A signal processing method for bowel sound signal monitoring system

Abstract

The present invention provides a signal processing method for a bowel sound signal monitoring system, and relates to the technical field of signal processing. The present invention effectively extracts a variety of feature information by real-time acquisition of pressure change data of the inner wall of the gastrointestinal tract and length change data of the gastrointestinal muscles, and combines adaptive filtering processing and spectrum analysis of bowel sound signals; on this basis, a comprehensive evaluation model is constructed to perform multi-parameter fusion analysis on the intestinal activity of the patient, and the output value of the comprehensive evaluation model is dynamically adjusted according to real-time changes; the method not only improves the detection accuracy of bowel sound signals, but also can provide a more comprehensive and accurate intestinal activity evaluation.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and in particular to a signal processing method for a bowel sound signal monitoring system. Background Art

[0002] In modern medicine, the monitoring and analysis of bowel sound signals have important clinical value in the diagnosis and treatment of gastrointestinal diseases. At present, the monitoring of bowel sound signals mainly relies on stethoscopes and audio analysis equipment. However, these traditional methods have some limitations, such as being greatly affected by subjective judgment, serious interference from environmental noise, and single signal processing methods. In recent years, with the rapid development of electronic technology and signal processing technology, bowel sound signal monitoring systems based on multi-sensor data fusion and advanced signal processing algorithms have gradually become a research hotspot. These systems usually integrate pressure sensors, displacement sensors, and acoustic sensors, and can provide a more comprehensive and accurate intestinal activity assessment by comprehensively monitoring and analyzing multiple physiological parameters.

[0003] In the prior art, the publication number is CN106021948A, and the name is a signal processing method for a bowel sound signal monitoring system. After the bowel sound data is quantized by a microphone into a digital signal with a sampling rate of Fs, it is processed by an adaptive filtering algorithm and segmented according to the window length. Each segment of the signal is applied with a fast Fourier transform algorithm to obtain its signal spectrum, which will first be used for intensity detection; thereafter, correlation detection and analysis are performed with the mid-spectrum model in the frequency band to be measured, and then it is sent to a comparator and compared with a reference value. Only when the correlation coefficient is large enough, it is considered that the bowel sound signal appears in the data window, and the output result is "1"; otherwise, "0" is output. The present invention will also give RIM, IMF and AMT based on this to evaluate the intestinal activity of the subject. The present invention is used for a wearable human bowel sound signal monitoring system, which can be implanted in the device hardware or placed in a hospital monitoring server. According to the bowel sound signal input into the system, the position of the subject's bowel sound is marked, thereby providing technical support for doctors to give reasonable postoperative rehabilitation care to patients undergoing abdominal surgery.

[0004] Although the existing bowel sound signal monitoring system has significantly improved the efficiency of signal acquisition and processing, there are still some shortcomings in practical applications. First, most of the existing systems only rely on a single type of sensor signal, such as pressure signals or acoustic signals, and lack joint analysis of multiple parameters, resulting in insufficient accuracy and robustness of the evaluation results.

[0005] Secondly, the current signal processing methods mostly use fixed parameter filtering and spectrum analysis algorithms, which are difficult to adapt to the individual differences of different patients and the real-time changes in physiological status. In addition, the existing systems often lack the comprehensive representation and dynamic adjustment mechanism of time domain and frequency domain characteristics, and cannot monitor and evaluate the changes in intestinal activity in detail; therefore, how to effectively integrate multiple physiological signals, adopt adaptive filtering and analysis methods, and build a dynamically adjusted comprehensive evaluation model has become a key issue in improving the performance of bowel sound signal monitoring systems.

[0006] The above information disclosed in the above Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0007] The purpose of the present invention is to provide a signal processing method for a bowel sound signal monitoring system to solve the problems raised in the above background technology.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A signal processing method for a bowel sound signal monitoring system, the specific steps comprising:

[0010] Step S1: collecting pressure change data of the inner wall of the gastrointestinal tract and length change data of the gastrointestinal muscles in real time, and transmitting the collected data to the signal processing unit in real time;

[0011] Step S2: receiving the collected pressure change data and performing low-pass filtering to obtain filtered pressure data; receiving the collected length change data and performing calibration to eliminate initial errors to obtain calibrated length data;

[0012] Step S3: Obtain filtered pressure data and calibrated length data, and calculate the muscle tension of the gastrointestinal tract;

[0013] Step S4: collecting bowel sound signals and processing them with an adaptive filtering algorithm to obtain filtered bowel sound signals; performing spectrum analysis on the filtered bowel sound signals and obtaining frequency component amplitudes through fast Fourier transform;

[0014] Step S5: determining an extraction time period immediately adjacent to the current time t, and in the extraction time period, extracting tension features, spectrum features, and time domain features from muscle tension, frequency component amplitude, and bowel sound signals respectively, and performing comprehensive representation on each of the extracted features;

[0015] Step S6: combining the tension features, spectrum features and time domain features represented comprehensively and analyzing and processing them to construct a comprehensive evaluation model for evaluating the intestinal activity of the patient;

[0016] Step S7: Determine the previous extraction time period adjacent to the current extraction time period, and calculate the output value of the comprehensive evaluation model of the previous extraction time period, combine and analyze the output values ​​of the comprehensive evaluation model of the current extraction time period with those of the previous extraction time period, and formulate an adjustment strategy for dynamically adjusting the output value of the comprehensive evaluation model corresponding to the current extraction time period.

[0017] Furthermore, the tension features, spectrum features and time domain features of the comprehensive representation are combined and analyzed to construct a comprehensive evaluation model for evaluating the intestinal activity of patients, including:

[0018] The output value of the comprehensive evaluation model within the extraction time period Td is defined as , the calculation formula is as follows:

[0019]

[0020] in, ; d1, d2, d3 are weight coefficients of corresponding parameters; d4 is the adjustment coefficient of normalization term, 0.12≤d4≤0.89, d5 is the adjustment coefficient of exponential function, 0.08≤d4≤0.93;

[0021] The value range of is (0, 1);

[0022] when The closer it is to 0, the lower the intestinal activity of the patient, and further monitoring or treatment measures are needed;

[0023] when When the value is in the middle range, it means that the patient's intestinal activity is within the normal range, but regular monitoring is required;

[0024] when The closer it is to 1, the higher the patient's intestinal activity is, indicating good intestinal function.

[0025] Furthermore, when <0.3, indicating In the interval close to 0, the corresponding signal output is a low-level signal, and the intestinal activity is low; the comprehensive tension value, comprehensive spectrum value, and comprehensive time domain value are all low, and the normalization term d4 and the index term d5 have a greater impact, which suppresses the overall score;

[0026]

[0027] when When In the middle range, the corresponding signal output is a normal level signal, and the intestinal activity is at a normal level; the comprehensive tension value, comprehensive spectrum value, and comprehensive time domain value are in the middle range;

[0028]

[0029] when When In the interval approaching 1, the corresponding signal output is a high-level signal, and the intestinal activity is at a high level; the comprehensive tension value, comprehensive spectrum value, and comprehensive time domain value are high, and the normalization term d4 and the index term d5 have little influence, and the inhibitory effect on the score is weak;

[0030] .

[0031] Furthermore, the output values ​​of the comprehensive evaluation model of the current extraction time period and the previous extraction time period are combined and analyzed to formulate an adjustment strategy for dynamically adjusting the output value of the comprehensive evaluation model corresponding to the current extraction time period, specifically including:

[0032] Define the previous extraction time period immediately adjacent to the current extraction time period Td as Td-1, the time sampling points of the extraction time period Td and the previous extraction time period Td-1 do not overlap, and the last time sampling point of the previous extraction time period Td-1 is continuous with the first time sampling point of the extraction time period Td;

[0033] Defining dynamic factors The calculation formula is:

[0034]

[0035] in, represents the output value of the comprehensive evaluation model in the previous extraction time period Td-1, express and The difference change; R represents the trend adjustment coefficient; f1 and f2 are the adjustment parameters of the corresponding parameters; 0.35≤f1≤0.78, 0.23≤f2≤0.66;

[0036] Set the trigger condition to :

[0037]

[0038] in, Indicates rounding up to an integer;

[0039] when Lower than When , make the following fine adjustments to d1, d2, d3, d4, and d5:

[0040] Increase d1 by 5% of its current value to increase the tension feature weight, and reduce d2 and d3 by 2% and 3% respectively, and reduce d4 and d5 by 3% and 4% respectively;

[0041] when When , make the following fine adjustments to d1, d2, d3, d4, and d5:

[0042] Increasing d1 by 3% of its current value increases the tension feature weight, and d2 and d3 decrease by 1% and 2% respectively, and d4 and d5 decrease by 1% and increase by 2% respectively.

[0043] Compared with the prior art, the beneficial effects of the present invention are: by real-time collection of pressure change data of the inner wall of the gastrointestinal tract and length change data of the gastrointestinal muscles, and combining adaptive filtering processing and spectrum analysis of bowel sound signals, a variety of feature information is effectively extracted; on this basis, by constructing a comprehensive evaluation model, a multi-parameter fusion analysis of the patient's intestinal activity is performed, and the output value of the comprehensive evaluation model is dynamically adjusted according to real-time changes; this method not only improves the detection accuracy of bowel sound signals, but also can provide a more comprehensive and accurate intestinal activity evaluation, and has important clinical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic diagram of the overall method flow of the present invention. DETAILED DESCRIPTION

[0045] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0046] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0047] Embodiment 1:

[0048] See also Figure 1 , the present invention provides a technical solution:

[0049] A signal processing method for a bowel sound signal monitoring system, the specific steps comprising:

[0050] Step S1: collecting pressure change data of the inner wall of the gastrointestinal tract and length change data of the gastrointestinal muscles in real time, and transmitting the collected data to the signal processing unit in real time;

[0051] Step S2: receiving the collected pressure change data and performing low-pass filtering to obtain filtered pressure data; receiving the collected length change data and performing calibration to eliminate initial errors to obtain calibrated length data;

[0052] Step S3: Obtain filtered pressure data and calibrated length data, and calculate the muscle tension of the gastrointestinal tract;

[0053] Step S4: collecting bowel sound signals and processing them with an adaptive filtering algorithm to obtain filtered bowel sound signals; performing spectrum analysis on the filtered bowel sound signals and obtaining frequency component amplitudes through fast Fourier transform;

[0054] Step S5: determining an extraction time period immediately adjacent to the current time t, and in the extraction time period, extracting tension features, spectrum features, and time domain features from muscle tension, frequency component amplitude, and bowel sound signals respectively, and performing comprehensive representation on each of the extracted features;

[0055] Step S6: combining the tension features, spectrum features and time domain features represented comprehensively and analyzing and processing them to construct a comprehensive evaluation model for evaluating the intestinal activity of the patient;

[0056] Step S7: Determine the previous extraction time period adjacent to the current extraction time period, and calculate the output value of the comprehensive evaluation model of the previous extraction time period, combine and analyze the output values ​​of the comprehensive evaluation model of the current extraction time period with those of the previous extraction time period, and formulate an adjustment strategy for dynamically adjusting the output value of the comprehensive evaluation model corresponding to the current extraction time period.

[0057] Further explanation: the acquisition of filtered pressure data and calibrated length data specifically includes:

[0058] Select and install a high-precision gastrointestinal pressure sensor and a length sensor, wherein the pressure sensor is used to measure the pressure change of the inner wall of the gastrointestinal tract, and the length sensor is used to measure the length change of the gastrointestinal muscle. The sensor transmits data to the signal processing unit in real time through a wireless transmission module;

[0059] The filter is a Butterworth low-pass filter, and the filter order N is adjusted according to the actual sampling rate;

[0060] The pressure change data and length change data collected at time t are recorded as P(t) and L(t), respectively, and the filtered pressure data is recorded as , The calculation formula is as follows:

[0061]

[0062] Where h(i) is the impulse response of the filter, which represents the weight coefficient of the filter at the i-th time offset; N is the order of the filter, which represents the window size of the filter; i represents the time offset relative to the current time point t, ranging from −N to N; the time offset i here is the discrete time step;

[0063] The length data after calibration at time t is recorded as , the calculation formula is as follows:

[0064]

[0065] in, is the initial length of the gastrointestinal muscle before data collection.

[0066] To further illustrate, the muscle tension of the gastrointestinal tract at time t is recorded as , the calculation formula is as follows:

[0067]

[0068] Where E is the elastic modulus, K is the stiffness coefficient, is the length change after calibration, is the pressure change after filtering, and They are the average length of the gastrointestinal muscles and the average pressure of the gastrointestinal lining of the patient at rest.

[0069] Further explanation: a high-precision microphone is used to collect bowel sound signals, and the bowel sound signal collected at time t is recorded as S(t), and the bowel sound signal S(t) after being processed by the adaptive filtering algorithm is recorded as ; The specific processing steps are as follows:

[0070] The adaptive filtering algorithm adopts LMS;

[0071] Initialize the following adaptive filter parameters and variables:

[0072] Filter coefficients The initial value of is set to zero;

[0073] Step size parameter The initial value of is set to 0.01; Controls the step size of filter coefficient updates;

[0074] Input signal buffer : Stores the data of the current and past N-1 sampling points, where N is the order of the filter;

[0075] For each new sampling point S(t), update the input signal buffer :

[0076]

[0077] Compute the output y(t) of the filter at time t:

[0078]

[0079] in, is the coefficient vector of the filter, represents its transpose;

[0080] Set the reference signal to d(t) and calculate the filtering error e(t);

[0081]

[0082] The reference signal d(t) is the expected output of the input bowel sound signal S(t);

[0083] Update the filter coefficient vector according to the filtering error e(t) ; The update formula is as follows:

[0084]

[0085] Filtered bowel sound signal As the output of the filter y(t):

[0086]

[0087] The advantage of the adaptive filtering algorithm is that it can adjust the filter coefficients in real time to adapt to signal changes and improve the filtering effect.

[0088] Further explanation: the filtered bowel sound signal is subjected to spectrum analysis, and the frequency component amplitude is obtained by fast Fourier transform, which specifically includes:

[0089] The filtered bowel sound signal According to the length of each Divide into several segments; set the bowel sound signal of each segment to ; n represents the sampling point index within the segment, ranging from 0 to ;

[0090] A window function is applied to each signal segment to reduce spectrum leakage. In this embodiment, a Hamming window w(n) is selected. Each segment of the bowel sound signal is Multiply by the window function w(n) to get the following windowed bowel sound signal value :

[0091]

[0092] For each windowed bowel sound signal Perform fast Fourier transform FFT to obtain the frequency component amplitude :

[0093]

[0094] Among them, n represents the sampling point index in the segment; k is the frequency index, ranging from 0 to , is the basis function of Fourier transform, and j is the imaginary unit.

[0095] It should be noted that n is used in the time domain, indicating the sampling points in each segmented signal; k is used in the frequency domain, indicating the amplitude of the frequency component obtained after FFT calculation; in the fast Fourier transform (FFT), the range of the time domain sampling point index n and the frequency domain frequency index k can be the same, because the essence of FFT is to convert a time domain signal of length N′ into a frequency domain signal of the same length N′; in short, the number of sampling points in the time domain and frequency domain are the same.

[0096] Further explanation: determining the extraction time period immediately adjacent to the current time t, extracting tension features, spectrum features and time domain features from muscle tension, frequency component amplitude and bowel sound signals in the extraction time period, specifically including:

[0097] Set the time period Td immediately before the current time t as the extraction time period, mark U time sampling points in the extraction time period Td, and form a mark set of time sampling points {1, 2, ..., z, ..., U}, where z represents the zth time sampling point;

[0098] In the extraction time period Td, the tension feature is extracted from the muscle tension calculated in step S3, and the tension feature includes the average tension, the maximum tension and the tension change rate in the extraction time period Td; and the average tension, the maximum tension and the tension change rate in the extraction time period Td are recorded as ;

[0099] The average tension The calculation formula is expressed as:

[0100]

[0101] in, is the tension value collected at the zth time sampling point; U is the total number of sampling points within the extraction time period Td;

[0102] The maximum tension within the time period Td will be extracted The calculation formula is expressed as:

[0103]

[0104] The tension change rate within the extraction time period Td The calculation formula is expressed as:

[0105]

[0106] is the tension value at the end of the extraction period Td; is the tension value at the beginning of the extraction time period Td;

[0107] During the extraction period, the frequency component amplitude calculated from step S4 Extract the spectrum features, the spectrum features include the spectrum peak value and the spectrum intensity change rate within the extraction time period Td; and the spectrum peak value and the spectrum intensity change rate within the extraction time period Td are respectively recorded as ;

[0108] The spectrum peak within the time period Td will be extracted The calculation formula is expressed as:

[0109]

[0110] The spectrum intensity change rate within the extracted time period Td The calculation formula is expressed as:

[0111]

[0112] is the spectrum intensity at the end of the extraction period Td; is the spectrum intensity at the beginning of the extraction time period Td;

[0113] During the extraction period, the The time domain features are extracted from the extraction time period Td, and the time domain features include the average signal strength and signal change rate in the extraction time period Td; and the average signal strength and signal change rate in the extraction time period Td are respectively recorded as ;

[0114] The average signal strength within the time period Td will be extracted The calculation formula is expressed as:

[0115]

[0116] in, It's at the time The value of the collected bowel sound signal;

[0117] The signal change rate within the time period Td will be extracted The calculation formula is expressed as:

[0118]

[0119] is the bowel sound signal value at the end of the extraction time period Td; is the bowel sound signal value at the beginning of the extraction time period Td;

[0120] All extracted features are standardized to eliminate the influence of different dimensions, and the standardized output values ​​are translated and scaled so that all extracted feature output values ​​are limited to the range of (0,1); the standardization method uses z-score standardization;

[0121] The tension characteristics within the extraction time period Td are comprehensively expressed as:

[0122]

[0123] in, is the comprehensive tension value extracted within the time period Td, a1, a2, and a3 are the weight coefficients of the corresponding parameters; ; a1, a2, and a3 are determined by the expert group through experimental data and will not be described in detail;

[0124] The spectrum characteristics within the extraction time period Td are comprehensively expressed as:

[0125]

[0126] in, To extract the comprehensive spectrum value within the time period Td, b1 and b2 are the weight parameters of the corresponding parameters. ; b1 and b2 are determined by the expert group through experimental data and will not be described in detail;

[0127] The time domain features within the extracted time period Td are comprehensively expressed as:

[0128]

[0129] in, To extract the comprehensive time domain value within the time period Td, c1 and c2 are the weight parameters of the corresponding parameters. ; c1 and c2 are specifically determined by the expert group based on experimental data and will not be elaborated on here.

[0130] It is further explained that the tension features, spectrum features and time domain features of the comprehensive representation are combined and analyzed to construct a comprehensive evaluation model for evaluating the intestinal activity of patients, which specifically includes:

[0131] The output value of the comprehensive evaluation model within the extraction time period Td is defined as , the calculation formula is as follows:

[0132]

[0133] in, ; d1, d2, d3 are weight coefficients of corresponding parameters; d4 is the adjustment coefficient of normalization term, 0.12≤d4≤0.89, d5 is the adjustment coefficient of exponential function, 0.08≤d4≤0.93;

[0134] The value range of is (0, 1); (0, 1) is obtained by normalization, scaling and offset means, which are existing preprocessing means and are not described in detail; under different value ranges, the technical effects are as follows:

[0135] when The closer it is to 0, the lower the intestinal activity of the patient, and further monitoring or treatment measures are needed;

[0136] Explanation of monitoring content:

[0137] Frequent monitoring: Increase the monitoring frequency and perform more frequent testing on the patient's intestinal activity; in this embodiment, the daily testing is adjusted to once an hour;

[0138] Detailed records: Record the patient’s detailed medical history, eating habits, bowel habits, medication use, etc.

[0139] Data analysis: Analyze the continuous monitoring data to find out the reasons for the reduced activity, such as improper diet, side effects of drugs, irregular lifestyle, etc.

[0140] Explanation of treatment measures:

[0141] Drug treatment: Prokinetic drugs: such as domperidone, mosapride, etc., help increase intestinal motility.

[0142] Laxatives: such as lactulose, polyethylene glycol, etc., help relieve constipation.

[0143] Antibiotics: If infection is present, consider appropriate antibiotics.

[0144] Dietary adjustments: High-fiber diet: Increase dietary fiber intake, such as eating more fruits, vegetables, whole grains, etc.

[0145] Drink more water: increasing water intake helps promote intestinal motility.

[0146] Avoid irritating foods: Avoid consuming too much spicy, greasy, and caffeinated food.

[0147] Lifestyle changes: Regular work and rest schedule: Maintain a regular work and rest schedule and avoid staying up late.

[0148] Moderate exercise: Increase the amount of physical exercise, such as walking, jogging, yoga, etc., to promote intestinal peristalsis.

[0149] Reduce stress and relax: Take effective stress-relief measures, such as meditation, deep breathing exercises, etc.

[0150] Other treatments: Physical therapy: such as abdominal massage, hot compress, etc., to help relieve intestinal tension and discomfort.

[0151] Psychological treatment: If necessary, psychological counseling or therapy can be considered to help reduce the impact of psychological stress on the intestines.

[0152] The above monitoring or treatment measures are selectively adjusted according to the actual medical staff and will not be elaborated on here;

[0153] when When the value is in the middle range, it means that the patient's intestinal activity is within the normal range, but regular monitoring is required;

[0154] Explanation of regularly monitored content:

[0155] Regular detection: Regular detection is performed at certain time intervals. In this embodiment, detection once every hour is adjusted to detection once every day.

[0156] Routine examination: Routine clinical examinations are performed, such as weight, blood pressure, blood routine, liver function, kidney function, etc.

[0157] Monitoring indicators: Continuously monitor indicators related to intestinal activity, such as bowel movement frequency, stool properties, abdominal pain, etc.

[0158] Data analysis: The collected data were analyzed to evaluate the changing trend of intestinal motility.

[0159] The content of regular monitoring is selectively adjusted according to the actual medical staff, so it will not be elaborated here;

[0160] when The closer it is to 1, the higher the intestinal activity of the patient, indicating good intestinal function;

[0161] Comprehensive tension value : Reflects the activity of intestinal smooth muscle; The higher the value, the more active the intestine is, but too high tension also indicates abnormality;

[0162] Comprehensive spectrum value : reflects the frequency and intensity of intestinal electrical activity, The higher the value, the more active the electrical activity;

[0163] Comprehensive time domain value : reflects the temporal regularity of the intestine, The higher the value, the better the regularity of intestinal movement;

[0164] when <0.3, indicating In the interval approaching 0, the corresponding signal output is a low-level signal, indicated by a red alarm light; the intestinal activity is at a low level; the comprehensive tension value, comprehensive spectrum value, and comprehensive time domain value are all low, and the normalization term d4 and the index term d5 have a greater impact, which suppresses the overall score;

[0165]

[0166] For every 1% increase in d1, Increased by 0.9%, other parameters remained unchanged;

[0167] For every 1% increase in d2, Increased by 0.8%, other parameters remained unchanged;

[0168] For every 1% increase in d3, Increased by 0.7%, other parameters remained unchanged;

[0169] For every 1% increase in d4, Reduce by 0.5%, other parameters remain unchanged;

[0170] For every 1% increase in d5, Reduced by 0.6%, other parameters remained unchanged;

[0171] when When In the middle range, the corresponding signal output is a normal level signal, indicated by a yellow alarm light, and the intestinal activity is at a normal level; the comprehensive tension value, comprehensive spectrum value, and comprehensive time domain value are in the middle range;

[0172]

[0173] For every 1% increase in d1, Increased by 0.8%, other parameters remained unchanged;

[0174] For every 1% increase in d2, Increased by 0.7%, other parameters remained unchanged;

[0175] For every 1% increase in d3, Increased by 0.6%, other parameters remained unchanged;

[0176] For every 1% increase in d4, Reduced by 0.4%, other parameters remained unchanged;

[0177] For every 1% increase in d5, Reduce by 0.5%, other parameters remain unchanged;

[0178] when When In the interval approaching 1, the corresponding signal output is a high-level signal, indicated by a green alarm light, and the intestinal activity is at a high level; the comprehensive tension value, comprehensive spectrum value, and comprehensive time domain value are high, and the normalization term d4 and the index term d5 have little influence, and the inhibitory effect on the score is weak;

[0179] .

[0180] For every 1% increase in d1, Increased by 0.6%, other parameters remained unchanged;

[0181] For every 1% increase in d2, Increase by 0.5%, other parameters remain unchanged;

[0182] For every 1% increase in d3, Increase by 0.4%, other parameters remain unchanged;

[0183] For every 1% increase in d4, Reduce by 0.3%, other parameters remain unchanged;

[0184] For every 1% increase in d5, The reduction was 0.4%, and other parameters remained unchanged.

[0185] Further explanation: the output values ​​of the comprehensive evaluation model of the current extraction time period and the previous extraction time period are combined and analyzed to formulate an adjustment strategy for dynamically adjusting the output value of the comprehensive evaluation model corresponding to the current extraction time period, specifically including:

[0186] Define the previous extraction time period immediately adjacent to the current extraction time period Td as Td-1, the time sampling points of the extraction time period Td and the previous extraction time period Td-1 do not overlap, and the last time sampling point of the previous extraction time period Td-1 is continuous with the first time sampling point of the extraction time period Td;

[0187] Defining dynamic factors The calculation formula is:

[0188]

[0189] in, represents the output value of the comprehensive evaluation model in the previous extraction time period Td-1, express and The difference change; R represents the trend adjustment coefficient; f1 and f2 are the adjustment parameters of the corresponding parameters; 0.35≤f1≤0.78, 0.23≤f2≤0.66; the specific values ​​of f1 and f2 are determined by the expert group through experimental data;

[0190] Set the trigger condition to :

[0191]

[0192] in, Indicates rounding up to an integer;

[0193] when Lower than When , make the following fine adjustments to d1, d2, d3, d4, and d5:

[0194] Increase d1 by 5% of its current value to increase the tension feature weight, and reduce d2 and d3 by 2% and 3% respectively, and reduce d4 and d5 by 3% and 4% respectively;

[0195] when When , make the following fine adjustments to d1, d2, d3, d4, and d5:

[0196] Increasing d1 by 3% of its current value increases the tension feature weight, and d2 and d3 decrease by 1% and 2% respectively, and d4 and d5 decrease by 1% and increase by 2% respectively.

[0197] The specific adjustment values ​​of d1, d2, d3, d4 and d5 mentioned above are subsequently adjusted based on experimental data by the expert group and will not be elaborated on here.

[0198] Embodiment 2:

[0199] In this embodiment, five test subjects were selected, namely patient A, patient B, patient C, patient D and patient E; the test subjects were all healthy individuals of different genders and ages to ensure the representativeness of the data; the test was conducted in a laboratory environment, and the experimental equipment included a pressure sensor, a length sensor and a bowel sound monitoring device; the test subjects took a standardized diet after an initial rest, and the bowel sound signals were monitored in the subsequent 24 hours; data collection and analysis were performed every hour during the test to ensure the continuity and reliability of the data;

[0200] 1) Real-time data collection of pressure change data of the gastrointestinal tract inner wall and length change data of the gastrointestinal tract muscles, and real-time transmission of the collected data to the signal processing unit; use high-precision pressure sensors and length sensors to ensure the accuracy and stability of data collection;

[0201] 2) Receive the collected pressure change data and perform low-pass filtering to obtain filtered pressure data; at the same time, receive the collected length change data and perform calibration to eliminate the initial error and obtain calibrated length data; low-pass filtering uses an adaptive filter, which can dynamically adjust the filtering parameters according to the actual situation to ensure the best filtering effect;

[0202] 3) Obtain filtered pressure data and calibrated length data to calculate the muscle tension of the gastrointestinal tract;

[0203] 4) Collect bowel sound signals and process them through adaptive filtering algorithms to obtain filtered bowel sound signals; perform spectrum analysis on the filtered bowel sound signals and obtain the frequency component amplitude through fast Fourier transform (FFT); the adaptive filter can dynamically adjust the filtering parameters to adapt to different environments and individual differences;

[0204] 5) Determine the extraction time period immediately adjacent to the current time t. In this extraction time period, respectively extract tension features, spectrum features and time domain features from muscle tension, frequency component amplitude and bowel sound signals, and perform comprehensive representation on these extracted features. Feature extraction is based on time domain analysis and frequency domain analysis, which can fully reflect intestinal activity.

[0205] 6) Combine and analyze the tension features, spectrum features and time domain features to construct a comprehensive evaluation model for evaluating the intestinal activity of patients;

[0206] 7) Determine the previous extraction time period that is adjacent to the current extraction time period, calculate the output value of the comprehensive evaluation model of the previous extraction time period, combine and analyze the output values ​​of the comprehensive evaluation model of the current extraction time period with the output values ​​of the previous extraction time period, and formulate an adjustment strategy for dynamically adjusting the output value of the comprehensive evaluation model corresponding to the current extraction time period; the dynamic adjustment strategy is based on the difference change amount and the trend adjustment coefficient, and can flexibly respond to changes in intestinal activity;

[0207] The test data is recorded in the following Excel table, which reflects the comprehensive evaluation model output value, comprehensive tension value, comprehensive spectrum value and comprehensive time domain value parameters of different test objects in different time periods;

[0208] Please refer to Table 1 below for specific experimental tables;

[0209] Table 1

[0210]

[0211] The table analysis is as follows:

[0212] By analyzing the table data, we can see the intestinal activity parameters and the output values ​​of the comprehensive evaluation model for each patient in different time periods; the beneficial effects of the present invention and its innovation and novelty will be reflected through detailed analysis below;

[0213] It can be seen from the table data that the comprehensive evaluation model output value of each patient in different time periods will vary; the following is the change in the comprehensive evaluation model output value for each patient in the three data collections.

[0214] Patient A: 0.65>0.66>0.67, showing a gradually increasing trend;

[0215] Patient B: 0.70>0.71>0.72, showing a gradually increasing trend;

[0216] Patient C: 0.60>0.61>0.62, showing a gradually increasing trend;

[0217] Patient D: 0.68>0.69>0.70, showing a gradually increasing trend;

[0218] Patient E: 0.63>0.64>0.65, showing a gradually increasing trend.

[0219] Through these data, it can be found that the output value of the comprehensive evaluation model shows a stable upward trend in each data collection time period; this shows that the dynamic adjustment strategy of the present invention can effectively capture the changes in intestinal activity, reflect the changes in intestinal activity in different time periods, and improve the detection accuracy of bowel sound signals;

[0220] Comprehensive tension value : When it increases, the larger the weight coefficient d1 in the formula, the more significant its influence on the output value of the comprehensive evaluation model; for every increase of 0.01N in tension value, the output value of the comprehensive evaluation model increases by 0.01-0.02 on average;

[0221] Comprehensive spectrum value : When the spectrum feature increases, the larger the weight coefficient d2 in the formula, the more significant the impact on the output value of the comprehensive evaluation model; for every increase of 0.01Hz in the spectrum value, the output value of the comprehensive evaluation model increases by 0.01-0.02 on average;

[0222] Comprehensive time domain value : When the time domain feature increases, the larger the coefficient d3 in the formula, the greater the impact on the output value of the comprehensive evaluation model; for every increase of 0.01s in the time domain value, the output value of the comprehensive evaluation model increases by 0.02-0.03 on average;

[0223] Through the adjustment coefficients d4 and d5 in the formula, the formula has an adaptive adjustment function, which can smooth the changes of various parameters and make the output value of the comprehensive evaluation model more stable and reliable;

[0224] The specific technical effects and numerical percentages of each interval are quantified as follows:

[0225] Comprehensive tension value Increase: From the first to the third acquisition, the tension value of patient A increased by 4.44% (from 0.45N to 0.47N), and the corresponding output value of the comprehensive evaluation model increased by 3.08% (from 0.65 to 0.67);

[0226] Comprehensive spectrum value Increase: The spectrum value of patient A increased by 5.26% (from 0.38 Hz to 0.40 Hz), and the corresponding output value of the comprehensive evaluation model increased by 3.08% (from 0.65 to 0.67);

[0227] Comprehensive time domain value Increase: The time domain value of patient A increased by 8.0% (from 0.25s to 0.27s), and the corresponding output value of the comprehensive evaluation model increased by 3.08% (from 0.65 to 0.67);

[0228] Through the above analysis, it can be concluded that the present invention has the following beneficial effects:

[0229] Improve the accuracy of bowel sound signal detection: Through dynamic adjustment strategies, it is possible to capture changes in intestinal activity in real time, making the detection of bowel sound signals more accurate;

[0230] Provide comprehensive and accurate intestinal activity assessment: The comprehensive evaluation model can combine tension characteristics, spectrum characteristics and time domain characteristics, and provide a more comprehensive and accurate intestinal activity assessment through adaptive adjustment function.

[0231] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0232] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0233] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0234] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A signal processing method for a bowel sound signal monitoring system, characterized in that: The specific steps include: Step S1: collecting pressure change data of the inner wall of the gastrointestinal tract and length change data of the gastrointestinal muscles in real time, and transmitting the collected data to the signal processing unit in real time; Step S2: receiving the collected pressure change data and performing low-pass filtering to obtain filtered pressure data; receiving the collected length change data and performing calibration to eliminate initial errors to obtain calibrated length data; Step S3: Obtain filtered pressure data and calibrated length data, and calculate the muscle tension of the gastrointestinal tract; Step S4: collecting bowel sound signals and processing them with an adaptive filtering algorithm to obtain filtered bowel sound signals; performing spectrum analysis on the filtered bowel sound signals and obtaining frequency component amplitudes through fast Fourier transform; Step S5: determining an extraction time period immediately adjacent to the current time t, and in the extraction time period, extracting tension features, spectrum features, and time domain features from muscle tension, frequency component amplitude, and bowel sound signals respectively, and performing comprehensive representation on each of the extracted features; Set the time period Td immediately before the current time t as the extraction time period, mark U time sampling points in the extraction time period Td, and form a mark set of time sampling points {1, 2, ..., z, ..., U}, where z represents the zth time sampling point; In the extraction time period Td, the tension feature is extracted from the muscle tension calculated in step S3, and the tension feature includes the average tension, the maximum tension and the tension change rate in the extraction time period Td; and the average tension, the maximum tension and the tension change rate in the extraction time period Td are recorded as ZL Td，max , ZL Td，bh ; During the extraction period, the frequency component amplitude F calculated in step S4 k The spectrum features are extracted from the spectral features, which include the spectrum peak value and the spectrum intensity change rate within the extraction time period Td; and the spectrum peak value and the spectrum intensity change rate within the extraction time period Td are respectively recorded as PF Td , PW Td ; During the extraction period, the S calculated from step S4 filtered (t) extracts the time domain features, which include the average signal strength and signal change rate in the extraction time period Td; and the average signal strength and signal change rate in the extraction time period Td are respectively recorded as XV Td ; The tension characteristics within the extraction time period Td are comprehensively expressed as: in, is the comprehensive tension value within the extraction time period Td, a1, a2, a3 are the weight coefficients of the corresponding parameters; The spectrum characteristics within the extraction time period Td are comprehensively expressed as: <h2 style=";text-align:left;direction:ltr">F<h2 style=";text-align:left;direction:ltr"> Td,combined <h2 style=";text-align:left;direction:ltr"> =b1·PF<h2 style=";text-align:left;direction:ltr"> Td <h2 style=";text-align:left;direction:ltr"> +b2·PQ<h2 style=";text-align:left;direction:ltr"> Td Among them, F Td，combined To extract the comprehensive spectrum value within the time period Td, b1 and b2 are the weight parameters of the corresponding parameters. The time domain features within the extraction time period Td are comprehensively expressed as: Among them, XQ Td，combined To extract the comprehensive time domain value within the time period Td, c1 and c2 are the weight parameters of the corresponding parameters. Step S6: combining the tension features, spectrum features and time domain features represented comprehensively and analyzing and processing them to construct a comprehensive evaluation model for evaluating the intestinal activity of the patient; Define the output value of the comprehensive evaluation model in the extraction time period Td as OP Td , the calculation formula is as follows: in, d1, d2, and d3 are weight coefficients of the corresponding parameters; d4 is the adjustment coefficient of the normalization term, 0.12≤d4≤0.89, and d5 is the adjustment coefficient of the exponential function, 0.08≤d4≤0.93; OP Td The valid value range of is (0, 1); When OP Td The closer it is to 0, the lower the intestinal activity of the patient, and further monitoring or treatment measures are needed; When OP Td When in the middle range, 0.3≤OP Td <0.7, indicating that the patient's intestinal motility is within the normal range, but regular monitoring is required; When OP Td The closer it is to 1, the higher the intestinal activity of the patient, indicating good intestinal function; When OP Td When <0.3, it means OP Ta In the interval approaching 0, the corresponding signal output is a low-level signal, and the intestinal activity is at a low level; the comprehensive tension value, comprehensive spectrum value, and comprehensive time domain value are all at a low level, and the normalization item d4 and the index item d5 have a greater impact, which inhibits the overall score; When 0.3≤OP Td When <0.7, it means OP Td In the middle range, the corresponding signal output is a normal level signal, and the intestinal activity is at a normal level; the comprehensive tension value, comprehensive spectrum value, and comprehensive time domain value are in the middle range; When 0.7≤OP Ta When <1.0, it means OP Td In the interval approaching 1, the corresponding signal output is a high-level signal, and the intestinal activity is at a high level; the comprehensive tension value, comprehensive spectrum value, and comprehensive time domain value are all at a high level, and the normalization term d4 and the index term d5 have little influence, and the inhibitory effect on the score is weak; Step S7: determining the previous extraction time period immediately adjacent to the current extraction time period, and calculating the output value of the comprehensive evaluation model of the previous extraction time period, combining and analyzing the output values ​​of the comprehensive evaluation model of the current extraction time period and the previous extraction time period, and formulating an adjustment strategy for dynamically adjusting the output value of the comprehensive evaluation model corresponding to the current extraction time period; Define the previous extraction time period immediately adjacent to the current extraction time period Td as Td-1, the time sampling points of the extraction time period Td and the previous extraction time period Td-1 do not overlap, and the last time sampling point of the previous extraction time period Td-1 is continuous with the first time sampling point of the extraction time period Td; The calculation formula for defining the dynamic factor k′ is: Among them, OP Td-1 represents the output value of the comprehensive evaluation model in the previous extraction time period Td-1, and ΔOP represents OP Td With OP Td-1 The difference change; R represents the trend adjustment coefficient; f1 and f2 are the adjustment parameters of the corresponding parameters; 0.35≤f1≤0.78, 0.23≤f2≤0.66; Set the trigger condition to M(k′): in, Indicates rounding up to an integer; When OP Td Lower than When , make the following fine adjustments to d1, d2, d3, d4, and d5: Increase d1 by 5% of its current value, so that the tension feature weight increases, and d2 and d3 decrease by 2% and 3% respectively, and d4 and d5 decrease by 3% and increase by 4% respectively; when When , make the following fine adjustments to d1, d2, d3, d4, and d5: Increasing d1 by 3% of its current value increases the tension feature weight, and d2 and d3 decrease by 1% and 2% respectively, and d4 and d5 decrease by 1% and increase by 2% respectively.

2. The signal processing method for a bowel sound signal monitoring system according to claim 1, characterized in that: The acquisition of filtered pressure data and calibrated length data specifically includes: The pressure change data and length change data collected at time t are recorded as P(t) and L(t), respectively, and the filtered pressure data is recorded as P filtered (t), P filtered (t) is calculated as follows: Among them, h(i) is the impulse response of the filter, which represents the weight coefficient of the filter at the i-th time offset, N is the order of the filter, which represents the window size of the filter; i represents the time offset relative to the current time point t, ranging from -N to N.

3. The signal processing method for a bowel sound signal monitoring system according to claim 2, characterized in that: The length data after calibration at time t is recorded as L calibrated (t), the calculation formula is as follows: L calibrated (t)=L(t)-L initial Among them, L initial is the initial length of the gastrointestinal muscle before data collection; The muscle tension of the gastrointestinal tract at time t is denoted as T muscle (t), the calculation formula is as follows: Where E is the elastic modulus, K is the stiffness coefficient, ΔL(t) is the length change after calibration, AP(t) is the pressure change after filtering, and L rest and P rest They are the average length of the gastrointestinal muscles and the average pressure of the gastrointestinal lining of the patient at rest.

4. The signal processing method for a bowel sound signal monitoring system according to claim 3, characterized in that: The bowel sound signal collected at time t is recorded as S(t), and the bowel sound signal S(t) obtained by the adaptive filtering algorithm is recorded as S filtered (t).

5. The signal processing method for a bowel sound signal monitoring system according to claim 4, characterized in that: The filtered bowel sound signal is subjected to spectrum analysis, and the frequency component amplitude is obtained by fast Fourier transform, including: The filtered bowel sound signal S filtered (t) Divide into several segments according to the length of each segment N′; set the bowel sound signal of each segment to S segment (n); n represents the sampling point index within the segment; Apply the window function to each signal segment and transform each bowel sound signal S segment (n) is multiplied by the window function w(n) to obtain the following windowed bowel sound signal value S windowed (n): S windowed (n)=S segment (n)·w(n) For each windowed bowel sound signal S windowed (n) Perform fast Fourier transform FFT to obtain the frequency component amplitude F k : Where n represents the sampling point index within the segment; k is the frequency index, is the basis function of Fourier transform, and j is the imaginary unit.

Images