Gastrointestinal function evaluation method and system

Abstract

The invention relates to the technical field of gastrointestinal data analysis, in particular to a gastrointestinal function evaluation method and system.The gastrointestinal function evaluation method comprises the following steps that gastrointestinal tract biological reaction information including an intragastric pressure signal and a peristaltic rate signal is collected and divided into independent time sequences, and according to the curvature change condition in each time sequence, the gastrointestinal function is evaluated; and analyzing the signal change trend and adjusting the fitting curve to generate curvature data of gastrointestinal reaction. According to the method, time sequence curvature change analysis is carried out on the intragastric pressure signal and the peristalsis rate signal, so that the dynamic trend and the potential abnormal state are disclosed more intuitively. Meanwhile, amplitude and frequency information of an abnormal fluctuation stage is extracted, so that acute changes of gastrointestinal reactions, such as violent pressure fluctuation or high-frequency wriggling, can be sensitively captured, and an important data basis is provided for state monitoring of gastrointestinal tracts. On the basis of considering external stimulation data such as feeding and movement, time delay characteristics of gastrointestinal reaction are analyzed, and comprehensive data support is provided for stress state monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of gastrointestinal data analysis, and in particular to a gastrointestinal function evaluation method and system. Background Art

[0002] Gastrointestinal data analysis collects and processes gastrointestinal data to analyze and understand the functional status of the digestive system. The field uses data analysis, machine learning, and bioinformatics to extract valuable information from various data sources (such as imaging data, biomarkers, and motor function data), covering multiple links such as data collection, preprocessing, modeling, analysis, and interpretation.

[0003] Among them, the gastrointestinal function assessment method aims to provide data support for the early diagnosis, disease assessment and treatment optimization of gastrointestinal diseases. The main assessment directions include gastrointestinal motility function, digestion and absorption efficiency, and the health status of the intestinal microbiota. Through accurate data analysis of gastrointestinal function, medical staff can more intuitively understand gastrointestinal function, identify potential lesion risks, and thus better formulate prevention and treatment measures. For example, gastrointestinal motility data can be used to monitor digestion speed, and biomarkers and microbiome data can be used to evaluate digestion, absorption and intestinal health.

[0004] Existing technologies mostly use static data collection and analysis methods in gastrointestinal function data analysis. In the case of sharp or high-frequency fluctuations, it is easy to fail to identify abnormal fluctuations in a timely manner due to the limitations of the data processing model, affecting the accurate judgment of potential risks. At the same time, in terms of monitoring gastrointestinal reactions caused by external stimuli, existing technologies are difficult to fully reflect the process characteristics of physiological changes when monitoring the gastrointestinal tract's stress response to stimuli such as eating and exercise. For the analysis of changes in the concentration of digestive enzymes and metabolites, including the dynamic changes in the detailed time dimension, it is difficult to accurately reflect the gradual changes in the digestion and absorption process, affecting the accuracy of the diagnosis. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a gastrointestinal function evaluation method and system.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a method for evaluating gastrointestinal function, comprising the following steps:

[0007] S1: Gastrointestinal biological response information including gastric pressure signal and peristaltic rate signal is collected and divided into independent time series. According to the change of curvature in each time series, the signal change trend is analyzed and the fitting curve is adjusted to generate the curvature data of gastrointestinal response;

[0008] S2: Based on the amplitude and frequency information of the curvature fluctuation in the curvature data of the gastrointestinal reaction, by analyzing the critical point of the curvature change, extracting the information of the abnormal fluctuation stage, combining the data of the physiological changes of the stomach caused by the external stimulation, determining the delay of the signal from the stimulation to the stability, and generating the delay characteristic data of the gastrointestinal reaction;

[0009] S3: Based on the time-delay characteristic data of the gastrointestinal reaction, classify the data of the characteristic change stage in the gastrointestinal reaction into multiple functional state sets, and quantify the transition frequency between states to generate a gastrointestinal state characteristic analysis result;

[0010] S4: Collect the concentration change data of digestive enzymes and metabolites, extract the time series characteristics of the concentration changes, compare the extracted time series characteristics of the concentration changes with the gastrointestinal state characteristic analysis results, and establish the gastrointestinal function evaluation results by comparing the concentration change amplitudes under each state.

[0011] As a further solution of the present invention, the step of acquiring the change trend of the analysis signal is specifically:

[0012] S111: Based on the collected intragastric pressure signal and peristaltic rate signal, the signals are quantified and mapped to a unified standard scale, and noise elimination is performed, and the denoised signal data is divided into independent time series according to the time window to obtain standardized time series data;

[0013] S112: According to the standardized time series data, the formula is used:

[0014]

[0015] Calculate the curvature value K and obtain a curvature value sequence;

[0016] Among them, y i is time point x i The signal value on y i+1 and i-1 is the adjacent time point x i+1 and x i-1 The signal value on, Δx is the time interval;

[0017] S113: According to the curvature value sequence, compared with the conventional curvature range, monitoring signal fluctuation by drawing a curvature change trend graph, gradually adjusting the fitting curve order to optimize the smoothness of the curvature change, and obtaining the curvature data of the gastrointestinal reaction.

[0018] As a further solution of the present invention, the step of obtaining the critical point of analyzing the curvature change is specifically:

[0019] S211: According to the amplitude and frequency information of the curvature fluctuation in the curvature data of the gastrointestinal reaction, the formula is used:

[0020]

[0021] Calculate the standardized curvature change amplitude ΔK and the standardized time interval T of adjacent extreme value points respectively, and obtain the curvature change amplitude and the time interval data of adjacent extreme value points;

[0022] Among them, K t+1 and K t They represent the curvature values ​​at time t+1 and time t respectively, D represents the distance change within the time period, t t+1 and t t represents the timestamps of adjacent extreme points, at times t+1 and t, respectively, and F represents the normalized frequency of the peristalsis rate;

[0023] S212: Analyze whether the value exceeds the normal fluctuation range according to the curvature change amplitude and the time interval data of adjacent extreme value points, determine the critical point of the curvature change and mark the abnormal fluctuation stage, and obtain the critical point information of the curvature change.

[0024] As a further solution of the present invention, the step of obtaining the time delay of the determined signal from stimulation to stabilization is specifically:

[0025] S221: extracting the difference between adjacent peak values ​​of intragastric pressure, the time interval between adjacent extreme values ​​of peristaltic rate and peristaltic frequency according to the critical point information of the curvature change, identifying the abnormal fluctuation stage, and obtaining an abnormal fluctuation data set;

[0026] S222: According to the abnormal fluctuation data set, the formula is used:

[0027]

[0028] Calculate the delay time τ of gastrointestinal reaction to obtain gastrointestinal physiological response data;

[0029] Wherein, ΔT′ represents the total reaction time of the gastrointestinal signal change caused by external stimulation, S represents the stable time period of the signal, M represents the intensity value of the external stimulation, E represents the frequency of the external stimulation, W represents the fluctuation value of the peristaltic rate, and D′ represents the fluctuation amplitude of the intragastric pressure;

[0030] S223: According to the gastrointestinal physiological response data, the data groups are divided according to average duration, high and low frequencies, and time stability to perform clustering operations and generate time-delay characteristic data of gastrointestinal reactions.

[0031] As a further solution of the present invention, the step of obtaining the plurality of function status sets is specifically as follows:

[0032] S311: Based on the time-delay characteristic data of the gastrointestinal reaction, according to the change of the reaction time and the peristaltic rate, identifying the stages of change and stability in the data, and obtaining data information of each characteristic change stage;

[0033] S312: Based on the data information of each characteristic change stage, with reference to the reaction time, frequency and stability data, each change stage is matched with a peristaltic state set to obtain a peristaltic state data set.

[0034] As a further solution of the present invention, the step of acquiring the transition frequency between the quantization states is specifically:

[0035] S321: Based on the data set of the peristaltic state, the formula is used:

[0036]

[0037] Calculate the transition frequency f from state set a to state set b ab , get the frequency data of transition between peristaltic states;

[0038] Among them, N ab represents the total number of transitions from state set a to state set b within a period of time, V ab represents the time variability of the transition from state a to state b, Q ab Represents the offset from state set a to state set b, N a represents the total number of occurrences of state set a, E a represents the energy of state a, D b and D a Respectively represent the duration of state sets b and a;

[0039] S322: Based on the transfer frequency data between the peristaltic states, a state characteristic matrix is ​​constructed according to the arrangement order of the states, each state set is mapped to the row and column positions of the matrix and the transfer frequency is filled in, the transfer paths of multiple frequencies are identified, and the gastrointestinal state characteristic analysis results are obtained.

[0040] As a further solution of the present invention, the step of obtaining the concentration variation range under each state is specifically as follows:

[0041] S411: extracting time series features based on the concentration change data of digestive enzymes and metabolites, identifying concentration fluctuation and stability features, and comparing them with the gastrointestinal state feature analysis results to obtain a concentration change feature set;

[0042] S412: According to the concentration change feature set, the formula is used:

[0043]

[0044] Calculate the average difference C between the concentration changes in state a and state b ab , quantify the difference in digestion or absorption capacity between states and obtain gastrointestinal function assessment results;

[0045] Among them, H represents the number of time points in the total observation period, c h,a and c h,b Respectively represent the concentration values ​​in states a and b at time h.

[0046] A gastrointestinal function evaluation system, the gastrointestinal function evaluation system is used to perform the above-mentioned gastrointestinal function evaluation method, the system comprising:

[0047] The biological signal acquisition module monitors and analyzes the curvature changes of the intragastric pressure signal and the peristaltic rate signal in real time based on the collected gastrointestinal biological reaction signals through independent time series division, adjusts the signal fitting curve, and generates the curvature data of the gastrointestinal reaction;

[0048] The curvature data analysis module identifies and extracts data of abnormal fluctuation stages by analyzing the critical points of curvature changes based on the curvature data of the gastrointestinal reaction, and at the same time, combines the data of physiological changes of the stomach caused by external stimulation to determine the time delay of the signal from being stimulated to restoring stability, and generates time delay characteristic data of the gastrointestinal reaction;

[0049] The time delay feature extraction module, based on the time delay feature data of the gastrointestinal reaction, classifies the data of the characteristic change stage in the gastrointestinal reaction into multiple state sets according to function, analyzes the data in each state set, quantifies the conversion frequency between states, and generates a gastrointestinal state feature analysis result;

[0050] The state feature classification module matches and analyzes the time series feature data of the changes in the concentrations of digestive enzymes and metabolites in each state based on the gastrointestinal state feature analysis results, identifies the reaction conditions of each gastrointestinal state, and generates state and concentration matching data;

[0051] The concentration change assessment module compares and evaluates the metabolite concentration change trend of the gastrointestinal function state based on the state and concentration matching data, identifies the concentration peak, decay rate and recovery level characteristics, quantifies the amplitude of concentration change, and generates a gastrointestinal function assessment result.

[0052] Compared with the prior art, the advantages and positive effects of the present invention are:

[0053] In the present invention, by analyzing the time series curvature changes of the intragastric pressure signal and the peristaltic rate signal, the dynamic trend and potential abnormal state are revealed more intuitively. At the same time, the amplitude and frequency information of the abnormal fluctuation stage are extracted, which can keenly capture the acute changes of gastrointestinal reactions, such as severe pressure fluctuations or high-frequency peristalsis, and provide important data basis for gastrointestinal state monitoring. On the basis of considering external stimulus data such as eating and exercise, the time delay characteristics of gastrointestinal reactions are analyzed, which also provides comprehensive dynamic characteristic data support for stress state monitoring. In addition, by extracting the time series characteristics of the concentrations of digestive enzymes and metabolites, the short-term fluctuations and long-term trends of the digestive process are identified, so that the digestion and absorption function of the gastrointestinal tract can be evaluated more comprehensively, so that the digestion efficiency of each physiological state can be accurately evaluated. Combined with the state transition frequency analysis, the understanding of the changes in gastrointestinal function under different physiological states is effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0055] Figure 2 A flow chart for analyzing signal change trends in the present invention;

[0056] Figure 3 A flow chart of analyzing the critical point of curvature change of the present invention;

[0057] Figure 4 A flow chart for determining the delay of a signal from stimulation to stabilization according to the present invention;

[0058] Figure 5 A flowchart of the present invention classified into a plurality of functional state sets;

[0059] Figure 6 A flow chart of the present invention for quantifying the frequency of transitions between states;

[0060] Figure 7 The flowchart of the present invention is to compare the concentration variation range under each state. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0062] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0063] See also Figure 1 The present invention provides a technical solution: a method for evaluating gastrointestinal function, comprising the following steps:

[0064] S1: Based on the collected gastrointestinal biological reaction information including intragastric pressure signal and peristaltic rate signal, the signal is quantified and processed, and the denoised data is divided into independent time series by mapping the signal value to a unified scale and eliminating noise. The curvature is calculated in each time series, and according to the change of curvature, the signal change trend is analyzed and the fitting curve is adjusted to generate the curvature data of gastrointestinal reaction;

[0065] S2: Based on the amplitude and frequency information of curvature fluctuation in the curvature data of gastrointestinal reactions, by calculating the difference value and time interval of adjacent extreme points, analyzing the critical point of curvature change, extracting information of abnormal fluctuation stage, which is the stage of rapid change of intragastric pressure, the stage of violent fluctuation of peristalsis rate, and the state of high-frequency peristalsis, combined with the data of physiological changes of the stomach caused by external stimuli including eating and exercise, determining the delay of the signal from stimulation to stability, clustering the reaction delay, and generating the delay characteristic data of gastrointestinal reactions;

[0066] S3: Based on the time-delay characteristic data of gastrointestinal reactions, the data of characteristic change stages in gastrointestinal reactions, including rapid changes, slow changes, and stable stages, are classified into multiple functional state sets, including normal peristalsis state, low peristalsis state, and high peristalsis state. The transition frequency between states is quantified to generate gastrointestinal state characteristic analysis results;

[0067] S4: Collect the concentration change data of digestive enzymes and metabolites, perform noise filtering and standardization, extract the time series characteristics of concentration changes, identify short-term fluctuations and long-term trends, compare the extracted time series characteristics of concentration changes with the gastrointestinal state characteristic analysis results, and compare the concentration change amplitudes under each state to evaluate the digestion and absorption capacity of the gastrointestinal tract and establish the gastrointestinal function evaluation results;

[0068] The curvature data of gastrointestinal reactions include the range of changes in curvature values, the frequency and amplitude of curvature fluctuations in the time series, and the distribution of curvature characteristics under different biological reaction states. The delay characteristic data of gastrointestinal reactions include the delay in changes in intragastric pressure after feeding stimulation, the delay in peristaltic rate response after exercise stimulation, and the clustering results of key time points between external stimulation and physiological response. The gastrointestinal state characteristic analysis results include the transition frequency between different states, the duration distribution of rapidly changing states, and the frequency distribution of stable and fluctuating states. The gastrointestinal function evaluation results include the fluctuation amplitude of digestive enzyme concentration under different gastrointestinal states, the long-term trend characteristics of changes in metabolite concentrations, and the graded evaluation of digestion and absorption capacity.

[0069] See also Figure 2 , the specific steps for obtaining the trend of signal change analysis are:

[0070] S111: Based on the collected intragastric pressure signal and peristaltic rate signal, the signals are quantified and mapped to a unified standard scale, and noise elimination is performed, and the denoised signal data is divided into independent time series according to the time window to obtain standardized time series data;

[0071] The signal is collected by patch-type abdominal surface sensors or intragastric catheter sensors, and the signal is quantified. The signal data is input into data analysis software (such as MATLAB or Python's NumPy library) to eliminate noise. In this process, the interference of noise signals is removed by selecting appropriate filter methods (such as bandpass filters) and the data is smoothed. After the signal cleaning is completed, the signal value is mapped to a unified scale of 0 to 1 by a normalization method. The intragastric pressure signal and the peristaltic rate signal are processed separately by a normalization function to ensure that the amplitudes of different signals are in the same quantitative standard for subsequent analysis and comparison. After the denoising and quantization processing is completed, the data is divided into multiple independent time series segments according to the time axis of the acquisition. The fixed time window method is used to process the data of each time series independently to ensure that the subsequent calculation accurately reflects the fluctuation characteristics of the signal in a specific time period.

[0072] S112: Based on the standardized time series data, the formula is:

[0073]

[0074] Calculate the curvature value K and obtain a curvature value sequence;

[0075] Among them, K is used to determine the intensity of signal changes in a specific period of time, helping to identify significant fluctuations in gastrointestinal function and facilitating the determination of abnormal conditions. i is time point x iThe signal value on the graph represents the intragastric pressure or peristaltic rate signal at a specific time point, which is collected in real time by gastrointestinal bioreactor equipment (such as pressure sensor). i+1 and i-1 is the adjacent time point x i+1 and x i-1 The signal value on the y is used to compare the signal change amplitude, the acquisition process and the i Similarly, Δx is the time interval in seconds, which can be directly obtained from the sampling rate of the device.

[0076] For example, if the signal value in the time series is [0.2, 0.5, 0.3] and the sampling time interval Δx = 0.1 seconds, calculate |y i+1 -2y i +y i-1 |:|0.5-2×0.2+0.3|=|0.4|.

[0077] Substituting into the formula:

[0078]

[0079] The results show that the current curvature value is 40.

[0080] S113: According to the curvature value sequence, compared with the conventional curvature range, observing the signal fluctuation by drawing a curvature change trend graph, gradually adjusting the fitting curve order to optimize the smoothness of the curvature change, and obtaining the curvature data of the gastrointestinal reaction;

[0081] According to the calculated curvature value K, the signal change trend is analyzed. If the healthy range is (20-30), the calculated K=40 is significantly higher than the healthy range (20-30), which can be judged that the signal has abnormal fluctuations during this period. In the analysis process, the calculated K value is first plotted into a curvature change graph according to the time series to observe the specific distribution of curvature changes in each period. Next, in order to accurately identify the trend, the curve needs to be fitted and adjusted: by selecting a suitable fitting function (such as polynomial fitting), gradually testing different orders, generating a fitting curve and comparing it with the original curvature data points, and judging the smoothness and accuracy of the fitting curve. In the fitting adjustment process, a lower-order fitting function is initially selected to ensure that the fitting curve can reflect the overall trend without overfitting. Gradually increase the fitting order, observe the changes in the smoothness and error of the curve, and find the order with smaller error and best fitting effect by calculating the difference between the curve and the original data points. If the error is too large or the fitting curve shows obvious excessive fluctuations, reduce the fitting order to maintain the actual change trend of the curve.

[0082] See also Figure 3 , the steps for obtaining the critical point of curvature change are as follows:

[0083] S211: Based on the amplitude and frequency information of curvature fluctuation in the curvature data of gastrointestinal reaction, the formula is used:

[0084]

[0085] Calculate the standardized curvature change amplitude ΔK and the standardized time interval T of adjacent extreme value points respectively, and obtain the curvature change amplitude and the time interval data of adjacent extreme value points;

[0086] Among them, K t+1 and K t They represent the curvature values ​​at time t+1 and time t respectively. D represents the distance change within the time period, which represents the distance of the inner wall movement between adjacent time points in the gastrointestinal tract. The data measured by the position sensor between the two points is used to extract the distance change value within the section. t t+1 and t t The timestamps of adjacent extreme points are directly extracted from the time records of pressure and motility rate signals at time t+1 and t respectively. F represents the standardized frequency of the motility rate. The average frequency is extracted from the time series data of historical gastrointestinal motility rate, and the motility frequency feature is obtained by Fourier transform combined with the scipy.fft method in Python.

[0087] For example, the curvature values ​​recorded at time t and t+1 are K t =0.8 and K t+1 =1.2, the corresponding time points are t=10 seconds and t+1=12 seconds, and the distance change D=0.5 and frequency F=0.25 are measured at the same time:

[0088]

[0089] The results show that the curvature variation amplitude is 0.8 and the normalized time interval is 8 seconds.

[0090] S212: Analyze whether the value exceeds the normal fluctuation range according to the curvature change amplitude and the time interval data of adjacent extreme value points, determine the critical point of the curvature change and mark the abnormal fluctuation stage, and obtain the critical point information of the curvature change;

[0091] Based on the specific values ​​of the calculated standardized curvature change amplitude ΔK=0.8 and the standardized time interval T=8, the baseline value range under normal gastrointestinal motility is compared, and whether ΔK and T exceed the set baseline range is analyzed to determine whether the critical point of curvature change has been reached. First, the ΔK value is compared in range. If the normal fluctuation range is 0.1≤ΔK≤0.5, if the current value 0.8 obviously exceeds the normal range, it is marked as an abnormal fluctuation stage; then the standardized time interval T is compared. If the time interval range corresponding to the normal motility frequency is 3≤T≤6, and the current T=8 is higher than the normal range, it is further confirmed that the motility rate frequency fluctuation during this period is abnormal. By comparing whether the two parameters ΔK and T are within the baseline range, it is determined whether a more significant curvature change critical point has occurred, and the abnormal characteristics of gastrointestinal motility are marked by the screened abnormal stage data.

[0092] See also Figure 4 , the acquisition steps for determining the delay from the stimulation to the stabilization of the signal are as follows:

[0093] S221: extracting the difference between adjacent peak values ​​of intragastric pressure, the time interval between adjacent extreme values ​​of peristaltic rate and peristaltic frequency according to the critical point information of curvature change, identifying the abnormal fluctuation stage, and obtaining an abnormal fluctuation data set;

[0094] First, the signal data is divided into two groups: pressure and peristaltic rate, and the peak difference and frequency change data are extracted respectively. The stage of rapid change in intragastric pressure can be defined as the difference between adjacent peak values ​​exceeding the threshold value of 1.0 within a time period. For example, the peak change from 2.3 to 3.4 reaches the condition of rapid pressure change; the stage of violent fluctuation in peristaltic rate is determined by extracting the time interval between adjacent extreme values. If the time interval between adjacent peaks is less than 70% of the benchmark time length, it is marked as a violent fluctuation stage, such as the peak time interval is shortened from 2 seconds to 1.2 seconds; the high-frequency peristaltic state is judged based on whether the peristaltic frequency exceeds 1.5 times the normal frequency within a certain period of time. If it exceeds the threshold, it is marked as a high-frequency state. Based on the above data judgment and marking results, the abnormal fluctuation stage information is extracted and organized.

[0095] S222: Based on the abnormal fluctuation data set, the formula is used:

[0096]

[0097] Calculate the delay time τ of gastrointestinal reaction to obtain gastrointestinal physiological response data;

[0098] Among them, ΔT′ represents the total reaction time of the gastrointestinal signal change caused by external stimulation, represents the time period from the start of external stimulation to the time when the gastrointestinal signal tends to be stable, and is usually obtained by monitoring the interval time of the signal change to stability through the sensor. S is the stable time period of the signal, which represents the time range for the gastrointestinal signal to return to stability after the external stimulation ends, and is extracted from the stable area in the signal graph. In the gastrointestinal signal data, a specific threshold can be set. For example, when the standard deviation S is less than 0.1, it is considered to be a stable stage. This threshold can be obtained by analyzing a large amount of gastrointestinal signal data under normal conditions. For example, the standard deviation distribution of the signal during daily gastrointestinal activities is statistically analyzed, and the standard deviation range of the stable period is determined by taking the lower 10% of this distribution as a reference value. M represents the intensity value of the external stimulus, such as the amount of food eaten or the intensity of exercise, which can be directly quantified and obtained through sensor measurement. The amount of food eaten can be monitored by the sensor through the volume change of the gastric contents, and the intensity of exercise can be measured by the acceleration sensor to measure the acceleration amplitude of the exercise. Taking eating as an example, if the weight of the food is 200 grams, then M can be set to 2 (assuming 100 grams is a standard unit). For exercise intensity, acceleration or heart rate rise value can be used to quantify. For example, when the heart rate rises by more than 20% during exercise, M can be set to 2. E represents the frequency of external stimulation, which is the number of stimulations per minute. For eating stimulation, the frequency is obtained through the eating time and frequency. For example, the number of eating or exercise frequency occurring in a unit time (such as per minute). When eating is performed 5 times within 10 minutes, E can be set to 0.5 times / minute. This frequency can be obtained by analyzing the time interval of the recorded data. W represents the fluctuation value of the peristaltic rate, which is used to represent the intensity of the change in the gastrointestinal peristaltic rate caused by stimulation. It is usually calculated by the variation amplitude of the peristaltic frequency data recorded by the gastrointestinal sensor. For example, the difference in the change of the peristaltic frequency is extracted from the output data of the peristaltic sensor, which can reflect the amplitude of the peristaltic acceleration or slowdown. D′ represents the fluctuation amplitude of the intragastric pressure, which represents the severity of the change in the intragastric pressure under stimulation. It can be obtained through continuous monitoring by the pressure sensor, that is, the difference between the maximum and minimum values. The value can be directly extracted from the pressure time series data in LabVIEW or MATLAB.

[0099] For example, in a measurement after a meal, ΔT′=50 seconds, S=10 seconds, M=4 (unit amount of food eaten), E=2 (frequency of stimulation per minute), W=0.5 (peristaltic rate fluctuation), D′=1.2 (pressure fluctuation amplitude), then:

[0100]

[0101] The results showed that the gastrointestinal tract stabilized after a delay of about 5.833 seconds after the feeding stimulation.

[0102] S223: dividing the data groups according to the average duration, high and low frequencies, and time stability based on the gastrointestinal physiological response data to perform clustering operations and generate time-delay characteristic data of gastrointestinal reactions;

[0103] First, each delay data is preprocessed to screen out abnormal values ​​and remove data that obviously deviates from the overall trend to ensure the accuracy of subsequent clustering results. Then, grouping and clustering are performed according to the characteristics of reaction delay, such as duration, stability and frequency. In the clustering operation, the average duration of all delay data is first calculated, and those above the average are classified as long delay groups, and those below the average are classified as short delay groups; on the basis of grouping, the peristaltic frequency of each group is further divided into high-frequency groups and those below the overall data frequency are divided into low-frequency groups according to the peristaltic frequency of each group. For the data in each group, its temporal stability is further calculated, and the data with a lower standard deviation is marked as a stable group. Usually, 10% lower than the overall standard deviation of the data is set as the stability division standard; if the data is continuously lower than the set standard deviation value for a period of time, it is judged as a stable group. For the clustering results of each group (long delay-high frequency, long delay-low frequency, short delay-high frequency, short delay-low frequency), the key characteristic values ​​such as average duration, average frequency and standard deviation are extracted, and the delay characteristic data of gastrointestinal reaction are sorted and generated.

[0104] See also Figure 5 , the steps for obtaining the multiple functional state sets are as follows:

[0105] S311: Based on the time-delay characteristic data of gastrointestinal reactions, according to the changes in reaction time and peristaltic rate, identifying the stages of change and stability in the data, and obtaining data information of each characteristic change stage;

[0106] First, the time-delay data features of gastrointestinal reactions are extracted, including reaction time and peristaltic rate change values, and the rapid change stage is identified. The obvious increase in peristaltic frequency or drastic increase in pressure fluctuations in a certain period of time is classified as a rapid change stage. For example, if the peristaltic rate is detected to increase from 0.5 to 1.2 within a few seconds and the peak value changes frequently, it is considered a rapid change; in the slow change stage, the gradual increase or decrease in peristaltic frequency in the time-delay data is analyzed, and the time period with smaller fluctuations is selected and the state of slow average rate change is recorded. For example, if the frequency rises from 0.6 to 0.8 and does not exceed 20% of the overall standard deviation, it is marked as a slow change stage; for the stable stage, by identifying the intervals with smaller curvature fluctuations or pressure fluctuations in the time series, the data segments with a standard deviation lower than 10% of the overall standard deviation are obtained, indicating that the signal fluctuates less in a stable state and there is no obvious increase or decrease in frequency. Finally, the characteristic data sets of the rapid change, slow change and stable stages are obtained through the analysis of the time-delay characteristic data.

[0107] S312: Based on the data information of each characteristic change stage, referring to the reaction time, frequency and stability data, each change stage is matched with the peristaltic state set to obtain the peristaltic state data set;

[0108] According to the division of rapid change, slow change and stable stages, each stage is classified into a functional state set, including normal peristalsis state, low peristalsis state and high peristalsis state, and one-to-one correspondence is made according to state characteristics. The data in the rapid change stage are classified into the high peristalsis state set, and by identifying the significant characteristics of its frequency change, such as the higher curvature fluctuation rate, it is divided into the high peristalsis set; the time delay feature data in the slow change stage corresponds to the low peristalsis state set, and the slow change data that gradually increases and decreases in this state indicates that the peristalsis rate is low, which is recorded in the low peristalsis set; the data in the stable stage are classified into the normal peristalsis state set, in which the stable time delay feature value represents the average frequency and stability of normal peristalsis, and the state data are sorted into the normal peristalsis set to form a data set of normal peristalsis, low peristalsis and high peristalsis states.

[0109] See also Figure 6 , the specific steps for obtaining the transition frequency between quantized states are:

[0110] S321: Based on the data set of the peristaltic state, the formula is used:

[0111]

[0112] Calculate the transition frequency f from state set a to state set b ab , get the frequency data of transition between peristaltic states;

[0113] Among them, N ab Indicates the total number of transitions from state set a to state set b within a period of time, obtained from the state transition events recorded in the delay characteristic data, V ab It represents the time variability of the transition from state a to state b, which is obtained by calculating the standard deviation of the transition time interval. The formula is: t′ k Indicates the time interval between each transfer. is the average value of all intervals, n is the total number of transfers, Q ab Represents the offset from state set a to state set b, which is used to reflect the overall deviation of state transfer in frequency. It is obtained by the root mean square error of state transfer frequency data to evaluate the frequency deviation of each state transfer. N a It represents the total number of occurrences of state set a, which is used to normalize the transition frequency to achieve the same magnitude of transition frequencies of different states. It is counted by the number of occurrences of the state label in the observation period. aIt represents the energy of state a, reflecting the dynamic strength of the state in the transition. The energy value is the mean of the absolute value of the amplitude of state a in its signal sequence, which can be obtained in Python's Numpy library or MATLAB environment. b and D a They represent the duration of state sets b and a respectively, and refer to the average residence time of the state during the observation period.

[0114] For example, the number of transitions N from the normal motility state to the low motility state during a specific observation period is ab =30, the time variation of the transfer V ab =2.5, offset Q ab =1.2, the total number of times state a is N a =100, energy E a =5, the duration of state b and state a are D b = 8 seconds and D a = 10 seconds, then:

[0115]

[0116] The results show that the transition frequency from the normal peristaltic state to the low peristaltic state is about 0.523. If the conventional transition frequency is 0.8, the calculation results show that the transition frequency is relatively low, indicating that the peristaltic state transition is relatively inactive.

[0117] S322: constructing a state feature matrix according to the state arrangement order based on the transition frequency data between peristaltic states, corresponding each state set to the row and column position of the matrix and filling in the transition frequency, identifying the transition paths of multiple frequencies, and obtaining the gastrointestinal state feature analysis results;

[0118] First, multiple state sets of gastrointestinal function are defined, such as normal, low-frequency, and high-frequency peristaltic states, and indexes are set for the rows and columns of the matrix according to the sequence of these states, so that each row and column of the matrix corresponds to a state. Next, the transition frequencies between the obtained state pairs are filled into the matrix according to the corresponding relationship between the states, for example, the frequency of transition from the normal state to the high-frequency state is filled into the corresponding position of the matrix. The transition frequency data of all state combinations are filled in sequence, so that each cell in the matrix represents the transition frequency from one state to another. After completion, the transition frequency values ​​in the matrix are sorted to identify the state combinations of high-frequency transitions and the stable state combinations of low-frequency transitions. Subsequently, by focusing on the high-frequency transition cells, these high-frequency state paths are analyzed in detail, and the relatively stable area of ​​the gastrointestinal state is confirmed through the low-frequency transition paths, and finally a state feature matrix reflecting the characteristic changes between the states is obtained to form the gastrointestinal state feature analysis results.

[0119] See also Figure 7, the steps for obtaining the concentration variation range under each state are as follows:

[0120] S411: Based on the concentration change data of digestive enzymes and metabolites, extract time series features, identify concentration fluctuation and stability features, compare with the gastrointestinal state feature analysis results, and obtain a concentration change feature set;

[0121] Based on the concentration change data of digestive enzymes and metabolites, the data is first collected and cleaned, and the transition frequencies f between different states are calculated. ab Compare and analyze with the concentration change trend. By observing the synchronization of the transition frequency and the concentration change curve, the fluctuation of the concentration value in the state transition with higher frequency can be identified, and the key state that may cause the change of gastrointestinal function can be identified. For example, the transition frequency is f ab =0.5 and the corresponding concentration change trend. If it is found that the concentration changes more when the transition frequency increases, the state transition may be accompanied by changes in digestive function. ab By comparing with the corresponding concentration change trend, the abnormal concentration fluctuation is marked in the state characteristics, which is used to further evaluate the gastrointestinal reaction characteristics under different states. Finally, the matching results of each state transition frequency and concentration fluctuation are sorted to form a concentration change feature set.

[0122] S412: Based on the concentration change feature set, the formula is used:

[0123]

[0124] Calculate the average difference C between the concentration changes in state a and state b ab , quantify the difference in digestion or absorption capacity between states and obtain gastrointestinal function assessment results;

[0125] Where H represents the number of time points in the total observation period, which is usually obtained based on the sampling frequency and observation duration, and c h,a and c h,b They represent the concentration values ​​in states a and b at time h respectively. The concentration values ​​can be directly obtained by sampling the experimental data.

[0126] For example, during the 5-second observation period, the digestive enzyme concentration value of the sample is collected once per second, and c h,a and c h,b The concentration sequence of , for the sampling concentration sequence in state a and state b is:

[0127] c 1,a , c 2,a , c 3,a , c 4,a , c 5,a ] and [c 1,b , c2,b , c 3,b , c 4,b , c 5,b ], then use the formula to calculate:

[0128]

[0129] Assumptions |c 1,a -c 1,b |=0.5、|c 2,a -c 2,b |=1.0、|c 3,a -c 3,b |=0.8,

[0130] c 4,a -c 4,b |=1.2、|c 5,a -c 5,b |=0.6, then:

[0131]

[0132] The results show that the average concentration change difference between state a and state b is 0.82. ab , to evaluate the digestion and absorption capacity of the gastrointestinal tract, first compare the concentration change range under different conditions with the benchmark value, which is usually obtained through historical data statistics to represent the concentration change range under normal gastrointestinal function. ab If the value is significantly higher than the baseline value, it may reflect that the digestion and absorption capacity of the gastrointestinal tract is abnormal during the state transition, such as insufficient digestion capacity leading to accumulation of metabolites in the gastrointestinal tract, or low absorption efficiency. Secondly, based on the time series analysis of concentration changes, determine the concentration change pattern of each state. For state transitions with large concentration changes, analyze whether they are accompanied by changes in peristaltic frequency. If high concentration changes are accompanied by low peristaltic states, it may indicate that the absorption process of the gastrointestinal tract in this state is obstructed. If high concentration changes are accompanied by high peristaltic frequency, it may be that the release of gastric acid or enzymes during digestion is abnormal. By establishing a concentration change evaluation model under different states, C ab Corresponding to the gastrointestinal digestion and absorption indicators, the states with large characteristic deviations are gradually screened out, and finally these characteristic data are recorded for comprehensive evaluation to help identify potential functional abnormalities and provide data support.

[0133] A gastrointestinal function evaluation system, the gastrointestinal function evaluation system is used to perform the above-mentioned gastrointestinal function evaluation method, the system comprises:

[0134] The biological signal acquisition module monitors and analyzes the curvature changes of the intragastric pressure signal and the peristaltic rate signal in real time based on the collected gastrointestinal biological reaction signals through independent time series division, adjusts the signal fitting curve, and generates the curvature data of the gastrointestinal reaction;

[0135] The curvature data analysis module is based on the curvature data of gastrointestinal reactions. By analyzing the critical points of curvature changes, it identifies and extracts data of abnormal fluctuation stages. At the same time, it combines the data of gastric physiological changes caused by external stimuli to determine the time delay from the signal being stimulated to the signal returning to stability, and generates time delay characteristic data of gastrointestinal reactions.

[0136] The time-delay feature extraction module is based on the time-delay feature data of gastrointestinal reactions. It classifies the data of the characteristic change stage in the gastrointestinal reactions into multiple state sets according to their functions. By analyzing the data in each state set, the conversion frequency between states is quantified to generate gastrointestinal state feature analysis results.

[0137] The state feature classification module matches and analyzes the time series feature data of the changes in the concentrations of digestive enzymes and metabolites in each state based on the gastrointestinal state feature analysis results, identifies the reaction conditions of each gastrointestinal state, and generates state and concentration matching data;

[0138] The concentration change assessment module compares and evaluates the metabolite concentration change trends of gastrointestinal function status based on the state and concentration matching data, identifies the concentration peak, decay rate and recovery level characteristics, quantifies the amplitude of concentration changes, and generates gastrointestinal function assessment results.

[0139] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for evaluating gastrointestinal function, characterized in that: The following steps are involved: Gastrointestinal biological response information including intragastric pressure signal and peristaltic rate signal is collected and divided into independent time series. According to the change of curvature in each time series, the signal change trend is analyzed and the fitting curve is adjusted to generate the curvature data of gastrointestinal response; Based on the amplitude and frequency information of the curvature fluctuation in the curvature data of the gastrointestinal reaction, by analyzing the critical point of the curvature change, extracting the information of the abnormal fluctuation stage, combining the data of the physiological changes of the stomach caused by external stimulation, determining the delay of the signal from stimulation to stability, and generating the delay characteristic data of the gastrointestinal reaction; Based on the time-delay characteristic data of the gastrointestinal reaction, classify the data of the characteristic change stage in the gastrointestinal reaction into multiple functional state sets, and quantify the transition frequency between states to generate gastrointestinal state characteristic analysis results; The concentration change data of digestive enzymes and metabolites are collected, and the time series characteristics of the concentration changes are extracted. The extracted time series characteristics of the concentration changes are compared with the gastrointestinal state characteristic analysis results. By comparing the concentration change amplitudes under each state, the gastrointestinal function evaluation results are established.

2. The gastrointestinal function evaluation method according to claim 1, characterized in that: The steps of obtaining the change trend of the analysis signal are specifically as follows: Based on the collected intragastric pressure signal and peristaltic rate signal, the signal is quantified and mapped to a unified standard scale, and noise elimination is performed. The denoised signal data is divided into independent time series according to the time window to obtain standardized time series data; According to the standardized time series data, the formula is adopted: Calculate the curvature value K and obtain a curvature value sequence; Among them, y i is time point x i The signal value on y i+1 and i-1 is the adjacent time point x i+1 and x i-1 The signal value on, Δx is the time interval; According to the curvature value sequence, compared with the conventional curvature range, the signal fluctuation is monitored by drawing a curvature change trend graph, and the fitting curve order is gradually adjusted to optimize the smoothness of the curvature change, so as to obtain the curvature data of the gastrointestinal reaction.

3. The gastrointestinal function evaluation method according to claim 2, characterized in that: The steps for obtaining the critical point of analyzing the curvature change are specifically as follows: According to the amplitude and frequency information of the curvature fluctuation in the curvature data of the gastrointestinal reaction, the formula is adopted: Calculate the standardized curvature change amplitude ΔK and the standardized time interval T of adjacent extreme value points respectively, and obtain the curvature change amplitude and the time interval data of adjacent extreme value points; Among them, K t+1 and K t They represent the curvature values ​​at time t+1 and time t respectively, D represents the distance change within the time period, t t+1 and t t represents the timestamps of adjacent extreme points, at times t+1 and t, respectively, and F represents the normalized frequency of the peristalsis rate; According to the curvature change amplitude and the time interval data of adjacent extreme value points, it is analyzed whether the value exceeds the normal fluctuation range, the critical point of the curvature change is determined and the abnormal fluctuation stage is marked to obtain the critical point information of the curvature change.

4. The gastrointestinal function evaluation method according to claim 3, characterized in that: The step of obtaining the time delay of the determined signal from stimulation to stabilization is specifically as follows: According to the critical point information of the curvature change, the difference between adjacent peak values ​​of the intragastric pressure, the time interval between adjacent extreme values ​​of the peristaltic rate and the peristaltic frequency are extracted to identify the abnormal fluctuation stage and obtain the abnormal fluctuation data set; According to the abnormal fluctuation data set, the formula is adopted: Calculate the delay time τ of gastrointestinal reaction to obtain gastrointestinal physiological response data; Wherein, ΔT′ represents the total reaction time of the gastrointestinal signal change caused by external stimulation, S represents the stable time period of the signal, M represents the intensity value of the external stimulation, E represents the frequency of the external stimulation, W represents the fluctuation value of the peristaltic rate, and D′ represents the fluctuation amplitude of the intragastric pressure; According to the gastrointestinal physiological response data, the data groups are divided according to average duration, high and low frequencies and time stability to perform clustering operations and generate time-delay characteristic data of gastrointestinal reactions.

5. The gastrointestinal function evaluation method according to claim 4, characterized in that: The steps of obtaining the classification into multiple function status sets are specifically as follows: Based on the time-delay characteristic data of the gastrointestinal reaction, according to the changes in reaction time and peristaltic rate, identifying the stages of change and stability in the data, and obtaining data information of each characteristic change stage; Based on the data information of each characteristic change stage, with reference to the reaction time, frequency and stability data, each change stage is matched with the peristaltic state set to obtain the peristaltic state data set.

6. The gastrointestinal function evaluation method according to claim 5, characterized in that: The steps for acquiring the transition frequency between the quantization states are specifically as follows: Based on the data set of the peristaltic state, the formula is adopted: Calculate the transition frequency f from state set a to state set b ab , get the frequency data of transition between peristaltic states; Among them, N ab represents the total number of transitions from state set a to state set b within a period of time, V ab represents the time variability of the transition from state a to state b, Q ab Represents the offset from state set a to state set b, N a represents the total number of occurrences of state set a, E a represents the energy of state a, D b and D a Respectively represent the duration of state sets b and a; According to the transfer frequency data between the peristaltic states, a state characteristic matrix is ​​constructed according to the arrangement order of the states, each state set is corresponded to the row and column positions of the matrix and the transfer frequency is filled in, the transfer paths of multiple frequencies are identified, and the gastrointestinal state characteristic analysis results are obtained.

7. The gastrointestinal function evaluation method according to claim 6, characterized in that: The steps of obtaining the concentration variation range under each state are specifically as follows: Based on the concentration change data of digestive enzymes and metabolites, time series features are extracted, concentration fluctuation and stability features are identified, and the concentration change feature set is obtained by comparing the results with the gastrointestinal state feature analysis results; According to the concentration change feature set, the formula is used: Calculate the average difference C between the concentration changes in state a and state b ab , quantify the difference in digestion or absorption capacity between states and obtain gastrointestinal function assessment results; Among them, H represents the number of time points in the total observation period, c h,a and c h,b Respectively represent the concentration values ​​in states a and b at time h.

8. A gastrointestinal function assessment system, characterized in that: The gastrointestinal function assessment method according to any one of claims 1 to 7, wherein the system comprises: The biological signal acquisition module monitors and analyzes the curvature changes of the intragastric pressure signal and the peristaltic rate signal in real time based on the collected gastrointestinal biological reaction signals through independent time series division, adjusts the signal fitting curve, and generates the curvature data of the gastrointestinal reaction; The curvature data analysis module identifies and extracts data of abnormal fluctuation stages by analyzing the critical points of curvature changes based on the curvature data of the gastrointestinal reaction, and at the same time, combines the data of physiological changes of the stomach caused by external stimulation to determine the time delay of the signal from being stimulated to restoring stability, and generates time delay characteristic data of the gastrointestinal reaction; The time delay feature extraction module, based on the time delay feature data of the gastrointestinal reaction, classifies the data of the characteristic change stage in the gastrointestinal reaction into multiple state sets according to function, analyzes the data in each state set, quantifies the conversion frequency between states, and generates a gastrointestinal state feature analysis result; The state feature classification module matches and analyzes the time series feature data of the changes in the concentrations of digestive enzymes and metabolites in each state based on the gastrointestinal state feature analysis results, identifies the reaction conditions of each gastrointestinal state, and generates state and concentration matching data; The concentration change assessment module compares and evaluates the metabolite concentration change trend of the gastrointestinal function state based on the state and concentration matching data, identifies the concentration peak, decay rate and recovery level characteristics, quantifies the amplitude of concentration change, and generates a gastrointestinal function assessment result.

Images