Bladder function training guide and control system and method based on multi-modal sensor

Abstract

The invention relates to the technical field of medical instruments, and particularly discloses a bladder function training guide and control system and method based on a multi-modal sensor, and the system comprises an integrated sensing module which monitors the body surface pressure of a bladder area through a pressure sensor array, obtains the geometric parameters of the bladder in combination with an ultrasonic probe, and transmits the parameters to a central processing module; and meanwhile, conductivity data of liquid in the bladder is acquired by using an impedance tomography technology. The bladder parameter inference module is fused with multi-source data to calculate real-time bladder wall tension and limit capacity to construct a personalized tension-capacity model. The urination demand evaluation module calculates a urination demand value based on the real-time tension, the model parameters, and the limit capacity. And the training plan optimization module is used for dynamically adjusting the initial training scheme according to the urination demand of the day. And the training guide and control module is used for executing the optimization plan and monitoring in real time, and triggering an acousto-optic vibration reminding or urethral catheterization device when a required value reaches a threshold value. The system realizes intelligent and personalized guidance and control of bladder function training through multi-modal data fusion and closed-loop control.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and particularly to a bladder function training and guiding control system and method based on multi-modal sensors. Background Art

[0002] In the field of the treatment and rehabilitation of urinary system diseases, bladder dysfunction is a common problem, seriously affecting the quality of life of patients. Bladder function training is crucial for improving bladder function and enhancing the patient's ability to urinate independently. However, traditional bladder function training often lacks accurate monitoring and scientific guidance, making it difficult to achieve personalized and effective training. The key to bladder function training lies in understanding the bladder state in real time and accurately, so as to formulate a reasonable training plan. In the past, clinicians mainly relied on the patient's subjective feelings and limited examination methods to evaluate bladder function. However, there are significant individual differences and inaccuracies in the patient's subjective descriptions, and although conventional examinations such as ultrasound can obtain some bladder information, the data of a single modality are difficult to comprehensively reflect the complex physiological state of the bladder. With the rapid development of sensor technology, the integrated application of multi-modal sensors in the medical field has become a trend. By combining various types of sensors, richer and more comprehensive physiological data can be obtained from different dimensions, providing strong support for precision medicine. The bladder function training and guiding control system and method based on multi-modal sensors have emerged as the times require. It uses a variety of sensors such as pressure sensors, ultrasonic probes, and electrical impedance tomography to obtain bladder-related information in all directions, which is of great significance for improving the scientific nature, accuracy, and personalization of bladder function training, and is expected to be widely used in the rehabilitation treatment of urinary system diseases, promoting the development of this field towards the direction of intelligence and precision.

[0003] There are many deficiencies in the existing technology in terms of bladder function training. In terms of bladder state monitoring, there is a lack of multi-modal data fusion monitoring means, resulting in an incomplete understanding of the bladder state, and thus the patient's urination needs cannot be evaluated and urination reminders cannot be given in a timely and accurate manner. Moreover, the common general training plans do not consider the training effect deviation caused by the patient's different physiological characteristics.

[0004] Therefore, the present invention proposes a bladder function training and guiding control system and method based on multi-modal sensors. Summary of the Invention

[0005] The present invention provides a bladder function training and guiding control system and method based on multi-modal sensors for...

[0006] The present invention provides a bladder function training and guiding control system based on multi-modal sensors, including: An integrated sensing module for detecting the body surface pressure in the patient's bladder area based on a pressure sensor array. Meanwhile, it measures the real-time geometric parameters of the bladder based on an ultrasound probe and obtains quantitative data on the liquid conductivity in the patient's bladder based on electrical impedance tomography; A bladder parameter inference module for calculating the real-time bladder wall tension and the number of limit capacity units based on the quantitative data of the liquid conductivity in the patient's bladder, the body surface pressure in the patient's bladder area, and the real-time geometric parameters of the bladder, and building a bladder wall tension - bladder capacity model for the patient; A urination demand assessment module for evaluating the real-time urination demand value of the patient based on the patient's real-time bladder wall tension, the bladder wall tension - bladder capacity model, and the number of limit capacity units; A training plan optimization module for optimizing the initial bladder function training plan based on the real-time urination demand value of the patient obtained on the same day to obtain an optimized bladder function training plan; A training guidance and control module for guiding and controlling the patient's bladder function training based on the optimized bladder function training plan. Meanwhile, when the urination demand value reaches the urination demand threshold in the optimized bladder function training plan, it triggers an audible and visual vibration to remind the patient to urinate or activates a urinary catheterization device.

[0007] Optionally, it further includes: A drinking reminder module for dynamically calculating the current water requirement based on the patient's body weight, activity level, environmental temperature and humidity, and the real-time bladder state model, and sending a real-time drinking reminder to the patient based on the current water requirement.

[0008] Optionally, the integrated sensing module includes: A pressure sensing sub-module for continuously detecting the real-time body surface pressure distribution value in the patient's bladder area based on a pressure sensor array as the body surface pressure; An ultrasound probe measurement sub-module for continuously emitting ultrasonic signals to the patient's bladder area based on an ultrasound probe and receiving the ultrasonic echo signals reflected from the bladder area, and determining the real-time geometric parameters of the bladder based on the ultrasonic echo signals received by the ultrasound probe in real time; A bladder impedance detection sub-module for applying a weak alternating current to multiple electrodes placed on the body surface of the patient's bladder area and measuring the potential difference on the body surface of the patient's bladder area to construct an electrical impedance tomography of the patient's bladder, and determining the quantitative data of the liquid conductivity in the patient's bladder based on the continuously obtained electrical impedance tomography.

[0009] Optionally, the bladder parameter inference module includes: A residual urine volume statistics sub-module for determining the number of residual urine volume units of the patient based on the quantitative data of the liquid conductivity in the patient's bladder; A bladder wall tension calculation sub-module for calculating the real-time bladder wall tension based on the body surface pressure in the patient's bladder area and the real-time geometric parameters of the bladder; A correlation analysis sub-module, which is used to perform a correlation analysis on the number of residual urine units and the real-time bladder wall tension of a patient within a preset period, introduce the instantaneous elastic modulus and delayed elastic modulus of the bladder, and use the three-element standard linear solid model to describe, so as to obtain the bladder wall tension-bladder capacity model of the patient; An ultimate capacity calculation sub-module, which is used to calculate the number of ultimate capacity units of the patient's bladder based on the change data of the conductivity of the liquid in the bladder, urination events, and the bladder wall tension-bladder capacity model of the patient within a long period.

[0010] Optionally, the bladder wall tension calculation sub-module includes: A bladder mechanical model and function establishment unit, which is used to obtain the range of body surface pressure in the patient's bladder area and the range of geometric limit parameters of the bladder, and establish the bladder elastic mechanics model and pressure transfer function of the patient based on the range of body surface pressure in the patient's bladder area and the range of geometric limit parameters of the bladder; A bladder pressure distribution conversion unit, which is used to convert the body surface pressure into the bladder internal pressure distribution based on the patient's bladder pressure transfer function; A bladder wall tension calculation unit, which is used to calculate the real-time bladder wall tension based on the patient's bladder elastic mechanics model and the geometric real-time parameters of the bladder.

[0011] Optionally, the ultimate capacity calculation sub-module includes: A non-linear model description unit, which is used to describe the non-linear relationship between the conductivity of the liquid in the bladder and the bladder volume during the bladder filling process based on the exponential decay model, so as to obtain the original conductivity-capacity dynamic model of the liquid in the bladder; A urination event recognition unit, which is used to identify all the conductivity sudden drop characteristics in the change data of the conductivity of the liquid in the bladder within a long period, calculate the occurrence probability of the urination event of each conductivity sudden drop characteristic based on the conductivity change data before and after all the conductivity sudden drop characteristics and the hidden Markov model, and screen out all the urination events from all the conductivity sudden drop characteristics based on the occurrence probability of the urination event; A model personalization description unit, which is used to optimize the volume decay coefficient and background noise compensation term in the original conductivity-capacity dynamic model based on the conductivity change data before and after all the urination events and the Markov chain Monte Carlo method, so as to obtain the conductivity-capacity dynamic model of the liquid in the patient's bladder; An ultimate capacity unit calculation unit, which is used to calculate the number of ultimate capacity units of the patient's bladder based on the conductivity-capacity dynamic model of the liquid in the patient's bladder, the ultimate value of the conductivity of the liquid in the patient's bladder, and the maximum value of the bladder wall tension of the patient.

[0012] Optionally, the method for calculating the number of limit capacity units by the limit capacity unit calculation unit based on the dynamic model of the conductivity-capacity of the liquid in the patient's bladder, the limit value of the conductivity of the liquid in the patient's bladder, and the maximum wall tension value of the patient's bladder includes: ; Wherein, is the number of limit capacity units of the patient's bladder, is the maximum wall tension value of the patient's bladder, is the maximum wall tension value of the bladder of a standard adult, is the Poisson's ratio of the bladder tissue, is the natural logarithm function, is the approximate radius of the patient's bladder in the non-filled state, is the wall thickness of the patient's bladder in the non-filled state, is the maximum conductivity that the liquid in the patient's bladder can reach, is the background noise compensation term in the conductivity-capacity dynamic model of the liquid in the patient's bladder, is the conductivity of the liquid in the patient's bladder in the non-filled state, is the volume attenuation coefficient in the conductivity-capacity dynamic model of the liquid in the patient's bladder.

[0013] Optionally, the urination demand assessment module includes: A real-time bladder capacity calculation sub-module, configured to determine the number of real-time bladder capacity units of the patient based on the real-time bladder wall tension of the patient and the bladder wall tension-bladder capacity model; A urination demand value quantification sub-module, configured to calculate the real-time urination demand value of the patient based on the number of real-time bladder capacity units of the patient, the number of limit capacity units, the real-time bladder wall tension of the patient, and the maximum wall tension value of the bladder.

[0014] Optionally, the training plan optimization module includes: An initial training plan generation sub-module, configured to generate an initial bladder function training plan for the patient based on the physiological sign data of the patient; A training plan optimization sub-module, configured to optimize the adjustable parameters in the initial bladder function training plan by reinforcement learning on the real-time urination demand value of the patient obtained on the current day and the occurrence times of all urination events of the patient, to obtain an optimized bladder function training plan.

[0015] The present invention provides a method for guiding and controlling bladder function based on a multi-modal sensor, including: Step 1: Detect the surface pressure of the patient's bladder area based on a pressure sensor array, and at the same time, measure the geometric real-time parameters of the bladder based on an ultrasonic probe, and obtain quantitative data on the conductivity of the liquid in the patient's bladder based on electrical impedance tomography; Step 2: Based on the quantitative data of the liquid conductivity in the patient's bladder, the body surface pressure in the patient's bladder area, and the real-time geometric parameters of the bladder, calculate the real-time bladder wall tension and the number of limit capacity units, and establish a bladder wall tension-bladder capacity model for the patient; Step 3: Based on the patient's real-time bladder wall tension, the bladder wall tension-bladder capacity model, and the number of limit capacity units, evaluate the patient's real-time urine excretion demand value; Step 4: Optimize the initial bladder function training plan based on the patient's real-time urine excretion demand value obtained on the same day to obtain an optimized bladder function training plan; Step 5: Conduct bladder function training control for the patient based on the optimized bladder function training plan. At the same time, when the urine excretion demand value reaches the urine excretion demand threshold in the optimized bladder function training plan, trigger an audible and visual vibration to remind the patient to urinate or activate the urinary catheterization device.

[0016] The beneficial effects of the present invention compared with the prior art are as follows: Other features and advantages of the present invention will be described in the following specification, and part of them will become obvious from the specification or be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in this application document.

[0017] The technical solutions of the present invention will be further described in detail below through the accompanying drawings and embodiments. Description of the Drawings

[0018] The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, but do not constitute a limitation to the present invention. In the accompanying drawings: Figure 1 It is a module diagram of a bladder function training control system based on multi-modal sensors in an embodiment of the present invention; Figure 2 It is a logic flowchart of a bladder function training control method based on multi-modal sensors in an embodiment of the present invention. Detailed Embodiments

[0019] The following describes the preferred embodiments of the present invention with reference to the accompanying drawings. It should be understood that the preferred embodiments described here are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0020] Provide an implementation manner of a bladder function training control system based on multi-modal sensors, as Figure 1 shown, including: An integrated sensing module is used to detect the body surface pressure in the patient's bladder area based on a pressure sensor array. At the same time, it measures the real-time geometric parameters of the bladder based on an ultrasonic probe and obtains quantitative data on the liquid conductivity in the patient's bladder based on electrical impedance tomography. Among them, the pressure sensor array is distributed on the body surface of the patient's bladder area to detect the change of body surface pressure in this area in real time. For example, when the bladder is full, the bladder wall generates pressure on the body surface, and the pressure sensor array can accurately sense this pressure change and convert it into an electrical signal or a digital signal.

[0021] At the same time, the ultrasonic probe uses the principle of ultrasonic reflection, just like performing real-time fluoroscopy on the bladder, to measure the real-time geometric parameters of the bladder, such as the length, width, height, etc. of the bladder, so as to reflect the morphological changes of the bladder.

[0022] In addition, the electrical impedance tomography technology applies a weak current to the human body and obtains quantitative data on the liquid conductivity in the patient's bladder according to the different conductivity characteristics of different tissues and liquids, so as to understand the electrical characteristics of the urine in the bladder. Combining these data provides comprehensive bladder status information for the subsequent module.

[0023] A bladder parameter inference module is used to calculate the real-time bladder wall tension and the number of limit volume units based on the quantitative data of the liquid conductivity in the patient's bladder, the body surface pressure in the patient's bladder area, and the real-time geometric parameters of the bladder, and build a bladder wall tension-bladder volume model for the patient, including: Based on the quantitative data of the liquid conductivity in the bladder, combined with specific algorithms and empirical formulas, the number of residual urine volume units of the patient can be determined to understand the approximate situation of the remaining urine in the bladder.

[0024] Then, based on the body surface pressure in the bladder area and the real-time geometric parameters of the bladder, using mechanical principles and mathematical models, the real-time bladder wall tension is calculated, and this value reflects the magnitude of the tension borne by the bladder wall in the current state.

[0025] Perform a correlation analysis on the number of residual urine volume units and the real-time bladder wall tension of the patient within a preset period (such as one day or one week). At the same time, introduce the instantaneous elastic modulus and the delayed elastic modulus of the bladder, and use the three-element standard linear solid model to describe the relationship between them, so as to build a personalized bladder wall tension-bladder volume model for the patient, including: ; In the formula, is the real-time bladder wall tension (Pa), is the bladder volume (m³), is the instantaneous elastic modulus (Pa), with a value of 4.2±1.5 kPa, is the delayed elastic modulus (Pa), with a value of 1.8±0.6 kPa, is the viscosity coefficient (Pa·s), representing the resistance coefficient of the bladder wall tissue to viscous flow under stress, reflecting the dynamic mechanical response characteristics during bladder filling. Based on clinical research data, the typical value of the viscosity coefficient of a healthy bladder is: ; This model can accurately depict the variation law of the wall tension of the patient's bladder in different volume states, providing a key basis for subsequent assessment of micturition demand.

[0026] By analyzing the data of the change in the electrical conductivity of the liquid in the bladder over a long period, combined with the record of micturition events and the established bladder wall tension - bladder volume model, the number of limit volume units of the patient's bladder is deduced. This value is crucial for judging the functional state of the bladder and formulating a reasonable training plan.

[0027] The micturition demand assessment module is used to evaluate the real - time micturition demand value of the patient based on the patient's real - time bladder wall tension, the bladder wall tension - bladder volume model, and the number of limit volume units, including: According to the patient's real - time bladder wall tension and the bladder wall tension - bladder volume model, determine the patient's current real - time bladder volume unit number, and clarify the actual filling degree of the bladder at this moment.

[0028] The micturition demand value quantification sub - module combines the real - time bladder volume unit number, the number of limit volume units, the patient's real - time bladder wall tension, and the maximum bladder wall tension value, and uses a specific calculation formula to accurately calculate the patient's real - time micturition demand value. For example, if the real - time bladder volume is close to the limit volume and the real - time bladder wall tension is high, close to the maximum bladder wall tension value, then the micturition demand value will increase accordingly, indicating that the patient has a strong micturition demand at this moment. In this way, the system can dynamically and accurately assess the micturition demand of the patient at different times, providing an important reference for the subsequent optimization and control of the training plan.

[0029] The training plan optimization module is used to optimize the initial bladder function training plan based on the real-time urine excretion demand value of the patient obtained on the same day, and obtain an optimized bladder function training plan. Among them, the initial training plan is formulated by the initial training plan generation sub-module based on the patient's physiological sign data (such as age, gender, physical condition, etc.), and is a preliminary and general training framework. The training plan optimization sub-module conducts reinforcement learning on the real-time urine excretion demand value of the patient obtained on the same day and the occurrence times of all the patient's urination events, deeply analyzes the patient's urination pattern and demand changes, and then optimizes and adjusts the adjustable parameters (such as urination interval time, bladder filling volume for each training, etc.) in the initial bladder function training plan to obtain an optimized bladder function training plan that better suits the actual situation of the patient on the same day. For example, if the real-time urine excretion demand value of the patient on a certain day shows that the patient's bladder function has recovered well, the urination interval time is relatively stable and long, the training plan optimization sub-module may appropriately extend the urination interval time in the initial plan to make the training plan more targeted and effective, so as to better promote the recovery of the patient's bladder function.

[0030] The training and control module is used to conduct bladder function training and control on the patient based on the optimized bladder function training plan. At the same time, when the urine excretion demand value reaches the urine excretion demand threshold in the optimized bladder function training plan, it triggers an audible and visual vibration to remind the patient to urinate or activates the urinary catheterization device. During the training process, it continuously monitors the patient's urine excretion demand value. Once the urine excretion demand value reaches the pre-set urine excretion demand threshold in the optimized bladder function training plan, the system will immediately trigger an audible and visual vibration to remind the patient to urinate in an intuitive and obvious way. For some patients who are unable to urinate independently, the system will also automatically activate the urinary catheterization device to ensure the smooth progress of the patient's bladder function training, while ensuring the patient's physical health and urination safety, and avoiding damage to the bladder caused by over-distension of the bladder or untimely urination, realizing the intelligent and personalized control of bladder function training.

[0031] In an alternative embodiment, it further includes: The drinking water reminder module is used to dynamically calculate the current water requirement based on the patient's weight, activity level, environmental temperature and humidity, and the real-time state model of the bladder, and issue a real-time drinking water reminder to the patient based on the current water requirement.

[0032] In this embodiment, the real-time bladder state model synthesizes the data obtained and analyzed by the integrated sensing module and the bladder parameter inference module, reflecting the current functional state, capacity, and urine production of the bladder. If the real-time bladder state model shows that the urine production rate in the bladder is slow, or the bladder function is in a specific stage of recovery, it may be necessary to appropriately adjust the water intake to promote urine production and bladder function recovery. For example, in order to maintain the normal filling and emptying cycle of the bladder, it may be necessary to increase the water intake to stimulate urine production in some cases, or to appropriately control the water intake when the risk of overfilling the bladder is high.

[0033] The calculation process of the current water requirement includes: Calculation of basic water requirement: According to general medical advice, the basic water requirement for an adult per kilogram of body weight per day is about 30 - 40 milliliters. Assuming the patient weighs 70 kilograms, we take the middle value of 35 milliliters per kilogram to calculate the basic water requirement. Then the basic water requirement = 70 kg × 35 ml / kg = 2450 ml. This is the basic daily water requirement based on the patient's body weight.

[0034] Adjustment for activity level: The activity level can be obtained by wearing devices such as a sports bracelet, and is measured by the number of steps and exercise intensity. If the patient takes more than 10,000 steps and the exercise intensity is medium (such as brisk walking) on a certain day, according to the empirical formula, for every additional 1000 steps, an additional 50 ml of water needs to be supplemented, and for every 30 minutes of medium-intensity exercise, an additional 100 ml of water needs to be supplemented.

[0035] The patient walked 12,000 steps on that day, and the additional number of steps is 12,000 - 10,000 = 2,000 steps. The corresponding increase in water requirement is 2000÷1000×50 = 100 ml. At the same time, the medium-intensity exercise lasted for 60 minutes, and the additional water requirement is 60÷30×100 = 200 ml. So the increase in water requirement due to the increase in activity level is 100 + 200 = 300 ml.

[0036] Adjustment for environmental temperature and humidity: Assume that the environmental temperature and humidity sensor obtains an environmental temperature of 30°C and a relative humidity of 40%. According to research, when the temperature rises by 1°C, an additional 5 ml of water needs to be ingested per kilogram of body weight per day; when the relative humidity drops by 10%, an additional 10 ml of water needs to be ingested per kilogram of body weight per day.

[0037] The temperature has increased by 30 - 25 = 5°C compared to the appropriate temperature (assumed to be 25°C). The additional water requirement due to the temperature increase is 70 kg × 5°C × 5 ml / kg / °C = 1750 ml. The relative humidity has decreased by 60% - 40% = 20% compared to the appropriate humidity (assumed to be 60%). The additional water requirement due to the humidity decrease is 70 kg × 20% ÷ 10% × 10 ml / kg = 1400 ml. Therefore, the increased water requirement due to environmental temperature and humidity changes is 1750 + 1400 = 3150 ml.

[0038] Adjustment of the real - time bladder state model: The real - time bladder state model is obtained by analyzing quantitative data on the electrical conductivity of the liquid in the bladder, the surface pressure in the bladder area, and real - time geometric parameters of the bladder. Suppose the model shows that the patient's bladder is currently in a slightly water - deficient state. To promote urine production and the recovery of normal bladder function, an additional 500 ml of water intake is required.

[0039] Calculate the current water requirement: Summing up the above items, the current water requirement = basic water requirement+water requirement adjustment due to activity+water requirement adjustment due to environmental temperature and humidity+water requirement adjustment due to the real - time bladder state model = 2450 + 300 + 3150 + 500 = 6400 ml.

[0040] After dynamically calculating the current water requirement based on the above multiple factors, the drinking reminder module will send a real - time drinking reminder to the patient according to this calculation result. The reminder methods may include but are not limited to mobile application push messages, vibration prompts of wearable devices, voice reminders, etc. For example, in the afternoon when the patient has a large amount of activity and the environmental temperature is high, the drinking reminder module calculates that the patient needs to intake 500 ml more water than usual, and then sends a vibration and displays a drinking reminder message through the patient's smart bracelet, reminding the patient to supplement water in time to maintain normal water metabolism in the body and good conditions for bladder function training. This way of dynamically calculating the water requirement based on multiple factors and giving real - time reminders helps the patient maintain reasonable water intake, better cooperate with bladder function training, and promote the recovery process.

[0041] In an alternative embodiment, the integrated sensing module includes: A pressure - sensing sub - module for continuously detecting the real - time surface pressure distribution value in the patient's bladder area based on a pressure sensor array as the surface pressure; this data provides important basis for subsequent analysis of the bladder's filling degree, pressure change trend, etc., helping the system understand the mechanical state of the bladder at different times.

[0042] The ultrasonic probe measurement sub-module is used to continuously transmit ultrasonic signals to the patient's bladder area by the ultrasonic probe and receive the ultrasonic echo signals reflected from the bladder area, and determine the real-time geometric parameters of the bladder based on the ultrasonic echo signals received by the ultrasonic probe in real time. For example, the length, width, and height of the bladder can be calculated, and then the volume of the bladder can be obtained. In addition, geometric features such as the thickness and shape of the bladder wall can also be analyzed. These real-time geometric parameters are crucial for accurately judging the morphological changes of the bladder and evaluating the functional status of the bladder, and provide intuitive bladder morphological information for the entire guidance and control system.

[0043] The bladder impedance detection sub-module is used to apply a weak alternating current to multiple electrodes placed on the body surface of the patient's bladder area and measure the potential difference on the body surface of the patient's bladder area, construct an impedance tomography of the patient's bladder, and determine the quantitative data of the liquid conductivity in the patient's bladder based on the continuously obtained impedance tomography.

[0044] This sub-module utilizes the characteristic that different tissues and liquids have different conductivities. By placing multiple electrodes on the body surface of the patient's bladder area and applying a weak alternating current to the human body, due to the difference in conductivity between the liquid (urine) in the bladder and the surrounding tissues, a potential difference will be generated on the body surface. The bladder impedance detection sub-module measures these potential differences and constructs an impedance tomography of the patient's bladder based on electrical principles and mathematical algorithms. The continuously obtained impedance tomography can reflect the change of the liquid conductivity in the bladder over time, and the sub-module determines the quantitative data of the liquid conductivity in the patient's bladder based on these data. For example, by analyzing the impedance tomography data, the composition and concentration changes of urine can be understood because the conductivity of urine with different compositions and concentrations is different. These quantitative data of liquid conductivity help to deeply understand the physiological state in the bladder and provide unique electrical dimension information for comprehensively evaluating bladder function.

[0045] These three sub-modules cooperate with each other to comprehensively obtain real-time information of the bladder from multiple dimensions such as pressure, geometric shape, and liquid conductivity, providing a rich and accurate data basis for subsequent modules such as bladder parameter inference and urination demand assessment, and are the key links for the accurate operation of the bladder function training guidance and control system based on multi-modal sensors.

[0046] In an alternative embodiment, the bladder parameter inference module includes: The residual urine volume statistical sub-module is used to determine the number of residual urine volume units of the patient based on the quantitative data of the liquid conductivity in the patient's bladder; the conductivity of urine is related to components such as electrolytes dissolved in it, and different urine volumes will cause changes in conductivity. Through a mathematical model or empirical formula established with a large amount of clinical data, the sub-module can convert the measured quantitative data of liquid conductivity into the number of residual urine volume units. For example, through research on many patients, it is found that under specific conditions, there is a certain functional relationship between urine conductivity and residual urine volume. Substituting the liquid conductivity data of the current patient into this function can obtain the number of residual urine volume units. Understanding the residual urine volume is of great significance for evaluating the bladder's emptying ability and formulating subsequent treatment and training plans. It reflects the amount of urine remaining in the bladder after the last urination and can assist in judging whether the bladder function is normal.

[0047] The bladder wall tension calculation sub-module is used to calculate the real-time bladder wall tension based on the body surface pressure in the patient's bladder area and the real-time geometric parameters of the bladder; the body surface pressure in the bladder area reflects the effect of the internal bladder pressure on the body surface, and the real-time geometric parameters of the bladder (such as shape, size, etc.) affect the distribution of pressure on the bladder wall. First, by establishing a mechanical model of the bladder, considering the elastic characteristics of the bladder and the pressure transmission law, the body surface pressure is converted into the internal bladder pressure, and combined with the geometric shape parameters of the bladder, the real-time bladder wall tension is calculated using mechanical principles and mathematical methods. For example, based on the theory of elasticity mechanics, assuming that the bladder is approximately a certain geometric shape, according to the body surface pressure and the parameters of this geometric shape, the tension borne by the bladder wall in the current state is calculated through the corresponding formula. The real-time bladder wall tension is an important indicator reflecting the bladder function state. Too high or too low tension may imply problems with the bladder and provide a key basis for subsequent evaluation of urination needs and formulation of training plans.

[0048] A correlation analysis sub-module is used to perform a correlation analysis on the number of residual urine units and the real-time bladder wall tension of a patient within a preset period, introduce the instantaneous elastic modulus and the delayed elastic modulus of the bladder, and use the three-element standard linear solid model to describe it, obtaining the bladder wall tension - bladder capacity model of the patient; these two moduli reflect the elastic response characteristics of the bladder tissue when stressed. The instantaneous elastic modulus reflects the elastic change of the bladder when stressed instantaneously, and the delayed elastic modulus describes the elastic change of the bladder during the continuous stress process. Using the three-element standard linear solid model to describe the relationship between the residual urine volume and the bladder wall tension, this model can more accurately depict the change law of the wall tension of the bladder in different urine volume states. Through the analysis and fitting of a large amount of data, a personalized bladder wall tension - bladder capacity model of the patient is obtained. This model is crucial for deeply understanding the physiological characteristics of the patient's bladder. It can help the system accurately infer the bladder capacity based on the bladder wall tension, providing an important theoretical basis for the assessment of micturition needs and the optimization of training plans, enabling the system to more precisely grasp the functional state of the patient's bladder.

[0049] A maximum capacity calculation sub-module is used to calculate the number of maximum capacity units of the patient's bladder based on the change data of the conductivity of the liquid in the bladder, micturition events, and the bladder wall tension - bladder capacity model within a long period. The change in the conductivity of the liquid in the bladder is closely related to the change in the urine volume in the bladder, and micturition events clearly mark the phased changes in the bladder capacity. By analyzing the change trend of the liquid conductivity within a long period, combined with the conductivity mutation situation when micturition events occur, and using the relationship reflected by the bladder wall tension - bladder capacity model, the number of maximum capacity units that the bladder can hold is calculated. For example, during multiple micturition processes, observe the change in the conductivity of the liquid before and after each micturition, as well as the corresponding bladder wall tension and capacity relationship, and determine the maximum capacity that the bladder can reach within the normal physiological range through specific algorithms and data analysis methods. Understanding the maximum capacity of the bladder is crucial for formulating a reasonable bladder function training plan. It can help the system set an appropriate micturition threshold, avoid damage to the bladder caused by overfilling, and at the same time provide an important reference for evaluating the recovery of the patient's bladder function.

[0050] In an alternative embodiment, the bladder wall tension calculation sub-module includes: The bladder mechanics model and function establishment unit is used to obtain the range of body surface pressure in the patient's bladder area and the range of geometric limit parameters of the bladder, and establish the bladder elastic mechanics model and pressure transfer function of the patient based on the range of body surface pressure in the patient's bladder area and the range of geometric limit parameters of the bladder; the range of body surface pressure reflects the change interval of the pressure exerted on the body surface by the bladder at different filling degrees, while the range of geometric limit parameters of the bladder covers parameters such as the shape and size of the bladder in extreme states such as maximum and minimum volumes. Based on these data, the unit uses mechanical principles and mathematical methods to establish the bladder elastic mechanics model of the patient. This model regards the bladder as a mechanical structure with elastic properties, simulating its deformation and stress distribution under different pressure actions. At the same time, a pressure transfer function is established, which describes the relationship between how the body surface pressure is transmitted and converted into the internal pressure of the bladder. For example, considering that the shape of the bladder is approximately an ellipsoid, according to the theory of elastic mechanics and the relationship between the measured body surface pressure and bladder geometric parameters, the corresponding mathematical model and function expression are constructed. These models and functions are the basis for accurately calculating the bladder wall tension subsequently, providing a theoretical framework for understanding the internal mechanical state of the bladder.

[0051] The bladder pressure distribution conversion unit is used to convert the body surface pressure into the internal bladder pressure distribution based on the patient's bladder pressure transfer function; since the body surface pressure is not directly equivalent to the internal bladder pressure, and the internal bladder pressure is not evenly distributed in different parts, this conversion process is required to obtain the true internal pressure distribution of the bladder. For example, assuming that the pressure transfer function indicates a non-linear relationship between the body surface pressure and the internal bladder pressure and is related to the specific geometric position of the bladder, this unit will use the pressure transfer function to perform accurate calculations based on the real-time body surface pressure data and the current geometric shape of the bladder, so as to obtain the pressure values at each position in the bladder and depict a detailed map of the internal bladder pressure distribution. The accurate internal bladder pressure distribution is crucial for subsequent calculation of the bladder wall tension, because the tension borne by different positions of the bladder wall is closely related to the pressure at that position.

[0052] The bladder wall tension calculation unit is used to calculate the real-time bladder wall tension based on the patient's bladder elastomechanics model and the real-time geometric parameters of the bladder. The real-time geometric parameters of the bladder (such as the current volume, shape change, etc.) determine the force-bearing area and direction of the bladder wall, while the bladder elastomechanics model provides the conversion relationship between pressure and tension. For example, according to the stress-strain relationship in elastomechanics, combined with the material properties of the bladder wall (described by the elastomechanics model) and the current pressure distribution (from the bladder pressure distribution conversion unit) and geometric shape, through specific calculation formulas, the real-time tension values of the bladder wall at various positions are obtained. These real-time bladder wall tension data are crucial for evaluating the functional state of the bladder, judging whether there is abnormal pressure load, and formulating targeted treatment and training plans, and are one of the important parameters in the entire bladder function training and control system.

[0053] In an alternative embodiment, the limit capacity calculation sub-module includes: The non-linear model description unit is used to describe the non-linear relationship between the conductivity of the liquid in the bladder and the bladder volume during bladder filling based on the exponential decay model, and obtain the original conductivity-volume dynamic model of the liquid in the bladder; during bladder filling, as the urine increases and the bladder volume increases, the conductivity of the liquid in the bladder will show non-linear changes due to reasons such as urine dilution, and the exponential decay model can better describe this relationship. For example, assuming that the conductivity of the liquid in the bladder is σ and the bladder volume is V, the exponential decay model may be expressed as: ; In the formula, is the real-time conductivity of the liquid in the bladder (S / m), is the conductivity of the liquid in the patient's bladder in the non-filled state (S / m), is the natural constant, is the volume decay coefficient, is the background noise compensation term (S / m).

[0054] Thus, the original conductivity-volume dynamic model of the liquid in the bladder is obtained. This model provides a basic framework for subsequent analysis, is used to initially establish the mathematical connection between the liquid conductivity and the bladder volume, and helps to understand the electrical and volume change laws of the bladder during filling.

[0055] The urination event recognition unit is used to identify all the sudden conductivity drop characteristics (the moments of sudden conductivity change and the conductivity change values before and after) in the conductivity change data of the liquid in the bladder within a long period, and calculate the urination event occurrence probability of each sudden conductivity drop characteristic based on the conductivity change data before and after all the sudden conductivity drop characteristics and the hidden Markov model, and screen out all the urination events among all the sudden conductivity drop characteristics, including: ; wherein, is the occurrence probability of the th conductivity sudden drop feature of a urination event, is the natural exponential function, is the event sensitivity parameter (taking a high value (0.25 - 0.3) during emergency monitoring and a medium value (0.1 - 0.15) during routine monitoring), is the change value of the conductivity before and after in the before - and - after conductivity change data of the th conductivity sudden drop feature ; is the total number of conductivity sudden drop features; Then, all conductivity sudden drop features with the occurrence probability of the urination event greater than the preset value are determined to have a urination event occurred.

[0056] The model personalization description unit is used to optimize the volume decay coefficient and the background noise compensation term in the original conductivity - volume dynamic model based on the before - and - after conductivity change data of all urination events and the Markov chain Monte Carlo method, so as to obtain the conductivity - volume dynamic model of the liquid in the patient's bladder; in this process, the volume decay coefficient and the background noise compensation term in the original model are mainly optimized. The volume decay coefficient affects the decay rate between the bladder volume and the liquid conductivity in the model, and the background noise compensation term is used to correct the noise interference in the measurement data. By continuously adjusting these two parameters, the model can better fit the actual situation of the patient individual, and obtain the personalized conductivity - volume dynamic model of the liquid in the patient's bladder. This personalized model can more accurately reflect the true relationship between the liquid conductivity in the patient's bladder and the bladder volume, and provide a more reliable basis for accurately calculating the limit volume.

[0057] The limit volume unit number calculation unit is used to calculate the limit volume unit number of the patient's bladder based on the conductivity - volume dynamic model of the liquid in the patient's bladder, the limit value of the liquid conductivity in the patient's bladder, and the maximum wall tension value of the patient's bladder. The limit value of the liquid conductivity in the bladder reflects the maximum or minimum conductivity that the liquid in the bladder can reach, and the maximum wall tension value of the bladder wall limits the maximum tension that the bladder can withstand within a safe range. By substituting these parameters into the personalized model and using relevant mathematical calculation methods (for example, it may involve solving specific equations or performing numerical simulations), the limit volume unit number of the patient's bladder is obtained. This limit volume unit number is of great significance for formulating a reasonable bladder function training plan and evaluating the patient's bladder function. It provides a key reference for setting a suitable bladder filling upper limit during the training process, and helps to avoid bladder damage caused by over - filling.

[0058] In an alternative embodiment, the method for calculating the number of limit capacity units by the limit capacity unit number calculation unit based on the intravesical fluid conductivity-capacity dynamic model of the patient, the limit value of the intravesical fluid conductivity of the patient, and the maximum wall tension value of the patient's bladder wall includes: ; Wherein, is the number of limit capacity units of the patient's bladder, which reflects the maximum capacity that the patient's bladder can hold under the current physiological state and is an important indicator for measuring bladder function, and is crucial for formulating a bladder function training plan, judging the bladder health status, etc.; is the maximum wall tension value of the patient's bladder wall, which represents the maximum tensile force that the bladder wall can withstand. During the filling process of the bladder, the tension borne by the bladder wall gradually increases. When this maximum value is reached, the bladder approaches its limit capacity. This parameter limits the overfilling of the bladder and ensures that the calculated limit capacity is within the physiological tolerance range of the bladder. For example, if the bladder wall material becomes fragile due to diseases, etc., Tmax will decrease accordingly, and the calculated number of limit capacity units will also decrease; is the maximum wall tension value of the standard adult bladder wall; is the Poisson's ratio of the bladder tissue. The Poisson's ratio describes the ratio of the transverse strain to the longitudinal strain when the material is unidirectionally tensioned or compressed. For the bladder tissue, the Poisson's ratio reflects the deformation characteristics of the bladder when it is stressed. For example, when the bladder wall is longitudinally stretched, the Poisson's ratio determines the degree of transverse contraction or expansion; is the natural logarithm function; is the approximate radius of the patient's bladder in the non-filled state (i.e., the radius of the corresponding sphere obtained by approximately converting the bladder volume into a sphere volume); is the bladder wall thickness of the patient's bladder in the non-filled state (the wall thickness of the existing adult bladder in the non-filled state can be used); is the maximum conductivity that the intravesical fluid of the patient can reach, which provides the electrical boundary condition for calculating the limit capacity; is the background noise compensation term in the intravesical fluid conductivity-capacity dynamic model of the patient. This term is used to correct the influence of these noises on the calculation and make the limit capacity calculated based on the fluid conductivity more accurate; is the intravesical fluid conductivity of the patient's bladder in the non-filled state; is the volume attenuation coefficient in the dynamic model of the conductivity - volume of the liquid in the patient's bladder. This coefficient describes the attenuation rate of the liquid conductivity as the bladder volume changes. It reflects the internal relationship between the bladder volume and the liquid conductivity in the model. By adjusting this coefficient, the model can better fit the bladder characteristics of the individual patient, thus accurately calculating the number of limit capacity units.

[0059] In an alternative embodiment, the micturition demand assessment module includes: A real - time bladder volume calculation sub - module, which is used to determine the number of real - time bladder volume units of the patient based on the real - time bladder wall tension of the patient and the bladder wall tension - bladder volume model; the sub - module uses this model, takes the real - time bladder wall tension as the input value, and calculates the number of real - time bladder volume units of the patient through the mathematical relationship defined by the model. This calculation method is based on the bladder characteristic model of the individual patient and can more accurately reflect the actual filling degree of the patient's current bladder compared with the general estimation method.

[0060] A micturition demand value quantification sub - module, which is used to calculate the real - time micturition demand value of the patient based on the number of real - time bladder volume units of the patient, the number of limit capacity units, the real - time bladder wall tension of the patient, and the maximum bladder wall tension value.

[0061] In this embodiment, the bladder wall tension - bladder volume model reflects the change law of the wall tension of the patient's bladder in different volume states and is a personalized mathematical model.

[0062] In this embodiment, the real - time bladder wall tension is obtained in real - time through the bladder wall tension calculation sub - module.

[0063] In this embodiment, the number of real - time bladder volume units reflects the actual filling amount of the bladder at this moment, the number of limit capacity units represents the maximum amount that the bladder can hold, the real - time bladder wall tension reflects the force borne by the current bladder wall, and the maximum bladder wall tension value is the maximum tensile force that the bladder wall can withstand. The sub - module integrates these factors through a specific calculation formula to quantify the micturition demand. For example, the formula that may be adopted is: ; In the formula, is the micturition demand value, is the number of real - time bladder volume units, is the number of limit capacity units, is the real - time bladder wall tension, is the maximum bladder wall tension value.

[0064] The closer the bladder is to its maximum capacity and the closer the bladder wall tension is to the maximum tension value, the higher the urination demand. Through such quantitative calculations, the obtained real-time urination demand value can more scientifically and accurately reflect the current actual urination demand level of the patient, providing a clear decision-making basis for the training plan optimization module and the training guidance and control module, such as deciding whether to remind the patient to urinate or activate the catheterization device, etc.

[0065] In an alternative embodiment, the training plan optimization module includes: An initial training plan generation sub-module for generating an initial bladder function training plan for the patient based on the patient's physiological sign data; the physiological sign data covers multiple aspects, such as the patient's age, gender, weight, underlying diseases, current physical condition, and previous bladder function test data, etc. These data comprehensively reflect the patient's overall health status and the basis of bladder function. For example, for a young patient without other serious underlying diseases, their physical recovery ability is relatively good, and when formulating a training plan, a relatively short urination interval training time and a higher bladder filling volume target may be set; while for an elderly patient or a patient with other diseases affecting bladder function, considering their physical tolerance, the training plan may be more conservative, with a longer urination interval and a relatively lower bladder filling volume each time. Through the analysis and evaluation of these physiological sign data, the sub-module uses pre-set rules or algorithms to formulate a preliminary bladder function training plan suitable for the patient's basic situation, which provides a basic framework for subsequent optimization.

[0066] A training plan optimization sub-module for optimizing the adjustable parameters in the initial bladder function training plan through reinforcement learning of the real-time urination demand value of the patient obtained on the current day and the occurrence times of all the patient's urination events, to obtain an optimized bladder function training plan.

[0067] In this scenario, the training plan optimization sub-module is like an intelligent agent. The adjustable parameters of the initial training plan (such as the urination interval time, the bladder filling volume for each training, etc.) are the "actions" it can take, and information such as the real-time urination demand value and the occurrence times of urination events constitutes the "environmental feedback". The sub-module gradually finds the optimal parameter settings by continuously adjusting these adjustable parameters, observing the "reward signals" such as the degree to which the patient's urination demand is met and whether it helps to improve bladder function, so as to obtain an optimized bladder function training plan. For example, if it is found that under the current urination interval time, the real-time urination demand value of the patient often rises rapidly in the later stage of the interval, indicating that the urination interval may be too long, the sub-module will appropriately shorten the urination interval time to better adapt to the patient's bladder function state and improve the training effect. This optimization method based on real-time data and reinforcement learning can make the training plan more in line with the actual needs of the patient and achieve personalized and precise bladder function training.

[0068] The present invention provides an implementation manner of a bladder function training and guiding method based on a multi-modal sensor, including: Step 1: Detect the body surface pressure of the patient's bladder area based on a pressure sensor array. At the same time, measure the real-time geometric parameters of the bladder based on an ultrasonic probe, and obtain quantitative data on the liquid conductivity in the patient's bladder based on electrical impedance tomography; Step 2: Based on the quantitative data on the liquid conductivity in the patient's bladder, the body surface pressure of the patient's bladder area, and the real-time geometric parameters of the bladder, calculate the real-time bladder wall tension and the number of limit capacity units, and establish a bladder wall tension - bladder capacity model for the patient; Step 3: Based on the patient's real-time bladder wall tension, the bladder wall tension - bladder capacity model, and the number of limit capacity units, evaluate the patient's real-time urine excretion demand value; Step 4: Optimize the initial bladder function training plan based on the patient's real-time urine excretion demand value obtained on the same day to obtain an optimized bladder function training plan; Step 5: Conduct bladder function training and guiding for the patient based on the optimized bladder function training plan. At the same time, when the urine excretion demand value reaches the urine excretion demand threshold in the optimized bladder function training plan, trigger an audible and visual vibration to remind the patient to urinate or start the urinary catheterization device.

[0069] The above method realizes the precise control and timely intervention of the patient's bladder function training, ensures the smooth progress of the training process, maintains the physical health of the patient, avoids damage to the bladder caused by untimely urination or excessive urine retention, and ensures the safety and effectiveness of the entire training process.

[0070] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and modifications.

Claims

1. A bladder function training and control system based on multi-modal sensors, characterized in that, Including: An integrated sensing module, which is used to detect the body surface pressure in the patient's bladder area based on a pressure sensor array. At the same time, it measures the real-time geometric parameters of the bladder based on an ultrasonic probe, and obtains quantitative data on the liquid conductivity in the patient's bladder based on electrical impedance tomography; A bladder parameter inference module, which is used to calculate the real-time bladder wall tension and the number of limit volume units based on the quantitative data of the liquid conductivity in the patient's bladder, the body surface pressure in the patient's bladder area, and the real-time geometric parameters of the bladder, and build a bladder wall tension-bladder volume model for the patient; A urination demand assessment module, which is used to evaluate the real-time urination demand value of the patient based on the patient's real-time bladder wall tension, the bladder wall tension-bladder volume model, and the number of limit volume units; A training plan optimization module, which is used to optimize the initial bladder function training plan based on the real-time urination demand value of the patient obtained on the same day, and obtain an optimized bladder function training plan; A training guidance and control module, which is used to conduct bladder function training guidance and control for the patient based on the optimized bladder function training plan. At the same time, when the urination demand value reaches the urination demand threshold in the optimized bladder function training plan, it triggers an audible and visual vibration to remind the patient to urinate or starts a urinary catheterization device.

2. The bladder function training guidance and control system based on a multi-modal sensor according to claim 1, wherein It also includes: A drinking reminder module, which is used to dynamically calculate the current water requirement based on the patient's weight, activity level, environmental temperature and humidity, and the real-time bladder state model, and send a real-time drinking reminder to the patient based on the current water requirement.

3. The bladder function training and control system based on multi-modal sensors according to claim 1, wherein The integrated sensing module includes: A pressure sensing sub-module, which is used to continuously detect the real-time body surface pressure distribution value in the patient's bladder area based on a pressure sensor array as the body surface pressure; An ultrasonic probe measurement sub-module, which is used to continuously emit ultrasonic signals to the patient's bladder area based on an ultrasonic probe and receive the ultrasonic echo signals reflected from the bladder area, and determine the real-time geometric parameters of the bladder based on the ultrasonic echo signals received by the ultrasonic probe in real time; A bladder impedance detection sub-module, which is used to apply a weak alternating current to multiple electrodes placed on the body surface of the patient's bladder area and measure the potential difference on the body surface of the patient's bladder area, construct an electrical impedance tomography of the patient's bladder, and determine the quantitative data of the liquid conductivity in the patient's bladder based on the continuously obtained electrical impedance tomography.

4. The bladder function training guidance and control system based on a multi-modal sensor according to claim 1, characterized in that, The bladder parameter inference module includes: A residual urine volume statistics sub-module, which is used to determine the number of residual urine volume units of the patient based on the quantitative data of the liquid conductivity in the patient's bladder; A bladder wall tension calculation sub-module, which is used to calculate the real-time bladder wall tension based on the body surface pressure in the patient's bladder area and the real-time geometric parameters of the bladder; A correlation analysis sub-module, which is used to conduct a correlation analysis on the number of residual urine volume units and the real-time bladder wall tension of the patient within a preset period, introduce the instantaneous elastic modulus and delayed elastic modulus of the bladder, and use the three-element standard linear solid model to describe it, and obtain the bladder wall tension-bladder volume model of the patient; A limit volume calculation sub-module, which is used to calculate the number of limit volume units of the patient's bladder based on the change data of the liquid conductivity in the bladder, urination events, and the bladder wall tension-bladder volume model of the patient within a long period.

5. The bladder function training and control system based on multimodal sensors according to claim 4, wherein, The bladder wall tension calculation sub-module includes: The bladder mechanics model and function establishment unit is used to obtain the body surface pressure range of the patient's bladder area and the geometric limit parameter range of the bladder, and establish the bladder elastic mechanics model and pressure transfer function of the patient based on the body surface pressure range of the patient's bladder area and the geometric limit parameter range of the bladder; The bladder pressure distribution conversion unit is used to convert the body surface pressure into the bladder internal pressure distribution based on the patient's bladder pressure transfer function; The bladder wall tension calculation unit is used to calculate the real-time bladder wall tension based on the patient's bladder elastic mechanics model and the geometric real-time parameters of the bladder.

6. The bladder function training and control system based on multimodal sensors according to claim 4, characterized in that, The limit capacity calculation sub-module includes: The non-linear model description unit is used to describe the non-linear relationship between the conductivity of the liquid in the bladder and the bladder volume during the bladder filling process based on the exponential decay model, and obtain the original conductivity-volume dynamic model of the liquid in the bladder; The urination event recognition unit is used to identify all the sudden drops in conductivity features in the conductivity change data of the liquid in the bladder within a long period, and calculate the occurrence probability of the urination event for each sudden drop in conductivity feature based on the conductivity change data before and after all the sudden drops in conductivity features and the hidden Markov model, and screen out all the urination events from all the sudden drops in conductivity features based on the occurrence probability of the urination event; The model personalization description unit is used to optimize the volume decay coefficient and background noise compensation term in the original conductivity-volume dynamic model based on the conductivity change data before and after all the urination events and the Markov chain Monte Carlo method, and obtain the conductivity-volume dynamic model of the liquid in the patient's bladder; The limit capacity unit number calculation unit is used to calculate the limit capacity unit number of the patient's bladder based on the conductivity-volume dynamic model of the liquid in the patient's bladder, the limit value of the conductivity of the liquid in the patient's bladder, and the maximum value of the bladder wall tension of the patient.

7. The bladder function training guidance and control system based on multi-modal sensors according to claim 6, wherein The method by which the limit capacity unit number calculation unit calculates the limit capacity unit number of the patient's bladder based on the conductivity-volume dynamic model of the liquid in the patient's bladder, the limit value of the conductivity of the liquid in the patient's bladder, and the maximum value of the bladder wall tension of the patient includes: ； In the formula, is the number of units of the limit capacity of the patient's bladder, is the maximum wall tension value of the patient's bladder, is the maximum wall tension value of the bladder of a standard adult, is the Poisson's ratio of the bladder tissue, is the natural logarithm function, is the approximate radius of the patient's bladder in the non-filled state, is the wall thickness of the patient's bladder in the non-filled state, is the maximum conductivity that the liquid in the patient's bladder can reach, is the background noise compensation term in the conductivity-capacity dynamic model of the liquid in the patient's bladder, is the conductivity of the liquid in the patient's bladder in the non-filled state, is the volume attenuation coefficient in the conductivity-capacity dynamic model of the liquid in the patient's bladder.

8. The bladder function training guidance and control system based on multi-modal sensors according to claim 1, characterized in that, The urination demand assessment module includes: The real-time bladder capacity calculation sub-module is used to determine the real-time bladder capacity unit number of the patient based on the patient's real-time bladder wall tension and the bladder wall tension-bladder capacity model; The urination demand value quantification sub-module is used to calculate the real-time urination demand value of the patient based on the real-time bladder capacity unit number of the patient, the limit capacity unit number, the patient's real-time bladder wall tension, and the maximum value of the bladder wall tension.

9. The bladder function training and guidance control system based on multimodal sensors according to claim 1, characterized in that, The training plan optimization module includes: The initial training plan generation sub-module is used to generate the initial bladder function training plan of the patient based on the physiological sign data of the patient; The training plan optimization sub-module is used to optimize the adjustable parameters in the initial bladder function training plan by performing reinforcement learning on the real-time urination demand value of the patient obtained on the current day and the occurrence times of all the urination events of the patient, and obtain the optimized bladder function training plan.

10. A bladder function training and control method based on multi-modal sensors, characterized in that, including: Step 1: Detect the body surface pressure of the patient's bladder area based on the pressure sensor array. At the same time, measure the geometric real-time parameters of the bladder based on the ultrasonic probe, and obtain the quantitative data of the conductivity of the liquid in the patient's bladder based on electrical impedance tomography; Step 2: Based on the quantitative data of the liquid conductivity in the patient's bladder, the body surface pressure in the patient's bladder area, and the real-time geometric parameters of the bladder, calculate the real-time bladder wall tension and the number of limit capacity units, and establish the patient's bladder wall tension-bladder capacity model; Step 3: Based on the patient's real-time bladder wall tension, the bladder wall tension-bladder capacity model, and the number of limit capacity units, evaluate the patient's real-time urine excretion demand value; Step 4: Optimize the initial bladder function training plan based on the patient's real-time urine excretion demand value obtained on the same day to obtain an optimized bladder function training plan; Step 5: Conduct bladder function training and control for the patient based on the optimized bladder function training plan. At the same time, when the urine excretion demand value reaches the urine excretion demand threshold in the optimized bladder function training plan, trigger an audible and visual vibration to remind the patient to urinate or activate the catheterization device.

Images