshold, perform a first mark on the arbitrary index parameter, if the arbitrary support index is not within the predetermined index threshold, perform a second mark on the arbitrary index parameter; compose the support evaluation result based on the first mark and the second mark.

[0086] Further, retrieve the arbitrary parameter sequence of the arbitrary status index, and analyze the arbitrary parameter sequence to obtain an arbitrary fitting spline, and the method further includes:

[0087] Construct a first sample parameter group according to any of the parameter sequences, and obtain a first verification parameter group by comparison; generate a first spline curve based on the first sample parameter group; use the first verification parameter group to verify the first spline curve to obtain a first verification result; when the first verification result meets a predetermined verification constraint, use the first spline curve as the arbitrary fitting spline.

[0088] Further, introduce a hypoxia evaluation function to evaluate and analyze the target state information to obtain a real-time predicted hypoxia coefficient. The method further includes:

[0089] Obtain the target individual characteristics of the target user; read the hypoxia probability evaluation function, and evaluate and analyze the target individual characteristics according to the hypoxia probability evaluation function to obtain a real-time predicted hypoxia probability; read the hypoxia degree evaluation function, and evaluate and analyze the target state information according to the hypoxia degree evaluation function to obtain a real-time predicted hypoxia degree; calculate the product of the real-time predicted hypoxia probability and the real-time predicted hypoxia degree according to the hypoxia evaluation function to obtain the real-time predicted hypoxia coefficient.

[0090] Further, the method further includes:

[0091] The expression of the hypoxia probability evaluation function is:

[0092]

[0093] where κ(x) refers to the real-time predicted hypoxia probability of the target user x, μ0, μ1, μ2, …, μ n refers to model parameters, and g1(x), g2(x), …, g n (x) refers to the target individual characteristics of the target user, and the target individual characteristics include multiple physical characteristics, multiple surgical characteristics, and multiple anesthesia characteristics.

[0094] Further, the method further includes:

[0095] The expression of the hypoxia degree evaluation function is:

[0096]

[0097] where D(x) refers to the real-time predicted hypoxia degree of the target user x, SP(x), RR(x), and HR(x) respectively refer to the real-time blood oxygen saturation, the real-time respiratory rate, and the real-time heart rate in the target state information, and a, b, and c refer to weight coefficients, and a + b + c = 1.

[0098] Further, the method further includes:

[0099] The multiple physical characteristics at least include a target age, a target body mass index, and a target inflammation index; the multiple surgical characteristics at least include a target surgical duration, a target surgical type, and a target surgical position; the multiple anesthesia characteristics at least include a target anesthesia method, a target postoperative infusion volume, and a target complication index.

[0100] Further, the method further includes:

[0101] The predetermined hypoxic intervention plan includes an oxygen inhalation concentration adjustment plan and a respiratory support parameter adjustment plan.

[0102] It should be noted that the above order of the embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above description of specific embodiments of this specification is given. The processes depicted in the drawings do not necessarily require the specific order and continuous sequence shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0103] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

[0104] This specification and the drawings are only exemplary descriptions of the present application and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A hypoxemia risk monitoring and early warning device for postoperative patients, characterized in that, Including: A real-time information acquisition module, configured to dynamically monitor the real-time status information of a target user through a sensor component; A support evaluation module, configured to retrieve a predetermined status support mechanism, and perform a support evaluation on the real-time status information according to the predetermined status support mechanism to obtain a support evaluation result; A calibration module, configured to calibrate the real-time status information according to the support evaluation result to obtain target status information; An evaluation and analysis module, configured to introduce a hypoxia evaluation function to evaluate and analyze the target status information to obtain a real-time predicted hypoxia coefficient; A threshold judgment module, configured to judge whether the real-time predicted hypoxia coefficient meets a predetermined hypoxia coefficient threshold; An early warning module, configured to issue a hypoxia risk early warning instruction if the real-time predicted hypoxia coefficient does not meet the predetermined hypoxia coefficient threshold; A risk intervention module, configured to start a predetermined hypoxia intervention plan based on the hypoxia risk early warning instruction to give risk intervention suggestions to the target user.

2. The hypoxemia risk monitoring and early warning device for postoperative patients according to claim 1, wherein The real-time information acquisition module includes: A blood oxygen sensor activation unit, configured to activate the blood oxygen sensor in the sensor component, wherein the blood oxygen sensor is fixed to a target finger tip of the target user; A blood oxygen saturation monitoring unit, configured to dynamically monitor the real-time blood oxygen saturation through the blood oxygen sensor; A respiratory sensor activation unit, configured to activate the respiratory sensor in the sensor component, and dynamically monitor the real-time respiratory rate and the real-time heart rate through the respiratory sensor; A real-time status information generation unit, configured to form the real-time status information based on the real-time blood oxygen saturation, the real-time respiratory rate, and the real-time heart rate.

3. The hypoxemia risk monitoring and early warning device for postoperative patients according to claim 1, wherein The support evaluation module includes: A status index extraction unit, configured to randomly extract any index parameter corresponding to any status index in the real-time status information; A parameter sequence retrieval and analysis unit, configured to retrieve an arbitrary parameter sequence of the arbitrary status index, and analyze the arbitrary parameter sequence to obtain an arbitrary fitting spline; A vertical distance calculation unit, configured to calculate an arbitrary vertical distance from the arbitrary index parameter to the arbitrary fitting spline; A support index analysis unit, configured to analyze the arbitrary vertical distance based on the predetermined status support mechanism to obtain an arbitrary support index; A parameter marking unit, configured to perform a first mark on the arbitrary index parameter if the arbitrary support index is within a predetermined index threshold, and perform a second mark on the arbitrary index parameter if the arbitrary support index is not within the predetermined index threshold; A support evaluation result generation unit, configured to form the support evaluation result based on the first mark and the second mark.

4. The hypoxemia risk monitoring and early warning device for postoperative patients according to claim 3, characterized in that, The support evaluation module includes: A sample parameter group construction unit, configured to construct a first sample parameter group according to the arbitrary parameter sequence, and compare to obtain a first verification parameter group; A spline curve generation unit, configured to generate a first spline curve based on the first sample parameter group; A spline curve verification unit, configured to verify the first spline curve by using the first verification parameter group to obtain a first verification result; A fitting spline determination unit, configured to use the first spline curve as the arbitrary fitting spline when the first verification result meets a predetermined verification constraint.

5. The low-oxygen risk monitoring and early warning device for postoperative patients according to claim 2, characterized in that, The evaluation and analysis module includes: An individual feature acquisition unit, configured to acquire target individual features of the target user; A hypoxemia probability evaluation unit, configured to read a hypoxemia probability evaluation function and perform evaluation and analysis on the target individual features according to the hypoxemia probability evaluation function to obtain a real-time predicted hypoxemia probability; A hypoxemia degree evaluation unit, configured to read a hypoxemia degree evaluation function and perform evaluation and analysis on the target status information according to the hypoxemia degree evaluation function to obtain a real-time predicted hypoxemia degree; A hypoxemia coefficient calculation unit, configured to perform a product calculation on the real-time predicted hypoxemia probability and the real-time predicted hypoxemia degree according to the hypoxemia evaluation function to obtain the real-time predicted hypoxemia coefficient.

6. The hypoxemia risk monitoring and early warning device for postoperative patients according to claim 5, wherein The expression of the hypoxemia probability evaluation function is: Among them, κ(x) refers to the real-time predicted hypoxia probability of the target user x, μ0, μ1, μ2, …, μ n refers to the model parameters, g1(x), g2(x), …, g n (x) refers to the target individual characteristics of the target user, and the target individual characteristics include multiple physical characteristics, multiple surgical characteristics, and multiple anesthetic characteristics.

7. The low-oxygen risk monitoring and early warning device for postoperative patients according to claim 6, characterized in that The expression of the hypoxemia degree evaluation function is: Wherein, D(x) refers to the real-time predicted hypoxemia degree of the target user x, SP(x), RR(x), and HR(x) respectively refer to the real-time blood oxygen saturation, the real-time respiratory rate, and the real-time heart rate in the target status information, and a, b, and c refer to weight coefficients, and a + b + c = 1.

8. The hypoxemia risk monitoring and early warning device for postoperative patients according to claim 6, wherein, The multiple physical characteristics at least include a target age, a target body mass index, and a target inflammation index; The multiple surgical characteristics at least include a target surgical duration, a target surgical type, and a target surgical position; The multiple anesthesia characteristics at least include a target anesthesia method, a target postoperative infusion volume, and a target complication index.

9. The hypoxemia risk monitoring and early warning device for postoperative patients according to claim 1, characterized in that, The predetermined hypoxemia intervention plan includes an oxygen inhalation concentration adjustment plan and a respiratory support parameter adjustment plan.

10. A method for monitoring and warning of hypoxia risk for postoperative patients, characterized in that, The method is executed by a hypoxemia risk monitoring and early warning device for postoperative patients according to any one of claims 1 to 9, and includes: Dynamically monitoring the real-time status information of the target user through a sensor assembly; Invoking a predetermined status support mechanism and performing a support evaluation on the real-time status information according to the predetermined status support mechanism to obtain a support evaluation result; Correcting the real-time status information according to the support evaluation result to obtain target status information; Introducing a hypoxemia evaluation function to perform evaluation and analysis on the target status information to obtain a real-time predicted hypoxemia coefficient; Determining whether the real-time predicted hypoxemia coefficient meets a predetermined hypoxemia coefficient threshold; If the real-time predicted hypoxemia coefficient does not meet the predetermined hypoxemia coefficient threshold, issuing a hypoxemia risk early warning instruction; Based on the hypoxemia risk early warning instruction, starting a predetermined hypoxemia intervention plan to give risk intervention suggestions to the target user.

Images