Intelligent multi-mode man-machine interaction stomach tube system

Abstract

The invention relates to the technical field of stomach tube interaction, and discloses an intelligent multi-mode man-machine interaction stomach tube system, which comprises a stomach state threshold evaluation module, a stomach state monitoring module, a stomach interaction module and a stomach tube verification module, and is characterized in that the stomach state threshold evaluation module comprises a patient correlation evaluation unit and a patient risk threshold evaluation unit; the stomach state monitoring module comprises a stomach comprehensive risk analysis unit and a first risk assessment unit, and the stomach interaction module comprises a fluctuation risk assessment unit and a second risk assessment unit. According to the system, historical patient data and current patient information are integrated, high-risk samples are screened through correlation coefficient calculation, a personalized risk threshold value is constructed, evaluation objectivity is improved, multi-dimensional stomach parameters are collected, the risk coefficient is synthesized to comprehensively reflect the stomach state, and the risk coefficient is used for evaluating the state of the stomach. The multi-modal data is converted into a visual stomach risk fluctuation coefficient and is dynamically displayed to assist medical personnel in quickly making decisions, so that the stomach condition of the patient is accurately monitored.

Description

Technical Field

[0001] The present invention relates to the field of gastric tube interaction technology, and in particular to an intelligent multimodal human-computer interaction gastric tube system. Background Art

[0002] Critically ill patients often require a gastric tube for enteral nutrition support. The gastric tube is inserted through the nasal cavity, passes through the pharynx and esophagus, and finally reaches the stomach. It is mainly used to extract gastric juice or inject nutrient solution to meet the patient's nutritional needs. However, enteral nutrition has many risks, such as aspiration, gastric mucosal damage, infection, and gastrointestinal dysfunction. The long-term placement of the gastric tube may further increase the risk of local irritation and complications due to contact with the stomach wall and esophageal wall. By dynamically monitoring key indicators such as intragastric pH and nutritional tolerance, the nutritional support plan can be optimized, the related risks can be reduced, and the safety of patients can be improved.

[0003] In the prior art, monitoring is usually performed by installing sensors on the stomach. For example, publication number "CN109846715A" discloses an intelligent gastric tube, which includes: a metal guide wire inserted into the gastric tube, and a CCD micro-image sensor installed at the front end of the gastric tube. By inserting the metal guide wire into the gastric tube, the hardness of the gastric tube is increased. At the same time, with the help of the CCD micro-image sensor at the front end, the insertion of the gastric tube can be monitored and guided under visual conditions; a micro pH probe is installed on the front wall of the tube to directly and dynamically monitor the pH value of the patient's gastric fluid in real time, providing a reliable basis for diagnosis and treatment; a flow sensor is installed at the tail end of the gastric tube to accurately measure the amount of gastric fluid entering and exiting the gastric tube and can also be used to control the speed of injecting the gastric tube contents; flushing tubes are installed inside and outside the gastric tube to flush and clean the inner and outer walls of the tube when necessary, which can reduce attachments on the tube wall, ensure the patency of the tube, and reduce the irritation of the gastric contents on the stomach wall. However, since a single sensor cannot directly measure the specific condition of the stomach during gastric monitoring, there may be deviations between the sampling results obtained at different sampling points, which can easily cause errors and lead to false warnings.

[0004] However, the existing technology still has the problem that a single sensor monitoring the stomach can easily cause false warnings. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides an intelligent multimodal human-computer interactive gastric tube system, which has the advantages of accurately monitoring the patient's gastric condition and solves the above-mentioned technical problems.

[0006] To achieve the above objectives, the present invention provides the following technical solutions: an intelligent multimodal human-computer interactive gastric tube system, comprising

[0007] The gastric status threshold assessment module is used to read all stored patient data and the current patient data, and calculate the correlation coefficient between all patients and the current patient. Based on the correlation coefficient calculated for all patient data, several patient data are selected to construct a patient risk set, and then the gastric risk threshold of the current patient is constructed based on the patient risk set. ;

[0008] Gastric status monitoring module, used to collect multimodal gastric coefficients of patients and construct a comprehensive gastric assessment risk coefficient , when the stomach comprehensively evaluates the risk factor Above the current patient gastric risk threshold When the first warning signal is issued, the first feedback strategy is implemented;

[0009] The stomach interaction module is used to read the multimodal stomach coefficients of the patient collected in the stomach status monitoring module and construct the current patient's stomach risk fluctuation coefficient ,At the same time, the patient's multimodal gastric coefficient is displayed to medical personnel, and it is determined whether to issue a second warning signal and implement a second feedback strategy;

[0010] Gastric tube calibration module, used to read the multimodal indicators of the gastric tube and comprehensively construct the gastric tube calibration coefficient , when the gastric tube calibration coefficient Exceeding the preset gastric tube verification threshold When the third warning signal is issued, the third feedback strategy is implemented;

[0011] The present invention integrates historical patient data with current patient information, screens high-risk samples through correlation coefficient calculation, constructs personalized risk thresholds, improves the objectivity of assessment, collects multi-dimensional gastric parameters, and integrates them into risk coefficients to comprehensively reflect the gastric status. It converts multimodal data into intuitive gastric risk fluctuation coefficients and dynamic displays, assisting medical personnel in making quick decisions, thereby ensuring accurate monitoring of the patient's gastric condition.

[0012] As a preferred technical solution of the present invention, the gastric state threshold assessment module includes a patient relevance assessment unit and a patient risk threshold assessment unit;

[0013] The patient correlation evaluation unit is used to calculate the comprehensive correlation coefficient between all patients and the current patient. The specific steps are as follows:

[0014] Step A1: Obtain the current patient's physical indicators and construct the Correlation coefficient between the patient's physical index and the current patient , the specific expression is as follows:

[0015]

[0016] in, , represents the first patient body indicators, represents the correlation coefficient between the physical index of the i-th patient and the current patient, Indicates the total number of body indicators, Indicates summation, Indicates the absolute value, Indicates the current patient's body indicators, , Indicates the total number of patient data;

[0017] Step A2: Obtain the patient's medical history and construct the Correlation coefficient between the patient and the current patient's medical history , the specific expression is as follows:

[0018]

[0019] in, Indicates the A collection of patients' medical histories, Represents the current patient's medical history set, Indicates the number of elements in the current patient's medical history set, Indicates the The number of elements in the patient's medical history set, Represents intersection. By quantifying the similarity of medical history, it can help medical staff quickly find cases with similar medical history to the current patient. It can also reflect to a certain extent the similarity between the current patient and other patients with gastric diseases.

[0020] Step A3: Based on Step A1 and Step A2, The comprehensive correlation coefficient between each patient and the current patient The specific expression is as follows:

[0021]

[0022] in, 、 They represent weight coefficients that sum to 1, Indicates the The comprehensive correlation coefficient between each patient and the current patient, Indicates the The correlation coefficient between the patient and the current patient's medical history, Indicates the The correlation coefficient between each patient and the current patient's physical indicators is calculated by traversing all patient data in turn and calculating the correlation coefficient between all patients and the current patient.

[0023] As a preferred technical solution of the present invention, the patient risk threshold evaluation unit is used to screen out a number of patient data based on the correlation coefficient calculated for all patient data to construct a patient risk set. The specific steps are as follows:

[0024] Step B1: Sort the correlation coefficients between all patients obtained in step A3 and the current patient from high to low;

[0025] Step B2: Calculate the first The comprehensive correlation coefficient between the patient and the current patient and the The difference between the comprehensive correlation coefficient of each patient and the current patient , and traverse and calculate element by element to obtain the set of difference values ​​of the comprehensive correlation coefficients of all patients and the current patient;

[0026] Step B3: Read the difference values ​​of the comprehensive correlation coefficients of all patients and the current patient The gastric risk coefficient of each patient is used to construct the patient risk set. Indicates a separation point;

[0027] Step B4: Construct the current patient's gastric risk threshold based on the patient risk set , the specific expression is as follows:

[0028]

[0029] in, Represents the patient risk set The patient's gastric comprehensive assessment risk factor is read by the doctor when the doctor intervenes. The multimodal gastric coefficients of each patient’s gastric tube were calculated. Comprehensive assessment of gastric risk factors for each patient , can reflect the patient's current stomach condition to a certain extent, and by screening the stomachs of several other patients similar to the current patient to comprehensively evaluate the risk factor, it can reflect the patient more accurately.

[0030] As a preferred technical solution of the present invention, the step B2 calculates the first The comprehensive correlation coefficient between the patient and the current patient and the The difference between the comprehensive correlation coefficient of each patient and the current patient The expression is as follows:

[0031]

[0032] in, Indicates the The comprehensive correlation coefficient between each patient and the current patient, Indicates the The comprehensive correlation coefficient between each patient and the current patient, , represents the sum, which can quantify the difference in similarity between two adjacent patients. This difference can be used to evaluate the rationality of the sorting, determine whether the current sorting accurately reflects the similarity between patients, and be used in subsequent steps;

[0033] The separation point in step B3 The specific steps to determine are as follows:

[0034] Step B3.1: Read the The comprehensive correlation coefficient between the patient and the current patient and the The difference between the comprehensive correlation coefficient of each patient and the current patient ;

[0035] Step B3.2: Determination Whether it exceeds the preset correlation threshold If exceeded, terminate step B3.2 and output the current As a dividing point , and retain the previous Construct a patient risk set based on the gastric risk coefficient of each patient. If the risk does not exceed the limit, proceed to step B3.3.

[0036] Step B3.3: After adding 1, traverse in sequence until the output separation point . Correlation analysis of multiple patients was performed to determine the The correlation between the patient and the current patient can be used to comprehensively assess the risk of the existing patient's stomach when the doctor intervenes, and the stability of the remaining patients as a reference can be guaranteed. The higher the correlation with the current patient, the more it can explain the similarity between the patient and the current patient. Through this analysis process, the previous patient who is most similar to the current patient can be screened out. patients, and by setting the correlation threshold It can be guaranteed that there will be a sufficient number of patients present.

[0037] As a preferred technical solution of the present invention, the stomach state monitoring module includes a stomach comprehensive risk analysis unit and a first risk assessment unit;

[0038] The gastric comprehensive risk analysis unit is used to collect the patient's multimodal gastric coefficients and construct a gastric comprehensive assessment risk coefficient , the specific steps are as follows:

[0039] Step C1: Obtain the patient's pH value at several moments based on the pH sensor in the gastric tube, and obtain the patient's daily pH fluctuation curve, and match the current patient's behavior affecting gastric pH with the patient's daily pH fluctuation curve, and determine the deviation value of the patient's gastric pH fluctuation value exceeding the preset pH range in the non-matching interval in the patient's daily pH fluctuation curve. and corresponding time By calculating the deviation value of the gastric pH fluctuation value of the patient in the non-matching interval that exceeds the preset range , and the corresponding time , achieving quantitative analysis of gastric pH abnormalities, which helps to more accurately judge the patient's gastric risk;

[0040] Among them, behaviors that affect gastric pH include eating, taking medicine, and medical treatment;

[0041] Step C2: Obtain the patient's resting intra-abdominal pressure through the pressure sensor integrated in the gastric tube and construct the intra-abdominal pressure deviation , the specific expression is as follows:

[0042]

[0043] in, Indicates the patient's resting intra-abdominal pressure. Indicates the maximum value of the standard intra-abdominal pressure range. By calculating the deviation ratio between the patient's intra-abdominal pressure and the standard value, it can intuitively reflect whether the patient's intra-abdominal pressure is abnormal;

[0044] Step C3: Constructing a comprehensive gastric risk assessment coefficient , the specific expression is as follows:

[0045]

[0046] in, and They represent weight coefficients that sum to 1, represents a natural constant, This value represents the deviation of the patient's gastric pH fluctuation value outside the preset pH range during the non-matching interval of the patient's daily pH fluctuation curve. By combining pH fluctuation and intra-abdominal pressure, and introducing weight coefficients and exponential functions, a scientific, comprehensive, and dynamic assessment of gastric risk can be achieved, providing strong support for medical decision-making.

[0047] As a preferred technical solution of the present invention, the first risk assessment unit is used to determine whether to issue a first warning signal and execute a first feedback strategy, specifically in the following steps:

[0048] When the stomach comprehensively assesses the risk factor Above the current patient gastric risk threshold When the first warning signal is issued, the first feedback strategy is implemented;

[0049] Comprehensive gastric assessment risk factor Not higher than the current patient's gastric risk threshold When , the judgment of the stomach status monitoring module for that day is terminated;

[0050] The first feedback strategy specifically includes: increasing the frequency of doctor inspections by 10% to 15%, adding a first warning mark to the current patient's electronic health record, and recording the comprehensive gastric assessment risk factor. By increasing the frequency of inspections and strengthening the dynamic monitoring of patients' gastric conditions, potential risks can be promptly identified, providing doctors with intuitive risk alerts, facilitating subsequent diagnosis and treatment decisions, providing data support for subsequent analysis, and facilitating the tracking of risk trends.

[0051] As a preferred technical solution of the present invention, the stomach interaction module includes a fluctuation risk assessment unit and a second risk assessment unit;

[0052] The fluctuation risk assessment unit is used to read the multimodal stomach coefficients of the patient collected in the stomach state monitoring module and construct the current patient's stomach risk fluctuation coefficient , the specific steps are as follows:

[0053] Step D1: Obtain the current patient's pH fluctuation coefficient , the specific expression is as follows:

[0054]

[0055] in, Indicates the maximum value of the patient's gastric pH exceeding the standard pH range in the non-matching interval of the daily pH fluctuation curve. If it does not exceed the standard pH range, it is 0. Indicates that the gastric pH of the patient in the non-matching interval of the daily pH fluctuation curve is lower than the minimum value of the standard pH range. If it is not lower than the standard pH range, it is 0;

[0056] Step D2: Obtain the intra-abdominal pressure deviation calculated by the comprehensive risk analysis unit at two adjacent sampling intervals, and calculate the intra-abdominal pressure fluctuation coefficient. The specific expression is as follows:

[0057]

[0058] in, express The intra-abdominal pressure fluctuation coefficient at each moment, express The patient's intra-abdominal pressure at rest at all times, express The patient's intra-abdominal pressure at rest at all times;

[0059] Step D3: Based on steps D1 to D2, comprehensively calculate the patient's current gastric fluctuation coefficient , the specific expression is as follows:

[0060]

[0061] in, and They represent weight coefficients that sum to 1, Indicates the maximum value of the intra-abdominal pressure fluctuation coefficient at all times. It directly reflects the abnormality of the acid-base environment in the patient's stomach. If it does not exceed the standard range, it will be recorded as 0, avoiding the interference of normal fluctuations and improving the accuracy of clinical judgment.

[0062] As a preferred technical solution of the present invention, the second risk assessment unit is used to display the patient's multimodal gastric coefficient to medical personnel, and determine whether to issue a second warning signal and execute a second feedback strategy;

[0063] The specific steps for determining whether to issue the second warning signal are as follows: if the patient's stomach fluctuation coefficient Exceeding the preset fluctuation value , a second warning signal is issued and a second feedback strategy is executed. If the patient's stomach fluctuation coefficient is Does not exceed the preset fluctuation value When the second warning signal is not issued;

[0064] The second feedback strategy specifically involves increasing physician inspection frequency by 20% to 35% and shortening the sampling and feedback interval for gastric tube sensors by 50%. In higher-risk situations, this further increase in inspection frequency strengthens monitoring of patients' gastric conditions, ensuring timely detection and resolution of potential issues. Furthermore, more frequent data collection allows for more refined gastric parameters, providing more real-time and accurate data support for risk assessment.

[0065] As the preferred technical solution of the present invention, the gastric tube calibration module comprehensively constructs the gastric tube calibration coefficient The specific steps are as follows:

[0066] Step E1: Record and identify the current patient's non-recumbent time ratio through the surveillance camera in the ward , the specific expression is as follows:

[0067]

[0068] in, Indicates the lying time. The camera is used in conjunction with the posture recognition model to identify whether the patient is in a lying position. When the recognition result is a lying position, the time is recorded until the patient is in a non-lying position. When it is recognized that there is no human body on the bed, the time recording stops. Several lying time periods are obtained, and the lying time period is summed up to get the lying time. , through the camera and the existing posture recognition model, the patient's lying and non-lying postures can be automatically identified, which reduces manual intervention and improves efficiency;

[0069] Step E2: Obtain the number of gastric tube sensor errors , and construct the gastric tube calibration coefficient , the specific expression is as follows:

[0070]

[0071] in, Indicates the percentage of time the patient spends in a non-recumbent position. Combining this percentage with sensor error data not only considers the patient's activity status but also incorporates the device's actual operating conditions, improving calibration accuracy.

[0072] As a preferred technical solution of the present invention, the third feedback strategy specifically reminds medical staff to check the gastric tube sensor and determine whether it needs to be replaced. At the appropriate time, the system will remind medical staff to check the gastric tube sensor and determine whether it needs to be replaced. This feedback strategy provides clear operational guidance, helping medical staff to promptly address issues and avoid potential risks.

[0073] Compared with the existing technology, the present invention provides an intelligent multimodal human-computer interactive gastric tube system with the following beneficial effects:

[0074] 1. The present invention integrates historical patient data with current patient information, screens high-risk samples through correlation coefficient calculation, constructs personalized risk thresholds, improves assessment objectivity, collects multi-dimensional gastric parameters, and integrates them into risk coefficients to comprehensively reflect gastric status. It converts multimodal data into intuitive gastric risk fluctuation coefficients and dynamic displays, assisting medical personnel in making quick decisions, thereby ensuring accurate monitoring of patients' gastric conditions.

[0075] 2. The present invention constructs a gastric tube calibration coefficient through multimodal indicators to ensure normal gastric tube function. A preset gastric tube calibration threshold is set. If it exceeds the limit, a third warning is triggered, reminding medical staff to check the gastric tube sensor and determine whether to replace it, thereby avoiding complications caused by gastric tube problems.

[0076] 3. This invention significantly improves the accuracy and reliability of gastrointestinal tube monitoring by integrating multiple monitoring functions, using MRI-compatible materials, having anti-blocking and self-cleaning functions, and supporting miniaturized data transmission. At the same time, by optimizing material selection and energy management, it reduces the cost of using the system and improves patient comfort and compliance. In addition, the intelligent monitoring function can provide clinicians with real-time and comprehensive physiological parameter data, helping medical staff to adjust treatment plans in a timely manner and improve treatment effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 Schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION

[0078] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0079] See also Figure 1 The gastric tube structure involved in the present invention includes a composite tube structure, a four-layer design: a main drainage cavity (2.1mm in diameter), a reflux prevention valve at the end; a sensor channel: an integrated fiber optic pressure sensor array (8-12), a miniature ultrasound probe (20MHz); a micro-cleaning gas path: connecting the CO2 gas storage capsule and the tube wall nozzle; an antibacterial outer sheath, and a medical-grade Bluetooth chip is used to transmit to the intelligent multimodal human-computer interactive gastric tube system. The intelligent multimodal human-computer interactive gastric tube system includes:

[0080] The gastric status threshold assessment module is used to read all stored patient data and the current patient data, and calculate the correlation coefficient between all patients and the current patient. Based on the correlation coefficient calculated for all patient data, several patient data are selected to construct a patient risk set, and then the gastric risk threshold of the current patient is constructed based on the patient risk set. ;

[0081] The gastric status threshold assessment module includes a patient relevance assessment unit and a patient risk threshold assessment unit;

[0082] The patient correlation evaluation unit is used to calculate the comprehensive correlation coefficient between all patients and the current patient. The specific steps are as follows:

[0083] Step A1: Obtain the current patient's physical indicators and construct the Correlation coefficient between the patient's physical index and the current patient , the specific expression is as follows:

[0084]

[0085] in, , represents the first patient body indicators, represents the correlation coefficient between the physical index of the i-th patient and the current patient, Indicates the total number of body indicators, Indicates summation, Indicates the absolute value, Indicates the current patient's body indicators, , Indicates the total number of patient data, and physical indicators such as the patient's height, weight, and other physiological parameters that can be directly measured to obtain specific values. The specific type of data can be set by the doctor, and includes but is not limited to physiological parameters such as blood pressure, blood sugar, and resting heart rate;

[0086] Step A2: Obtain the patient's medical history and construct the Correlation coefficient between the patient and the current patient's medical history , the specific expression is as follows:

[0087]

[0088] in, Indicates the A collection of patients' medical histories, Represents the current patient's medical history set, Indicates the number of elements in the current patient's medical history set, Indicates the The number of elements in the patient's medical history set, Represents the intersection. In the statistics, only the history of stomach-related diseases is counted in the past medical history, and medical staff cannot view it during the statistics. For details, see Table a below:

[0089] Table a

[0090] Past medical history Mechanisms affecting the stomach Helicobacter pylori infection Causes chronic gastritis and ulcers, which can lead to gastric cancer in the long term Long-term NSAIDs / aspirin Inhibit prostaglandin synthesis and destroy the gastric mucosal barrier History of alcoholism Directly damage the mucosal epithelium and increase gastric acid secretion Smoking history Increase gastric acid secretion, reduce mucosal blood flow, and reduce repair capacity Autoimmune diseases Crohn's disease: inflammatory infiltrate; celiac disease: gluten-induced mucosal damage History of gastric surgery Anatomical changes (eg, bile reflux), decreased absorption area diabetes Autonomic neuropathy causing gastroparesis (delayed gastric emptying)

[0091] The information in Table a is only for example and does not cover all medical histories of stomach-related diseases;

[0092] Step A3: Based on Step A1 and Step A2, The comprehensive correlation coefficient between each patient and the current patient The specific expression is as follows:

[0093]

[0094] in, 、 They represent weight coefficients that sum to 1, Indicates the The comprehensive correlation coefficient between each patient and the current patient, Indicates the The correlation coefficient between the patient and the current patient's medical history, Indicates the The correlation coefficient between each patient and the current patient's physical indicators is calculated by traversing all patient data in turn and obtaining the correlation coefficient between all patients and the current patient;

[0095] For example, in a calculation process, the physiological parameters considered are weight and blood pressure (including systolic and diastolic pressure). The specific calculation process is: , the set of all diseases of patient i = {history of alcoholism, history of smoking, history of gastric surgery}, the current patient's disease collection For {history of smoking, autoimmune disease, history of gastric surgery}, at this time , , , (This data example is only used to demonstrate the calculation results and does not mean that the actual results will be exactly the same as this example).

[0096] The patient risk threshold assessment unit is used to select a number of patient data based on the correlation coefficient calculated for all patient data to construct a patient risk set. The specific steps are as follows:

[0097] Step B1: Sort the correlation coefficients between all patients obtained in step A3 and the current patient from high to low;

[0098] Step B2: Calculate the first The comprehensive correlation coefficient between the patient and the current patient and the The difference between the comprehensive correlation coefficient of each patient and the current patient , and traverse and calculate element by element to obtain the set of difference values ​​of the comprehensive correlation coefficients of all patients and the current patient;

[0099] Step B3: Read the difference values ​​of the comprehensive correlation coefficients of all patients and the current patient The gastric risk coefficient of each patient is used to construct the patient risk set. Indicates a separation point;

[0100] Step B4: Construct the current patient's gastric risk threshold based on the patient risk set , the specific expression is as follows:

[0101]

[0102] in, Represents the patient risk set The patient's gastric comprehensive assessment risk factor is read by the doctor when the doctor intervenes. The multimodal gastric coefficient of each patient's gastric tube is calculated and recorded manually when the doctor intervenes. Comprehensive assessment of gastric risk factors for each patient , which can reflect the patient's current stomach condition to a certain extent.

[0103] In step B2, the first The comprehensive correlation coefficient between the patient and the current patient and the The difference between the comprehensive correlation coefficient of each patient and the current patient The expression is as follows:

[0104]

[0105] in, Indicates the The comprehensive correlation coefficient between each patient and the current patient, Indicates the The comprehensive correlation coefficient between each patient and the current patient, , Indicates the sum, by The comprehensive correlation coefficient between the patient and the current patient is The difference between the comprehensive correlation coefficient of each patient and the current patient can reflect the correlation difference between the patients arranged in sequence. When this difference is too large, it will lead to Too large, so that a breakpoint appears, making it possible to determine the similarity breakpoint;

[0106] Separation point in step B3 The specific steps to determine are as follows:

[0107] Step B3.1: Read the The comprehensive correlation coefficient between the patient and the current patient and the The difference between the comprehensive correlation coefficient of each patient and the current patient ;

[0108] Step B3.2: Determination Whether it exceeds the preset correlation threshold If exceeded, terminate step B3.2 and output the current As a dividing point , and retain the previous Construct a patient risk set based on the gastric risk coefficient of each patient. If the risk does not exceed the limit, proceed to step B3.3.

[0109] Step B3.3: After adding 1, traverse in sequence until the output separation point , by analyzing the correlation of multiple patients to determine the The correlation between the patient and the current patient can be used to comprehensively assess the risk of the existing patient's stomach when the doctor intervenes, and the stability of the remaining patients as a reference can be guaranteed. The higher the correlation with the current patient, the more it can explain the similarity between the patient and the current patient. Through this analysis process, the previous patient who is most similar to the current patient can be screened out. patients, and by setting the correlation threshold It can be guaranteed that there will be a sufficient number of patients present.

[0110] Gastric status monitoring module, used to collect multimodal gastric coefficients of patients and construct a comprehensive gastric assessment risk coefficient , when the stomach comprehensively evaluates the risk factor Above the current patient gastric risk threshold When the first warning signal is issued, the first feedback strategy is implemented;

[0111] The gastric status monitoring module includes a gastric comprehensive risk analysis unit and a first risk assessment unit;

[0112] The gastric comprehensive risk analysis unit is used to collect the patient's multimodal gastric coefficients and construct a comprehensive gastric assessment risk coefficient , the specific steps are as follows:

[0113] Step C1: Obtain the patient's pH value at several moments based on the pH sensor in the gastric tube, and obtain the patient's daily pH fluctuation curve, and match the current patient's behavior affecting gastric pH with the patient's daily pH fluctuation curve, and determine the deviation value of the patient's gastric pH fluctuation value exceeding the preset pH range in the non-matching interval in the patient's daily pH fluctuation curve. and corresponding time , the deviation value here is the absolute value of the difference, where the time refers to the specific time length in hours;

[0114] Among them, behaviors that affect gastric pH include eating, taking medicine, and medical treatment;

[0115] Step C2: Obtain the patient's resting intra-abdominal pressure through the pressure sensor integrated in the gastric tube and construct the intra-abdominal pressure deviation , the specific expression is as follows:

[0116]

[0117] in, Indicates the patient's resting intra-abdominal pressure. Indicates the maximum value of the standard intra-abdominal pressure range;

[0118] In another embodiment, step C2 can also construct the gastric residual volume, gastric motility and the patient's resting intra-abdominal pressure in the same way, and take their average or weighted average as the intra-abdominal pressure deviation ;

[0119] Step C3: Constructing a comprehensive gastric risk assessment coefficient , the specific expression is as follows:

[0120]

[0121] in, and They represent weight coefficients that sum to 1, represents a natural constant, It indicates the deviation value of the patient's gastric pH fluctuation value in the non-matching interval of the patient's daily pH fluctuation curve that exceeds the preset pH range.

[0122] The first risk assessment unit is used to determine whether to issue a first warning signal and execute a first feedback strategy. The specific steps are as follows:

[0123] When the stomach comprehensively assesses the risk factor Above the current patient gastric risk threshold When the first warning signal is issued, the first feedback strategy is implemented;

[0124] Comprehensive gastric assessment risk factor Not higher than the current patient's gastric risk threshold When , the judgment of the stomach status monitoring module for that day is terminated;

[0125] The first feedback strategy specifically includes: increasing the frequency of doctor inspections by 10% to 15%, adding a first warning mark to the current patient's electronic health record, and recording the comprehensive gastric assessment risk factor. By increasing the inspection frequency and strengthening the dynamic monitoring of the patient's gastric condition, potential risks can be discovered in a timely manner, providing doctors with intuitive risk prompts, facilitating subsequent diagnosis and treatment decisions, providing data support for subsequent analysis, and facilitating tracking of risk change trends.

[0126] The gastric status monitoring module also includes, intra-abdominal pressure monitoring: high-precision pressure sensors are embedded at the end of the gastric tube and the abdomen to measure the intra-abdominal pressure in real time, providing real-time intra-abdominal pressure data to help medical staff detect abnormal intra-abdominal pressure in a timely manner and prevent complications such as intra-abdominal hypertension; gastric residual volume detection: combining multiple sensors, the residual amount of gastric fluid is calculated through a multi-sensor fusion algorithm to improve the accuracy and reliability of gastric residual volume measurement and reduce the risk of aspiration caused by gastric retention; gastric motility monitoring: accelerometers and electromyography sensors are embedded in the side wall of the gastric tube to monitor gastric peristalsis and gastrointestinal motility in real time, providing real-time data on gastric motility, helping to evaluate gastrointestinal function and promptly detect motility disorders such as gastroparesis; emptying dynamics analysis: by analyzing the time series of data from multiple sensors, the emptying dynamics of gastric contents are evaluated to provide a basis for nutritional support and drug therapy and optimize treatment plans;

[0127] The stomach interaction module is used to read the multimodal stomach coefficients of the patient collected in the stomach status monitoring module and construct the current patient's stomach risk fluctuation coefficient ,At the same time, the patient's multimodal gastric coefficient is displayed to medical personnel, and it is determined whether to issue a second warning signal and implement a second feedback strategy;

[0128] The gastric interaction module includes a fluctuation risk assessment unit and a second risk assessment unit;

[0129] The fluctuation risk assessment unit is used to read the multimodal gastric coefficients of the patient collected in the gastric status monitoring module and construct the current patient's gastric risk fluctuation coefficient , the specific steps are as follows:

[0130] Step D1: Obtain the current patient's pH fluctuation coefficient , the specific expression is as follows:

[0131]

[0132] in, Indicates the maximum value of the patient's gastric pH exceeding the standard pH range in the non-matching interval of the daily pH fluctuation curve. If it does not exceed the standard pH range, it is 0. Indicates that the gastric pH of the patient in the non-matching interval of the daily pH fluctuation curve is lower than the minimum value of the standard pH range. If it is not lower than the standard pH range, it is 0. Directly reflects the abnormal degree of the patient's gastric acid-base environment. If it does not exceed the standard range, it will be recorded as 0, avoiding the introduction of interference from normal fluctuations and improving the accuracy of clinical judgment;

[0133] Step D2: Obtain the intra-abdominal pressure deviation calculated by the comprehensive risk analysis unit at two adjacent sampling intervals, and calculate the intra-abdominal pressure fluctuation coefficient. The specific expression is as follows:

[0134]

[0135] in, express The intra-abdominal pressure fluctuation coefficient at each moment, express The patient's intra-abdominal pressure at rest at all times, express The patient's resting intra-abdominal pressure reflects the instantaneous fluctuation trend of abdominal pressure. The relative rate of change is more sensitive to sudden increases or decreases in pressure and is particularly suitable for monitoring acute risks.

[0136] In another embodiment, step D2 can also construct the gastric residual volume, gastric motility and the patient's resting intra-abdominal pressure in the same way, and take their average or weighted average as Intra-abdominal pressure fluctuation coefficient at each moment;

[0137] Step D3: Based on steps D1 to D2, comprehensively calculate the patient's current gastric fluctuation coefficient , the specific expression is as follows:

[0138]

[0139] in, and They represent weight coefficients that sum to 1, It represents the maximum value of the intra-abdominal pressure fluctuation coefficient at all times.

[0140] The second risk assessment unit is used to display the patient's multimodal gastric coefficient to medical personnel, determine whether to issue a second warning signal, and implement a second feedback strategy;

[0141] The specific steps for determining whether to issue the second warning signal are as follows: if the patient's stomach fluctuation coefficient Exceeding the preset fluctuation value , a second warning signal is issued and a second feedback strategy is executed. If the patient's stomach fluctuation coefficient is Does not exceed the preset fluctuation value When the second warning signal is not issued;

[0142] The second feedback strategy is specifically: increase the frequency of doctor inspections by 20% to 35%, shorten the sampling feedback interval of gastric tube sensors by 50%, and in higher-risk situations, further increase the inspection frequency and strengthen close monitoring of the patient's gastric condition to ensure timely detection and treatment of potential problems. Through more frequent data collection, more detailed gastric parameters (such as pH value, intra-abdominal pressure, etc.) are obtained to provide more real-time and accurate data support for risk assessment.

[0143] Gastric tube calibration module, used to read the multimodal indicators of the gastric tube and comprehensively construct the gastric tube calibration coefficient , when the gastric tube calibration coefficient Exceeding the preset gastric tube verification threshold When the third warning signal is issued, the third feedback strategy is implemented.

[0144] Gastric tube calibration module comprehensively constructs gastric tube calibration coefficient The specific steps are as follows:

[0145] Step E1: Record and identify the current patient's non-recumbent time ratio through the surveillance camera in the ward , the specific expression is as follows:

[0146]

[0147] in, Indicates the duration of lying posture. The camera is combined with the posture recognition model (the posture recognition model is an existing video analysis model. As long as it can realize the judgment of lying posture and non-lying posture, it will not be elaborated here) to identify whether the patient is in a lying posture. When the recognition result is lying posture, the time is recorded until the non-lying posture appears. When it is recognized that there is no human body on the bed, the time recording stops. Several lying posture periods are obtained, and the lying posture duration is obtained by summing up the several lying posture periods within the day. ;

[0148] Step E2: Obtain the number of gastric tube sensor errors , and construct the gastric tube calibration coefficient , the specific expression is as follows:

[0149]

[0150] in, Indicates the proportion of time the patient is currently in a non-recumbent position. Through the camera and the existing posture recognition model, the patient's lying and non-recumbent positions can be automatically identified, reducing manual intervention and improving efficiency.

[0151] The gastric tube of the present invention uses MRI-compatible materials to miniaturize the sensor and data processing unit. The target tube diameter is ≤14Fr, meeting clinical use standards, improving the system's portability and patient comfort. Furthermore, the gastric acid fuel cell (Mg-Cu primary cell) uses gastric acid as an electrolyte to provide a stable energy supply. This solves the problem of inconvenient traditional battery replacement and extends the system's operating time. It has anti-blocking and self-cleaning functions and supports miniaturized data transmission, significantly improving the accuracy and reliability of gastrointestinal tube monitoring. At the same time, by optimizing material selection and energy management, it reduces the cost of the system and improves patient comfort and compliance. Furthermore, the intelligent monitoring function can provide real-time, comprehensive physiological parameter data to the clinic, helping medical staff to adjust treatment plans in a timely manner and improve treatment effectiveness.

[0152] The third feedback strategy specifically involves reminding medical staff to check the gastric tube sensor and determine whether it needs to be replaced. The system will remind medical staff to check the gastric tube sensor and determine whether it needs to be replaced. This feedback strategy provides clear operational guidance, helping medical staff to deal with problems in a timely manner and avoid potential risks.

[0153] In addition, anti-blocking and self-cleaning functions can be considered. For example, anti-blocking design: a special channel structure is set on the gastric tube to prevent gastric contents from blocking the gastric tube. This ensures the continuity of monitoring and treatment and reduces the difficulty of nursing.

[0154] For example, the self-cleaning function: an integrated micro pump or airflow system regularly flushes the gastric tube to prevent dirt accumulation, thereby increasing the life of the gastric tube and reducing the frequency of replacement due to blockage;

[0155] The sensor layout that may be involved in the present invention is as follows:

[0156] Pressure sensor: Location: at the end of the gastric tube and in the abdomen. Function: to measure intra-abdominal pressure.

[0157] Capacitive sensor: Location: Side wall of gastric tube. Function: Measure capacitance changes of gastric fluid.

[0158] Ultrasound sensor: Location: Side wall of gastric tube. Function: Measure the depth and volume of gastric fluid.

[0159] Optical sensor: Location: Side wall of gastric tube. Function: Measure the optical properties of gastric fluid.

[0160] Accelerometer: Location: Side wall of gastric tube. Function: Monitor gastric motility.

[0161] Myoelectric sensor: Location: Side wall of or around the gastric tube. Function: Monitor gastrointestinal motility.

[0162] In addition, the specific operation process is to insert a gastric tube with a sensor into the patient's stomach, and upload the data to an external device (such as a mobile device or a monitor) via Bluetooth, Wi-Fi, or an external interface. Inserting the gastric tube: insert the gastric tube into the patient's stomach according to the conventional method, and start the software on the external device at the same time; real-time monitoring: the external device displays the intra-abdominal pressure, gastric residual volume, gastric motility frequency and gastrointestinal motility status in real time; provide real-time prompt information, such as "intra-abdominal pressure is normal", "gastric residual volume is too high", "gastric motility is normal" or "gastrointestinal motility is weakened"; adjustment and confirmation: adjust the position of the gastric tube or the patient's position according to the prompt information; fix the gastric tube: after completing the operation, fix the gastric tube and end the process;

[0163] At the same time, after the external device monitors the real-time transmission data to the mobile terminal, a dynamic trend curve is automatically generated: a trend chart of changes in intra-abdominal pressure and gastric retention volume is automatically generated, and an alarm threshold is preset (such as intra-abdominal pressure>12mmHg);

[0164] Example 1:

[0165] This example provides specific data for constructing a comprehensive gastric risk assessment coefficient, see Table 1 below:

[0166] Table 1

[0167] parameter Specific values #timg# 0 #timg# 15h #timg# 0.45 #timg# 0.55 #timg# 12mmhg #timg# 8 #timg# 3.32

[0168] At this time, the calculation , <Current patient gastric risk threshold =3.32, no first warning signal is issued, and the gastric status monitoring module is terminated for that day;

[0169] Example 2:

[0170] This example provides specific data for constructing the current patient's gastric risk fluctuation coefficient, see Table 2 below:

[0171] Table 2

[0172] parameter Specific values #timg# 3.6 #timg# 1.5 #timg# 1.4 #timg# 0.38 #timg# 0.62 #timg# 1.5

[0173] At this time, we calculate , , at this time it exceeds the preset fluctuation value =1.5, a second warning signal is issued and the second feedback strategy is implemented, increasing the frequency of doctor inspections by 27% and shortening the sampling feedback interval of the gastric tube sensor by 50%;

[0174] Example 3:

[0175] This example provides specific data for constructing the gastric tube calibration coefficient, see Table 3 below:

[0176] parameter Specific values #timg# 18h #timg# 5 #timg# 4.5

[0177] At this time, the calculation , at this time remind medical staff to check the sensor of the gastric tube and determine whether to replace it;

[0178] The above embodiments only provide one or more feasible solutions, which do not represent the optimal solution. The threshold value is set to facilitate comparison. The size of the threshold value depends on the amount of sample data and the number of bases set by technicians in this field for each group of sample data. As long as it does not affect the proportional relationship between the parameter and the quantized value, and the size of the weight can be determined by technicians in this field based on each sample data and multiple rounds of experiments, the above formulas are all calculated by removing the dimension and taking the numerical value.

[0179] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. Intelligent multimodal human-computer interactive gastric tube system, characterized by: include, The gastric status threshold assessment module is used to read all stored patient data and the current patient data, and calculate the correlation coefficient between all patients and the current patient. Based on the correlation coefficient calculated for all patient data, several patient data are selected to construct a patient risk set, and then the gastric risk threshold of the current patient is constructed based on the patient risk set. ; Gastric status monitoring module, used to collect multimodal gastric coefficients of patients and construct a comprehensive gastric assessment risk coefficient , when the stomach comprehensively evaluates the risk factor Above the current patient gastric risk threshold When the first warning signal is issued, the first feedback strategy is implemented; The stomach interaction module is used to read the multimodal stomach coefficients of the patient collected in the stomach status monitoring module and construct the current patient's stomach risk fluctuation coefficient ,At the same time, the patient's multimodal gastric coefficient is displayed to medical personnel, and it is determined whether to issue a second warning signal and implement a second feedback strategy; Gastric tube calibration module, used to read the multimodal indicators of the gastric tube and comprehensively construct the gastric tube calibration coefficient , when the gastric tube calibration coefficient Exceeding the preset gastric tube verification threshold When the third warning signal is issued, the third feedback strategy is implemented.

2. The intelligent multimodal human-computer interactive gastric tube system according to claim 1, characterized in that: The gastric state threshold assessment module includes a patient relevance assessment unit and a patient risk threshold assessment unit; The patient correlation evaluation unit is used to calculate the comprehensive correlation coefficient between all patients and the current patient, and the specific steps are as follows: Step A1: Get the current patient's physical indicators and construct the Correlation coefficient between the patient's physical index and the current patient , the specific expression is as follows; in, , represents the first patient body indicators, represents the correlation coefficient between the physical index of the i-th patient and the current patient, Indicates the total number of body indicators, Indicates summation, Indicates the absolute value, Indicates the current patient's body indicators, , Indicates the total number of patient data; Step A2: Obtain the patient's medical history and construct the Correlation coefficient between the patient and the current patient's medical history , the specific expression is as follows; in, Indicates the A collection of patients' medical histories, Represents the current patient's medical history set, Indicates the number of elements in the current patient's medical history set, Indicates the The number of elements in the patient's medical history set, Indicates intersection; Step A3: Based on Step A1 and Step A2, The comprehensive correlation coefficient between each patient and the current patient The specific expression of is as follows; in, 、 They represent weight coefficients that sum to 1, Indicates the The comprehensive correlation coefficient between each patient and the current patient, Indicates the The correlation coefficient between the patient and the current patient's medical history, Indicates the The correlation coefficient between each patient and the current patient's physical indicators is calculated by traversing all patient data in turn and calculating the correlation coefficient between all patients and the current patient.

3. The intelligent multimodal human-computer interactive gastric tube system according to claim 2, characterized in that: The patient risk threshold evaluation unit is used to screen out a number of patient data based on the correlation coefficient calculated from all patient data to construct a patient risk set. The specific steps are as follows: Step B1, sorting the correlation coefficients between all patients obtained in step A3 and the current patient from high to low; Step B2, calculate the first The comprehensive correlation coefficient between the patient and the current patient and the The difference between the comprehensive correlation coefficient of each patient and the current patient , and traverse and calculate element by element to obtain the set of difference values ​​of the comprehensive correlation coefficients of all patients and the current patient; Step B3, read the difference values ​​of the comprehensive correlation coefficients of all patients and the current patient The gastric risk coefficient of each patient is used to construct the patient risk set. Indicates a separation point; Step B4: Construct the current patient's gastric risk threshold based on the patient risk set , the specific expression is as follows; in, Represents the patient risk set The patient's gastric comprehensive assessment risk factor is read by the doctor when the doctor intervenes. The multimodal gastric coefficients of each patient's gastric tube were calculated.

4. The intelligent multimodal human-computer interactive gastric tube system according to claim 3, characterized in that: Step B2 calculates the first The comprehensive correlation coefficient between the patient and the current patient and the The difference between the comprehensive correlation coefficient of each patient and the current patient The expression is as follows; in, Indicates the The comprehensive correlation coefficient between each patient and the current patient, Indicates the The comprehensive correlation coefficient between each patient and the current patient, , Indicates summation; The separation point in step B3 The specific steps to determine are as follows; Step B3.1, read The comprehensive correlation coefficient between the patient and the current patient and the The difference between the comprehensive correlation coefficient of each patient and the current patient ; Step B3.2, determine Whether it exceeds the preset correlation threshold If exceeded, terminate step B3.2 and output the current As a dividing point , and retain the previous Construct a patient risk set based on the gastric risk coefficient of each patient. If the risk does not exceed the limit, proceed to step B3.

3. Step B3.3, After adding 1, traverse in sequence until the output separation point .

5. The intelligent multimodal human-computer interactive gastric tube system according to claim 4, characterized in that: The stomach state monitoring module includes a stomach comprehensive risk analysis unit and a first risk assessment unit; The gastric comprehensive risk analysis unit is used to collect the patient's multimodal gastric coefficients and construct a gastric comprehensive assessment risk coefficient , the specific steps are as follows; Step C1: Obtain the patient's pH value at several moments based on the pH sensor in the gastric tube, and obtain the patient's daily pH fluctuation curve, and match the patient's behavior affecting the gastric pH with the patient's daily pH fluctuation curve, and determine the deviation value of the patient's gastric pH fluctuation value exceeding the preset pH range in the non-matching interval of the patient's daily pH fluctuation curve. and corresponding time ; Among them, behaviors that affect gastric pH include eating, taking medicine, and medical treatment; Step C2: Obtain the patient's resting intra-abdominal pressure through the pressure sensor integrated in the gastric tube and construct the intra-abdominal pressure deviation. , the specific expression is as follows; in, Indicates the patient's resting intra-abdominal pressure. Indicates the maximum value of the standard intra-abdominal pressure range; Step C3: Constructing a comprehensive gastric risk assessment coefficient , the specific expression is as follows; in, and They represent weight coefficients that sum to 1, represents a natural constant, It indicates the deviation value of the patient's gastric pH fluctuation value in the non-matching interval of the patient's daily pH fluctuation curve that exceeds the preset pH range.

6. The intelligent multimodal human-computer interactive gastric tube system according to claim 1, characterized in that: The first risk assessment unit is used to determine whether to issue a first warning signal and execute a first feedback strategy, specifically in the following steps: When the stomach comprehensively assesses the risk factor Above the current patient gastric risk threshold When the first warning signal is issued, the first feedback strategy is implemented; Comprehensive gastric assessment risk factor Not higher than the current patient's gastric risk threshold When , the judgment of the stomach status monitoring module for that day is terminated; The first feedback strategy specifically includes: increasing the frequency of doctor inspections by 10% to 15%, adding a first warning mark to the current patient's electronic health record, and recording the comprehensive gastric assessment risk factor. and timestamp.

7. The intelligent multimodal human-computer interactive gastric tube system according to claim 5, characterized in that: The stomach interaction module includes a fluctuation risk assessment unit and a second risk assessment unit; The fluctuation risk assessment unit is used to read the multimodal stomach coefficients of the patient collected in the stomach state monitoring module and construct the current patient's stomach risk fluctuation coefficient , the specific steps are as follows; Step D1, obtain the current patient's pH fluctuation coefficient , the specific expression is as follows; in, Indicates the maximum value of the patient's gastric pH exceeding the standard pH range in the non-matching interval of the daily pH fluctuation curve. If it does not exceed the standard pH range, it is 0. Indicates that the gastric pH of the patient in the non-matching interval of the daily pH fluctuation curve is lower than the minimum value of the standard pH range. If it is not lower than the standard pH range, it is 0; Step D2, obtaining the intra-abdominal pressure deviation calculated by the comprehensive risk analysis unit at two adjacent sampling intervals, and calculating the intra-abdominal pressure fluctuation coefficient, the specific expression is as follows; in, express The intra-abdominal pressure fluctuation coefficient at each moment, express The patient's intra-abdominal pressure at rest at all times, express The patient's intra-abdominal pressure at rest at all times; Step D3, based on steps D1 to D2, comprehensively calculate the current patient's stomach fluctuation coefficient , the specific expression is as follows; in, and They represent weight coefficients that sum to 1, It represents the maximum value of the intra-abdominal pressure fluctuation coefficient at all times.

8. The intelligent multimodal human-computer interactive gastric tube system according to claim 7, characterized in that: The second risk assessment unit is used to display the patient's multimodal gastric coefficient to medical personnel, and determine whether to issue a second warning signal and execute a second feedback strategy; The specific steps for determining whether to issue the second warning signal are as follows: if the patient's stomach fluctuation coefficient Exceeding the preset fluctuation value , a second warning signal is issued and a second feedback strategy is executed. If the patient's stomach fluctuation coefficient is Does not exceed the preset fluctuation value When the second warning signal is not issued; The second feedback strategy specifically includes increasing the frequency of doctor inspections by 20% to 35% and shortening the sampling feedback interval of the gastric tube sensor by 50%.

9. The intelligent multimodal human-computer interactive gastric tube system according to claim 1, characterized in that: The gastric tube calibration module comprehensively constructs the gastric tube calibration coefficient The specific steps are as follows; Step E1: Record and identify the current patient's non-recumbent time ratio through the surveillance camera in the ward , the specific expression is as follows; in, Indicates the lying time. The camera is used in conjunction with the posture recognition model to identify whether the patient is in a lying position. When the recognition result is a lying position, the time is recorded until the patient is in a non-lying position. When it is recognized that there is no human body on the bed, the time recording stops. Several lying time periods are obtained, and the lying time period is summed up to get the lying time. ; Step E2: Get the number of gastric tube sensor errors , and construct the gastric tube calibration coefficient , the specific expression is as follows; in, Indicates the current percentage of the patient's non-lying time.

10. The intelligent multimodal human-computer interactive gastric tube system according to claim 9, characterized in that: The third feedback strategy is specifically to remind medical staff to check the sensor of the gastric tube and determine whether to replace it.

Images