A patient postoperative eating intolerance risk assessment system based on bowel sound signal analysis and a method thereof

Abstract

This invention discloses a system and method for assessing the risk of postoperative food intolerance in patients based on bowel sound signal analysis, belonging to the field of medical monitoring technology. It includes a data acquisition module for acquiring bowel sound signals from the target patient after surgery, identifying audio signals corresponding to food intolerance from the acquired bowel sound signals, and marking these audio signals as reference signals. A data processing module is used to process the reference signals. The data processing module includes a processing unit and an analysis unit. The processing unit performs correlation processing on the reference signals, including noise reduction and filtering. The analysis unit analyzes the change path of the bowel sound signals. This invention provides multi-dimensional and dynamic signal analysis, and accurately identifies the risk of food intolerance in patients, providing medical staff with a scientific basis for developing personalized intervention measures, thereby preventing complications, reducing medical costs, and improving patient comfort.

Description

Technical Field

[0001] This invention relates to the field of medical monitoring technology, and in particular to a system and method for assessing the risk of postoperative food intolerance in patients based on bowel sound signal analysis. Background Technology

[0002] Enteral nutrition is a crucial supportive method for the postoperative recovery of critically ill patients. It provides essential nutrients, promotes the recovery of intestinal function, and reduces the risk of complications, playing a key role in clinical practice. Bowel sounds, as an objective indicator reflecting the state of intestinal peristalsis, are directly related to the patient's tolerance to enteral nutrition. Timely identification of tolerance differences helps avoid adverse events such as abdominal distension and vomiting caused by feeding intolerance.

[0003] Regarding this research, application number CN202510322928.X provides a risk warning system based on enteral nutrition management for patients. This technical solution includes a patient monitoring module, a scoring calculation module, a nutrition control module, a warning notification module, a nutrition plan adjustment module, and a continuous observation and feedback module. This invention addresses the problems of untimely symptom monitoring, low efficiency and error-prone regulation of enteral nutrition supply, and imperfect warning mechanisms in existing technologies for critically ill patients with acute respiratory distress syndrome. This technical solution solves the main problems currently existing in the enteral nutrition management of critically ill patients with acute respiratory distress syndrome.

[0004] Another application, CN202311377981.7, provides a method, apparatus, and storage medium for constructing a feature recognition model for bowel sound signals. This technical solution approximates the acoustic features of bowel sound signals using Chebyshev polynomials, effectively capturing the detailed features of the signal and reducing the impact of noise on the spectral fitting. Based on this, by combining a trained physiological sound recognition model and a trained sound recognition model, and utilizing the similarity of physiological sounds (i.e., a physiological sound pre-training model and a general sound pre-training model), it can capture and understand bowel sound signals from different perspectives, improving the robustness of recognition. The resulting bowel sound recognition model has good generalization performance and strong robustness when processing bowel sound signals with different features and diversity, while effectively reducing noise interference and improving recognition accuracy.

[0005] However, the above-mentioned technical solutions mainly rely on the doctor's subjective auscultation judgment, making it difficult to achieve objective data quantification. This makes it impossible to systematically transform a series of parameters such as the frequency, intensity, and duration of bowel sounds into analytical objects, resulting in the assessment results being easily influenced by the operator's experience, which is not conducive to accurately assessing the risk of food intolerance. Summary of the Invention

[0006] In view of the problems existing in the field of medical monitoring technology, the present invention is proposed.

[0007] Therefore, one of the objectives of this invention is to provide a system and method for assessing the risk of postoperative food intolerance in patients based on bowel sound signal analysis. Through multi-dimensional and dynamic signal analysis and accurate identification of the risk of food intolerance in patients, it provides medical staff with a scientific basis for developing personalized intervention measures, thereby preventing complications, reducing medical costs, and improving patient comfort.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution: On one hand, the present invention provides a patient postoperative food intolerance risk assessment system based on bowel sound signal analysis, comprising: The data acquisition module is used to acquire bowel sound signals of the target patient after surgery, identify the audio signal corresponding to food intolerance from the acquired bowel sound signals, and mark the audio signal as a reference signal. The data processing module is used to process the reference signal; the data processing module includes a processing unit and an analysis unit. The processing unit is used to perform correlation processing on the reference signal, and the correlation processing includes noise reduction and filtering. The analysis unit includes analyzing the change path of the bowel sound signal, and dividing the change path into a pre-change path and a post-change path; the pre-change path includes the range in which the signal waveform of the bowel sound signal changes steadily; The post-change link includes the interval from the first identification of the reference signal in the bowel sound signal until the reference signal disappears; Acquire the final audio signal in the interval segment corresponding to the preceding change link, obtain the time corresponding to the final audio signal, obtain the time corresponding to the appearance of the reference signal based on the time, calculate the interval between the two time periods, and analyze the change characteristics of the final audio signal toward the reference signal with one second as an analysis node in the interval. A data fusion processing module, which includes a data acquisition unit; The acquisition unit is used to perform relevant acquisition based on the final audio signal, and the acquisition steps include: The selection is based on the time corresponding to the acquired end audio signal, including selecting bowel sound signals at least 10 minutes before the time, and marking the time corresponding to this bowel sound signal to the time corresponding to the end audio signal as a reference time interval; Relevant data is collected during the reference time interval. The relevant data includes the peak values ​​of the collected bowel sound signals. The collected peak values ​​include at least 6 peak values ​​that reach a preset peak value, which is a peak value between 45 and 48 dB.

[0009] In a preferred embodiment of the present invention, the data fusion processing module further includes an extraction unit and an evaluation unit; The extraction unit is used to perform feature extraction, which includes extracting features from the six signal peaks, extracting the time interval between adjacent signal peaks, and marking the time interval as a risk time interval. The feature extraction also includes extracting the peak difference between adjacent signal peaks; The assessment unit responds to the extraction unit and is used to assess the risk of food intolerance based on the extracted features. When assessing the risk of food intolerance for a target patient at a future time, if the peak value of the collected bowel sound signal is the same as the preset peak value, and the time interval between adjacent collected signal peak values ​​is shorter than the risk time interval, the system determines that the target patient has the risk of food intolerance; otherwise, it does not determine the risk.

[0010] In a preferred embodiment of the present invention, the analysis unit analyzes the variation characteristics of the final audio signal toward the reference signal at a rate of one analysis node per second, and performs the analysis according to the following calculation formula: ; In the formula, Indicates the rate of change of signal amplitude; Indicates the relationship with the first The amplitude of the final audio signal corresponding to each analysis node; Indicates the relationship with the first +1 amplitude of the final audio signal corresponding to the analysis node; Indicates the time interval between adjacent analysis nodes; In a preferred embodiment of the present invention, the analysis is further performed according to the following calculation formula: ; In the formula, Indicates the rate of change of signal frequency; Indicates the relationship with the first The frequency of the final audio signal corresponding to each analysis node; Indicates the relationship with the first +1 analysis node corresponding to the frequency of the final audio signal; This indicates the time interval between adjacent analysis nodes.

[0011] In a preferred embodiment of the present invention, if the peak value of the collected bowel sound signal is the same as the preset peak value, and the time interval between adjacent collected signal peak values ​​is shorter than the risk time interval, but the peak value difference between adjacent signal peak values ​​is less than 2%, then the system determines that the target patient does not have the risk of food intolerance; otherwise, it does not determine this.

[0012] In a preferred embodiment of the present invention, if, based on the results of the analysis of the variation characteristics, the system determines that the target patient does not have the risk of food intolerance because the peak difference between adjacent signal peaks is less than 2%, then a correlation analysis is performed. The correlation analysis includes analyzing the similarity between the final audio signal and the reference signal at different analysis nodes. If the similarity is low, the determination that the target patient does not have the risk of food intolerance is maintained; otherwise, if the similarity is high, this determination is not maintained.

[0013] In a preferred embodiment of the present invention, the relevant analysis is performed according to the following calculation formula: ; In the formula, Indicates the first The correlation coefficient between the final audio signal of each analysis node and the reference signal; Indicates the first The sampling sequence of the final audio signal of each analysis node. This indicates the first in this sequence. One sampling point; This represents the sampling sequence of the reference signal. This indicates the first in this sequence. One sampling point; This represents the number of sampling points; Indicates the first Sampling sequence of the last segment of audio signal of each analysis node The mean; Sampling sequence representing the reference signal The mean.

[0014] In a preferred embodiment of the present invention, if the system determines that the target patient does not have the risk of food intolerance because the peak difference between adjacent signal peaks is less than 2%, then the relevant distinction is made based on the 2%, and the steps are as follows: If the time interval between adjacent signal peaks is shorter than the risk time interval, the duration corresponding to the shorter risk time interval is obtained, and this duration is regarded as the target duration. Based on the target duration, the duration with a peak difference below 2% is obtained; this duration is marked as the occupied duration. When the risk assessment of food intolerance is performed on the target patient in the future, if the peak difference between adjacent signal peaks is also below 2%, but the duration of the difference is less than the duration of the difference, the system determines that the target patient has changed from having no risk of food intolerance to having the risk of food intolerance. If the peak difference between adjacent signal peaks is also below 2%, but the duration of time spent below 2% is longer than the specified duration, then the system determines that the target patient does not have the risk of food intolerance, and the state corresponding to not having the risk of food intolerance is a stable state.

[0015] On the other hand, the present invention provides a method for applying a patient postoperative food intolerance risk assessment system based on bowel sound signal analysis as described above, comprising the following steps: Obtain bowel sound signals after surgery in the target patient, identify the audio signals corresponding to food intolerance from the obtained bowel sound signals, and mark the audio signals as reference signals; The reference signal is subjected to correlation processing, which includes noise reduction and filtering; The variation path of bowel sound signals is analyzed and divided into an earlier variation path and a later variation path; the earlier variation path includes the interval where the signal waveform of the bowel sound signal changes steadily. The post-change link includes the interval from the first identification of the reference signal in the bowel sound signal until the reference signal disappears; Acquire the final audio signal in the interval segment corresponding to the preceding change link, obtain the time corresponding to the final audio signal, obtain the time corresponding to the appearance of the reference signal based on the time, calculate the interval between the two time periods, and analyze the change characteristics of the final audio signal toward the reference signal with one second as an analysis node in the interval. Based on the final audio signal, relevant data acquisition is performed, and the acquisition steps include: The selection is based on the time corresponding to the acquired end audio signal, including selecting bowel sound signals at least 10 minutes before the time, and marking the time corresponding to this bowel sound signal to the time corresponding to the end audio signal as a reference time interval; Relevant data is collected during the reference time interval. The relevant data includes the peak values ​​of the collected bowel sound signals. The collected peak values ​​include at least 6 peak values ​​that reach a preset peak value, which is a peak value between 45 and 48 dB. The data fusion processing module also includes an extraction unit and an evaluation unit; The extraction unit is used to perform feature extraction, which includes extracting features from the six signal peaks, extracting the time interval between adjacent signal peaks, and marking the time interval as a risk time interval. The feature extraction also includes extracting the peak difference between adjacent signal peaks; The assessment unit responds to the extraction unit and is used to assess the risk of food intolerance based on the extracted features. When assessing the risk of food intolerance for a target patient at a future time, if the peak value of the collected bowel sound signal is the same as the preset peak value, and the time interval between adjacent collected signal peak values ​​is shorter than the risk time interval, the system determines that the target patient has the risk of food intolerance; otherwise, it does not determine the risk.

[0016] Beneficial effects

[0017] 1. This invention not only analyzes the peak value of bowel sound signals, but also comprehensively considers multiple dimensions such as signal waveform, frequency change rate, and time interval. It improves the accuracy of risk assessment through algorithm models. For example, the system analyzes the change characteristics of the final audio signal towards the reference signal by calculating the signal amplitude change rate and frequency change rate, thereby more accurately judging the risk of food intolerance. 2. The present invention can provide personalized assessment based on the patient's specific situation. Taking into account the physiological differences between different patients, for example, the system extracts features such as the time interval and peak difference between adjacent signal peaks, and combines the patient's historical data and preset peak values ​​to provide the patient with customized risk assessment results. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic diagram of the modular structure of the patient postoperative food intolerance risk assessment system based on bowel sound signal analysis according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the method flow according to an embodiment of the present invention; Figure 3This is the flowchart of the method for acquiring the final audio signal in this embodiment of the invention; The diagram is labeled as follows: 110 - Data acquisition module; 120 - Data processing module; 1201 - Processing unit; 1202 - Analysis unit; 130 - Data fusion processing module; 1301 - Acquisition unit; 1302 - Extraction unit; 1303 - Evaluation unit. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0020] Because existing technical solutions mainly rely on doctors' subjective auscultation judgment, it is difficult to achieve objective data quantification, which is not conducive to accurately assessing the risk of food intolerance.

[0021] Based on this, the present invention proposes a patient postoperative food intolerance risk assessment system and method based on bowel sound signal analysis. Through multi-dimensional and dynamic signal analysis and accurate identification of patients' food intolerance risk, it provides medical staff with a scientific basis for developing personalized intervention measures, thereby preventing complications, reducing medical costs, and improving patient comfort.

[0022] The present solution will be further described in detail below through embodiments and in conjunction with the accompanying drawings.

[0023] Reference Figures 1 to 3 As one embodiment of the present invention, this embodiment provides a patient postoperative food intolerance risk assessment system based on bowel sound signal analysis, comprising: The data acquisition module 110 is used to acquire bowel sound signals after surgery of the target patient, identify the audio signal corresponding to food intolerance from the acquired bowel sound signals, and mark the audio signal as a reference signal. In this embodiment, the system acquires the bowel sound signal of the patient after surgery through a sensor and identifies characteristic audio signals related to food intolerance from it; By collecting signals, abnormalities can be detected at an early stage, avoiding the lag of relying on subjective symptom reports; Data processing module 120 is used to process the reference signal; data processing module 10 includes processing unit 1201 and analysis unit 1202; Processing unit 1201 is used to perform correlation processing on the reference signal, including noise reduction and filtering; The analysis unit 1202 includes analyzing the change path of the bowel sound signal and dividing the change path into the pre-change path and the post-change path; the pre-change path includes the range in which the signal waveform of the bowel sound signal is in a stable change range; It should be noted that the range of stable changes is defined as signal fluctuations between 10 and 50 dB; among them, approximately 10 to 30 dB (sound pressure level) corresponds to slight intestinal peristalsis, approximately 30 to 50 dB corresponds to intestinal activity after normal eating, and more than 50 dB may indicate intestinal hyperactivity (such as diarrhea or early intestinal obstruction). The post-change link includes the interval from the first identification of the reference signal in the bowel sound signal to the disappearance of the reference signal; By distinguishing the stages of signal change, the timing of the appearance of food intolerance-related characteristics can be clarified, thereby improving the pertinence of risk assessment. Acquire the final audio signal in the interval corresponding to the previous change link, obtain the time corresponding to the final audio signal, obtain the time corresponding to the appearance of the reference signal based on the time, calculate the interval between the two time periods, and analyze the change characteristics of the final audio signal toward the reference signal with one second as an analysis node within the interval. By analyzing the rate of change (amplitude and frequency) of the signal per second, the real-time dynamics of intestinal peristalsis can be captured, avoiding the limitations of static analysis.

[0024] It should be noted that analyzing the changes in the final audio signal towards the reference signal at a rate of one second allows for a detailed capture of the characteristics of this change process, thus providing more accurate data support for assessing the risk of postoperative food intolerance in patients. By calculating the characteristic parameters of the signal at different time points, the changes in the signal can be quantified, and then the association between the signal and food intolerance can be analyzed.

[0025] Data fusion processing module 130, which includes acquisition unit 1301; Acquisition unit 1301 is used to perform correlation acquisition based on the final audio signal. The acquisition steps include: The selection is based on the time corresponding to the acquired end audio signal, including selecting bowel sound signals at least 10 minutes before the time, and marking the time corresponding to this bowel sound signal to the time corresponding to the end audio signal as a reference time interval; Relevant data are collected during the reference time interval. The relevant data includes the peak values ​​of the collected bowel sound signals. The collected peak values ​​include at least 6 peak values ​​that reach a preset peak value, which is a peak value between 45 and 48 dB. By combining peak intensity, time interval, and variability, misjudgments caused by a single parameter can be avoided, thus improving the comprehensiveness of risk assessment. The system's applicability is enhanced by adapting to different patient baselines through a preset peak range (45–48 dB).

[0026] The data fusion processing module 130 also includes an extraction unit 1302 and an evaluation unit 1303; The extraction unit 1302 is used to perform feature extraction, which includes extracting features from 6 signal peaks, extracting the time interval between adjacent signal peaks, and marking the time interval as a risk time interval. Feature extraction also includes extracting the peak difference between adjacent signal peaks; The assessment unit 1303 is a response extraction unit used to assess the risk of food intolerance based on the extracted features. When assessing the risk of food intolerance for a target patient at a future time, if the peak value of the collected bowel sound signal is the same as the preset peak value, and the time interval between adjacent collected signal peak values ​​is shorter than the risk time interval, the system determines that the target patient has the risk of food intolerance; otherwise, it does not determine the risk.

[0027] In the analysis unit, the variation characteristics of the final audio signal towards the reference signal are analyzed with each second as an analysis node, and the analysis is performed according to the following calculation formula: ; In the formula, This represents the rate of change of signal amplitude (used to reflect how quickly the signal amplitude changes between adjacent analysis nodes). Indicates the relationship with the first The amplitude of the final audio signal corresponding to each analysis node; Indicates the relationship with the first +1 amplitude of the final audio signal corresponding to the analysis node; This represents the time interval between adjacent analysis nodes (fixed to 1 second here). In addition to the above, analysis is also performed based on the following calculation formula: ; In the formula, This represents the rate of change of signal frequency (used to reflect the rate of change of signal frequency between adjacent analysis nodes). Indicates the relationship with the first The (primary) frequency of the final audio signal corresponding to each analysis node (reflecting the speed of signal vibration at this moment); Indicates the relationship with the first +1 analysis node corresponding to the (primary) frequency of the final audio signal; Indicates the time interval between adjacent analysis nodes; It should be noted that the two calculation formulas mentioned above are used to analyze the amplitude change rate and frequency change rate of bowel sound signals, respectively. The two are complementary and correlated by jointly describing the dynamic characteristics of the signal. Both calculation formulas use time intervals. Using the denominator, the rate of change of the uniformly quantified signal characteristics over time can be used to distinguish the types of intestinal dysfunction (such as intensity mutations or rhythm disorders) by synchronously analyzing the dynamic trends of amplitude and frequency, thus avoiding the limitations of single-parameter analysis.

[0028] Abnormal intestinal motility may manifest as changes in both amplitude and frequency. For example, patients with food intolerance may experience a sudden increase in the amplitude (spasm) and an increase in the frequency (tachymosm) of bowel sounds due to intestinal stimulation.

[0029] Combining the two calculation formulas can comprehensively capture such complex features and improve the sensitivity of risk assessment; By calculating the rate of change every second, the system can monitor the evolution trend of signal characteristics in real time.

[0030] If the amplitude and frequency of change increase simultaneously, it may indicate a sharp deterioration in intestinal function, requiring immediate intervention; if only a single parameter is abnormal, it may be a temporary fluctuation, and a comprehensive judgment should be made in combination with other characteristics.

[0031] In the early stages of postoperative bowel function recovery, bowel sounds may be characterized by low amplitude but high frequency (insufficient motility).

[0032] The system can distinguish between "inefficient peristalsis" and "obstruction risk" by comparing the relative magnitudes of the rate of change of amplitude and the rate of change of frequency, thus guiding precise treatment.

[0033] When assessing the risk of food intolerance, if the rate of change in amplitude is significantly higher than the rate of change in frequency, it may indicate an overreaction of the intestine to stimulation (such as spasms); conversely, it may indicate metabolic factors (such as electrolyte imbalances). This grading mechanism helps in developing differentiated intervention strategies.

[0034] Based on the extracted peak difference, if the peak value of the collected bowel sound signal is the same as the preset peak value, and the time interval between adjacent signal peak values ​​is shorter than the risk time interval, but the peak difference between adjacent signal peak values ​​is less than 2%, then the system determines that the target patient does not have the risk of food intolerance; otherwise, it does not determine the risk. In reality, for some patients with intestinal diseases who are in a stable phase, their intestinal function recovers to a relatively normal state after treatment or self-adjustment. At this time, the peak variation of bowel sound signals will also be smaller, reflecting relatively stable intestinal peristalsis. For example, after effective drug treatment, patients with ulcerative colitis have their condition controlled, intestinal inflammation reduced, and intestinal peristalsis returns to normal, generally without triggering intolerance reactions after eating.

[0035] Based on the results of the change characteristics analysis, if the system determines that the target patient does not have the risk of food intolerance because the peak difference between adjacent signal peaks is less than 2%, then a correlation analysis is performed. The correlation analysis includes analyzing the similarity between the final audio signal and the reference signal at different analysis nodes. If the similarity is low, the determination that the target patient does not have the risk of food intolerance is maintained; otherwise, if the similarity is high, this determination is not maintained.

[0036] Based on the above, relevant analysis is performed according to the following calculation formula: ; In the formula, Indicates the first The correlation coefficient between the final audio signal of each analysis node and the reference signal (the value ranges from [-1, 1]). Indicates the first The sampling sequence of the final audio signal of each analysis node. This indicates the first in this sequence. One sampling point; This represents the sampling sequence of the reference signal. This indicates the first in this sequence. One sampling point; The number of sampling points (the number of sampling points is the number of times the final audio signal and the reference signal are sampled within the time period corresponding to an analysis node); Indicates the first Sampling sequence of the last segment of audio signal of each analysis node The mean; The calculation formula is ; Sampling sequence representing the reference signal The mean; The calculation formula is ; If the peak difference is less than 2%, the risk is eliminated; however, further analysis of the correlation coefficient between the final audio signal and the reference signal (calculated by formula) is required to confirm the similarity in order to maintain or correct the judgment; through the dual verification of peak difference and correlation coefficient, false positive results caused by brief signal fluctuations can be avoided.

[0037] If the system determines that the target patient does not have the risk of food intolerance because the peak difference between adjacent signal peaks is less than 2%, then the relevant distinction is made based on 2%, as follows: If the time interval between adjacent signal peaks is shorter than the risk time interval, the duration corresponding to the shorter risk time interval is obtained and this duration is regarded as the target duration. Based on the target duration, obtain the duration where the peak difference is below 2%; mark the duration as occupied duration; When assessing the risk of food intolerance in a target patient at a future time, if the peak difference between adjacent signal peaks is also below 2%, but the duration of the difference is less than the duration of the difference, the system determines that the target patient has changed from having no risk of food intolerance to having the risk of food intolerance. If the peak difference between adjacent signal peaks is also below 2%, but the duration of time spent below 2% is longer than the total duration of time spent in the system, the system determines that the target patient does not have the risk of food intolerance, and the state corresponding to not having the risk of food intolerance is a stable state.

[0038] In this embodiment, the duration of peak difference ratio (occupancy time) is continuously monitored to determine whether the risk status is stable, providing a basis for intervention timing.

[0039] Based on the above, this application provides medical staff with a scientific basis for developing personalized intervention measures through multi-dimensional and dynamic signal analysis and accurate identification of patients' food intolerance risk, thereby preventing complications, reducing medical costs, and improving patient comfort.

[0040] This embodiment, in conjunction with the aforementioned patient postoperative food intolerance risk assessment system based on bowel sound signal analysis, also proposes a working method for applying this system, as follows: S10: Acquire bowel sound signals after surgery in the target patient, identify the audio signals corresponding to food intolerance from the acquired bowel sound signals, and mark the audio signals as reference signals; S20: Perform correlation processing on the reference signal, including noise reduction and filtering; S30: Analyze the variation path of bowel sound signals, and divide the variation path into the pre-variation path and the post-variation path; the pre-variation path includes the range where the signal waveform of the bowel sound signal is in a stable variation range; The post-change link includes the interval from the first identification of the reference signal in the bowel sound signal to the disappearance of the reference signal; S40: Acquire the last audio signal in the interval segment corresponding to the previous change link, obtain the time corresponding to the last audio signal, obtain the time corresponding to the appearance of the reference signal based on the time, calculate the interval between the two time periods, and analyze the change characteristics of the last audio signal toward the reference signal with one second as an analysis node in the interval. S50: Perform relevant data acquisition based on the final audio signal. The acquisition steps include: S501: Select based on the time corresponding to the acquired end audio signal, including selecting bowel sound signals at least 10 minutes before the time, and marking the time corresponding to this bowel sound signal to the time corresponding to the end audio signal as a reference time interval; S502: Collect relevant data during the reference time interval. The relevant data includes the peak values ​​of the collected bowel sound signals. The collected peak values ​​include at least 6 peak values ​​that reach a preset peak value. The preset peak value is a peak value between 45 and 48 dB. S60: Perform feature extraction, which includes feature extraction from 6 signal peaks, extracting the time interval between adjacent signal peaks, and marking the time interval as a risk time interval; Feature extraction also includes extracting the peak difference between adjacent signal peaks; S70: Based on the extracted features, a risk assessment of food intolerance is performed. When a risk assessment of food intolerance is performed on a target patient at a future time, if the peak value of the collected bowel sound signal is the same as the preset peak value, and the time interval between adjacent collected signal peak values ​​is shorter than the risk time interval, the system determines that the target patient has a risk of food intolerance; otherwise, no assessment is made.

[0041] In summary, this invention provides a scientific and accurate solution for assessing the risk of postoperative food intolerance in patients through multi-dimensional and dynamic bowel sound signal analysis, thereby improving the quality of medical care and reducing medical costs.

[0042] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A patient postoperative food intolerance risk assessment system based on bowel sound signal analysis, characterized in that, include: The data acquisition module is used to acquire bowel sound signals of the target patient after surgery, identify the audio signal corresponding to food intolerance from the acquired bowel sound signals, and mark the audio signal as a reference signal. The data processing module is used to process the reference signal; the data processing module includes a processing unit and an analysis unit. The processing unit is used to perform correlation processing on the reference signal, and the correlation processing includes noise reduction and filtering. The analysis unit includes analyzing the change path of the bowel sound signal, and dividing the change path into a pre-change path and a post-change path; the pre-change path includes the range in which the signal waveform of the bowel sound signal changes steadily; The post-change link includes the interval from the first identification of the reference signal in the bowel sound signal until the reference signal disappears; Acquire the final audio signal in the interval segment corresponding to the preceding change link, obtain the time corresponding to the final audio signal, obtain the time corresponding to the appearance of the reference signal based on the time, calculate the interval between the two time periods, and analyze the change characteristics of the final audio signal toward the reference signal with one second as an analysis node in the interval. A data fusion processing module, which includes a data acquisition unit; The acquisition unit is used to perform relevant acquisition based on the final audio signal, and the acquisition steps include: The selection is based on the time corresponding to the acquired end audio signal, including selecting bowel sound signals at least 10 minutes before the time, and marking the time corresponding to this bowel sound signal to the time corresponding to the end audio signal as a reference time interval; Relevant data is collected during the reference time interval. The relevant data includes the peak values ​​of the collected bowel sound signals. The collected peak values ​​include at least 6 peak values ​​that reach a preset peak value, which is a peak value between 45 and 48 dB.

2. The postoperative food intolerance risk assessment system for patients based on bowel sound signal analysis as described in claim 1, characterized in that, The data fusion processing module also includes an extraction unit and an evaluation unit; The extraction unit is used to perform feature extraction, which includes extracting features from the six signal peaks, extracting the time interval between adjacent signal peaks, and marking the time interval as a risk time interval. The feature extraction also includes extracting the peak difference between adjacent signal peaks; The assessment unit responds to the extraction unit and is used to assess the risk of food intolerance based on the extracted features. When the target patient is assessed for food intolerance at a future time, if the peak value of the collected bowel sound signal is the same as the preset peak value, and the time interval between adjacent collected signal peak values ​​is shorter than the risk time interval, the system determines that the target patient has the risk of food intolerance. Conversely, no judgment is made.

3. The postoperative food intolerance risk assessment system for patients based on bowel sound signal analysis as described in claim 1, characterized in that, In the analysis unit, the variation characteristics of the final audio signal toward the reference signal are analyzed at a rate of one second, based on the following calculation formula: ； In the formula, Indicates the rate of change of signal amplitude; Indicates the relationship with the first The amplitude of the final audio signal corresponding to each analysis node; Indicates the relationship with the first +1 amplitude of the final audio signal corresponding to the analysis node; This indicates the time interval between adjacent analysis nodes.

4. The postoperative food intolerance risk assessment system for patients based on bowel sound signal analysis as described in claim 3, characterized in that, It also includes analysis based on the following calculation formula: ； In the formula, Indicates the rate of change of signal frequency; Indicates the relationship with the first The frequency of the final audio signal corresponding to each analysis node; Indicates the relationship with the first +1 analysis node corresponding to the frequency of the final audio signal; This indicates the time interval between adjacent analysis nodes.

5. The postoperative food intolerance risk assessment system for patients based on bowel sound signal analysis as described in claim 2, characterized in that, Based on the extracted peak difference, if the peak value of the collected bowel sound signal is the same as the preset peak value, and the time interval between adjacent collected signal peak values ​​is shorter than the risk time interval, but the peak difference between adjacent signal peak values ​​is less than 2%, then the system determines that the target patient does not have the risk of food intolerance; otherwise, it does not determine this.

6. The postoperative food intolerance risk assessment system for patients based on bowel sound signal analysis as described in claim 5, characterized in that, Based on the results of the change characteristics analysis, if the system determines that the target patient does not have the risk of food intolerance because the peak difference between adjacent signal peaks is less than 2%, then a correlation analysis is performed. The correlation analysis includes analyzing the similarity between the final audio signal and the reference signal at different analysis nodes. If the similarity is low, the determination that the target patient does not have the risk of food intolerance is maintained. Conversely, if the degree of similarity is high, this determination result is not maintained.

7. The postoperative food intolerance risk assessment system for patients based on bowel sound signal analysis as described in claim 6, characterized in that, The relevant analysis is performed based on the following calculation formula: ； In the formula, Indicates the first The correlation coefficient between the final audio signal of each analysis node and the reference signal; Indicates the first The sampling sequence of the final audio signal of each analysis node. This indicates the first in this sequence. One sampling point; This represents the sampling sequence of the reference signal. This indicates the first in this sequence. One sampling point; This represents the number of sampling points; Indicates the first Sampling sequence of the last segment of audio signal of each analysis node The mean; Sampling sequence representing the reference signal The mean.

8. The postoperative food intolerance risk assessment system for patients based on bowel sound signal analysis as described in claim 6, characterized in that, If the system determines that the target patient does not have the risk of food intolerance because the peak difference between adjacent signal peaks is less than 2%, then the relevant distinction is made based on the 2%, as follows: If the time interval between adjacent signal peaks is shorter than the risk time interval, the duration corresponding to the shorter risk time interval is obtained, and this duration is regarded as the target duration. Based on the target duration, the duration in which the peak difference is below 2% is obtained; Mark the duration as the occupied duration; When the risk assessment of food intolerance is performed on the target patient in the future, if the peak difference between adjacent signal peaks is also below 2%, but the duration of the difference is less than the duration of the difference, the system determines that the target patient has changed from having no risk of food intolerance to having the risk of food intolerance. If the peak difference between adjacent signal peaks is also below 2%, but the duration of time spent below 2% is longer than the specified duration, then the system determines that the target patient does not have the risk of food intolerance, and the state corresponding to not having the risk of food intolerance is a stable state.

9. A method applied to a patient postoperative food intolerance risk assessment system based on bowel sound signal analysis as described in claim 1, characterized in that, Includes the following steps: Obtain bowel sound signals after surgery in the target patient, identify the audio signals corresponding to food intolerance from the obtained bowel sound signals, and mark the audio signals as reference signals; The reference signal is subjected to correlation processing, which includes noise reduction and filtering; The variation path of bowel sound signals is analyzed and divided into an earlier variation path and a later variation path; the earlier variation path includes the interval where the signal waveform of the bowel sound signal changes steadily. The post-change link includes the interval from the first identification of the reference signal in the bowel sound signal until the reference signal disappears; Acquire the final audio signal in the interval segment corresponding to the preceding change link, obtain the time corresponding to the final audio signal, obtain the time corresponding to the appearance of the reference signal based on the time, calculate the interval between the two time periods, and analyze the change characteristics of the final audio signal toward the reference signal with one second as an analysis node in the interval. Based on the final audio signal, relevant data acquisition is performed, and the acquisition steps include: The selection is based on the time corresponding to the acquired end audio signal, including selecting bowel sound signals at least 10 minutes before the time, and marking the time corresponding to this bowel sound signal to the time corresponding to the end audio signal as a reference time interval; Relevant data is collected during the reference time interval. The relevant data includes the peak values ​​of the collected bowel sound signals. The collected peak values ​​include at least 6 peak values ​​that reach a preset peak value, which is a peak value between 45 and 48 dB.

10. The method as described in claim 9, applied to a patient postoperative food intolerance risk assessment system based on bowel sound signal analysis, characterized in that... The data fusion processing module also includes an extraction unit and an evaluation unit; The extraction unit is used to perform feature extraction, which includes extracting features from the six signal peaks, extracting the time interval between adjacent signal peaks, and marking the time interval as a risk time interval. The feature extraction also includes extracting the peak difference between adjacent signal peaks; The assessment unit responds to the extraction unit and is used to assess the risk of food intolerance based on the extracted features. When assessing the risk of food intolerance for a target patient at a future time, if the peak value of the collected bowel sound signal is the same as the preset peak value, and the time interval between adjacent collected signal peak values ​​is shorter than the risk time interval, the system determines that the target patient has the risk of food intolerance; otherwise, it does not determine the risk.

Images