Cardiac bypass intraoperative decision support system based on fuzzy logic control
Through a decision support system based on fuzzy logic control, the problem of insufficient data integration and decision recommendations in cardiac bypass surgery is solved, more efficient decision support and user experience is achieved, and the accuracy and safety of the surgery are improved.
Patent Information
- Application Number
- CN202411068998.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-11-19
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The intracardiac bypass decision support system has shortcomings in data integration and interoperability, decision recommendation generation, real-time and responsiveness, and user experience, which affects its widespread adoption and effectiveness in clinical practice.
Adopt a decision support system based on fuzzy logic control, through real-time physiological parameter monitoring and data collection module, fuzzy logic inference engine and multi-parameter data fusion technology, combining fuzzy logic algorithms and deep learning, provide personalized decision-making suggestions and design intuitive user interface.
It improves the accuracy and personalization of surgical decisions, enhances the system's real-time response capabilities, provides intuitive and operational decision support, reduces the cognitive burden of medical staff, and improves the safety and effectiveness of the surgery.
Smart Images

Figure CN118983055A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a decision support system during heart bypass surgery, in particular to a decision support system during heart bypass surgery based on fuzzy logic control. Background Art
[0002] Intraoperative decision support systems for coronary artery bypass grafting (CABG) are crucial in modern medicine, aiming to help surgeons make more accurate clinical decisions in complex and changing surgical environments. However, despite the significant potential advantages these systems offer in theory, they still face many deficiencies and drawbacks in actual applications, which affect their widespread adoption and effectiveness in clinical practice. The first major problem is insufficient data integration and interoperability. In coronary artery bypass grafting, data from multiple sources are involved, including patients' physiological monitoring data (heart rate, blood pressure, blood oxygen saturation), imaging data (preoperative and intraoperative cardiovascular imaging), electronic medical record data, and real-time input from the surgical team. However, due to the poor interoperability between different medical devices and systems, data cannot be seamlessly integrated and processed in real time. The lack of standardized data interfaces and protocols further limits the decision support system's ability to obtain comprehensive and timely information. In addition, poor data integration can also lead to delayed and inaccurate information, which in turn affects the accuracy and safety of surgical decisions.
[0003] Secondly, the intraoperative decision support system for coronary bypass surgery also has deficiencies in the generation and updating of decision recommendations. Many existing systems rely on predefined rule sets or simple algorithms, which are usually based on historical data and literature. However, there are many complex and unforeseen situations during coronary bypass surgery, which often exceed the processing capabilities of static rules. Although machine learning and artificial intelligence technologies have been gradually introduced in recent years, these technologies are still in their early stages and face problems such as insufficient model training, incomplete data sets, and poor model interpretability. In addition, the training of machine learning models requires a large amount of high-quality annotated data, but due to reasons such as medical data privacy protection and difficulty in obtaining data, it is impossible to obtain enough data to train high-performance models. Third, insufficient real-time and responsiveness are also a major challenge for intraoperative decision support systems for coronary bypass surgery. During surgery, it is necessary to respond quickly to emergencies, but the computing and processing capabilities of existing systems are limited and often cannot provide effective decision recommendations in the shortest time. This lag stems from the complexity of the system architecture, inefficient algorithms, or hardware limitations. In order to improve the response speed, the system sometimes has to simplify the model or reduce the accuracy of data processing, which will lead to inaccurate recommendations and affect the safety and effectiveness of the operation. Fourth, deficiencies in user experience and usability are also an important factor affecting the effectiveness of intraoperative decision support systems for heart bypass surgery. Many existing systems lack sufficient consideration of the actual needs of surgeons and surgical teams in their design. The user interface is often too complex or not intuitive enough, making it difficult for surgical teams to quickly obtain the required information under high-pressure environments. In addition, the suggestions provided by the system often lack contextual explanations and operability, and surgeons may choose to ignore the system's prompts because the suggestions are unclear or unreliable. The complexity of the system and the unfriendly user experience increase the cognitive burden of medical staff, which in turn affects its application effect in surgery. Summary of the invention
[0004] The purpose of the present invention is to provide a decision support system for coronary artery bypass surgery based on fuzzy logic control, thereby solving some of the drawbacks and shortcomings pointed out in the background technology.
[0005] The present invention solves the above-mentioned technical problems by adopting the following technical solutions:
[0006] Real-time physiological parameter monitoring and data collection module, used to monitor and collect the patient's electrocardiogram ECG, arterial blood pressure, pulse oxygen saturation physiological indicators through sensors and data acquisition equipment;
[0007] A fuzzy logic inference engine, designed and controlled to handle the uncertainty and ambiguity of input parameters, and to set fuzzy rules based on the experience of cardiac surgery experts and medical guidelines;
[0008] Multi-parameter data fusion technology uses fuzzy logic algorithms to fuse multiple physiological parameters to form a comprehensive assessment of the patient's status, and develops algorithms to adjust the membership functions and rule weights in the fuzzy logic controller;
[0009] The intraoperative decision support module provides drug adjustment and surgical operation decision support based on fuzzy logic output, and designs an intuitive user interface to provide decision support information and operation suggestions to the surgical team.
[0010] Furthermore, the real-time physiological parameter monitoring and data collection module construction process is as follows:
[0011] S1. Build a multimodal sensing technology integration module to collect ECG, arterial blood pressure, and pulse oxygen saturation physiological indicators through the device. The data is processed by the function:
[0012]
[0013] Where x, y, z represent the collected ECG, blood pressure, and oxygen saturation, respectively, and a, b, c are adjustment parameters;
[0014] S2, intelligent data preprocessing unit, using signal processing formula:
[0015]
[0016] Where t represents time, ω represents signal frequency, and d is the adjustment factor;
[0017] S3, context-aware data fusion module, applies the integral adjustment formula:
[0018]
[0019] Where s represents the input physiological signal strength, p represents the surgical progress, and k and m are parameters for adjusting the sensitivity of the surgical stage;
[0020] S4, real-time visual feedback system, using feedback formula:
[0021]
[0022] Where r represents the intensity of the decision result output by the fuzzy logic controller, and n is the visual display intensity factor.
[0023] Furthermore, the fuzzy logic reasoning engine implementation method includes:
[0024] S1. Build a decision support system for heart bypass surgery, including a hybrid fuzzy logic framework, combining type-1 and type-2 fuzzy logic, and adjusting the logic processing mode according to the data collected from sensors including blood pressure, heart rate and highly fuzzy clinical observation data to respond to emergencies during surgery;
[0025] S2, integrated expert system, which integrates the knowledge and experience of experts in the field of cardiac surgery to form a fuzzy logic rule base;
[0026] S3, integrates NLP technology to analyze and interpret the latest cardiac surgery research and medical guidelines, update and improve decision-making logic, and ensure that every decision during surgery is based on the most advanced medical knowledge.
[0027] Furthermore, the decision support system first constructs a hybrid fuzzy logic framework that combines type-1 and type-2 fuzzy logic through a multivariable function:
[0028]
[0029] where x, y, z represent the blood pressure, heart rate and highly ambiguous clinical observation data collected from sensors, a, b are adjustment parameters, and σ x represents the standard deviation of x;
[0030] Dynamic fuzzy rule adjustment uses differential equations:
[0031]
[0032] Where t represents time, ω(t) represents the dynamic changes of the data collected during surgery, and c, d, and e are adjustment factors;
[0033] Real-time data fusion and processing through integral formula:
[0034]
[0035] Where s(t) and r(t) represent the uncertainty coefficients of sensor data and clinical observation data, respectively, T represents the time of the entire surgical process, and f is the weight factor.
[0036] Furthermore, the integrated expert system implementation method includes:
[0037] S1, a module that uses natural language processing technology and machine learning to extract knowledge related to cardiac surgery, and uses differential equations to process and analyze text data:
[0038]
[0039] Where t represents the time or sequence position, a and b are adjustment parameters, and λ is the attenuation coefficient;
[0040] S2. Use knowledge graph technology to encode the extracted knowledge and convert it into a module of fuzzy logic rules, and use multivariable functions to construct and reason about entity relationships in the knowledge graph:
[0041]
[0042] Where x, y, z represent the attribute values of different entities, and c is the coefficient;
[0043] S3, a module that implements expert feedback loop and optimizes rules, adjusts fuzzy logic rules according to surgeon feedback, and uses a high-order integral formula:
[0044]
[0045] Where v represents the feedback strength, d is the adjustment coefficient, and u is the integral variable.
[0046] Furthermore, the integrated NLP technology uses high-order differential equations to update the rules in the fuzzy logic engine in real time:
[0047]
[0048] Where x represents the timestamp of data extracted from the latest research literature, y represents the key parameters extracted from the data, and a is the adjustment coefficient;
[0049] Then build the feedback processing module and apply the integral formula to optimize the fuzzy logic rules:
[0050]
[0051] Where t represents the time of feedback collected from the surgical team, and b is the adjustment factor;
[0052] Finally, a data fusion module is constructed, which uses a multivariate function to fuse clinical decision support and research data:
[0053]
[0054] Where x, y, z represent the efficiency of the treatment plan, the intensity of the patient's response and the reliability of the scientific evidence, respectively, and c is the coefficient.
[0055] Furthermore, the multi-parameter data fusion technology includes: constructing an adjustment module of a real-time dynamic membership function, and adopting an adaptive algorithm based on deep learning:
[0056]
[0057] Where x represents the blood pressure data collected from the physiological monitoring device, y represents the heart rate data, z represents the blood oxygen saturation, and a and b are adjustment coefficients;
[0058] Reinforcement learning optimization module for fuzzy logic rules, using the integral formula:
[0059]
[0060] Where t represents time, s represents the intensity of surgical team input collected from the feedback system, and c is the adjustment factor;
[0061] Multi-parameter data fusion and decision making module, applying multivariable functions:
[0062]
[0063] Where x, y, z represent cardiac output pressure, blood oxygen saturation and other key physiological parameters, respectively, and d is the adjustment parameter.
[0064] Furthermore, the intraoperative decision support module construction process is as follows:
[0065] S1. Build a real-time decision support module based on fuzzy logic and use a dynamic adjustment algorithm to directly drive surgical operation recommendations:
[0066]
[0067] where x and y represent the data collected from physiological monitoring devices, including blood pressure and heart rate, respectively, z represents the surgical stage, and a is the adjustment coefficient;
[0068] S2. Construct an adaptive adjustment module for personalized fuzzy logic rules, using an adaptive learning function:
[0069]
[0070] Where x represents the patient's real-time physiological parameters, y represents the influencing factor of the patient's historical data, and b is the adjustment coefficient.
[0071] The beneficial effects of the present invention are as follows: first, the accuracy and personalization of surgical decisions are improved. By integrating fuzzy logic control to process physiological parameters and clinical data, real-time analysis is performed and personalized decision-making suggestions are provided. The advantage of fuzzy logic is that it can handle uncertainty and ambiguity, so that the system can generate more accurate suggestions when faced with complex and changeable surgical environments. Compared with traditional fixed rule systems, fuzzy logic-based systems can better adapt to individual differences and provide tailored treatment plans for each patient. Secondly, the system greatly improves the real-time response capability during surgery. Since the fuzzy logic controller can quickly process data from multiple monitoring devices, the system can monitor the patient's physiological state in real time during surgery and provide timely adjustment suggestions when necessary. BRIEF DESCRIPTION OF THE DRAWINGS
[0072] Figure 1 This is a flow chart of the intraoperative decision support system for cardiac bypass surgery based on fuzzy logic control of the present invention.
[0073] Figure 2 This is a diagram of the real-time physiological parameter monitoring and data collection module of the present invention.
[0074] Figure 3 Construct a flow chart for the fuzzy logic reasoning engine of the present invention.
[0075] Figure 4 This is a diagram of the intraoperative decision support module of the present invention. DETAILED DESCRIPTION
[0076] The specific implementation modes of the present invention will be described in detail below in conjunction with the accompanying drawings.
[0077] Combined with Figure 1 The construction process of the intraoperative decision support system for cardiac bypass surgery based on fuzzy logic control includes: The real-time physiological parameter monitoring and data collection module uses high-precision sensors, including electrocardiogram (ECG) sensors, arterial blood pressure monitors and pulse oxygen saturation probes, to capture and transmit the patient's physiological data in real time. These sensors are connected to the central data processing unit, and the real-time data stream is transmitted through a high-bandwidth connection to ensure the data is processed without delay and with high reliability.
[0078] The fuzzy logic inference engine uses this real-time data as input. The engine is configured with a set of fuzzy rules developed by cardiac surgeons and data scientists. These rules can process and interpret the inherent uncertainty of medical data, including small fluctuations in heart rate and brief jumps in blood pressure, to generate recommendations for drug dosage adjustments or physiological support measures.
[0079] In terms of multi-parameter data fusion technology, the system uses advanced algorithms to integrate data from different sensors, which not only includes mapping the data to a unified health status indicator, but also uses machine learning technology to dynamically adjust the shape of the membership function and the weight of the rules. This dynamic adjustment is based on real-time surgical performance and historical data model training results, enabling the system to adapt to the physiological responses of each unique patient.
[0080] The intraoperative decision support module is the front-end display layer of the system, which converts the output of the fuzzy logic engine into intuitive graphics and text suggestions and displays them on the monitor in the operating room.
[0081] Embodiment 1:
[0082] View attached Figure 2 In the intraoperative decision support system for cardiac bypass surgery based on fuzzy logic control, the multimodal sensing technology integration module is used to collect and process physiological parameters, and multiple sensing devices are integrated to collect the patient's electrocardiogram (ECG), arterial blood pressure and pulse oxygen saturation key physiological indicators in real time. ECG sensors, arterial blood pressure monitoring devices and pulse oxygen saturation sensors are selected. All devices have real-time data transmission capabilities to ensure the real-time and accuracy of the data. The data processing function is used:
[0083]
[0084] Parameter definition:
[0085] x: voltage value measured by electrocardiogram (ECG) (mV);
[0086] y: arterial blood pressure (mmHg);
[0087] z: pulse oxygen saturation (%);
[0088] a, b, c: adjustment parameters. The specific values need to be adjusted according to the actual application. The value range is a = 0.1 to 0.5, b = 1 to 5, c = 0.01 to 0.1.
[0089] The data collected by the sensor are set as follows: x=0.8 mV; y=120 mmHg; z=98%.
[0090] Apply the median values a=0.3, b=3, c=0.05 in the above parameter range and substitute them into the data processing function to obtain:
[0091]
[0092] Calculate the parts:
[0093] 0.8 3 / 2 =0.717
[0094] sin 120 = 0.866
[0095] ln(1+98)=4.595
[0096] 98 2 =9604
[0097] but:
[0098]
[0099] f(x,y,z)=0.2151+311.28+0.000239≈311.495339
[0100] This value will be used to assess the patient's immediate physiological state and serve as an input for the fuzzy logic engine decision-making, further providing recommendations for intraoperative medication adjustments and surgical procedures.
[0101] The embodiment then constructs an intelligent data pre-processing unit which plays a key role in processing the physiological signal data collected from the multi-modal sensing technology integration module to ensure data quality and provide accurate input to the fuzzy logic engine, applying the signal processing formula:
[0102]
[0103] Parameter definition:
[0104] t: time, in seconds;
[0105] ω: signal frequency, in Hertz, selected according to the characteristics of the ECG signal or other physiological signals;
[0106] d: Adjustment factor, used to adjust the strength of the processed signal, ranging from 0.1 to 1.0, depending on the original amplitude of the signal and the expected output amplitude.
[0107] Assume that during a heart bypass surgery, a patient's ECG signal is monitored in real time by a sensor. At a specific time point t = 0.5 seconds, the signal frequency ω is 5 Hz, and the adjustment factor d is selected as 0.5, which is used as an example of data preprocessing.
[0108] Substitute into the data processing function for calculation:
[0109]
[0110] cos(ωt 2 )=cos(5×0.5 2 )=cos(1.25)≈0.3171
[0111] therefore:
[0112] g(t,ω)=0.5·0.7788·0.3171≈0.1236
[0113] This result shows that the ECG signal strength after being processed by the intelligent data pre-processing unit is optimized and adjusted to suit the needs of the fuzzy logic engine.
[0114] Then the embodiment applies the context-aware data fusion module to comprehensively process multiple physiological signals by applying the integral adjustment formula and combine it with the surgical progress to provide status assessment and decision support. This module can dynamically adapt to changes in the patient's status during the operation to ensure the accuracy and real-time nature of the decision. Apply the integral adjustment formula:
[0115]
[0116] Parameter definition:
[0117] s: preprocessed ECG signal strength;
[0118] p: duration of the operation, in minutes;
[0119] k and m: parameters for adjusting the sensitivity of the surgical stage, ranging from k = 0.1 to 1.0, m is the key time point during the surgery, and m is 50% when the surgery is halfway through.
[0120] Assume that during a heart bypass surgery, the physiological signal strength s collected by the sensor is 0.8, and the surgery has been going on for 30 minutes, at which time p = 30. Set the adjustment parameter k = 0.5, and the key time point of the surgery m = 20 minutes.
[0121] When t is from 0 to 30:
[0122] e -k(t-m) =e -0.5(t-20)
[0123] so:
[0124]
[0125] Using numerical integration methods:
[0126] The integration interval (0,30) is divided into 100 small intervals, each with a length of 0.3 minutes. The function value is calculated for each small interval and accumulated to approximate the integration result.
[0127] Calculate the function value on each small interval, in the i-th interval:
[0128] t i =i·0.3
[0129]
[0130]
[0131] Accumulate the function values of all intervals and multiply them by the interval length 0.3 to get the integral approximation.
[0132] Approximate calculation results:
[0133]
[0134] This result represents the comprehensive physiological signal strength assessment calculated by the context-aware data fusion module at 30 minutes into the surgery.
[0135] Next, the embodiment uses the feedback formula to convert the output of the fuzzy logic controller into easy-to-understand visual information to help the surgical team understand the patient's status and decision-making suggestions. Through the feedback formula:
[0136]
[0137] Parameter definition:
[0138] r: the intensity of the decision result output by the fuzzy logic controller, which is calculated based on the comprehensive evaluation value of the physiological parameters;
[0139] n: Visual display intensity factor, ranging from 0.1 to 2.0, used to adjust the sensitivity of visual feedback to adapt to different surgical scenarios and team preferences.
[0140] The decision result intensity r of the fuzzy logic controller output was set to 1.5, and the visual display intensity factor n was selected to 1.5 according to the experience and requirements of the surgical team.
[0141] First, calculate the integral part Use Simpson's integral method to calculate the integral value:
[0142] Divide the integral interval (0,1.5) into 100 small intervals, each with a length of 0.015, and calculate the function value for each small interval And accumulate to approximate the integral result.
[0143] Points
[0144] Substitute the integral value into the feedback formula to calculate:
[0145] v(r)=1.5·(0.45) 2 =1.5 0.2025 = 0.30375
[0146] This feedback value v(r)=0.30375 will be displayed on the monitor in the operating room through the real-time visual feedback system and presented to the surgical team in the form of a dynamic dashboard or bar chart.
[0147] Embodiment 2:
[0148] View attached Figure 3 In the embodiment, in the intraoperative decision support system of cardiac bypass surgery based on fuzzy logic control, a hybrid fuzzy logic framework is adopted to combine type-1 and type-2 fuzzy logic together to process physiological data and clinical observations from multiple sensors, and the hybrid fuzzy logic framework is applied:
[0149]
[0150] Parameter definition:
[0151] x: arterial blood pressure collected from the sensor (unit: mmHg);
[0152] y: heart rate collected from the sensor (unit: bpm);
[0153] z: patient's subjective pain score, ranging from 0 to 10;
[0154] a, b: adjustment parameters used to weigh the importance of different signals for decision making, with a value range of 1 to 5 and b of 0.5 to 2.0;
[0155] σ x : Represents the standard deviation of blood pressure data, calculated based on the patient's historical data.
[0156] During the operation, the data collected by the sensor are:
[0157] Blood pressure x = 120 mmHg, standard deviation of blood pressure σ x =15; heart rate y = 80 bpm; clinical observation data z = 4; use adjustment parameters a = 3, b = 1.0, substitute into the function for calculation:
[0158] Calculate each component:
[0159]
[0160] Substituting these results into the function:
[0161] f(x,y,z)=3·1.266×10 -14 0.99992+1.0 0.2107
[0162] f(x,y,z)≈3.798×10 -14 +0.2107≈0.2107
[0163] The embodiment further constructs a dynamic fuzzy rule adjustment module, using a differential equation:
[0164]
[0165] Parameter definition:
[0166] t: time, in minutes;
[0167] ω(t): dynamic changes of data collected during surgery, including blood pressure fluctuations and heart rate changes;
[0168] c, d, e: adjustment factors used to control the sensitivity of the differential equation to rule adjustments. The value range is c = 0.01 to 0.1, d = 0.1 to 1.0, e = 0.1 to 1.0.
[0169] The changes of blood pressure fluctuation ω(t) monitored in real time by the sensor over time are as follows:
[0170] At t = 0 minutes, ω(0) = 120 mmHg;
[0171] At t = 1 minute, ω(1) = 122 mmHg;
[0172] At t = 2 minutes, ω(2) = 125 mmHg.
[0173] Using these data, calculate and
[0174]
[0175] Substitute into the differential equation for calculation, and select adjustment factors c = 0.05, d = 0.5, e = 0.8:
[0176] g(t,ω)=0.05·1+0.5·3+0.8·125
[0177] g(t,ω)=0.05+1.5+100
[0178] g(t,ω)=101.55
[0179] By calculating g(t,ω)=101.55, the system dynamically adjusts the fuzzy rules to optimize decision support during surgery.
[0180] The last embodiment uses the real-time data fusion and processing module to comprehensively analyze the uncertainty of sensor data and clinical observation data using the integral formula:
[0181]
[0182] Parameter definition:
[0183] s(t): sensor data, including real-time strength of electrocardiogram and blood pressure physiological signals;
[0184] r(t): uncertainty coefficient of clinical observation data, which is the subjective assessment of the patient's response during surgery;
[0185] T: time of the entire surgical procedure (in minutes);
[0186] f: Weight factor, used to adjust the impact of different signals on the comprehensive evaluation results, ranging from 0.1 to 1.0.
[0187] During a coronary bypass surgery, the sensor data and clinical observation data are as follows:
[0188] s(t)=0.8+0.1·sin(0.2t) represents the sensor data intensity (fluctuating between 0.7 and 0.9);
[0189] r(t) = 0.05 + 0.02·cos(0.1t) represents the uncertainty of clinical observation data (fluctuating between 0.03 and 0.07);
[0190] Total operation time T = 60 minutes;
[0191] The weighting factor f=0.5 is chosen.
[0192] The inner integral is evaluated numerically using Simpson's integration method, At each time t, the cumulative uncertainty from 0 to the current time is calculated, and the external integral is numerically calculated to calculate the contribution of each time point:
[0193]
[0194] The total time (0,60) is divided into 600 small intervals, each with a length of 0.1 minutes; s(t) and the cumulative uncertainty are calculated in each interval, and the integral contribution of the interval is calculated; the integral contributions of all intervals are accumulated and multiplied by the weight factor f.
[0195] Approximate calculation results:
[0196] At the end of t = T. Inner integral
[0197] External integral:
[0198]
[0199] The calculated result h(s,r)≈1.22 represents the comprehensive physiological status evaluation value during the operation.
[0200] Embodiment 3:
[0201] In the intraoperative decision support system for cardiac bypass surgery based on fuzzy logic control, the integrated expert system module integrates the knowledge and experience of experts in the field of cardiac surgery to form a dynamic fuzzy logic rule base, uses natural language processing (NLP) technology and machine learning algorithms to extract knowledge from cardiac surgery literature, and uses differential equations:
[0202]
[0203] Parameter definition:
[0204] t: sequence position, used to represent a time series point in a data stream or text data;
[0205] a, b: adjustment parameters, used to adjust the impact of each item on the output, with a value range of 0.1 to 1.0 and b of 0.1 to 0.5;
[0206] λ: Attenuation coefficient, used to control the influence range of the exponential term, with a value range of λ = 0.01 to 0.1.
[0207] A key information sequence extracted from a large amount of literature is assumed to be as follows to represent the information intensity or complexity corresponding to the time series position t=1, 2, 3.
[0208] Using parameter values a=0.5, b=0.2, λ=0.05:
[0209] Compute the first term:
[0210] t=1,a·t 3 ·e -λt =0.5·1 3 ·e -0.05×1 =0.5·1·0.9512≈0.4756
[0211] t=2,a·t 3 ·e -λt =0.5·2 3 ·e -0.05×2 =0.5·8·0.9048≈3.6192
[0212] t=3,a·t 3 ·e -λt =0.5·3 3 ·e -0.05×3 =0.5·27·0.8607≈11.6174
[0213] Calculate the second term:
[0214]
[0215] Calculate the comprehensive output:
[0216] f1=0.4756+0.1571≈0.6327
[0217] f(2)=3.6192+0.0886≈3.7078
[0218] f(3)=11.6174+0.0499≈11.6673
[0219] By calculating the result f(t), the system can determine the validity and relevance of the knowledge points extracted from the literature and reports.
[0220] The embodiment further uses graph technology to encode the extracted expert knowledge and convert it into fuzzy logic rules that are directly applied to surgical decision support. Use multivariate functions to construct and reason about the relationship between entities in the knowledge graph:
[0221]
[0222] Parameter definition:
[0223] x: changes in blood pressure of patients during surgery (unit: mmHg);
[0224] y: represents the number or citations of empirical rules of experts in the field of cardiac surgery;
[0225] z: patient's postoperative recovery speed (unit: day);
[0226] c: coefficient, used to adjust the sensitivity of the function to different variables, ranging from 1 to 5.
[0227] In the heart bypass surgery knowledge graph, the following entity relationships need to be analyzed:
[0228] Blood pressure change x = 10 mmHg;
[0229] The number of citations of empirical rules y = 30;
[0230] The recovery rate observed clinically was z = 3 days.
[0231] Substitute the coefficient c = 2 into the formula for calculation:
[0232] Calculate each component:
[0233]
[0234] ln(y+e x )=ln(30+e 10 )=ln(30+22026.47)≈10.0
[0235]
[0236] Calculate the comprehensive output:
[0237] g(x,y,z)=2·11013.23·10.0·0.9
[0238] g(x,y,z)≈198238.14
[0239] The calculated result g(x,y,z)≈198238.14 indicates the relationship strength and influence of these variable combinations in the current knowledge graph.
[0240] The embodiment further implements the expert feedback loop and optimizes the rules by collecting and analyzing the feedback from the surgeons, dynamically adjusting the fuzzy logic rules to optimize the surgical decision, and applying the high-order integral formula:
[0241]
[0242] Parameter definition:
[0243] v: feedback intensity, evaluation provided by the surgical team regarding the performance of the current decision support system;
[0244] d: adjustment coefficient, used to control the impact of feedback in rule optimization, ranging from 0.1 to 2.0;
[0245] u,w: Integral variables used to accumulate and analyze feedback effects.
[0246] The surgical team's feedback intensity v = 3 on the decision support system is derived from the evaluation of the effectiveness and timeliness of the system's suggestions. The adjustment coefficient d = 0.5 is selected.
[0247] The integral interval (0,3) is divided into 30 small intervals, each with a length of 0.1. The function value is calculated for each interval and accumulated to approximate the integral result.
[0248] When u=v, inner integral:
[0249]
[0250] External integral:
[0251]
[0252] The calculated result h(v)≈1.05 indicates the influence of feedback strength on the optimization of fuzzy logic rules.
[0253] Embodiment 4:
[0254] In the example, the natural language processing (NLP) technology module is integrated into the intraoperative decision support system for cardiac bypass surgery based on fuzzy logic control. By analyzing and parsing the latest cardiac surgery research and medical guidelines, the decision logic in the system is updated in real time. Application of high-order differential equations:
[0255]
[0256] Parameter definition:
[0257] x: the timestamp of the data extracted from the latest research literature, indicating the frequency or timeliness of data updates;
[0258] y: key parameters extracted from the literature, including the effectiveness of newly discovered surgical techniques or drug therapies;
[0259] a: Adjustment coefficient, used to control the impact of new knowledge on rule updating, with a value range of 0.1 to 1.0.
[0260] The NLP module extracts the following data from the latest research literature:
[0261] Data timestamp x=2, indicating that the data was released within the last two months;
[0262] Key Parameters (30 degrees), indicating the effectiveness of anesthetic drugs under specific conditions;
[0263] Select adjustment coefficient a=0.5;
[0264] set up Find the third derivative
[0265] Compute the first-order derivative:
[0266]
[0267] Compute the second-order derivative:
[0268]
[0269] Compute the third-order derivative:
[0270]
[0271] Substituting x=2, a=0.5:
[0272]
[0273] The calculated result f(x,y)≈1200.7 indicates the influence of new knowledge on the updating of fuzzy logic rules.
[0274] Next, the embodiment constructs a feedback processing module, optimizes the fuzzy logic rules by analyzing the real-time feedback of the surgical team, and uses the integral formula to process and evaluate the feedback data, thereby dynamically adjusting the rules in the system to ensure flexibility and accuracy in different clinical scenarios. The integral formula is applied:
[0275]
[0276] Parameter definition:
[0277] t: represents the feedback time collected from the surgical team (unit: minutes),
[0278] b: Adjustment factor, used to control the impact of feedback in rule optimization, ranging from 0.1 to 1.0.
[0279] In the heart bypass surgery, the surgical team provided an evaluation of the decision support system performance within the postoperative feedback time t = 30 minutes. The adjustment factor b = 0.5 was selected, and the integral value was calculated using the Simpson integral method. The integral interval (0, 30) was divided into 300 small intervals, each with a length of 0.1 minutes. The function value was calculated for each interval. And accumulate.
[0280] Set the approximate calculation results:
[0281]
[0282] Substitute into the formula to calculate:
[0283] g(30)=0.5·150=75
[0284] The calculated result g(t)=75 indicates the influence of feedback time on the optimization of fuzzy logic rules.
[0285] The final example constructs a data fusion module to fuse different data sources of multiple variables, provide a comprehensive assessment of the patient's condition and treatment recommendations, and apply multivariate functions:
[0286]
[0287] Parameter definition:
[0288] x: represents the efficiency of the treatment plan, expressed as a relative efficiency value ranging from 0 to 10;
[0289] y: represents the intensity of the patient's response, quantifying the patient's postoperative recovery speed or drug response, ranging from 0 to 10;
[0290] z: represents the reliability of scientific evidence, quantified as an evidence quality score or confidence value, ranging from 0 to 1;
[0291] c: coefficient, used to adjust the importance of different data sources on decision-making, ranging from 0.1 to 5.0.
[0292] Integrate the following data:
[0293] The efficiency of the treatment plan is x = 8;
[0294] The intensity of the patient's response y = 4;
[0295] The reliability of scientific evidence is z=0.8.
[0296] Select coefficient c = 2 and substitute it into the function to calculate:
[0297] Calculate each component:
[0298] log(x+1)=log(8+1)=log(9)≈2.1972
[0299] x 2 +y 2 =8 2 +4 2 =64+16=80
[0300]
[0301] Calculate the comprehensive output:
[0302] h(x,y,z)=2·0.0275·1.027×10 -10
[0303] h(x,y,z)≈5.646×10 -12
[0304] The calculated result is h(x,y,z)≈5.646×10 -12 Indicates the strength and influence of data fusion in the current situation.
[0305] Embodiment 5:
[0306] Embodiment In a cardiac bypass surgery intraoperative decision support system based on fuzzy logic control, an adjustment module of a real-time dynamic membership function is constructed, and an adaptive algorithm based on deep learning is used to process and fuse a variety of physiological data. The formula is as follows:
[0307]
[0308] Parameter definition:
[0309] x: represents the blood pressure data collected from the physiological monitoring device (unit: mmHg);
[0310] y: represents heart rate data (unit: bpm);
[0311] z: represents blood oxygen saturation (unit: %);
[0312] a and b: adjustment coefficients, used to adjust the sensitivity of the function, with a value range of 0.1 to 1.0 and b of 10 to 100.
[0313] Data collected by physiological monitoring equipment: blood pressure x=120 mmHg; heart rate y=80 bpm; blood oxygen saturation z=98%.
[0314] Select adjustment coefficient a = 0.5, b = 50, calculate the third-order derivative of the internal term of the function, and assume:
[0315]
[0316] Compute the third-order derivative:
[0317]
[0318] therefore:
[0319] f(x,y,z)=a·6=0.5·6=3
[0320] The calculated result f(x, y, z)=3 represents the fusion result based on the current physiological monitoring data.
[0321] Embodiment Next, a reinforcement learning optimization module of fuzzy logic rules is constructed, and the fuzzy logic rules are optimized using integral formulas to adapt to the ever-changing needs and feedback during surgery. The formula is:
[0322]
[0323] Parameter definition:
[0324] t: represents time (unit: second);
[0325] s: represents the intensity of surgical team input collected from the feedback system (unit: feedback intensity score);
[0326] c: Adjustment factor, used to adjust the influence of the integral result, the value range is c = 0.1 to 1.0.
[0327] During the set operation, the system received the following feedback data:
[0328] Feedback time t = 60 seconds;
[0329] Feedback strength s = 2;
[0330] The adjustment factor c=0.5.
[0331] The value of the function g(t,s) is optimized by fuzzy logic rules:
[0332] Substituting into the formula:
[0333]
[0334] The integration range is 0 to 60 seconds, s=2, c=0.5.
[0335] Compute the integral:
[0336]
[0337] Use Simpson integration method, select integration step h = 1 second, segmentation interval [0,60], and calculate the integral value at each step:
[0338]
[0339] Substituting in the adjustment factor c:
[0340] g(60,2)=0.5·8.45=4.225
[0341] The calculated result g(60,2)=4.225 represents the reinforcement value of the fuzzy logic rule obtained by the reinforcement learning optimization based on the surgical team feedback.
[0342] The final embodiment constructs a multi-parameter data fusion and decision-making module to improve the accuracy of decision-making and the safety of surgery. Using a multivariable function:
[0343]
[0344] Among them, x represents cardiac output pressure, y represents blood oxygen saturation, z represents other key physiological parameters, and d is the adjustment parameter.
[0345] The following data was collected:
[0346] Cardiac output pressure x = 120 mmHg;
[0347] Blood oxygen saturation y = 95%;
[0348] Other key physiological parameters z = 0.5;
[0349] Adjust parameter d=1.
[0350] calculate
[0351]
[0352] Calculate log(y+1):
[0353] log(y+1)=log(95+1)=log(96)≈1.982
[0354] Compute arctan(z):
[0355] arctan(z)=arctan(0.5)≈0.464
[0356] Substitute the formula to calculate the fusion value:
[0357]
[0358] h(x,y,z)=(46.8) 2
[0359] h(x,y,z)≈2194.2
[0360] The calculated result h(x,y,z)≈2194.2 represents the output value of the comprehensive decision support system based on the provided physiological parameters during the operation.
[0361] Embodiment 6:
[0362] View attached Figure 4 In the embodiment, in the intraoperative decision support system of cardiac bypass surgery based on fuzzy logic control, a real-time decision support module is constructed, and the surgical operation suggestion is directly driven by the dynamic adjustment algorithm, and the dynamic adjustment algorithm is applied:
[0363]
[0364] Parameter definition:
[0365] x: represents the blood pressure data collected from the physiological monitoring device (unit: mmHg);
[0366] y: represents the heart rate data collected from the physiological monitoring device (unit: bpm);
[0367] z: represents the surgical stage (unit: time, minutes);
[0368] a: Adjustment coefficient, used to adjust the influence of the function, with a value range of 0.1 to 1.0.
[0369] During a heart bypass surgery, the following data are collected by the physiological monitoring equipment:
[0370] Blood pressure x = 120 mmHg;
[0371] Heart rate y = 80 bpm;
[0372] surgical phase z = 30 minutes;
[0373] Adjustment coefficient a=0.5.
[0374] Calculate the value of the function f(x,y,z) to provide surgical operation suggestions, calculate the second-order derivative of the internal term of the function, and assume:
[0375]
[0376] Find the second derivative
[0377] Compute the first-order derivative:
[0378] Compute the second-order derivative:
[0379] Merge Items:
[0380] Substitute the data for calculation:
[0381] Substitute x=120, y=80, z=30
[0382] calculate
[0383] (The value is very small and can be ignored)
[0384] Calculate sin(30)=0.5, substitute the adjustment coefficient a=0.5, and calculate the second-order derivative:
[0385]
[0386] result:
[0387] f(120,80,30)=0.5·0=0
[0388] The calculated result f(x, y, z)=0 indicates that under the current circumstances, the dynamic adjustment based on real-time monitoring data and the surgical stage has very little impact on the surgical operation recommendations.
[0389] The embodiment further constructs an adaptive adjustment module of personalized fuzzy logic rules, and dynamically adjusts the fuzzy logic rules according to the patient's real-time physiological parameters and historical data through an adaptive learning function, thereby optimizing the decision support during the operation. The formula is as follows:
[0390]
[0391] Parameter definition:
[0392] x: represents the patient’s current blood pressure;
[0393] y: historical data of the patient’s average blood pressure;
[0394] b: adjustment coefficient, used to adjust the influence of the function, ranging from 0.1 to 1.0;
[0395] During a heart bypass surgery, the system collects the following data:
[0396] Real-time physiological parameter x = 120 mmHg (blood pressure);
[0397] Historical data impact factor y = 1.2 (average blood pressure impact factor calculated based on historical data);
[0398] Adjustment coefficient b=0.5.
[0399] Calculated using Simpson's integral method:
[0400]
[0401] Select the step size h = 1 second, the segmentation interval is [0,120], and calculate the integral value at each step.
[0402] The formula for Simpson's integral method is:
[0403]
[0404] In this example,
[0405] Calculate the value of f(t) at each step size t:
[0406]
[0407] Substitute the adjustment factor b and calculate the final result:
[0408] g(120,1.2)=0.5·0.06=0.03
[0409] The calculated result g(120,1.2)=0.03 represents the fuzzy logic rule adjustment value based on the current patient physiological parameters and historical data.
[0410] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.
Claims
1. A fuzzy logic-based decision support system for cardiac bypass surgery, characterized by include: Real-time physiological parameter monitoring and data collection module, used to monitor and collect the patient's electrocardiogram ECG, arterial blood pressure, pulse oxygen saturation physiological indicators through sensors and data acquisition equipment; A fuzzy logic inference engine, designed and controlled to handle the uncertainty and ambiguity of input parameters, and to set fuzzy rules based on the experience of cardiac surgery experts and medical guidelines; Multi-parameter data fusion technology uses fuzzy logic algorithms to fuse multiple physiological parameters to form a comprehensive assessment of the patient's status, and develops algorithms to adjust the membership functions and rule weights in the fuzzy logic controller; The intraoperative decision support module provides drug adjustment and surgical operation decision support based on fuzzy logic output, and designs an intuitive user interface to provide decision support information and operation suggestions to the surgical team.
2. The fuzzy logic control-based intraoperative decision support system for cardiac bypass surgery according to claim 1 is characterized in that The real-time physiological parameter monitoring and data collection module construction process is as follows: S1. Build a multimodal sensing technology integration module to collect ECG, arterial blood pressure, and pulse oxygen saturation physiological indicators through the device. The data is processed by the function: Where x, y, z represent the collected ECG, blood pressure, and oxygen saturation, respectively, and a, b, c are adjustment parameters; S2, intelligent data preprocessing unit, using signal processing formula: Where t represents time, ω represents signal frequency, and d is the adjustment factor; S3, context-aware data fusion module, applies the integral adjustment formula: Where s represents the input physiological signal strength, p represents the surgical progress, and k and m are parameters for adjusting the sensitivity of the surgical stage; S4, real-time visual feedback system, using feedback formula: Where r represents the intensity of the decision result output by the fuzzy logic controller, and n is the visual display intensity factor.
3. The intraoperative decision support system for cardiac bypass surgery based on fuzzy logic control according to claim 1 is characterized in that The fuzzy logic reasoning engine implementation method comprises: S1. Build a decision support system for heart bypass surgery, including a hybrid fuzzy logic framework, combining type-1 and type-2 fuzzy logic, and adjusting the logic processing mode according to the data collected from sensors including blood pressure, heart rate and highly fuzzy clinical observation data to respond to emergencies during surgery; S2, integrated expert system, which integrates the knowledge and experience of experts in the field of cardiac surgery to form a fuzzy logic rule base; S3. Integrate NLP technology to analyze and interpret the latest cardiac surgery research and medical guidelines, update and improve decision-making logic, and ensure that every decision during surgery is based on online updated medical knowledge.
4. The fuzzy logic control-based intraoperative decision support system for cardiac bypass surgery according to claim 3 is characterized in that The decision support system first constructs a hybrid fuzzy logic framework that combines type-1 and type-2 fuzzy logic through a multivariable function: where x, y, z represent the blood pressure, heart rate and highly ambiguous clinical observation data collected from sensors, a, b are adjustment parameters, and σ x represents the standard deviation of x; Dynamic fuzzy rule adjustment uses differential equations: Where t represents time, ω(t) represents the dynamic changes of the data collected during surgery, and c, d, and e are adjustment factors; Real-time data fusion and processing through integral formula: Where s(t) and r(t) represent the uncertainty coefficients of sensor data and clinical observation data, respectively, T represents the time of the entire surgical process, and f is the weight factor.
5. The intraoperative decision support system for cardiac bypass surgery based on fuzzy logic control according to claim 3 is characterized in that The integrated expert system implementation method comprises: S1, a module that uses natural language processing technology and machine learning to extract knowledge related to cardiac surgery, and uses differential equations to process and analyze text data: Where t represents the time or sequence position, a and b are adjustment parameters, and λ is the attenuation coefficient; S2. Use knowledge graph technology to encode the extracted knowledge and convert it into a module of fuzzy logic rules, and use multivariable functions to construct and reason about entity relationships in the knowledge graph: Where x, y, z represent the attribute values of different entities, and c is the coefficient; S3, a module that implements expert feedback loop and optimizes rules, adjusts fuzzy logic rules according to surgeon feedback, and uses a high-order integral formula: Where v represents the feedback strength, d is the adjustment coefficient, and u is the integral variable.
6. The fuzzy logic control-based intraoperative decision support system for cardiac bypass surgery according to claim 3, characterized in that The integrated NLP technology uses high-order differential equations to update the rules in the fuzzy logic engine in real time: Where x represents the timestamp of data extracted from the latest research literature, y represents the key parameters extracted from the data, and a is the adjustment coefficient; Then build the feedback processing module and apply the integral formula to optimize the fuzzy logic rules: Where t represents the time of feedback collected from the surgical team, and b is the adjustment factor; Finally, a data fusion module is constructed, which uses a multivariate function to fuse clinical decision support and research data: Where x, y, z represent the efficiency of the treatment plan, the intensity of the patient's response and the reliability of the scientific evidence, respectively, and c is the coefficient.
7. The fuzzy logic control-based intraoperative decision support system for cardiac bypass surgery according to claim 1, characterized in that The multi-parameter data fusion technology includes: constructing an adjustment module of a real-time dynamic membership function, and adopting an adaptive algorithm based on deep learning: Where x represents the blood pressure data collected from the physiological monitoring device, y represents the heart rate data, z represents the blood oxygen saturation, and a and b are adjustment coefficients; Reinforcement learning optimization module for fuzzy logic rules, using the integral formula: Where t represents time, s represents the intensity of surgical team input collected from the feedback system, and c is the adjustment factor; Multi-parameter data fusion and decision making module, applying multivariable functions: Where x, y, z represent cardiac output pressure, blood oxygen saturation and other key physiological parameters, respectively, and d is the adjustment parameter.
8. The fuzzy logic control-based intraoperative decision support system for cardiac bypass surgery according to claim 1, characterized in that The intraoperative decision support module construction process is as follows: S1. Build a real-time decision support module based on fuzzy logic and use a dynamic adjustment algorithm to directly drive surgical operation recommendations: where x and y represent the data collected from physiological monitoring devices, including blood pressure and heart rate, respectively, z represents the surgical stage, and a is the adjustment coefficient; S2. Construct an adaptive adjustment module for personalized fuzzy logic rules, using an adaptive learning function: Where x represents the patient's real-time physiological parameters, y represents the influencing factor of the patient's historical data, and b is the adjustment coefficient.
Citation Information
Cited By
Energy-saving illumination control method and system for urban illumination LED lamp group
CN119403009A
Intelligent regulation and control method for monitoring period of DevOps platform
CN119938313A
A devops platform monitoring cycle intelligent regulation method
CN119938313B
