A biofeedback-based adaptive regulation system for intestinal function

Abstract

The present invention discloses an intestinal function adaptive regulation system based on biofeedback, which relates to the technical field of intestinal function regulation, and includes a signal acquisition module, a data analysis module, an adjustment feedback module and a user interaction module; the signal acquisition module adopts a multi-sensor fusion algorithm and a multi-source sensor to acquire intestinal function status information; wherein the multi-source sensor includes a pressure sensor, a peristalsis sensor, and a neural sensor. The intestinal function adaptive regulation system based on biofeedback can comprehensively collect biological signals of the intestine, including pressure, peristalsis, nervous system signals and chemical substance secretion signals; this multi-dimensional monitoring method can more accurately reflect the true state of intestinal function, avoid the limitations of a single sensor, and provide a rich and accurate data basis for subsequent analysis and regulation.

Description

Technical Field

[0001] The present invention relates to the technical field of intestinal function regulation, and in particular to an intestinal function adaptive regulation system based on biofeedback. Background Art

[0002] With the change of people's lifestyle and the adjustment of dietary structure, intestinal health issues have received increasing attention; intestinal dysfunction not only affects the body's digestion and absorption, but may also lead to a series of health problems, such as constipation, diarrhea, and abdominal pain; currently, the regulation of intestinal function mainly relies on drug therapy and dietary adjustment, but these methods often have certain limitations; drug therapy may have side effects, and long-term use may cause intestinal flora imbalance; although dietary adjustment is relatively safe, the effect is often not significant enough, and patients need to have a high degree of self-discipline; in addition, existing intestinal function regulation methods are usually based on experience and general principles, lacking accurate assessment and targeted regulation of individual intestinal function;

[0003] In recent years, biofeedback technology has been widely used in the medical field. Biofeedback is a technology that helps patients self-regulate their physiological functions by monitoring human physiological signals and feeding them back to patients. In terms of intestinal function regulation, biofeedback technology can help patients understand the functional state of the intestine by monitoring biological signals such as intestinal pressure and peristalsis, and improve the autonomous regulation ability of the intestine through training and regulation. However, the existing intestinal function regulation system based on biofeedback still has some shortcomings. The accuracy and comprehensiveness of current signal acquisition need to be improved. This problem limits the widespread application and development of the intestinal function regulation system based on biofeedback. Summary of the invention

[0004] In order to solve the above technical problems, the present invention is implemented through the following technical solutions: a biofeedback-based intestinal function adaptive regulation system, including a signal acquisition module, a data analysis module, a regulation feedback module and a user interaction module;

[0005] The signal acquisition module uses a multi-sensor fusion algorithm and multi-source sensors to acquire intestinal function status information; wherein the multi-source sensors include pressure sensors, peristalsis sensors, neural sensors, and chemical sensors; neural sensors are used to detect intestinal nervous system signals; chemical sensors are used to detect the secretion of chemical substances in the intestine;

[0006] The data analysis module analyzes and processes the collected status information, extracts key feature information, determines the intestinal function status, obtains the judgment result, and provides a decision basis for the adjustment feedback module;

[0007] The regulation feedback module uses a collective regulation method to adaptively regulate the intestinal function according to the judgment result of the data analysis module to obtain a regulation feedback effect. The collective regulation method includes electrical stimulation, magnetic stimulation, and biofeedback training to improve the accuracy and effectiveness of regulation.

[0008] The user interaction module is used to interact with the user, provide real-time intestinal function status information and regulation feedback effect display, so that the user can understand his or her intestinal health status and perform self-regulation and management;

[0009] Through the above technical solutions, the problem that the accuracy and comprehensiveness of signal acquisition in the current biofeedback-based intestinal function adaptive regulation system needs to be improved can be effectively solved, providing more reliable technical support for the monitoring and regulation of intestinal health.

[0010] Preferably, the signal acquisition module includes:

[0011] The acquisition unit uses multi-source sensors to collect intestinal biological signals based on a multi-sensor fusion algorithm to improve the accuracy and comprehensiveness of signal acquisition and collect intestinal biological signals from different angles. The biological signals include pressure signals, peristalsis signals, nervous system signals, and chemical substance secretion signals;

[0012] Among them, the multi-source sensor is based on micro-electromechanical system technology to detect abnormal conditions in the intestine, including detecting tiny pressure changes, transient peristaltic abnormalities, abnormal nervous system signals and abnormal secretion of chemical substances; based on micro-electromechanical system technology, small-sized and high-precision sensors can be manufactured, which can more accurately perceive the biological signals of the intestine.

[0013] Preferably, the multi-source sensor is based on micro-electromechanical system technology, and the process of detecting abnormal conditions in the intestine is as follows:

[0014] The pressure sensor based on MEMS technology uses micromachining technology to set sensitive elements on a tiny chip; its core part includes a sensitive film, with a fixed electrode below the sensitive film and direct contact with the pressure environment in the intestine above; when the pressure in the intestine changes, the sensitive film deforms, thereby changing the capacitance value between the sensitive film and the fixed electrode; and the capacitance change is converted into a voltage change by means of a Wheatstone bridge circuit, and then amplified and filtered to obtain an analog electrical signal corresponding to the intestinal pressure; among them, based on the detection of tiny pressure changes, the sensitive film of the pressure sensor is based on silicon material and is manufactured with a circular film through micromachining technology to improve its responsiveness to tiny pressure changes; in terms of signal processing, a high-precision analog-to-digital converter is used to convert the analog electrical signal into a digital signal, and then the difference between adjacent sampling points is calculated to determine whether the pressure has changed slightly; when the difference exceeds the set pressure threshold, it is determined that a tiny pressure change abnormality has been detected, including a sudden increase or decrease in local pressure in the intestine, which is related to problems such as abnormal intestinal peristalsis and intestinal obstruction;

[0015] The peristaltic sensor based on MEMS technology uses a micromechanical structure to sense the mechanical deformation caused by intestinal peristalsis. The sensor contains sensitive elements: micro strain gauges and micro cantilever beams. The sensitive elements are strained under the deformation of the intestinal wall caused by intestinal peristalsis, causing their resistance values ​​to change. Like the pressure sensor, the peristaltic sensor also integrates a signal conditioning circuit to convert the resistance change of the sensitive element into an electrical signal. The electrical signal is amplified and filtered to reflect the intensity, frequency and direction of intestinal peristalsis. When the peristaltic frequency suddenly accelerates or slows down, the peristaltic amplitude increases or decreases abnormally, and the duration is within the range of 2s-16s, it is judged as a short-term peristalsis. Abnormal conditions; indicating intestinal spasms and indigestion problems; by comparing with the preset normal peristaltic pattern, the abnormal pattern is identified using a neural network-based pattern recognition method; after identifying the abnormality, the relevant abnormal characteristics are recorded, including the time when the abnormality occurred, the duration, the frequency and amplitude of the abnormal peristalsis; in order to accurately detect short-term abnormal peristalsis, the peristaltic sensor is designed to focus on the response ability to high-frequency signals; its sampling frequency is set high to capture rapid changes in the intestinal peristalsis process; normal intestinal peristalsis has a certain rhythmicity, and the peristaltic sensor can calculate the average frequency and amplitude of the peristalsis by monitoring the peristaltic signal over a period of time;

[0016] MEMS neural sensors are used to detect enteric nervous system signals. They are electrically coupled to enteric nerve tissue through microelectrode arrays to sense the electrical activity generated by nerve cells. The microelectrodes of the neural sensors are made of platinum and polypyrrole materials, and are treated with oxygen plasma, modified with nerve growth factor, and modified with carbon nanotubes to improve their compatibility with nerve tissue and signal conduction efficiency. The weak electrical signals generated and propagated by the action potential of nerve cells are detected and amplified and filtered. Since the enteric nervous system signals are very weak and have complex time and frequency domain characteristics, the neural sensors need to have high gain and low noise characteristics to ensure that the neural signals can be accurately acquired. When detecting intestinal tract, the collected nerve signals are first analyzed in time domain and frequency domain. The spectral characteristic model of normal intestinal nerve signals is established by calculating the power spectrum density, average frequency and peak frequency of the signals. When the spectral characteristic power of the nerve signals is detected to be abnormally increased or decreased, or the frequency distribution is changed, it is judged as an abnormality of the nervous system signal. The function of the nervous system can also be identified by analyzing the firing pattern of the nerve signals, including abnormal changes in firing frequency and disordered firing rhythm. These abnormalities may be related to intestinal neuropathy and imbalanced nerve regulation. Advanced signal processing algorithms such as wavelet analysis and principal component analysis are used to extract features and recognize patterns of nerve signals to improve the accuracy and reliability of abnormality detection.

[0017] The micro-electromechanical system chemical sensor is used to detect the secretion of chemical substances in the intestine. The chemical sensor has a chemical sensitive membrane. The chemical sensitive membrane is made of a polyvinyl chloride-based membrane material that responds to the change in pH in the intestine. When the chemical substances in the intestine interact with the chemical sensitive membrane, in the chemical sensor based on the electrochemical principle, the chemical substances react with the chemical sensitive membrane, resulting in a change in the electrode surface potential. The change is converted into an electrical signal by the conversion circuit inside the chemical sensor, and then output after amplification and analog-to-digital conversion. Before detection, the chemical sensor is calibrated to determine its response characteristics to chemical substances of different concentrations. By measuring in a standard solution of chemical substances of known concentration, a calibration curve between the sensor output signal and the concentration of the chemical substances is established. During actual detection in the intestine, the signal collected by the sensor is converted into a corresponding chemical substance concentration value according to the calibration curve. By setting a concentration threshold within the normal physiological range, when the concentration of the chemical substance is detected to exceed the threshold range, it is judged as an abnormal secretion of the chemical substance. For example, if excessive gastric acid secretion in the intestine causes the pH value to be lower than the normal range, the chemical sensor can detect this abnormality in time and record the relevant abnormal characteristics, including the type of abnormal chemical substance, the concentration change amplitude, and the duration of the abnormality.

[0018] This information is of great significance for diagnosing intestinal diseases such as gastric ulcers and indigestion. At the same time, the interactions and proportional relationships between different chemicals can also be comprehensively analyzed through multi-source chemical sensors to more comprehensively evaluate the chemical environment and functional status of the intestine.

[0019] Preferably, the multi-sensor fusion algorithm collects and analyzes the biological signal characteristics of different sensors based on a method combining weighted average and principal component analysis to obtain more accurate intestinal function status information, and the process is as follows:

[0020] Pressure sensors, peristaltic sensors, neural sensors and chemical sensors are deployed in the intestine and its surrounding areas to collect biological signals in the intestine in real time, including pressure signals, peristaltic signals, nervous system signals and chemical substance secretion signals; each sensor continuously samples the corresponding biological signals at a sampling frequency range of 100Hz-1000Hz, converts the analog signals into digital signals and transmits them to the subsequent processing unit;

[0021] The weight coefficients were determined in advance according to the importance of each sensor in reflecting the functional status of the intestine: based on experimental research and data analysis, the pressure sensor was given a weight of 0.3, the peristaltic sensor was given a weight of 0.3, the neural sensor was given a weight of 0.2, and the chemical sensor was given a weight of 0.2;

[0022] For the signal values ​​of different sensors collected at the same time, weighted sum is performed according to their respective weight coefficients; let the pressure signal value be P t , the creep signal value is M t , the neural signal value is N t , the chemical signal value is D t , t represents the sampling time, and the fused signal F t The calculation formula is: t =0.3P t +0.3M t +0.2N t +0.2D t ;

[0023] Through weighted average processing, the contribution of each sensor is comprehensively considered to obtain a fusion signal that can preliminarily reflect the overall functional status of the intestine;

[0024] The continuous signal data collected in the past 6 minutes or 30 minutes are processed by weighted average processing to form a fusion signal sequence into a data matrix X, where each row represents the fusion signal value at a sampling moment, and each column represents the same type of signal at different moments, including pressure, peristalsis, mental and chemical weighted signals; the data matrix X is decentralized, that is, the mean of the column is subtracted from each element to obtain the centralized data matrix X'; then the covariance matrix C of X' is calculated, and the element c in the covariance matrix ij It represents the covariance between the i-th signal and the j-th signal, reflecting the degree of linear correlation between the two signals;

[0025] Calculate the eigenvalue λ of the covariance matrix C i and the corresponding eigenvector v i ; The eigenvalue indicates the amount of information contained in each principal component, and the eigenvector determines the direction of the principal component; the eigenvalues ​​are arranged in descending order, and the eigenvectors corresponding to the eigenvalues ​​with a cumulative contribution rate of more than 85% are selected to form the principal component transformation matrix V, where each column of the principal component transformation matrix is ​​a principal component vector, which is the projection direction of the original signal in the new orthogonal coordinate system;

[0026] The original fusion signal data matrix X' is projected into the principal component space through the principal component transformation matrix V to obtain the principal component matrix Y, and the calculation formula is Y=X'V; each column of the principal component matrix Y is a principal component, where the first column of principal components contains the largest variance information in the original data, the second column is second, and so on.

[0027] The principal component data after dimensionality reduction is subjected to feature extraction and statistical features are calculated, including the mean, variance and slope of the principal component; the extracted features are compared and matched with the pre-established feature pattern library of normal intestinal function status and different disease states, and the support vector machine in the pattern recognition algorithm is used for classification and judgment to obtain the pattern recognition results; among them, if the extracted features are consistent with the feature patterns under the intestinal obstruction disease state, it is judged that the intestine has obstruction; if they are consistent with the feature patterns in the normal digestive process, it is considered that the intestinal function is in a normal state;

[0028] According to the pattern recognition results, the intestinal function status is evaluated to determine whether the intestinal function is normal. If it is abnormal, the abnormal type and severity are determined, among which the abnormal types include abnormal peristalsis, neural regulation imbalance, and chemical environment disorder; the intestinal function status information obtained by the evaluation is output to the user or medical personnel in the form of digital indicators plus graphic display plus text description, so that corresponding treatment or intervention measures can be taken in time, and it also provides a decision-making basis for the subsequent adaptive regulation system; if the intestinal peristalsis is judged to be too slow, the regulation feedback module can start the corresponding stimulation plan based on this information, and can promote intestinal peristalsis through electrical stimulation or biofeedback training;

[0029] By designing a sensor fusion algorithm, the data collected by different sensors are fused and processed to improve the accuracy and comprehensiveness of signal acquisition.

[0030] Preferably, abnormal conditions in the intestine are detected, and the collected biological signals are analyzed and processed in real time by setting thresholds and using pattern recognition algorithms to discover signal characteristics of abnormal intestinal function, record them as abnormal characteristics, and trigger corresponding individual difference adaptation mechanisms based on the abnormal characteristics.

[0031] Preferably, the process of obtaining the abnormal characteristics is as follows:

[0032] For pressure signals, the range of normal pressure in the intestine is determined based on clinical research and experimental data: in the fasting state, the pressure in the small intestine is between 2-5 mmHg, and after eating, the pressure in the small intestine is between 5-10 mmHg; the pressure in the large intestine is between 10-20 mmHg; based on the range of normal pressure in the intestine, the corresponding pressure threshold range is set: the upper threshold of the small intestine pressure threshold is set to 12 mmHg, and the lower threshold is set to 0 mmHg; the upper threshold of the large intestine pressure threshold is set to 25 mmHg, and the lower threshold is set to 0 mmHg. When the collected pressure signal exceeds the pressure threshold range, it indicates that there is an abnormal pressure condition, including increased pressure due to intestinal obstruction, or decreased pressure due to intestinal paralysis;

[0033] For peristaltic signals, determine the frequency and intensity range of normal intestinal peristalsis: Under normal circumstances, the frequency of small intestinal peristalsis is 3-12 times per minute. In the fasting state, the normal range of small intestinal peristaltic wave intensity is 5-15 mmHg. After eating, the small intestinal peristaltic wave intensity pressure range is 10-30 mmHg. In the resting state, the peristaltic wave intensity pressure range of the large intestine is 3-10 mmHg. When group peristalsis occurs, the peristaltic wave intensity pressure range is 15-40 mmHg. According to the normal intestinal peristalsis, the peristaltic wave intensity pressure range is 15-40 mmHg. The frequency and intensity range of peristalsis, upper and lower thresholds of peristalsis frequency and peristalsis intensity are set: the upper limit of frequency is 15 times / minute, and the lower limit is 1 time / minute; the lower limit of peristaltic wave intensity of small intestine is 3mmHg, and the upper limit is 40mmHg; the lower limit of peristaltic intensity threshold of large intestine is 2mmHg, and the upper limit is 50mmHg; if the collected peristaltic signal exceeds these thresholds, it may indicate abnormal peristalsis, such as faster peristalsis frequency during spasm, slower peristalsis frequency or weakened peristalsis intensity when power is insufficient;

[0034] For nerve signals, determine the electrical signal characteristics of normal enteric nervous system activity, including the frequency and amplitude of the signal; under normal circumstances, the main frequency range of enteric nerve signals is 1-10Hz, and the amplitude is in the range of 10-100μV; set the threshold range of electrical signal characteristics: the lower threshold of the frequency threshold range is 0.5Hz, and the upper threshold is 15Hz; the lower threshold of the amplitude threshold range is 5μV, and the upper threshold is 150μV; when the frequency and amplitude of the nerve signal exceed the normal threshold, it may indicate abnormal function of the nervous system, including abnormal signal emission caused by neuropathy;

[0035] For chemical signals, the pH value in the intestine is normally between 6 and 7.5 in the small intestine and between 5.5 and 7 in the large intestine. The thresholds are set according to these normal pH ranges: the lower threshold of the small intestinal pH value is 5.5 and the upper threshold is 8; the lower threshold of the large intestinal pH value is 5.0 and the upper threshold is 7.5. If the pH value is less than 5 or greater than 8, it is considered abnormal, which may be related to excessive gastric acid secretion or acid-base imbalance in the intestine. For other chemical substances, such as intestinal hormones and digestive enzymes, the corresponding thresholds are also set by studying their normal physiological concentration ranges.

[0036] The multi-source sensor continuously collects biological signals in the intestine at a set sampling frequency of 550Hz; after the sensor converts the collected analog signals into digital signals, it compares the signal value of each sampling point with the corresponding threshold; for each pressure signal sampling value, it determines whether it is higher than the upper pressure threshold or lower than the lower pressure threshold; for peristaltic signals, it checks whether its frequency and intensity exceed the set threshold range; for neural signals, it analyzes whether its frequency and amplitude are outside the normal threshold; for chemical signals, it determines whether its concentration value exceeds the set pH threshold; if the signal value of the sampling point exceeds the corresponding threshold, then the signal at that moment is preliminarily marked as abnormal, recorded as an abnormal signal segment, and feature extraction is performed; the time domain features, frequency domain features and time-frequency domain features of the signal are extracted; among which, the time domain features include mean, variance, slope, and zero-crossing rate, the frequency domain features include power spectrum density, main frequency, and frequency band energy, and the time-frequency domain features include wavelet coefficients obtained by wavelet transform;

[0037] The extracted features are input into a pre-trained pattern recognition model to obtain recognition results, wherein the pattern recognition model adopts a variety of algorithms, including support vector machines, multi-layer perceptrons in neural networks, convolutional neural networks, and decision trees; based on the recognition results, the specific signal features of intestinal dysfunction are determined and recorded as abnormal features.

[0038] Preferably, before using multi-source sensors to collect intestinal biological signals, the sensors are calibrated and optimized by establishing a standard test environment and using an automatic calibration algorithm to ensure that they can maintain stable performance under different environmental conditions and improve the accuracy and reliability of the sensors.

[0039] Preferably, the process of establishing a standard test environment is as follows:

[0040] The temperature was controlled at 37°C ± 0.5°C and the humidity was 40%-60%. Copper foil was used as an electromagnetic shielding material to create an electromagnetic interference-free space. Silica gel was used to simulate the mechanical properties of the intestinal wall. A solution containing appropriate amounts of sodium chloride, potassium chloride, and sodium bicarbonate was used to make its pH value close to the normal physiological range in the intestine, simulating the chemical environment in the intestine. This ensured that the sensor was calibrated in a chemical environment similar to that of the actual intestine, improving the accuracy of detecting chemical substance secretion signals. The sensor was placed in a standard environment to collect reference signals without external stimulation.

[0041] Then, the parameters are adjusted and optimized: for pressure sensors, the sensitivity coefficient and offset are adjusted according to the standard pressure source; for peristaltic sensors, the parameters are adjusted using a simulated intestinal peristalsis device to accurately detect the peristaltic frequency and intensity; for neural sensors, the gain and bandwidth are adjusted by simulating a neural electrical activity signal generator; for chemical sensors, a calibration curve is established using a standard chemical solution of known concentration; actual testing is performed using a standard test signal or a simulated environment, and if the requirements are not met, the parameters are readjusted and iteratively calibrated;

[0042] Among them, for pressure sensors, standard pressure sources with different pressure values ​​are used for retesting, and the measurement error and accuracy of the sensor are calculated; for peristaltic sensors, mechanical devices that simulate intestinal peristalsis are used for verification to check the sensor's detection accuracy of different peristaltic patterns; for neural sensors, signal generators that simulate neural electrical activity are used for testing to evaluate the sensor's response to neural signals of different frequencies and amplitudes; for chemical sensors, standard chemical solutions of known concentrations are used for verification to ensure that the error between the sensor's measurement results and the actual concentration is within the allowable range; sensor performance and output signals are continuously monitored, user feedback is collected, and recalibration is triggered in case of abnormalities.

[0043] Preferably, the individual difference adaptation mechanism is constructed: by establishing an individual intestinal function model, it is used to adjust the sensor parameters and acquisition methods according to the physiological characteristics and intestinal function characteristics of different patients; it is also used to adjust the sensor parameters and acquisition methods according to the abnormal characteristics of intestinal dysfunction; for patients with more sensitive intestines, a low-intensity sensor acquisition method is used to avoid excessive stimulation of the intestine; for patients with more complex intestinal functions, the number and types of sensors are increased to improve the comprehensiveness of signal acquisition;

[0044] It is able to learn and analyze the intestinal biological signal data of different patients, and optimize the system's signal acquisition and processing strategies to adapt to individual differences.

[0045] Preferably, the process of constructing the individual difference adaptation mechanism is as follows:

[0046] For different patient groups, collect the basic physiological characteristics of the patients, including age, gender, height, weight, and body mass index. These data can preliminarily reflect the overall physiological state of the patients and may have an impact on intestinal function. Collect the intestinal function-related data of the patients, including intestinal motility habits, defecation regularity, intestinal sensitivity, whether there is a history of intestinal diseases, and family history of intestinal diseases. Among them, intestinal motility habits include daily motility frequency, intensity, and rhythm; defecation regularity includes the number of bowel movements per day and stool characteristics; intestinal sensitivity includes tolerance to food and drugs, and whether there is a history of intestinal diseases including inflammatory bowel disease and irritable bowel syndrome; at the same time, collect the biological signal data of the patient's intestine through multi-source sensors, and collect multiple times at different time periods of fasting and after eating to obtain comprehensive intestinal function status information;

[0047] A machine learning algorithm is used to construct an individual intestinal function model. A recurrent neural network is used to take the collected patient physiological characteristics and intestinal function data as input, and the model is trained to learn the relationship between these data and the intestinal function status, and a mathematical model of the physiological characteristics and intestinal function data and the intestinal function status is established; the collected patient data is divided into a training set and a validation set, and the model is trained using the training set data. By adjusting the parameters of the model, the model is made to accurately predict the intestinal function status; during the training process, the validation set data is used to evaluate the performance of the model, and the accuracy, recall rate, and mean square error indicators of the model prediction are calculated; according to the evaluation results, the structure of the model is adjusted, such as increasing or decreasing the number of layers and nodes of the neural network or optimizing the training algorithm, such as adjusting the learning rate and the number of iterations, to improve the performance and generalization ability of the model; after multiple iterations of training and optimization, the final individual intestinal function model is obtained;

[0048] For patients with more sensitive intestines, the individual intestinal function model will adjust the sensor parameters after judging that the sensitivity is high based on their physiological characteristics and intestinal function data; this includes reducing the measurement range of the pressure sensor so that it can more accurately detect lower-intensity pressure changes and avoid discomfort to the intestine caused by excessive pressure stimulation; at the same time, the sampling frequency of the peristaltic sensor is reduced to reduce interference with intestinal peristalsis and adopt a low-intensity sensor collection method;

[0049] For patients with more complex intestinal functions, the model analyzes their complex intestinal functional manifestations, including abnormal peristalsis and disordered secretion of chemical substances, and decides to increase the number and types of sensors; including adding additional chemical sensors to the original sensors to more accurately monitor the secretion of multiple intestinal hormones or special chemicals; or adding highly sensitive neural sensors to better capture subtle changes in nervous system signals, thereby obtaining more comprehensive intestinal function information;

[0050] When abnormal intestinal function characteristics are detected, the individual intestinal function model adjusts the sensor parameters and acquisition methods in real time according to the type and severity of the abnormal characteristics. Among them, the abnormal characteristics include abnormality, abnormal peristalsis, disordered nerve signals, and chemical imbalance. The real-time adjustment of sensor parameters and acquisition methods includes: if bradyintestinal peristalsis is detected, the model will increase the sampling frequency of the peristaltic sensor to more carefully monitor the peristaltic change trend; at the same time, adjust the sensitivity of the pressure sensor, because bradyintestinal peristalsis may cause changes in pressure distribution in the intestine; if it is an abnormal nerve signal, the model will optimize the signal amplification factor and filtering parameters of the nerve sensor to improve the detection ability of abnormal nerve signals; for abnormal secretion of chemical substances, the model will guide the chemical sensor to switch to a working mode that is more suitable for detecting the abnormal chemical substance: change the detection range or adjust the detection sensitivity, and adjust the acquisition time interval to ensure that the dynamic changes in the concentration of the chemical substance can be accurately captured.

[0051] The present invention provides a biofeedback-based intestinal function adaptive regulation system, which has the following beneficial effects:

[0052] The intestinal function adaptive regulation system based on biofeedback can comprehensively collect intestinal biological signals, including pressure, peristalsis, nervous system signals and chemical substance secretion signals, through the synergistic effect of pressure sensors, peristalsis sensors, neural sensors and chemical sensors. This multi-dimensional monitoring method can more accurately reflect the true state of intestinal function, avoid the limitations of a single sensor, and provide a rich and accurate data basis for subsequent analysis and regulation. Moreover, through sensors based on micro-electromechanical system technology, which have the advantages of high precision, small size and fast response, they can detect tiny pressure changes in the intestine, short-term peristalsis abnormalities, abnormal nervous system signals and abnormal chemical substance secretion. This helps to detect intestinal function problems at an early stage and provides a key basis for timely intervention. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a flowchart of a biofeedback-based intestinal function adaptive regulation system of the present invention. DETAILED DESCRIPTION

[0054] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0055] like Figure 1As shown, the present invention provides a technical solution: a biofeedback-based intestinal function adaptive regulation system, including a signal acquisition module, a data analysis module, a regulation feedback module and a user interaction module;

[0056] The signal acquisition module uses a multi-sensor fusion algorithm and multi-source sensors to collect intestinal function status information; the multi-source sensors include pressure sensors, peristalsis sensors, neural sensors, and chemical sensors; neural sensors are used to detect intestinal nervous system signals; chemical sensors are used to detect the secretion of chemical substances in the intestine;

[0057] The data analysis module analyzes and processes the collected status information, extracts key feature information, determines the intestinal function status, obtains the judgment result, and provides decision-making basis for the adjustment feedback module;

[0058] The regulation feedback module uses a collective regulation method to adaptively regulate intestinal function according to the judgment results of the data analysis module to obtain a regulation feedback effect. The collective regulation method includes electrical stimulation, magnetic stimulation, and biofeedback training to improve the accuracy and effectiveness of regulation.

[0059] The user interaction module is used to interact with users, provide real-time intestinal function status information and regulation feedback effect display, so that users can understand their intestinal health status and conduct self-regulation and management;

[0060] Through the above technical solutions, the problem that the accuracy and comprehensiveness of signal acquisition in the current biofeedback-based intestinal function adaptive regulation system needs to be improved can be effectively solved, providing more reliable technical support for the monitoring and regulation of intestinal health.

[0061] The signal acquisition module includes:

[0062] The acquisition unit uses multi-source sensors to collect intestinal biological signals based on a multi-sensor fusion algorithm to improve the accuracy and comprehensiveness of signal acquisition and collect intestinal biological signals from different angles. The biological signals include pressure signals, peristalsis signals, nervous system signals, and chemical substance secretion signals;

[0063] Among them, the multi-source sensor is based on micro-electromechanical system technology to detect abnormal conditions in the intestine, including detecting tiny pressure changes, short-term peristaltic abnormalities, abnormal nervous system signals, and abnormal secretion of chemical substances; based on micro-electromechanical system technology, small-sized and high-precision sensors can be manufactured, which can more accurately perceive the biological signals of the intestine.

[0064] The multi-source sensor is based on micro-electromechanical system technology. The process of detecting abnormal conditions in the intestine is as follows:

[0065] The pressure sensor based on MEMS technology uses micromachining technology to set sensitive elements on a tiny chip; its core part includes a sensitive film, with a fixed electrode below the sensitive film and direct contact with the pressure environment in the intestine above; when the pressure in the intestine changes, the sensitive film deforms, thereby changing the capacitance value between the sensitive film and the fixed electrode; and the capacitance change is converted into a voltage change by means of a Wheatstone bridge circuit, and then amplified and filtered to obtain an analog electrical signal corresponding to the intestinal pressure; among them, based on the detection of tiny pressure changes, the sensitive film of the pressure sensor is based on silicon material and is manufactured with a circular film through micromachining technology to improve its responsiveness to tiny pressure changes; in terms of signal processing, a high-precision analog-to-digital converter is used to convert the analog electrical signal into a digital signal, and then the difference between adjacent sampling points is calculated to determine whether the pressure has changed slightly; when the difference exceeds the set pressure threshold, it is determined that a tiny pressure change abnormality has been detected, including a sudden increase or decrease in local pressure in the intestine, which is related to problems such as abnormal intestinal peristalsis and intestinal obstruction;

[0066] The peristaltic sensor based on MEMS technology uses a micromechanical structure to sense the mechanical deformation caused by intestinal peristalsis. The sensor contains sensitive elements: micro strain gauges and micro cantilever beams. The sensitive elements are strained under the deformation of the intestinal wall caused by intestinal peristalsis, causing their resistance values ​​to change. Like the pressure sensor, the peristaltic sensor also integrates a signal conditioning circuit to convert the resistance change of the sensitive element into an electrical signal. The electrical signal is amplified and filtered to reflect the intensity, frequency and direction of intestinal peristalsis. When the peristaltic frequency suddenly accelerates or slows down, the peristaltic amplitude increases or decreases abnormally, and the duration is within the range of 2s-16s, it is judged as a short-term peristalsis. Abnormal conditions; indicating intestinal spasms and indigestion problems; by comparing with the preset normal peristaltic pattern, the abnormal pattern is identified using a neural network-based pattern recognition method; after identifying the abnormality, the relevant abnormal characteristics are recorded, including the time when the abnormality occurred, the duration, the frequency and amplitude of the abnormal peristalsis; in order to accurately detect short-term abnormal peristalsis, the peristaltic sensor is designed to focus on the response ability to high-frequency signals; its sampling frequency is set high to capture rapid changes in the intestinal peristalsis process; normal intestinal peristalsis has a certain rhythmicity, and the peristaltic sensor can calculate the average frequency and amplitude of the peristalsis by monitoring the peristaltic signal over a period of time;

[0067] MEMS neural sensors are used to detect enteric nervous system signals. They are electrically coupled to enteric nerve tissue through microelectrode arrays to sense the electrical activity generated by nerve cells. The microelectrodes of the neural sensors are made of platinum and polypyrrole materials, and are treated with oxygen plasma, modified with nerve growth factor, and modified with carbon nanotubes to improve their compatibility with nerve tissue and signal conduction efficiency. The weak electrical signals generated and propagated by the action potential of nerve cells are detected and amplified and filtered. Since the enteric nervous system signals are very weak and have complex time and frequency domain characteristics, the neural sensors need to have high gain and low noise characteristics to ensure that the neural signals can be accurately acquired. When detecting intestinal tract, the collected nerve signals are first analyzed in time domain and frequency domain. The spectral characteristic model of normal intestinal nerve signals is established by calculating the power spectrum density, average frequency and peak frequency of the signals. When the spectral characteristic power of the nerve signals is detected to be abnormally increased or decreased, or the frequency distribution is changed, it is judged as an abnormality of the nervous system signal. The function of the nervous system can also be identified by analyzing the firing pattern of the nerve signals, including abnormal changes in firing frequency and disordered firing rhythm. These abnormalities may be related to intestinal neuropathy and imbalanced nerve regulation. Advanced signal processing algorithms such as wavelet analysis and principal component analysis are used to extract features and recognize patterns of nerve signals to improve the accuracy and reliability of abnormality detection.

[0068] The micro-electromechanical system chemical sensor is used to detect the secretion of chemical substances in the intestine. The chemical sensor has a chemical sensitive membrane. The chemical sensitive membrane is made of a polyvinyl chloride-based membrane material that responds to the change in pH in the intestine. When the chemical substances in the intestine interact with the chemical sensitive membrane, in the chemical sensor based on the electrochemical principle, the chemical substances react with the chemical sensitive membrane, resulting in a change in the electrode surface potential. The change is converted into an electrical signal by the conversion circuit inside the chemical sensor, and then output after amplification and analog-to-digital conversion. Before detection, the chemical sensor is calibrated to determine its response characteristics to chemical substances of different concentrations. By measuring in a standard solution of chemical substances of known concentration, a calibration curve between the sensor output signal and the concentration of the chemical substances is established. During actual detection in the intestine, the signal collected by the sensor is converted into a corresponding chemical substance concentration value according to the calibration curve. By setting a concentration threshold within the normal physiological range, when the concentration of the chemical substance is detected to exceed the threshold range, it is judged as an abnormal secretion of the chemical substance. For example, if excessive gastric acid secretion in the intestine causes the pH value to be lower than the normal range, the chemical sensor can detect this abnormality in time and record the relevant abnormal characteristics, including the type of abnormal chemical substance, the concentration change amplitude, and the duration of the abnormality.

[0069] This information is of great significance for diagnosing intestinal diseases such as gastric ulcers and indigestion. At the same time, the interactions and proportional relationships between different chemicals can also be comprehensively analyzed through multi-source chemical sensors to more comprehensively evaluate the chemical environment and functional status of the intestine.

[0070] The multi-sensor fusion algorithm collects and analyzes the biological signal characteristics of different sensors based on a combination of weighted average and principal component analysis to obtain more accurate intestinal function status information. The process is as follows:

[0071] Pressure sensors, peristaltic sensors, neural sensors and chemical sensors are deployed in the intestine and its surrounding areas to collect biological signals in the intestine in real time, including pressure signals, peristaltic signals, nervous system signals and chemical substance secretion signals; each sensor continuously samples the corresponding biological signal at a sampling frequency of 550, converts the analog signal into a digital signal and transmits it to the subsequent processing unit;

[0072] The weight coefficients were determined in advance according to the importance of each sensor in reflecting the functional status of the intestine: based on experimental research and data analysis, the pressure sensor was given a weight of 0.3, the peristaltic sensor was given a weight of 0.3, the neural sensor was given a weight of 0.2, and the chemical sensor was given a weight of 0.2;

[0073] For the signal values ​​of different sensors collected at the same time, weighted sum is performed according to their respective weight coefficients; let the pressure signal value be P t , the creep signal value is M t , the neural signal value is N t , the chemical signal value is D t , t represents the sampling time, and the fused signal F t The calculation formula is: t =0.3P t +0.3M t +0.2N t +0.2D t ;

[0074] Through weighted average processing, the contribution of each sensor is comprehensively considered to obtain a fusion signal that can preliminarily reflect the overall functional status of the intestine;

[0075] The continuous signal data collected in the past 18 minutes are processed by weighted average to form a fusion signal sequence into a data matrix X, where each row represents the fusion signal value at a sampling moment, and each column represents the same type of signal at different moments, including pressure, peristalsis, mental and chemical weighted signals; the data matrix X is decentralized, that is, the mean of the column is subtracted from each element to obtain the centralized data matrix X'; then the covariance matrix C of X' is calculated, and the element c in the covariance matrixij It represents the covariance between the i-th signal and the j-th signal, reflecting the degree of linear correlation between the two signals;

[0076] Calculate the eigenvalue λ of the covariance matrix C i and the corresponding eigenvector v i ; The eigenvalue indicates the amount of information contained in each principal component, and the eigenvector determines the direction of the principal component; the eigenvalues ​​are arranged in descending order, and the eigenvectors corresponding to the eigenvalues ​​with a cumulative contribution rate of more than 85% are selected to form the principal component transformation matrix V, where each column of the principal component transformation matrix is ​​a principal component vector, which is the projection direction of the original signal in the new orthogonal coordinate system;

[0077] The original fusion signal data matrix X' is projected into the principal component space through the principal component transformation matrix V to obtain the principal component matrix Y, and the calculation formula is Y=X'V; each column of the principal component matrix Y is a principal component, where the first column of principal components contains the largest variance information in the original data, the second column is second, and so on.

[0078] It should be further explained that, in the specific implementation process, by adopting a method combining weighted average and principal component analysis, the biological signal characteristics of different sensors can be effectively integrated, key feature information can be extracted, and the intestinal functional status can be accurately judged; this algorithm can reduce the dimension of the data, reduce noise interference, and improve the accuracy and reliability of data analysis; at the same time, by comparing and matching with the pre-established feature pattern library of normal intestinal functional status and different disease states, and using pattern recognition algorithms for classification and judgment, it can quickly and accurately identify the type and severity of intestinal dysfunction, and provide clear guidance for regulatory feedback.

[0079] The principal component data after dimensionality reduction is subjected to feature extraction and statistical features are calculated, including the mean, variance and slope of the principal component; the extracted features are compared and matched with the pre-established feature pattern library of normal intestinal function status and different disease states, and the support vector machine in the pattern recognition algorithm is used for classification and judgment to obtain the pattern recognition results; among them, if the extracted features are consistent with the feature patterns under the intestinal obstruction disease state, it is judged that the intestine has obstruction; if they are consistent with the feature patterns in the normal digestive process, it is considered that the intestinal function is in a normal state;

[0080] According to the pattern recognition results, the intestinal function status is evaluated to determine whether the intestinal function is normal. If it is abnormal, the abnormal type and severity are determined, among which the abnormal types include abnormal peristalsis, neural regulation imbalance, and chemical environment disorder; the intestinal function status information obtained by the evaluation is output to the user or medical personnel in the form of digital indicators plus graphic display plus text description, so that corresponding treatment or intervention measures can be taken in time, and it also provides a decision-making basis for the subsequent adaptive regulation system; if the intestinal peristalsis is judged to be too slow, the regulation feedback module can start the corresponding stimulation plan based on this information, and can promote intestinal peristalsis through electrical stimulation or biofeedback training;

[0081] It should be further explained that, in the specific implementation process, the data collected by different sensors are fused and processed by designing a sensor fusion algorithm to improve the accuracy and comprehensiveness of signal acquisition.

[0082] Abnormal conditions in the intestine are detected. By setting thresholds and using pattern recognition algorithms, the collected biological signals are analyzed and processed in real time. The signal characteristics of abnormal intestinal function are found and recorded as abnormal characteristics. The corresponding individual difference adaptation mechanism is triggered based on the abnormal characteristics.

[0083] The process of obtaining abnormal characteristics is as follows:

[0084] For pressure signals, the range of normal intestinal pressure is determined based on clinical research and experimental data: in the fasting state, the pressure in the small intestine is between 3.5 mmHg, and after eating, the pressure in the small intestine is 7.5 mmHg; the pressure in the large intestine is between 15 mmHg; based on the range of normal intestinal pressure, the corresponding pressure threshold range is set: the small intestine pressure threshold is 6Hg; the large intestine pressure threshold is 12.5 mmHg. When the collected pressure signal exceeds the pressure threshold range, it indicates that there is an abnormal pressure condition, including increased pressure due to intestinal obstruction, or decreased pressure due to intestinal paralysis.

[0085] For peristaltic signals, determine the frequency and intensity range of normal intestinal peristalsis: Under normal circumstances, the frequency of small intestinal peristalsis is 8 times per minute. In the fasting state, the normal range of small intestinal peristaltic wave intensity is 10 mmHg. After eating, the pressure range of small intestinal peristaltic wave intensity changes is 15 mmHg. In the resting state, the pressure range of large intestinal peristaltic wave intensity changes is 6.5 mmHg. When group peristalsis occurs, the pressure range of peristaltic wave intensity changes is 27.5 mmHg. According to the frequency and intensity range of normal intestinal peristalsis, the upper and lower thresholds of peristaltic frequency and the threshold of peristaltic intensity are set to 8 times / minute; the peristaltic wave intensity of the small intestine is 21.5 mmHg; the peristaltic intensity threshold of the large intestine is 26 mmHg. If the collected peristaltic signal exceeds these thresholds, it may indicate abnormal peristalsis, such as faster peristaltic frequency during spasm, slower peristaltic frequency or weakened peristaltic intensity when power is insufficient.

[0086] For nerve signals, determine the electrical signal characteristics of normal enteric nervous system activity, including the frequency and amplitude of the signal; under normal circumstances, the main frequency of enteric nerve signals is 5.5Hz, and the amplitude is 55μV; set the threshold of electrical signal characteristics to 7.75Hz; the amplitude threshold is 77.5μV; when the frequency and amplitude of nerve signals exceed the normal threshold, it may indicate abnormal nervous system function, including abnormal signal emission caused by neuropathy;

[0087] For chemical signals, the pH value in the intestine is normally 6.75 in the small intestine and 6.25 in the large intestine. Thresholds are set based on these normal pH ranges: the lower threshold of the small intestinal pH value is 5.5 and the upper threshold is 8; the lower threshold of the large intestinal pH value is 5.0 and the upper threshold is 7.5. If the pH value is less than 5 or greater than 8, it is considered abnormal, which may be related to excessive gastric acid secretion or acid-base imbalance in the intestine. For other chemical substances, such as intestinal hormones and digestive enzymes, corresponding thresholds are also set by studying their normal physiological concentration ranges.

[0088] The multi-source sensor continuously collects biological signals in the intestine at a set sampling frequency of 550Hz; after the sensor converts the collected analog signals into digital signals, it compares the signal value of each sampling point with the corresponding threshold; for each pressure signal sampling value, it determines whether it is higher than the upper pressure threshold or lower than the lower pressure threshold; for peristaltic signals, it checks whether its frequency and intensity exceed the set threshold range; for neural signals, it analyzes whether its frequency and amplitude are outside the normal threshold; for chemical signals, it determines whether its concentration value exceeds the set pH threshold; if the signal value of the sampling point exceeds the corresponding threshold, then the signal at that moment is preliminarily marked as abnormal, recorded as an abnormal signal segment, and feature extraction is performed; the time domain features, frequency domain features and time-frequency domain features of the signal are extracted; among which, the time domain features include mean, variance, slope, and zero-crossing rate, the frequency domain features include power spectrum density, main frequency, and frequency band energy, and the time-frequency domain features include wavelet coefficients obtained by wavelet transform;

[0089] The extracted features are input into a pre-trained pattern recognition model to obtain a recognition result, wherein the pattern recognition model uses a variety of algorithms, including support vector machines, multi-layer perceptrons in neural networks, convolutional neural networks, and decision trees; it is trained by a large number of normal and abnormal intestinal biological signal samples, and can learn the characteristic patterns corresponding to different abnormal conditions; it should be further explained that in the specific implementation process, the extracted neural signal features are input into the trained neural network model, and the model determines whether the signal belongs to a specific abnormal nervous system pattern based on the input features, including excessive nerve excitability and nerve conduction block; the model outputs a classification result, indicating the probability that the signal segment belongs to a normal or abnormal type; if the classification result shows that the abnormal probability exceeds the set confidence threshold of 0.8, it is determined that the signal segment is abnormal, and its abnormal characteristics are further determined; according to the recognition result, the specific signal characteristics of intestinal dysfunction are determined and recorded as abnormal characteristics;

[0090] It should be further explained that, in the specific implementation process, if the pattern recognition model determines that the intestinal peristalsis is abnormal and the classification result shows the spasm type, then the abnormal characteristics can be described as detailed information such as "the peristalsis frequency is significantly higher than the normal range, such as 17.5 times / minute, the peristaltic wave intensity is uneven and some peaks are too high, and the duration is about X minutes"; for abnormal nerve signals, the abnormal characteristics may include "the main frequency of the nerve signal is offset to a specific range, such as 12.5Hz, the amplitude fluctuates greatly and irregular spikes appear, and the signal release rhythm is disordered"; these abnormal characteristics will be recorded and used to trigger the individual difference adaptation mechanism later; the intestinal biological signals are continuously monitored in real time, and the above threshold judgment, feature extraction and pattern recognition processes are repeated continuously; because the intestinal functional state may change over time, the previous detection The abnormal situation detected may return to normal, or new abnormal situation may appear; with the accumulation of monitoring data, the judgment of abnormal characteristics is updated and corrected; it should be further explained that in the specific implementation process, if it is initially judged to be a short-term peristaltic abnormality, but subsequent monitoring finds that the abnormality persists and has a tendency to worsen, the description of the abnormal characteristics will be adjusted according to the new data, including updating the abnormal duration, frequency and intensity change trend; at the same time, if it is found that the previous abnormal judgment may be misjudged, such as signal abnormality caused by temporary interference of the sensor, it will be corrected according to the subsequent recovery of normal signals to ensure the accuracy of the abnormal characteristics; this can provide a more reliable basis for the individual difference adaptation mechanism, so as to more accurately adjust the sensor parameters and acquisition methods to adapt to the dynamic changes in intestinal function.

[0091] The system can analyze and process the collected biological signals in real time, and promptly detect the signal characteristics of abnormal intestinal function by setting thresholds and adopting pattern recognition algorithms. This real-time monitoring and analysis capability helps to detect intestinal problems at an early stage, provides a guarantee for timely intervention measures, and improves the timeliness and effectiveness of intestinal function regulation.

[0092] Before using multi-source sensors to collect intestinal biological signals, the sensors are calibrated and optimized by establishing a standard test environment and using an automatic calibration algorithm to ensure that they can maintain stable performance under different environmental conditions and improve the accuracy and reliability of the sensors.

[0093] The process of establishing a standard test environment is as follows:

[0094] The temperature was controlled at 18.75℃ and the humidity was 50%. Copper foil was used as electromagnetic shielding material to create an electromagnetic interference-free space. Silicone material was used to simulate the mechanical properties of the intestinal wall. A solution containing appropriate amounts of sodium chloride, potassium chloride, and sodium bicarbonate was used to make its pH value close to the normal physiological range in the intestine to simulate the chemical environment in the intestine. This ensured that the sensor was calibrated in a chemical environment similar to that of the actual intestine, thereby improving the accuracy of detecting chemical substance secretion signals. The sensor was placed in a standard environment to collect reference signals without external stimulation.

[0095] Then, the parameters are adjusted and optimized: for pressure sensors, the sensitivity coefficient and offset are adjusted according to the standard pressure source; for peristaltic sensors, the parameters are adjusted using a simulated intestinal peristalsis device to accurately detect the peristaltic frequency and intensity; for neural sensors, the gain and bandwidth are adjusted by simulating a neural electrical activity signal generator; for chemical sensors, a calibration curve is established using a standard chemical solution of known concentration; actual testing is performed using a standard test signal or a simulated environment, and if the requirements are not met, the parameters are readjusted and iteratively calibrated;

[0096] Among them, for pressure sensors, standard pressure sources with different pressure values ​​are used for retesting, and the measurement error and accuracy of the sensor are calculated; for peristaltic sensors, mechanical devices that simulate intestinal peristalsis are used for verification to check the sensor's detection accuracy of different peristaltic patterns; for neural sensors, signal generators that simulate neural electrical activity are used for testing to evaluate the sensor's response to neural signals of different frequencies and amplitudes; for chemical sensors, standard chemical solutions of known concentrations are used for verification to ensure that the error between the sensor's measurement results and the actual concentration is within the allowable range; sensor performance and output signals are continuously monitored, user feedback is collected, and recalibration is triggered in case of abnormalities.

[0097] Constructing individual difference adaptation mechanism: By establishing an individual intestinal function model, it is used to adjust sensor parameters and acquisition methods according to the physiological characteristics and intestinal function characteristics of different patients; it is also used to adjust sensor parameters and acquisition methods according to the abnormal characteristics of intestinal dysfunction; for patients with more sensitive intestines, a low-intensity sensor acquisition method is used to avoid excessive stimulation of the intestine; for patients with more complex intestinal functions, the number and types of sensors are increased to improve the comprehensiveness of signal acquisition;

[0098] It is able to learn and analyze the intestinal biological signal data of different patients, and optimize the system's signal acquisition and processing strategies to adapt to individual differences.

[0099] The process of constructing individual difference adaptation mechanisms is as follows:

[0100] For different patient groups, collect the basic physiological characteristics of the patients, including age, gender, height, weight, and body mass index. These data can preliminarily reflect the overall physiological state of the patients and may have an impact on intestinal function. Collect the intestinal function-related data of the patients, including intestinal motility habits, defecation regularity, intestinal sensitivity, whether there is a history of intestinal diseases, and family history of intestinal diseases. Among them, intestinal motility habits include daily motility frequency, intensity, and rhythm; defecation regularity includes the number of bowel movements per day and stool characteristics; intestinal sensitivity includes tolerance to food and drugs, and whether there is a history of intestinal diseases including inflammatory bowel disease and irritable bowel syndrome; at the same time, collect the biological signal data of the patient's intestine through multi-source sensors, and collect multiple times at different time periods of fasting and after eating to obtain comprehensive intestinal function status information;

[0101] A machine learning algorithm is used to construct an individual intestinal function model. A recurrent neural network is used to take the collected patient physiological characteristics and intestinal function data as input, and the model is trained to learn the relationship between these data and the intestinal function status, and a mathematical model of the physiological characteristics and intestinal function data and the intestinal function status is established; the collected patient data is divided into a training set and a validation set, and the model is trained using the training set data. By adjusting the parameters of the model, the model is made to accurately predict the intestinal function status; during the training process, the validation set data is used to evaluate the performance of the model, and the accuracy, recall rate, and mean square error indicators of the model prediction are calculated; according to the evaluation results, the structure of the model is adjusted, such as increasing or decreasing the number of layers and nodes of the neural network or optimizing the training algorithm, such as adjusting the learning rate and the number of iterations, to improve the performance and generalization ability of the model; after multiple iterations of training and optimization, the final individual intestinal function model is obtained;

[0102] For patients with more sensitive intestines, the individual intestinal function model will adjust the sensor parameters after judging that the sensitivity is high based on their physiological characteristics and intestinal function data; this includes reducing the measurement range of the pressure sensor so that it can more accurately detect lower-intensity pressure changes and avoid discomfort to the intestine caused by excessive pressure stimulation; at the same time, the sampling frequency of the peristaltic sensor is reduced to reduce interference with intestinal peristalsis and adopt a low-intensity sensor collection method;

[0103] For patients with more complex intestinal functions, the model analyzes their complex intestinal functional manifestations, including abnormal peristalsis and disordered secretion of chemical substances, and decides to increase the number and types of sensors; including adding additional chemical sensors to the original sensors to more accurately monitor the secretion of multiple intestinal hormones or special chemicals; or adding highly sensitive neural sensors to better capture subtle changes in nervous system signals, thereby obtaining more comprehensive intestinal function information;

[0104] When abnormal intestinal function characteristics are detected, the individual intestinal function model adjusts the sensor parameters and acquisition methods in real time according to the type and severity of the abnormal characteristics. Among them, the abnormal characteristics include abnormality, abnormal peristalsis, disordered nerve signals, and chemical imbalance. The real-time adjustment of sensor parameters and acquisition methods includes: if bradyintestinal peristalsis is detected, the model will increase the sampling frequency of the peristaltic sensor to more carefully monitor the peristaltic change trend; at the same time, adjust the sensitivity of the pressure sensor, because bradyintestinal peristalsis may cause changes in pressure distribution in the intestine; if it is an abnormal nerve signal, the model will optimize the signal amplification factor and filtering parameters of the nerve sensor to improve the detection ability of abnormal nerve signals; for abnormal secretion of chemical substances, the model will guide the chemical sensor to switch to a working mode that is more suitable for detecting the abnormal chemical substance: change the detection range or adjust the detection sensitivity, and adjust the acquisition time interval to ensure that the dynamic changes in the concentration of the chemical substance can be accurately captured.

[0105] It should be further explained that, in the specific implementation process, by establishing an individual intestinal function model, the system can adjust the sensor parameters and acquisition methods according to the physiological characteristics and intestinal function characteristics of different patients, as well as the specific circumstances of intestinal dysfunction; for patients with more sensitive intestines, a low-intensity sensor acquisition method is used to avoid excessive stimulation of the intestine; for patients with more complex intestinal function, the number and types of sensors are increased to more comprehensively understand the intestinal function status; this personalized adjustment method can better adapt to the needs of different patients and improve the targetedness and effectiveness of the adjustment.

[0106] The sensor uses advanced materials and surface treatment processes. For example, the microelectrodes of the neural sensor are made of platinum and polypyrrole materials, and are treated with oxygen ions, modified with nerve growth factor and carbon nanotubes to improve the performance and biocompatibility of the sensor. The use of electromagnetic shielding materials creates a space free of electromagnetic interference, ensuring the accuracy of signal acquisition. These advanced materials and technologies improve the reliability and stability of the system, providing a strong guarantee for intestinal function monitoring and regulation.

[0107] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. A biofeedback-based intestinal function adaptive regulation system, characterized in that: It includes a signal acquisition module, a data analysis module, an adjustment feedback module and a user interaction module; The signal acquisition module uses a multi-sensor fusion algorithm and multi-source sensors to acquire intestinal function status information; wherein the multi-source sensors include pressure sensors, peristalsis sensors, neural sensors, and chemical sensors; neural sensors are used to detect intestinal nervous system signals; chemical sensors are used to detect the secretion of chemical substances in the intestine; The data analysis module analyzes and processes the collected status information, extracts key feature information, determines the intestinal function status, and obtains a determination result; The regulating feedback module uses a collective regulating method to adaptively regulate the intestinal function according to the judgment result of the data analysis module to obtain a regulating feedback effect. The collective regulating method includes electrical stimulation, magnetic stimulation, and biofeedback training. The user interaction module is used to interact with the user and provide real-time intestinal function status information and regulation feedback effect display; The multi-sensor fusion algorithm collects and analyzes the biological signal characteristics of different sensors based on a method combining weighted average and principal component analysis to obtain intestinal function status information. The process is as follows: Pressure sensors, peristaltic sensors, neural sensors and chemical sensors are deployed in the intestine and its surrounding areas to collect biological signals in the intestine in real time, including pressure signals, peristaltic signals, nervous system signals and chemical substance secretion signals; each sensor continuously samples the corresponding biological signals at a sampling frequency range of 100Hz-1000Hz, converts the analog signals into digital signals and transmits them to the subsequent processing unit; The weight coefficients were determined in advance according to the importance of each sensor in reflecting the functional status of the intestine: based on experimental research and data analysis, the pressure sensor was given a weight of 0.3, the peristaltic sensor was given a weight of 0.3, the neural sensor was given a weight of 0.2, and the chemical sensor was given a weight of 0.2; For the signal values ​​of different sensors collected at the same time, weighted sum is performed according to their respective weight coefficients; assuming that the pressure signal value is , the peristaltic signal value is , the neural signal value is , the chemical signal value is , t represents the sampling time, and the fused signal The calculation formula is: ; The continuous signal data collected in the past 6 minutes or 30 minutes are processed by weighted average processing to form a fusion signal sequence into a data matrix X, where each row represents the fusion signal value at a sampling moment, and each column represents the same type of signal at different moments, including pressure, peristalsis, nerve and chemical weighted signals; the data matrix X is decentralized, that is, the mean of the column is subtracted from each element to obtain the centralized data matrix X'; then the covariance matrix C of X' is calculated, and the element c in the covariance matrix ij It represents the covariance between the i-th signal and the j-th signal, reflecting the degree of linear correlation between the two signals; Calculate the eigenvalue λ of the covariance matrix C i and the corresponding eigenvector v i ; The eigenvalue indicates the amount of information contained in each principal component, and the eigenvector determines the direction of the principal component; the eigenvalues ​​are arranged in descending order, and the eigenvectors corresponding to the eigenvalues ​​with a cumulative contribution rate of more than 85% are selected to form the principal component transformation matrix V, where each column of the principal component transformation matrix is ​​a principal component vector, which is the projection direction of the original signal in the new orthogonal coordinate system; The original fusion signal data matrix X' is projected into the principal component space through the principal component transformation matrix V to obtain the principal component matrix Y, which is calculated as follows: ; Each column of the principal component matrix Y is a principal component, where the first column of principal components contains the largest variance information in the original data, the second column is second, and so on; The principal component data after dimensionality reduction is subjected to feature extraction and statistical features are calculated, including the mean, variance and slope of the principal component; the extracted features are compared and matched with the pre-established feature pattern library of normal intestinal function status and different disease states, and the support vector machine in the pattern recognition algorithm is used for classification and judgment to obtain the pattern recognition results; among them, if the extracted features are consistent with the feature patterns under the intestinal obstruction disease state, it is judged that the intestine has obstruction; if they are consistent with the feature patterns in the normal digestive process, it is considered that the intestinal function is in a normal state; Based on the pattern recognition results, the intestinal function status is evaluated to determine whether the intestinal function is normal. If it is abnormal, the type and severity of the abnormality are determined. The abnormality types include abnormal peristalsis, neural regulation imbalance, and chemical environment disorder. The intestinal function status information obtained from the evaluation is output to the user or medical staff in the form of digital indicators, graphical display, and text description.

2. The intestinal function adaptive regulation system based on biofeedback according to claim 1, characterized in that: The signal acquisition module comprises: The acquisition unit uses multi-source sensors to collect intestinal biological signals based on a multi-sensor fusion algorithm. The biological signals include pressure signals, peristalsis signals, nervous system signals, and chemical substance secretion signals. Among them, the multi-source sensor is based on micro-electromechanical system technology to detect abnormal conditions in the intestine, including detecting small pressure changes, transient peristaltic abnormalities, abnormal nervous system signals, and abnormal secretion of chemical substances.

3. The intestinal function adaptive regulation system based on biofeedback according to claim 2, characterized in that: The multi-source sensor is based on micro-electromechanical system technology, and the process of detecting abnormal conditions in the intestine is as follows: The pressure sensor based on micro-electromechanical system technology uses micro-machining technology to set sensitive elements on a tiny chip; its core part includes a sensitive film, a fixed electrode below the sensitive film, and direct contact with the pressure environment in the intestine above; when the pressure in the intestine changes, the sensitive film deforms, thereby changing the capacitance value between the sensitive film and the fixed electrode; and the capacitance change is converted into a voltage change by means of a Wheatstone bridge circuit, and then amplified and filtered to obtain an analog electrical signal corresponding to the intestinal pressure; among them, based on the detection of tiny pressure changes, the sensitive film of the pressure sensor is based on silicon material and is manufactured with a circular film through micro-machining technology; in terms of signal processing, an analog-to-digital converter is used to convert the analog electrical signal into a digital signal, and then the difference between adjacent sampling points is calculated to determine whether the pressure has changed slightly; when the difference exceeds the set pressure threshold, it is determined that a tiny pressure change abnormality has been detected, including a sudden increase or decrease in local pressure in the intestine; The peristaltic sensor based on micro-electromechanical system technology uses micro-mechanical structure to sense the mechanical deformation caused by intestinal peristalsis. The sensor contains sensitive elements: micro strain gauges and micro cantilever beams. The sensitive elements are strained under the deformation of the intestinal wall caused by intestinal peristalsis, which causes their resistance value to change. Like the pressure sensor, the peristaltic sensor also integrates a signal conditioning circuit to convert the resistance change of the sensitive element into an electrical signal. The electrical signal is amplified and filtered to reflect the intensity, frequency and direction of intestinal peristalsis. When the peristaltic frequency suddenly speeds up or slows down, the peristaltic amplitude increases or decreases abnormally, and the duration is within the range of 2s-16s, it is judged as a short-term abnormal peristalsis, indicating that the intestine has spasms and indigestion problems. By comparing with the preset normal peristaltic pattern, the abnormal pattern is identified using a neural network-based pattern recognition method. After the abnormality is identified, the relevant abnormal characteristics are recorded, including the time when the abnormality occurred, the duration, the frequency and amplitude of the abnormal peristalsis. The MEMS neural sensor is used to detect intestinal nervous system signals. It electrically couples with intestinal nerve tissue through a microelectrode array to sense the electrical activity generated by nerve cells. The microelectrodes of the neural sensor are made of platinum and polypyrrole materials, and are treated with oxygen plasma, modified with nerve growth factor and modified with carbon nanotubes. The weak electrical signals generated and propagated by the action potential of nerve cells are detected, and then amplified and filtered. When detecting abnormalities in nervous system signals, the collected nerve signals are first analyzed in the time domain and frequency domain. The frequency spectrum characteristic model of normal intestinal nerve signals is established by calculating the power spectrum density, average frequency and peak frequency of the signal. When it is detected that the frequency spectrum characteristic power of the nerve signal increases or decreases abnormally, or the frequency distribution changes, it is judged as an abnormality in the nervous system signal. The micro-electromechanical system chemical sensor is used to detect the secretion of chemical substances in the intestine. The chemical sensor has a chemical sensitive membrane; the chemical sensitive membrane is made of a polyvinyl chloride-based membrane material that responds to changes in the pH value in the intestine; when the chemical substances in the intestine interact with the chemical sensitive membrane, in the chemical sensor based on the electrochemical principle, the chemical substances react with the chemical sensitive membrane, resulting in a change in the electrode surface potential; the change is converted into an electrical signal by the conversion circuit inside the chemical sensor, and then output after amplification and analog-to-digital conversion; before detection, the chemical sensor is calibrated to determine its response characteristics to chemical substances of different concentrations; by measuring in a standard solution of chemical substances of known concentration, a calibration curve between the sensor output signal and the chemical substance concentration is established; during actual detection in the intestine, the signal collected by the sensor is converted into a corresponding chemical substance concentration value according to the calibration curve; by setting a concentration threshold within the normal physiological range, when the concentration of the chemical substance is detected to exceed the threshold range, it is judged as an abnormal secretion of the chemical substance.

4. The intestinal function adaptive regulation system based on biofeedback according to claim 3, characterized in that: Abnormal conditions in the intestine are detected, and the collected biological signals are analyzed and processed in real time by setting thresholds and using pattern recognition algorithms. The signal characteristics of abnormal intestinal function are found and recorded as abnormal characteristics. The corresponding individual difference adaptation mechanism is triggered based on the abnormal characteristics.

5. The intestinal function adaptive regulation system based on biofeedback according to claim 4, characterized in that: The process of obtaining the abnormal characteristics is as follows: For pressure signals, based on clinical research and experimental data, the range of normal pressure in the intestine is determined: in the fasting state, the pressure in the small intestine is between 2-5 mmHg, and after eating, the pressure in the small intestine is between 5-10 mmHg; the pressure in the large intestine is between 10-20 mmHg; based on the range of normal pressure in the intestine, the corresponding pressure threshold range is set: the upper threshold of the small intestine pressure threshold is set to 12 mmHg, and the lower threshold is set to 0 mmHg; the upper threshold of the large intestine pressure threshold is set to 25 mmHg, and the lower threshold is set to 0 mmHg; For peristaltic signals, determine the frequency and intensity range of normal intestinal peristalsis: Under normal circumstances, the frequency of small intestinal peristalsis is 3-12 times per minute. In the fasting state, the normal range of small intestinal peristaltic wave intensity is 5-15 mmHg. After eating, the small intestinal peristaltic wave intensity pressure changes in the range of 10-30 mmHg. In the resting state, the peristaltic wave intensity pressure of the large intestine changes in the range of 3-10 mmHg. When group peristalsis occurs, the peristaltic wave intensity pressure changes in the range of 15-40 mmHg; according to the frequency and intensity range of normal intestinal peristalsis, set the upper and lower thresholds of peristaltic frequency and the threshold of peristaltic intensity: the upper limit of frequency is 15 times / minute, and the lower limit is 1 time / minute; the lower limit threshold of small intestinal peristaltic wave intensity is 3 mmHg, and the upper limit threshold is 40 mmHg; the lower limit threshold of large intestinal peristaltic intensity threshold is 2 mmHg, and the upper limit threshold is 50 mmHg; For nerve signals, determine the electrical signal characteristics of normal enteric nervous system activity, including the frequency and amplitude of the signal; under normal circumstances, the main frequency range of enteric nerve signals is 1-10Hz, and the amplitude is in the range of 10-100μV; set the threshold range of electrical signal characteristics: the lower limit threshold of the frequency threshold range is 0.5Hz, and the upper limit threshold is 15Hz; the lower limit threshold of the amplitude threshold range is 5μV, and the upper limit threshold is 150μV; For chemical signals, the pH value in the intestine is normally between 6 and 7.5 in the small intestine and between 5.5 and 7 in the large intestine. The thresholds are set according to these normal pH ranges: the lower threshold of the small intestinal pH value is 5.5 and the upper threshold is 8; the lower threshold of the large intestinal pH value is 5.0 and the upper threshold is 7.

5. The multi-source sensor continuously collects biological signals in the intestine at a set sampling frequency of 550Hz. After the sensor converts the collected analog signals into digital signals, it compares the signal value of each sampling point with the corresponding threshold. For each pressure signal sampling value, it determines whether it is higher than the upper pressure threshold or lower than the lower pressure threshold. For peristaltic signals, it checks whether its frequency and intensity exceed the set threshold range. For neural signals, it analyzes whether its frequency and amplitude are outside the normal threshold. For chemical signals, determine whether their concentration value exceeds the set pH threshold; If the signal value of the sampling point exceeds the corresponding threshold, the signal at that moment is initially marked as abnormal, recorded as an abnormal signal segment, and feature extraction is performed; the time domain features, frequency domain features, and time-frequency domain features of the signal are extracted; among which, the time domain features include mean, variance, slope, and zero-crossing rate, the frequency domain features include power spectrum density, main frequency, and frequency band energy, and the time-frequency domain features include wavelet coefficients obtained by wavelet transform; The extracted features are input into a pre-trained pattern recognition model to obtain a recognition result, and based on the recognition result, specific signal features of intestinal dysfunction are determined and recorded as abnormal features.

6. The intestinal function adaptive regulation system based on biofeedback according to claim 5, characterized in that: Before using multi-source sensors to collect intestinal biological signals, the sensors are calibrated and optimized by establishing a standard test environment and using an automatic calibration algorithm to ensure that they can maintain stable performance under different environmental conditions.

7. The intestinal function adaptive regulation system based on biofeedback according to claim 6, characterized in that: The process of establishing a standard test environment is as follows: The temperature was controlled at 37°C ± 0.5°C and the humidity was 40%-60%. Copper foil was used as an electromagnetic shielding material to create an electromagnetic interference-free space. Silica gel was used to simulate the mechanical properties of the intestinal wall. A solution containing sodium chloride, potassium chloride, and sodium bicarbonate was used to make its pH value close to the normal physiological range in the intestine to simulate the chemical environment in the intestine. The sensor was placed in a standard environment to collect a reference signal without external stimulation. Then, the parameters are adjusted and optimized: for pressure sensors, the sensitivity coefficient and offset are adjusted according to the standard pressure source; for peristaltic sensors, the parameters are adjusted using a simulated intestinal peristalsis device to accurately detect the peristaltic frequency and intensity; for neural sensors, the gain and bandwidth are adjusted by simulating a neural electrical activity signal generator; for chemical sensors, a calibration curve is established using a standard chemical solution of known concentration; actual testing is performed using a standard test signal or a simulated environment, and if the requirements are not met, the parameters are readjusted and iteratively calibrated; Among them, for pressure sensors, standard pressure sources with different pressure values ​​are used for retesting, and the measurement error and accuracy of the sensor are calculated; for peristaltic sensors, mechanical devices that simulate intestinal peristalsis are used for verification to check the sensor's detection accuracy of different peristaltic patterns; for neural sensors, signal generators that simulate neural electrical activity are used for testing to evaluate the sensor's response to neural signals of different frequencies and amplitudes; for chemical sensors, standard chemical solutions of known concentrations are used for verification to ensure that the error between the sensor's measurement results and the actual concentration is within the allowable range; sensor performance and output signals are continuously monitored, user feedback is collected, and recalibration is triggered in case of abnormalities.

8. The intestinal function adaptive regulation system based on biofeedback according to claim 7, characterized in that: Constructing the individual difference adaptation mechanism: by establishing an individual intestinal function model, it is used to adjust the sensor parameters and acquisition methods according to the physiological characteristics and intestinal function characteristics of different patients; it is also used to adjust the sensor parameters and acquisition methods according to the abnormal characteristics of intestinal dysfunction; for patients with more sensitive intestines, a low-intensity sensor acquisition method is used; for patients with more complex intestinal functions, the number and types of sensors are increased.

9. The intestinal function adaptive regulation system based on biofeedback according to claim 8, characterized in that: The process of constructing individual difference adaptation mechanisms is as follows: For different patient groups, we collect the basic physiological characteristics of the patients, including age, gender, height, weight, and body mass index. We also collect the intestinal function-related data of the patients, including intestinal motility habits, defecation regularity, intestinal sensitivity, history of intestinal diseases, and family history of intestinal diseases. At the same time, we collect the biological signal data of the patients' intestines through multi-source sensors, and collect data multiple times at different time periods when fasting and after eating. An individual intestinal function model is constructed using a machine learning algorithm. A recurrent neural network is used to take the collected patient physiological characteristics and intestinal function data as input, and the model is trained to learn the relationship between these data and the intestinal function state, and a mathematical model of the physiological characteristics and intestinal function data and the intestinal function state is established; the collected patient data is divided into a training set and a validation set, and the model is trained using the training set data, and the parameters of the model are adjusted; during the training process, the performance of the model is evaluated using the validation set data, and the accuracy, recall rate, and mean square error indicators of the model prediction are calculated; based on the evaluation results, the structure of the model is adjusted or the training algorithm is optimized to improve the performance and generalization ability of the model; After multiple iterations of training and optimization, the final individual intestinal function model was obtained; For patients with more sensitive intestines, the individual intestinal function model will adjust the sensor parameters after judging that the sensitivity is high based on their physiological characteristics and intestinal function data; this includes reducing the measurement range of the pressure sensor so that it can more accurately detect lower-intensity pressure changes and avoid discomfort to the intestine caused by excessive pressure stimulation; at the same time, the sampling frequency of the peristaltic sensor is reduced to reduce interference with intestinal peristalsis and adopt a low-intensity sensor collection method; For patients with more complex intestinal functions, the model analyzes their complex intestinal functional manifestations, including abnormal peristalsis and disordered secretion of chemical substances, and decides to increase the number and type of sensors; including adding additional chemical sensors on the basis of the original sensors to more accurately monitor the secretion of multiple intestinal hormones or special chemicals; or adding highly sensitive neural sensors; When abnormal intestinal function characteristics are detected, the individual intestinal function model adjusts the sensor parameters and acquisition methods in real time according to the type and severity of the abnormal characteristics; the abnormal characteristics include abnormalities, abnormal peristalsis, disordered nerve signals, and chemical imbalance; real-time adjustment of sensor parameters and acquisition methods includes: if slow intestinal peristalsis is detected, the model will increase the sampling frequency of the peristaltic sensor; at the same time, adjust the sensitivity of the pressure sensor; if it is an abnormal nerve signal, the model will optimize the signal amplification factor and filtering parameters of the nerve sensor. For abnormal secretion of chemical substances, the model will guide the chemical sensor to switch to a working mode that is more suitable for detecting the abnormal chemical substance: change the detection range or adjust the detection sensitivity, and adjust the acquisition time interval.

Images