A labor process energy consumption monitoring and individualized energy requirement management system

Abstract

The application discloses a kind of childbirth process energy consumption monitoring and individualized energy demand management system, including multimodal perception layer, transmission layer, computing layer and application decision layer, multimodal perception layer includes multimodal data acquisition module and non-invasive metabolism monitoring module, and the physiological sign parameter of pregnant woman is collected using wearable sensor array in multimodal data acquisition module, and non-invasive metabolism monitoring module carries out collection to tissue oxygenation data, body fluid distribution and metabolism data and oxygen consumption and carbon dioxide production data, computing layer establishes individual basal metabolic rate model by intelligent computing module, and calculates total energy consumption of labor, and then balanced state is calculated by dynamic energy demand prediction model, and application decision layer generates individualized nutrition supplement scheme according to the data of computing layer, triggers multi-level early warning and pushes clinical decision suggestion, and the application realizes real-time accurate monitoring of childbirth process energy consumption, individualized energy demand prediction and dynamic nutrition supplement.

Description

Technical Field

[0001] This invention belongs to the field of smart medical technology, specifically relating to a monitoring and individualized energy demand management system for energy consumption during childbirth. Background Technology

[0002] In current clinical delivery management scenarios, the monitoring of maternal energy consumption and the management of energy replenishment during labor face many key challenges, seriously affecting delivery quality and maternal and infant safety:

[0003] 1. Insufficient monitoring accuracy: Current clinical practice mainly relies on the experience of medical staff to estimate the energy consumption of postpartum women. There is a lack of accurate monitoring equipment that can capture the dynamic changes in the labor process in real time, which leads to a significant deviation between energy consumption assessment and actual needs, and cannot provide a scientific basis for nutritional intervention.

[0004] 2. Individual differences are ignored: There are significant individual differences in basal metabolic rate, labor progress speed, physical strength reserve level and stress response among different mothers. However, existing technologies mostly adopt standardized assessment models, making it difficult to develop appropriate energy demand plans based on individual characteristics, resulting in frequent problems of insufficient or excessive supplementation.

[0005] 3. Delayed and rigid nutritional supplementation: Traditional nutritional supplementation plans are mostly fixed templates that cannot be dynamically adjusted according to the mother's real-time energy consumption data and the stage of labor. This often results in a disconnect between energy supplementation and actual needs, making it difficult to effectively support the mother's physical exertion during childbirth.

[0006] 4. Difficulty in integrating multi-dimensional data: Energy consumption during childbirth is closely related to multiple physiological parameters such as heart rate, respiration, uterine contraction intensity, body temperature, and activity level. Existing technologies lack the ability to efficiently integrate and analyze multi-source data in real time, making it impossible to comprehensively and systematically assess energy consumption status.

[0007] 5. Lack of risk warning mechanism: When the mother experiences risks such as insufficient energy reserves, excessive consumption or metabolic disorders, the existing system cannot identify and issue warnings in a timely manner, nor can it conduct scientific risk level assessments. This may lead to increased maternal fatigue, prolonged labor, or even affect maternal and infant safety. To address this, we propose an energy consumption monitoring and individualized energy demand management system for the childbirth process. Summary of the Invention

[0008] The purpose of this invention is to provide a monitoring and individualized energy demand management system for energy consumption during childbirth, enabling real-time and accurate monitoring of energy consumption during childbirth, prediction of individualized energy demand, and dynamic nutritional supplementation. At the same time, it constructs a multi-level early warning mechanism to effectively solve problems such as insufficient accuracy, poor individual adaptability, and delayed supplementation in traditional management.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a monitoring and individualized energy demand management system for labor process energy consumption, comprising a multimodal perception layer, a transmission layer, a computing layer, and an application decision layer. The multimodal perception layer includes a multimodal data acquisition module and a non-invasive metabolic monitoring module. The multimodal data acquisition module uses a wearable sensor array to collect various physiological parameters of the pregnant woman. The non-invasive metabolic monitoring module collects tissue oxygenation data, body fluid distribution and metabolic data, and oxygen consumption and carbon dioxide production data. The transmission layer integrates the data collected by the multimodal perception layer through a gateway device and encrypts and transmits the integrated data to the computing layer. The computing layer establishes an individual basal metabolic rate (BMR) model through an intelligent computing module and calculates the total energy consumption during labor. Then, it calculates the balance state through a dynamic energy demand prediction model. The application decision layer generates individualized nutritional supplementation plans based on the data from the computing layer, triggers multi-level early warnings, and pushes clinical decision suggestions. Data visualization, operation feedback, and collaborative management of the labor process are achieved through medical / nursing / partner terminals.

[0010] Preferably, the wearable sensor array includes a heart rate variability sensor (HRV), a respiratory rate and depth sensor, a skin temperature and core body temperature sensor, an electromyography (EMG) sensor, and a motion acceleration sensor.

[0011] Preferably, the non-invasive metabolic monitoring module includes a near-infrared spectroscopy sensor, an impedance analyzer, and a micro gas exchange sensor.

[0012] Preferably, the steps for establishing the individual basal metabolic rate (BMR) model are as follows:

[0013] S1. Input parameters: Mother's height, age, current weight, pre-pregnancy BMI, gestational weight gain, and fetal ultrasound development parameters;

[0014] S2. Substitute the base value into the BMR formula and dynamically adjust it through the individual correction coefficient K to obtain the personalized BMR value;

[0015] S3. Calculate the total energy consumption during labor. .

[0016] Preferably, the BMR formula is:

[0017] ;

[0018] In the formula: W is the mother's current weight, H is the mother's height, A is the mother's age, and K is the individual correction coefficient.

[0019] Preferably, the total energy consumption of the production process The calculation formula is as follows:

[0020] ;

[0021] In the formula: Basic metabolic energy consumption, Energy consumption for uterine contractions Energy consumption for breathing For thermal regulation energy consumption, For activity energy consumption, This is for stress-related energy consumption.

[0022] Preferably, the dynamic energy demand prediction model is as follows:

[0023] Total energy consumption during labor Input is fed into an LSTM neural network model, which outputs a predicted energy demand value for the next 1-2 hours. The formula is as follows:

[0024]

[0025] In the formula: For long short-term memory neural network models, This is a historical energy consumption sequence for the production process. This is the feature vector of labor progress. It is an individual physiological characteristic vector (including personalized BMR, energy reserve rate, and fatigue risk level). This is a feature fusion operation (concatenating multi-dimensional data into an LSTM input vector). The correction factor for the labor stage is 1.0 for the first stage of labor, 1.2 for the second stage, and 0.8 for the third stage, to adapt to the differences in energy consumption intensity at different stages of labor.

[0026] Preferably, the application decision layer calculates cumulative energy consumption based on real-time statistics. With cumulative energy intake Calculate the energy gap Assess energy reserve rate (energy converted from glycogen and fat), and determine fatigue risk level based on energy deficit and reserve rate; optimize nutrient ratio based on predicted energy demand, maternal digestive and absorptive capacity (medical history collection), dietary preferences, and stage of labor progress, and based on predicted energy demand. The infusion system automatically adjusts the infusion rate.

[0027] Preferably, the energy gap The calculation formula is:

[0028] .

[0029] Compared with the prior art, the beneficial effects of the present invention are:

[0030] (1) Through the wearable sensor array and non-invasive metabolic monitoring module of the multimodal perception layer, the data of the parturient physiological signs, tissue oxygenation, body fluid metabolism and gas exchange are collected in a comprehensive manner. Combined with the data cleaning and standardization processing of the transmission layer, the limitations of traditional experience estimation are broken through, and the real-time and accurate quantification of energy consumption during labor is realized, providing reliable data support for subsequent energy management.

[0031] (2) Based on individual parameters such as the mother's height, age, weight, and pre-pregnancy BMI, a personalized basal metabolic rate (BMR) model is established through an intelligent computing module. Combined with the characteristics of labor progress and individual physiological reserves, the predicted energy demand value is dynamically generated. At the same time, taking into full account the mother's digestive and absorption capacity, dietary preferences, and other factors, a differentiated nutritional supplementation plan is formulated, which completely solves the problem of neglecting individual differences in traditional standardized plans.

[0032] (3) Based on the LSTM neural network model, the energy demand for the next 1-2 hours is predicted. The decision-making layer is linked with the precision infusion system and the postpartum terminal to realize the automatic adjustment of the intravenous infusion rate and the intelligent reminder of oral nutritional supplementation. Through real-time calculation of energy deficit and assessment of reserve rate, the nutrient ratio (precise proportion of carbohydrates, protein and fat) is dynamically optimized to ensure that energy supplementation is dynamically matched with the energy consumption of labor and effectively maintain the postpartum woman's physical strength.

[0033] (4) Establish an energy balance monitoring mechanism and a multi-level early warning system. Through comprehensive assessment of energy deficit, reserve rate and metabolic indicators, accurately determine the fatigue risk level, trigger yellow, orange and red three-level early warnings and push targeted decision-making suggestions. At the same time, through the visual interaction between medical staff terminals and maternal terminals, realize the real-time sharing of physiological data, energy consumption status, nutrition plan and early warning information, help medical staff and maternal patients to jointly manage the labor process, and reduce problems such as prolonged labor and maternal and infant risks caused by insufficient energy.

[0034] (5) Integrating physiological parameters, energy consumption data, nutritional program implementation status, and early warning records throughout the entire labor process forms a traceable and complete data chain. This not only provides closed-loop support for individual childbirth management but also accumulates rich data for clinical research, promoting the transformation of labor energy management from experience-based to scientific and intelligent. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the architecture of the entire system of the present invention;

[0036] Figure 2 This is a schematic diagram of the architecture of the multimodal sensing layer in this invention;

[0037] Figure 3 This is a schematic diagram of the architecture of the mid-computation layer;

[0038] Figure 4This is a schematic diagram of the architecture of the intelligent nutrition management submodule in this invention;

[0039] Figure 5 This is a schematic diagram of the architecture of the early warning and decision support submodule in this invention. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] Please see Figure 1 This invention provides a technical solution: a monitoring and individualized energy demand management system for labor process energy consumption, comprising a multimodal perception layer, a transmission layer, a computing layer, and an application decision layer. The multimodal perception layer includes a multimodal data acquisition module and a non-invasive metabolic monitoring module. The multimodal data acquisition module uses a wearable sensor array to collect various physiological parameters of the pregnant woman. The non-invasive metabolic monitoring module collects tissue oxygenation data, body fluid distribution and metabolic data, and oxygen consumption and carbon dioxide production data. The transmission layer integrates the data collected by the multimodal perception layer through a gateway device and encrypts and transmits the integrated data to the computing layer. The computing layer establishes an individual basal metabolic rate (BMR) model through an intelligent computing module and calculates the total energy consumption during labor. Then, it calculates the balance state through a dynamic energy demand prediction model. The application decision layer generates individualized nutritional supplementation plans based on the data from the computing layer, triggers multi-level early warnings, and pushes clinical decision suggestions. Data visualization, operation feedback, and collaborative management of the labor process are achieved through medical / nursing / partner terminals.

[0042] Please see Figure 2 The wearable sensor array includes a heart rate variability sensor (HRV), a respiratory rate and depth sensor, a skin temperature and core body temperature sensor, an electromyography (EMG) sensor, and a motion acceleration sensor.

[0043] Heart rate variability sensor (HRV): monitors heart rate fluctuations, extracts the LF / HF ratio, and reflects stress state and autonomic nervous activity;

[0044] Respiratory rate and depth sensor: Real-time acquisition of respiratory rate and tidal volume, and calculation of respiratory work;

[0045] Skin temperature and core body temperature sensors: Simultaneously monitor body surface temperature and core body temperature (accuracy ±0.1℃) to provide data for thermal regulation energy consumption calculation;

[0046] Electromyography (EMG) sensor: Collects electromyographic signals from muscle groups related to uterine contractions (such as the rectus abdominis and pelvic floor muscles) to identify the intensity, frequency, and duration of uterine contractions;

[0047] Motion accelerometer: Captures changes in the mother's body position (lying still, turning over, walking) and activity intensity, and quantifies the amount of activity through the root mean square value of acceleration.

[0048] Please see Figure 2 The non-invasive metabolic monitoring module includes a near-infrared spectroscopy sensor, an impedance analyzer, and a miniature gas exchange sensor. The module independently collects metabolic data and transmits it to the transmission layer via Wi-Fi 6, aligning it with wearable sensor data based on timestamps.

[0049] Near-infrared spectroscopy sensor: penetrates tissue to monitor blood oxygen saturation and tissue oxygenation status, indirectly reflecting metabolic level;

[0050] Impedance analyzer: Measures human body impedance through electrodes on the limbs to assess fluid distribution and muscle mass, and assists in judging metabolic status;

[0051] Miniature gas exchange sensor: monitors oxygen intake (VO2) and carbon dioxide expulsion (VCO2) to indirectly calculate energy consumption.

[0052] Transport Layer: Data Transmission and Preprocessing Module

[0053] Gateway device

[0054] Data aggregation: Receives data streams from various sensors and monitoring modules in the perception layer to integrate multi-source data;

[0055] Data cleaning: Remove outliers (such as data that exceeds the physiologically reasonable range), fill in missing values ​​(using linear interpolation), and smooth the data using the moving average method;

[0056] Protocol conversion: Convert the proprietary protocol of the perception layer to the HL7FHIR standard protocol to ensure compatibility with the computing layer;

[0057] Resume interrupted download: When the network is interrupted, local cached data (maximum cache capacity 10GB) will be automatically resumed after the network is restored.

[0058] Interaction logic: Receive data from the perception layer, preprocess it, standardize and encapsulate it, and transmit it to the edge nodes and cloud servers of the computing layer.

[0059] Please see Figure 3 Computation layer: Intelligent computing module

[0060] 1. Individual Basal Metabolic Rate (BMR) Calculation Submodule

[0061] Parameter entry: Receives clinical data such as the mother's pre-pregnancy BMI, gestational weight gain, and fetal ultrasound parameters;

[0062] Establishment of individual basal metabolic rate (BMR) model: Substitute into the basic BMR formula, dynamically adjust through individual correction coefficient K (integrating pre-pregnancy BMI, weight gain, and fetal development) to output personalized BMR value.

[0063] Interaction logic: Receive preprocessed weight data and clinically entered parameters from the transmission layer, calculate personalized BMR and push it to the total energy consumption calculation submodule during labor;

[0064] The steps for establishing an individual basal metabolic rate (BMR) model are as follows:

[0065] Step 1: Input parameters: Mother's height, age, current weight, pre-pregnancy BMI, gestational weight gain, and fetal ultrasound development parameters;

[0066] Step 2: Substitute the values ​​into the BMR formula to calculate the base value, and dynamically adjust it using the individual correction coefficient K to obtain the personalized BMR value;

[0067] Step 3: Calculate the total energy consumption during labor. .

[0068] The BMR formula is:

[0069] ;

[0070] In the formula: W is the mother's current weight, H is the mother's height, A is the mother's age, and K is the individual correction coefficient.

[0071] 2. Total Energy Consumption Calculation Submodule

[0072] Itemized energy consumption calculation: Extract EMG signals, uterine contraction data, respiratory data, etc. from the transmission layer, substitute them into the itemized energy consumption formula, and calculate five types of energy consumption, including uterine contraction energy consumption and respiratory energy consumption.

[0073] Total energy consumption summation: Integrating basal metabolic energy consumption and individual energy consumption, the total energy consumption during labor is obtained. ).

[0074] Interaction logic: Receives BMR data and preprocessed physiological parameters from the sensory layer, performs component calculations, sums the results, and outputs the output. The data is pushed to the dynamic demand forecasting submodule and the energy balance monitoring submodule.

[0075] Total energy consumption during labor The calculation formula is as follows:

[0076] ;

[0077] In the formula: Basic metabolic energy consumption, Energy consumption for uterine contractions Energy consumption for breathing For thermal regulation energy consumption, For activity energy consumption, This is for stress-related energy consumption.

[0078] Basal metabolic energy consumption :

[0079] ;

[0080] Where: T: cumulative labor time (unit: h).

[0081] Energy consumption of uterine contractions :

[0082] ;

[0083] In the formula: The average intensity of the electromyographic signal (μV) is given by F, the frequency of uterine contractions (times / minute) is given by D, and the duration of a single uterine contraction is given by D (seconds). : Contraction energy consumption coefficient (fixed value 0.012kcal / (μV・time・second));

[0084] Respiratory energy consumption :

[0085] ;

[0086] In the formula: Respiratory rate (breaths / minute). Tidal volume (liters, derived from respiratory depth sensor). Work of breathing coefficient (fixed value 0.03 kcal / (breaths·L)).

[0087] Thermal regulation energy consumption :

[0088] ;

[0089] In the formula: The mother's core body temperature (unit: °C, monitored in real time by a core body temperature sensor). The optimal physiological body temperature for childbirth (a fixed constant, taken as 37.0℃) is used. It is the absolute difference between core body temperature and optimal body temperature (reflecting the degree of deviation from body temperature). The body surface area of ​​the mother (unit: m²) 2 Derived from height H (cm) and weight W (kg):

[0090] , Thermal conductivity (a constant, valued at 1.8 kcal / (°C·m)) 2 •h), adapting to the heat exchange characteristics of the childbirth scenario), T: the duration of the corresponding body temperature state (unit: h, i.e. the cumulative duration of this body temperature range during labor).

[0091] Activity energy consumption :

[0092] ;

[0093] In the formula: The root mean square value of the acceleration (unit: m / s²) 2 The data is collected in real time by a motion acceleration sensor to reflect the intensity of activity (approximately 0.05 for resting, 0.3 for turning over, and 1.2 for walking). The activity metabolic equivalent (dimensionless, values ​​are assigned based on activity type: 1.0 for resting, 1.5 for turning over, 3.0 for walking around the bedside, and 4.0 for voluntary exertion). : The cumulative duration of the corresponding activity (unit: hours). Activity energy consumption correction factor (fixed constant, valued at 0.85 kcal / (kg・m / s)) 2 •h•MET), adapted to the energy conversion efficiency of human activities in childbirth scenarios.

[0094] Stress energy consumption :

[0095] ;

[0096] In the formula: The ratio of low-frequency to high-frequency heart rate variability (reflecting stress level). Stress hormone metabolic coefficient (fixed value 5.2 kcal).

[0097] 3. Dynamic Demand Prediction Submodule (LSTM Neural Network)

[0098] Data input: Receive historical energy consumption data (first 30 minutes), labor progress data (cervical dilation rate, fetal head descent rate), and individual physiological reserve parameters;

[0099] Forecast output: Based on time series analysis, predict energy demand for the next 1-2 hours. .

[0100] Interaction logic: Receive total energy consumption data and clinical labor data, perform model calculations, output predicted values, and push them to the nutrition plan generation submodule.

[0101] 4. Energy Balance Monitoring Submodule

[0102] Core functions:

[0103] Income and expenditure statistics: Real-time statistics of cumulative energy consumption With cumulative energy intake (Oral / intravenous supplementation data from application layer feedback);

[0104] Gap and Reserve Assessment: Calculating the Energy Gap Assess energy reserve rate (energy conversion from glycogen and fat) and determine fatigue risk level.

[0105] Interaction logic: Receive total energy consumption data and intake data fed back from the application decision layer, calculate the balance state, and push it to the early warning submodule and the nutrition plan generation submodule.

[0106] The dynamic energy demand forecasting model is as follows:

[0107] Total energy consumption during labor Input is fed into an LSTM neural network model, which outputs a predicted energy demand value for the next 1-2 hours. The formula is as follows:

[0108] ;

[0109] In the formula: For long short-term memory neural network models, This is a historical energy consumption sequence for the production process. This is the feature vector of labor progress. It is an individual physiological characteristic vector (including personalized BMR, energy reserve rate, and fatigue risk level). This is a feature fusion operation (concatenating multi-dimensional data into an LSTM input vector). The correction factor for the labor stage is 1.0 for the first stage of labor, 1.2 for the second stage, and 0.8 for the third stage, to adapt to the differences in energy consumption intensity at different stages of labor.

[0110] The application decision-making level calculates cumulative energy consumption based on real-time statistics. With cumulative energy intake Calculate the energy gap Assess energy reserve rate (energy converted from glycogen and fat), and determine fatigue risk level based on energy deficit and reserve rate; optimize nutrient ratio based on predicted energy demand, maternal digestive and absorptive capacity (medical history collection), dietary preferences, and stage of labor progress, and based on predicted energy demand. The infusion system automatically adjusts the infusion rate as follows:

[0111] Please see Figure 4 Intelligent nutrition management submodule

[0112] Based on dynamic energy demand forecasts Based on fatigue risk level, postpartum dietary preferences and digestive capacity, optimize nutrient ratio (carbohydrate 50%-60%, protein 15%-20%, fat 20%-25%), and provide recommendations on oral nutritional supplements (types and dosages) and intravenous infusion rates.

[0113] Intake control: In conjunction with the precision infusion system, the glucose infusion rate is automatically adjusted (2-10 mg / kg·min); oral supplementation reminders are pushed through the postpartum terminal, and the actual intake is recorded as follows: When the fatigue risk level is low or medium, oral supplementation can be carried out according to the nutrient ratio; when the fatigue risk level is high, the precision infusion system is linked to automatically adjust the glucose infusion rate and use the infusion system for automatic supplementation.

[0114] Interaction logic: Receive demand forecasts and balance monitoring results, generate personalized plans, push them to medical staff terminals, maternal terminals and precision infusion systems, receive execution feedback (actual intake), and send back to the calculation layer to update the balance status.

[0115] Please see Figure 5 Early warning and decision support submodule

[0116] Energy gap The calculation formula is:

[0117] ;

[0118] ;

[0119] In the formula: The rate of intravenous glucose infusion (unit: g / h, monitored in real time by the precision infusion system). The cumulative time of intravenous infusion (unit: h) is given, and 4 is the glucose energy conversion coefficient (a fixed value; 1g of glucose can provide about 4kcal of energy when completely metabolized, which meets the clinical nutritional standards). This refers to the types and quantities of oral nutritional supplements (such as low-sugar electrolyte drinks, high-protein gels, etc.). For the first Cumulative intake of oral supplements (unit: g or mL; liquid supplements are converted to mass by "volume × density", with a default density of 1 g / mL). For the first The energy density of oral supplements (unit: kcal / g, such as electrolyte drinks about 0.15 kcal / g and high-protein gel about 4.0 kcal / g, determined by the product's nutrition facts label).

[0120] Fatigue risk level assessment:

[0121] ;

[0122] Energy reserve rate = (glycogen reserves + energy converted from fat) / expected remaining production energy consumption × 100%

[0123] Glycogen reserves: Pre-pregnancy weight × 1.8 kcal / kg (default); Fat conversion energy: (Current weight - Pre-pregnancy weight) × 7 kcal / kg (default)

[0124] Multi-level early warning: based on preset thresholds (Level 1: ≥ Threshold 80%; Level 2: Energy reserve rate ≤ 30%; Level 3: >500kcal or metabolic disorder), triggering a yellow / orange / red level three warning (audio-visual prompts + push notifications from the medical staff's APP);

[0125] Intelligent decision-making: Based on the warning level, action suggestions are output (e.g., increase the monitoring frequency for a level 1 warning, and adjust the nutrition plan and assess midwifery intervention for a level 3 warning).

[0126] It receives energy balance monitoring results and metabolic index data, determines the warning level, pushes warning signals and decision suggestions to medical staff terminals, receives feedback from medical staff on implementation, and optimizes subsequent decisions.

[0127] Terminal Interaction Submodule

[0128] Healthcare terminals (computer / tablet APP):

[0129] It displays physiological parameters, energy consumption data, energy balance status, early warning information, and nutrition plans in real time, and supports manual adjustment of plans and recording of labor progress;

[0130] Maternal terminal (wristband / mobile APP):

[0131] Receive reminders for oral nutritional supplements, display fatigue risk level (only low / medium / high is shown, not professional data), and provide breathing relaxation guidance.

[0132] Interaction logic: Receive data from the computing layer / application decision layer for visualization, and receive user operation feedback and send it back to the corresponding module.

[0133] Data storage and traceability submodule

[0134] It stores physiological parameters, energy consumption data, nutritional plans, early warning records, and maternal and infant outcome data throughout the entire labor process, and supports querying and tracing by maternal ID and time range, providing data support for clinical research.

[0135] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A monitoring and individualized energy demand management system for energy consumption during childbirth, characterized in that, The system comprises a multimodal perception layer, a transmission layer, a computation layer, and an application decision layer. The multimodal perception layer includes a multimodal data acquisition module and a non-invasive metabolic monitoring module. The multimodal data acquisition module uses a wearable sensor array to collect various physiological parameters of the pregnant woman. The non-invasive metabolic monitoring module collects data on tissue oxygenation, body fluid distribution and metabolism, and oxygen consumption and carbon dioxide production. The transmission layer integrates the data collected by the multimodal perception layer through a gateway device and encrypts the integrated data before transmitting it to the computation layer. The computation layer establishes an individual basal metabolic rate (BMR) model through an intelligent computing module, calculates the total energy consumption during labor, and then calculates the balance state through a dynamic energy demand prediction model. The application decision layer generates individualized nutritional supplementation plans based on the data from the computation layer, triggers multi-level warnings, and pushes clinical decision suggestions. Data visualization, operational feedback, and collaborative management of the labor process are achieved through medical / nursing / maternal terminals.

2. The energy consumption monitoring and individualized energy demand management system for childbirth as described in claim 1, characterized in that, The wearable sensor array includes a heart rate variability sensor (HRV), a respiratory rate and depth sensor, a skin temperature and core body temperature sensor, an electromyography (EMG) sensor, and a motion acceleration sensor.

3. The energy consumption monitoring and individualized energy demand management system for childbirth as described in claim 1, characterized in that, The non-invasive metabolic monitoring module includes a near-infrared spectroscopy sensor, an impedance analyzer, and a miniature gas exchange sensor.

4. The energy consumption monitoring and individualized energy demand management system for childbirth as described in claim 1, characterized in that, The steps for establishing the individual basal metabolic rate (BMR) model are as follows: S1. Input parameters: Mother's height, age, current weight, pre-pregnancy BMI, gestational weight gain, and fetal ultrasound development parameters; S2. Substitute the base value into the BMR formula and dynamically adjust it through the individual correction coefficient K to obtain the personalized BMR value; S3. Calculate the total energy consumption during labor. .

5. The energy consumption monitoring and individualized energy demand management system for childbirth as described in claim 4, characterized in that, The BMR formula is: ； In the formula: W is the mother's current weight, H is the mother's height, A is the mother's age, and K is the individual correction coefficient.

6. The energy consumption monitoring and individualized energy demand management system for childbirth as described in claim 4, characterized in that, Total energy consumption during labor The calculation formula is as follows: ； In the formula: Basic metabolic energy consumption, Energy consumption for uterine contractions Energy consumption for breathing For thermal regulation energy consumption, For activity energy consumption, This is for stress-related energy consumption.

7. The energy consumption monitoring and individualized energy demand management system for childbirth as described in claim 1, characterized in that, The dynamic energy demand prediction model is specifically as follows: Total energy consumption during labor Input is fed into an LSTM neural network model, which outputs a predicted energy demand value for the next 1-2 hours. The formula is as follows: In the formula: For long short-term memory neural network models, This is a historical energy consumption sequence for the production process. This is the feature vector of labor progress. It is an individual physiological characteristic vector (including personalized BMR, energy reserve rate, and fatigue risk level). This is a feature fusion operation (concatenating multi-dimensional data into an LSTM input vector). The correction factor for the labor stage is 1.0 for the first stage of labor, 1.2 for the second stage, and 0.8 for the third stage, to adapt to the differences in energy consumption intensity at different stages of labor.

8. The energy consumption monitoring and individualized energy demand management system for childbirth as described in claim 1, characterized in that, The application decision-making layer calculates cumulative energy consumption in real time. With cumulative energy intake Calculate the energy gap Assess the energy reserve rate and determine the fatigue risk level based on the energy gap and reserve rate; Based on predicted energy requirements, maternal digestive and absorptive capacity (medical history), dietary preferences, and stage of labor, the nutrient ratio was optimized, and the predicted energy requirements were used as a basis for further optimization. The infusion system automatically adjusts the infusion rate.

9. A monitoring and individualized energy demand management system for labor process energy consumption according to claim 9, characterized in that, The energy gap The calculation formula is: 。

Images