Optical fiber sensing cardiovascular postoperative wound intelligent monitoring and nursing system

Abstract

The invention discloses an optical fiber sensing cardiovascular postoperative wound intelligent monitoring and nursing system, and relates to the technical field of medical monitoring and nursing, and the system comprises an optical fiber sensing array module which is customized for different cardiovascular operations and comprises distributed and point type optical fibers; the multi-parameter data acquisition module is used for receiving a signal by using a two-channel demodulator, performing high-frequency data acquisition after interference removal, and synchronizing patient information and vital signs; the wound state analysis module is used for adjusting a threshold value according to the individual condition of the patient and judging the wound state by using an improved support vector machine; the intelligent nursing suggestion module generates a nursing scheme according to the state; according to the user interaction module, the medical care terminal displays data and states, and the patient terminal has voice reminding and AR guidance; and the data storage and update module adopts a three-level storage architecture and periodically updates an algorithm and a nursing scheme library. According to the system, the wound monitoring accuracy and continuity are improved, the medical care management efficiency is optimized, the autonomous nursing ability of a patient is enhanced, and the risk of postoperative wound complications is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical monitoring and nursing, and particularly relates to an optical fiber sensing postoperative cardiovascular wound intelligent monitoring and nursing system. BACKGROUND

[0002] After cardiovascular surgery (such as coronary artery bypass surgery and heart valve replacement surgery), the wound healing process is easily affected by multiple factors such as pressure, temperature and exudation. If timely monitoring or improper nursing is not performed, complications such as inflammation, infection and delayed healing may occur, and in severe cases, even a second surgery may be required. Current clinical postoperative wound monitoring mainly relies on manual observation and regular examination. Medical staff checks the degree of swelling of the wound by naked eye, judges the pressure condition by manual pressing, and evaluates the amount of exudation when regularly changing the dressing. This method has significant limitations: on the one hand, manual observation cannot capture the changes in microscopic parameters such as subcutaneous interstitial fluid pressure and exudation refractive index, which are often early signs of inflammation or infection. For example, when the subcutaneous pressure abnormally rises, the naked eye cannot detect it, but it may have caused local tissue ischemia. On the other hand, regular examination cannot achieve real-time continuous monitoring. The first to third days after surgery is a high-risk period for exudation and pressure abnormalities. If the interval between two examinations is too long, the key moment of abnormality may be missed, resulting in delayed treatment of complications.

[0003] Although existing wound monitoring devices can partially replace manual observation, they still have technical defects. Most electronic sensing devices have weak anti-electromagnetic interference ability. Patients after cardiovascular surgery usually have bedside devices such as electrocardiogram monitors and infusion pumps. Electromagnetic signals can seriously interfere with the accuracy of sensor data, resulting in large deviations in pressure and temperature values, which cannot be used as a basis for nursing. Some devices can only monitor a single parameter (such as only measuring surface temperature), and cannot achieve multi-parameter collaborative analysis. For example, only temperature rise is found, but exudation changes are ignored, making it difficult to accurately determine whether the wound has inflammation. In addition, existing devices lack individual adaptability and do not consider the impact of body mass index (BMI) and underlying diseases (such as diabetes) on wound healing. For example, diabetic patients have a lower inflammation threshold, but the device still uses a universal threshold, which can easily lead to misjudgment or missed judgment of inflammation. At the same time, there is no effective alternative when the sensor fails. Once the device fails, it needs to wait for maintenance personnel to handle it, and during this period, wound monitoring is completely interrupted.

[0004] The postoperative care plan and management also have deficiencies. The traditional care plan is mostly based on fixed clinical guidelines and is not dynamically adjusted in combination with real-time wound data of the patient. For example, the amount of exudate has exceeded the safe range, but the dressing is still replaced according to the regular cycle, causing the wound to be in a humid environment for a long time and increasing the healing burden. The management of multiple patients relies on manual recording and scheduling, and medical staff need to check the wound data of each patient, sort out the nursing records, which is low in efficiency, and it is difficult to quickly identify emergency situations (such as infection risk), resulting in confusion in the priority of triage. In addition, after the clinical guidelines are updated, the care plan cannot be timely synchronized to the actual operation, and some medical staff still follow the old plan. For example, although the new high-absorbent dressing has been widely used, some nursing processes still recommend traditional dressings, which affects the nursing effect. At the same time, the patient's participation is low, and postoperative care mostly relies on medical staff, so the patient cannot independently master the wound state and also cannot obtain accurate home care guidance, resulting in a decline in wound care quality after discharge. SUMMARY

[0005] The optical fiber sensing cardiovascular postoperative wound intelligent monitoring and nursing system provided by the present application solves the problems mentioned in the prior art.

[0006] In order to achieve the above purpose, the present application adopts the following technical scheme: an optical fiber sensing cardiovascular postoperative wound intelligent monitoring and nursing system, comprising the following modules: The optical fiber sensing array module: a customized plan for coronary artery bypass surgery, heart valve replacement surgery and congenital heart disease repair surgery; double helix distributed optical fiber + 3 point optical fiber for coronary artery bypass surgery, single helix distributed optical fiber + 2 point optical fiber for heart valve replacement surgery; the sensor is polyimide packaging; The multi-parameter data acquisition module: a double-channel optical fiber demodulator is configured, and after receiving the optical signal, differential amplification and 50Hz notch filter processing are performed; the patient's basic information and real-time vital signs are obtained synchronously through medical-grade Bluetooth; The wound state analysis module: an individual difference and multi-parameter fusion judgment model is constructed, the abnormal threshold is adjusted according to the patient's BMI and basic diseases, and features such as pressure gradient, temperature fluctuation value and exudate refractive index deviation are extracted; an improved support vector machine algorithm is used, combined with clinical case training, to judge the wound state as normal healing, mild inflammation, excessive exudation, pressure abnormality and infection risk; The intelligent nursing suggestion module: a three-dimensional nursing plan is generated based on the wound state, a disinfection reminder is pushed for normal healing, cold compress and antibacterial dressing are used for mild inflammation, high-absorbent dressing and drainage operation are recommended for excessive exudation, pressure adjustment is linked for pressure abnormality, and an emergency plan is triggered and medical contact is prompted for infection risk; The user interaction module: the medical staff end touch screen displays real-time data, judgment results and historical data, and supports threshold adjustment and plan customization; the patient end APP has voice reminders, AR guidance, pain scores and photo labeling functions, and supports offline data storage; Data storage and update module: adopt local and cloud and three-level storage architecture of blockchain; daily automatic incremental backup, monthly update model feature weight based on new case using gradient descent algorithm, update nursing plan library combined with the latest literature every year.

[0007] Further, it also includes a dynamic analysis module of liquid penetration components. Based on the refractive index data obtained by the multi-parameter data acquisition module, combined with the liquid penetration temperature and the collection time, the liquid penetration amount Q in unit time is calculated through the formula , wherein Q is the liquid penetration amount in the period from t1 to t2, c is the optical fiber sensing coefficient, t1 is the monitoring starting time, t2 is the monitoring ending time, n(t) is the liquid penetration refractive index at time t, n0 is the normal interstitial fluid refractive index, T(t) is the wound surface temperature at time t, and the integral term reflects the cumulative effect of refractive index deviation and temperature in the monitoring period; the module adds an indirect analysis function of liquid penetration components, which calculates the protein concentration in the liquid penetration through the correlation between the refractive index deviation and the temperature.

[0008] Further, it also includes a self-adaptive adjustment module of dressing pressure. The module receives the pressure gradient data of the wound state analysis module, combines the patient's body size parameters and real-time body position, and realizes dynamic adjustment of pressure through inflatable dressing belt; the module first calculates the average pressure of the wound area through the formula , wherein is the average pressure, L is the total length of the wound, n is the number of monitoring points of the distributed optical fiber, is the pressure value of the i-th monitoring point, and is the length of the wound segment corresponding to the i-th monitoring point; the target pressure is set to 12 kPa when lying, 15 kPa when sitting, and 18 kPa when standing. Further, it also includes a pain and inflammation correlation analysis module. The module performs multi-dimensional correlation modeling on the patient's pain data obtained by the user interaction module and the temperature fluctuation value, liquid penetration amount, and refractive index deviation data of the wound state analysis module; a pain and inflammation mapping matrix is constructed: sharp pain and temperature fluctuation value > 1℃ and liquid penetration refractive index deviation > 0.015 correspond to "acute inflammation"; dull pain and liquid penetration amount > 4mL / h and pressure gradient > 4kPa / cm correspond to "exudation compression pain"; burning pain and temperature > 38℃ and refractive index deviation > 0.02 correspond to "infective pain"; the module calculates the difference between the pain score and the corresponding threshold value in real time.

[0009] Further, it also includes a wound healing trend prediction module. Based on the historical monitoring data in the data storage and update module, the improved long short-term memory algorithm is used to predict the wound state in the next 7 days; the daily feature parameters are extracted to construct a prediction vector:

[0010] Further, it also includes a wound healing trend prediction module. Based on the historical monitoring data in the data storage and update module, the improved long short-term memory algorithm is used to predict the wound state in the next 7 days; the daily feature parameters are extracted to construct a prediction vector: ,​ 、 , Δn(t), is calculated by the formula The healing trend index H(t) is calculated, where H(t) is the healing trend index at time t, k is a trend coefficient, and the derivative term reflects the rate of change of the characteristic parameter over time. The healing trend is divided into four levels according to H(t): rapid healing, normal healing, slow healing, and delayed risk. When H(t) is ≤-0.3 for 3 consecutive days, the system automatically generates an expert consultation request and pushes it to the department director's end. The intelligent nursing recommendations are also increased with the measures of "strengthening nutritional support" and "increasing monitoring frequency". The module updates the LSTM model weights based on 500 and new healing cases every month to assist medical staff in early intervention of potential healing problems.

[0011] Further, it also includes a sensor failure self-diagnosis module that monitors the optical signal intensity and signal-to-noise ratio of the optical fiber sensing array module in real time, determines the failure type: signal intensity < -40 dBm is "optical fiber breakage", data transmission rate < 5 Hz is "connection loose", signal-to-noise ratio < 15 dB is "signal attenuation"; for distributed optical fibers, the fault point is located by an optical time domain reflectometer, and the relative position of the fault point and the wound is marked; if the fault point is located in the non-wound area, the redundant optical fiber channel is automatically enabled; for point-type optical fibers, if a single fiber fails, the monitoring data of the adjacent two fibers is used to calculate by linear interpolation, and if the middle point-type optical fiber fails, the pressure and temperature values of the upper and lower optical fibers are weighted and averaged; the fault information is synchronized to the medical staff end and the graded maintenance guidance is pushed: optical fiber breakage pushes temporary repair steps and contact engineer on-site time, connection loose pushes optical fiber connector re-plug and signal test method; the module also supports remote operation and maintenance.

[0012] Further, it also includes a multi-patient centralized management module that adds two levels of management interfaces, ward level and department level, on the medical staff end: the ward level interface displays the patient list by ward number, each column labeled with patient name, surgery type, postoperative days, and current wound status, and clicking on the patient name can view real-time monitoring data and nursing records; the department level interface calculates the incidence of wound complications by surgery type, generates trend charts, and marks the high-incidence complication types; the module introduces AI-assisted triage function, sets the triage priority based on the patient's wound state severity, underlying disease risk, and postoperative days, and automatically recommends the doctor's visit order; when the same type of abnormality occurs in 3 consecutive patients in the same ward, the system automatically analyzes the common factors and pushes the investigation suggestions; the module also supports exporting statistical reports to assist department quality management and clinical research.

[0013] Further, it also includes a patient activity state association module that collects patient activity data through an integrated three-axis acceleration sensor and gyroscope, and combines the wound pressure changes to subdivide the activity state into: static, light activity, moderate activity, and heavy activity; the module builds an activity intensity and wound pressure association model, and calculates the pressure peak safety threshold under different activity states: static state ≤ 15 kPa, light activity ≤ 20 kPa, moderate activity ≤ 25 kPa, and heavy activity ≤ 30 kPa; when the pressure peak under a certain activity state exceeds the threshold for three consecutive times, the intelligent nursing suggestion module pushes an activity restriction reminder and adjusts the pressure buffering coefficient of the bandage; the module also supports associated rehabilitation exercise plans, and recommends appropriate activity intensity according to the postoperative days, and real-time monitors the pressure changes when the patient performs the rehabilitation exercise.

[0014] Further, it also includes an intelligent disinfection opportunity judgment module that combines the temperature data, exudate volume data, pressure stability, patient activity state, and environmental data of the wound state analysis module to build a disinfection suitability evaluation system; sets the disinfection suitability index calculation rules: temperature ≤ 37℃, exudate volume ≤ 2mL / h, pressure stability, patient static, and environmental temperature and humidity meet the standards, total score 100; index 80-100 points for “suitable disinfection”, 50-79 points for “more suitable”, and < 50 points for “unsuitable”; when the index ≥ 80 and the distance from the last disinfection is more than 12 hours, the user interaction module pushes a disinfection reminder to the patient end and displays a disinfection operation guidance video through the AR function; if the index is 50-79 points and the distance from the last disinfection is more than 24 hours, a “more suitable disinfection” reminder is pushed; if the environmental humidity > 60%, the drying time after disinfection is automatically increased; the module also records the suitability index and effect of each disinfection to improve the disinfection effect and patient cooperation.

[0015] Further, it also includes a clinical guideline dynamic adaptation module that builds a guideline grabbing, parsing, arbitration, and updating full-process automation mechanism: every quarter, it grabs the latest postoperative wound care guidelines from authoritative channels such as UpToDate, the official website of the Chinese Nursing Association, and Journal of Cardiothoracic Surgery, extracts key recommendations through natural language processing technology, and structures them according to the classification structure of wound assessment, nursing measures, complication handling, and consumable selection; for conflicting contents of different guidelines, the module calls the local case efficacy data in the data storage and update module, calculates the efficiency of the two schemes, and adopts the scheme with higher efficiency; if the efficiency is close, it is arbitrated in combination with the patient population characteristics; the structured guideline content is automatically updated to the scheme generation logic of the intelligent nursing suggestion module, and a guideline update notification is pushed to the medical and nursing ends; the module also supports department customization, such as the cardiac surgery department that can manually add department internal nursing specifications, and the system nursing suggestions conform to international guidelines and are adapted to local clinical practice.

[0016] Compared with the existing technology, the beneficial effects of the present application are: The monitoring accuracy and continuity are greatly improved, and the early abnormality identification capability is significantly enhanced. The optical fiber sensing array module adopts an anti-electromagnetic interference design and can work stably in an environment with electrocardiogram monitoring equipment and the like, thereby avoiding data deviation caused by electromagnetic signals. The distributed and point type optical fibers are combined, which can not only monitor the wound surface temperature and circumferential pressure, but also capture microscopic parameters such as subcutaneous interstitial fluid pressure and exudate refractive index, thereby achieving full coverage of multidimensional data. The multi-parameter data acquisition module continuously acquires data at a high frequency, thereby avoiding abnormality omission caused by periodic inspection, for example, a slight increase in exudate volume can be found in real time, thereby providing a basis for early intervention of inflammation. The sensor failure self-diagnosis module can quickly locate the fault and enable a redundancy scheme, thereby ensuring uninterrupted monitoring. Compared with the problem of frequent interruption of existing equipment, the data integrity and reliability are significantly improved.

[0017] The nursing scheme is more personalized and timely, and the complication risk is greatly reduced. The wound state analysis module can adjust the abnormal threshold according to individual differences such as patient BMI and whether or not the patient has diabetes, for example, the inflammation temperature threshold of a diabetic patient is appropriately reduced, thereby avoiding misjudgment caused by a universal threshold. The intelligent nursing suggestion module generates a targeted scheme based on the real-time wound state, for example, when there is excessive exudate, high-absorbency dressings are recommended and the replacement frequency is increased, and when the pressure is abnormal, the bandaging parameters are adjusted, rather than following a fixed scheme. The clinical guideline dynamic adaptation module regularly updates the nursing logic, thereby ensuring that the scheme is always synchronized with the latest clinical evidence, and avoiding the impact on the effect caused by a lagging scheme. The healing trend prediction module can identify the risk of delayed healing in advance, thereby assisting medical staff to intervene as early as possible and reducing the probability of occurrence of serious complications such as infection and secondary surgery.

[0018] The medical management efficiency and patient participation are significantly improved, and the overall nursing quality is optimized. The multi-patient centralized management module supports two-level management at the ward level and the department level, and helps medical staff to quickly identify emergency situations and reasonably arrange the order of outpatient visits through color marking of the wound state and AI-assisted triage, thereby greatly reducing the workload of manual recording and scheduling. The AR nursing guidance and voice reminders enable patients to independently complete standard nursing operations, for example, the disinfection range and steps can be viewed through AR, thereby avoiding the impact on healing caused by improper operation. The patient end APP supports offline storage of data and feedback of nursing feelings, and medical staff can master the home nursing situation of patients in real time, thereby solving the problem of decreased nursing quality after discharge. Overall, the system optimizes the whole process of monitoring, nursing and management, thereby not only reducing the work burden of medical staff, but also improving the postoperative experience of patients, and providing more reliable protection for cardiovascular postoperative wound healing. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 A schematic block diagram of the optical fiber sensing cardiovascular postoperative wound intelligent monitoring and nursing system proposed in the present application; Figure 2 a comparative chart for wound parameters of different monitoring methods; Figure 3 a comparative chart for dressing pressure adaptation under different body positions; Figure 4 a comparative chart for wound healing trend index change. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0021] In the description of the present application, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation of the present application.

[0022] In addition, the terms "first", "second" are only for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited. In addition, the terms "mounting", "connecting", "connecting" should be broadly understood, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be the communication between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances, and the present application will be further described in detail with reference to the drawings.

[0023] Reference Figures 1 to 4 An optical fiber sensing postoperative wound intelligent monitoring and nursing system, comprising the following modules: Optical fiber sensor array module, customized sensor arrangement for different types of cardiovascular surgery (coronary artery bypass graft, heart valve replacement, congenital heart disease repair), double helix distributed optical fiber (pitch 2 cm) for coronary artery bypass graft surgery (mid-sternal incision, length 15-20 cm), monitoring circumferential pressure (0-50 kPa) and surface temperature (35-39℃), synchronously implanting 3 point optical fibers (located at the upper, middle and lower segments of the incision subcutaneously 0.8 cm apart) to monitor interstitial fluid pressure (0-30 kPa) and exudate refractive index (1.33-1.38); single helix distributed optical fiber (pitch 1.5 cm) for heart valve replacement surgery (right lateral chest incision, length 8-12 cm), point optical fiber implantation 2 (0.6 cm subcutaneously at both ends of the incision); the sensor is packaged with polyimide (thickness 0.05 mm), the diameter is ≤0.2 mm, the response time is ≤10 ms, and it has anti-electromagnetic interference ability (shielding effectiveness ≥40 dB); Multi-parameter data acquisition module, configure double-channel optical fiber demodulator (pressure resolution 0.1 kPa, temperature resolution 0.1℃, refractive index resolution 0.001), after receiving the optical signal of the sensor array, through the differential amplification circuit (gain 100 times, common mode rejection ratio ≥80 dB) and 50Hz notch filter processing, remove the electromagnetic interference of electrocardio monitoring equipment, collect pressure, temperature, refractive index data at 10Hz frequency; synchronously obtain patient's basic information (age, type of surgery, body mass index, whether combined with diabetes / hypertension) and real-time vital signs (heart rate 60-100 times / min, systolic pressure 90-140 mmHg, diastolic pressure 60-90 mmHg) through medical-grade Bluetooth (BLE5.0, transmission distance 10m), data acquisition delay ≤500ms; Wound state analysis module, build an "individual difference-multi-parameter fusion" judgment model, first adjust the abnormal threshold according to the patient's body mass index (BMI <18.5 / 18.5-23.9 / ≥24) and basic diseases (diabetes patients' inflammation temperature threshold is reduced to 37.2℃, exudate volume threshold is reduced to 3mL / h), then extract data features: pressure gradient (pressure difference / distance between adjacent monitoring points, normal ≤3kPa / cm), temperature fluctuation value (maximum value-minimum value within 5 minutes, normal ≤0.5℃), exudate refractive index deviation (actual value-1.33, normal ≤0.01); use improved support vector machine algorithm (introduce radial basis kernel function), combine 10000+ clinical cases to train the model, judge the wound state as normal healing, mild inflammation (temperature 37.2-38℃+exudate volume 3-5mL / h), excessive exudation (exudate volume >5mL / h), pressure abnormality (average pressure <8kPa or >22kPa), infection risk (temperature >38℃+refractive index deviation >0.02); The intelligent nursing suggestion module generates a "state-material-operation" three-dimensional nursing scheme based on the wound state: normal healing pushes daily disinfection reminders (2 times a day, recommends povidone-iodine disinfectant, concentration 0.5%), with a disinfection range diagram (5 cm area around the wound edge); mild inflammation suggests local cold compress (15 minutes each time, interval 4 hours, cold compress temperature 4-6℃, not directly contacting the wound), combined with antibacterial dressing (silver ion dressing, replacement cycle 48 hours); excessive exudation prompts the use of high-absorbency dressing (alginate dressing, absorption capacity ≥20 mL / g), increases the replacement frequency (once every 24 hours), and synchronously guides the exudate drainage operation (press the edge of the dressing to promote absorption); when the pressure is abnormal, the linkage bandaging pressure adjustment module is recommended, and the type of bandaging tape (elastic bandage / air-filled bandaging tape) and the tightness parameter (pressure control at 10-20 kPa) are recommended; the infection risk state automatically triggers the emergency nursing scheme, pushes the antibiotic external use suggestion (mupirocin ointment, 0.1mm thick), and prompts the contact with medical staff; The user interaction module uses a 12-inch touch screen (resolution 1920x1080) on the medical side to display real-time monitoring data (pressure / temperature / exudate volume curve, trend graph updated every hour), wound state determination results (color labeling: green normal, yellow warning, red emergency) and historical data (checkable for the past 14 days, support parameter filtering), support manual adjustment of threshold (need to input medical staff ID for verification) and custom nursing scheme (save as a template for subsequent call); The patient side mobile APP (supports Android 9.0+ / iOS 13.0+) has voice reminders (pushed 10 minutes before nursing time, adjustable volume), AR nursing guidance (scan the wound area to display virtual operation steps, accuracy ±1cm), pain score entry (visual scale, 0-10 points corresponding to expression icons), wound photo automatic labeling (identify wound edge, mark abnormal areas such as redness) functions, support offline data storage (save monitoring records within 24 hours when offline); Data storage and update module adopts "local + cloud + blockchain" three-level storage architecture: local server (capacity >= 2TB, RAID5 redundant backup, read / write speed >= 100MB / s) stores patient data in the past 3 months, cloud database (Ali cloud medical special cloud, conforms to the third level of network security protection standard) stores the whole cycle data, and the blockchain node (consortium chain, including hospital, equipment manufacturer and regulatory agency node) stores the key data (wound state judgment result, nursing scheme change record) to realize non-tamperable; Automatic backup (incremental backup mode, save storage space) at 2am every day, based on 500+ new clinical case data every month, the feature weight of support vector machine model is updated through gradient descent algorithm (such as the weight of refractive index of diabetic patient exudate is increased by 15%), and the nursing scheme library is updated every year (new type of dressing use suggestion, minimally invasive wound nursing process) based on the latest literature such as "cardiovascular surgery postoperative wound nursing guideline" and "infection control standard".

[0024] In the application, the exudate component dynamic analysis module is also included, which is based on the refractive index data obtained by the multi-parameter data acquisition module, combined with the exudate temperature and the collection time, and the exudate amount Q in unit time is calculated through the formula , wherein Q is the exudate amount (unit: mL) in the period from t1 to t2, c is the optical fiber sensing coefficient (0.02 mL・℃⁻ 1 ・min⁻ 1 , which is determined by sensor calibration experiment), t1 is the monitoring starting time (unit: min), t2 is the monitoring ending time (unit: min), n(t) is the exudate refractive index at t time, n0 is the normal interstitial fluid refractive index (1.33), T(t) is the wound surface temperature at t time (unit: ℃), and the integral term reflects the cumulative effect of refractive index deviation and temperature in the monitoring period; The module adds indirect analysis function of exudate component, which calculates the protein concentration in exudate (refractive index increases by 0.005 per liter, corresponding to the increase of 1g / dL of protein concentration) through the correlation between refractive index deviation and temperature, when the protein concentration is greater than 3g / dL and Q is greater than 5mL / h, it is determined that "high protein exudate" exists, which prompts that there may be tissue damage aggravation, and the intelligent nursing suggestion module automatically adjusts the scheme to "increase the dressing change frequency to once every 12 hours, and synchronously collect the exudate sample for inspection"; At the same time, the module supports setting the exudate amount threshold according to the type of operation (coronary artery bypass surgery after 1-3 days <=6mL / h, heart valve replacement surgery <=4mL / h), if Q exceeds the corresponding threshold for 2 hours, the red early warning is pushed to the medical staff end, which improves the accuracy and timeliness of exudate monitoring.

[0025] This invention also includes a bandage pressure adaptive adjustment module. This module receives pressure gradient data from the wound condition analysis module and combines it with the patient's body shape parameters (chest diameter, subcutaneous fat thickness) and real-time position (identifying supine / sitting / standing positions via an integrated tilt sensor, with an accuracy of ±5°). It then uses an inflatable bandage (with 8 independent inflatable chambers, each corresponding to one distributed fiber optic monitoring point) to achieve dynamic pressure adjustment. The module first uses a formula... Calculate the average pressure in the wound area ,in Where is the average pressure (in kPa), L is the total wound length (in cm), and n is the number of monitoring points in the distributed optical fiber. Let be the pressure value (in kPa) at the i-th monitoring point. The length of the wound segment corresponding to the i-th monitoring point (in cm); target position when supine. The pressure is set at 12 kPa (uniform force on the chest wall), 15 kPa in the sitting position (pressure is dispersed due to gravity), and 18 kPa in the standing position (maximum pressure loss); when When the pressure is 5 kPa below the target value, the corresponding inflation chamber is inflated at a rate of 8 kPa / min; when the pressure is 5 kPa above the target value, the chamber is deflated at a rate of 5 kPa / min. During the adjustment process, the pressure gradient is monitored in real time to ensure that the pressure difference between adjacent chambers is ≤3 kPa / cm, preventing excessive local pressure (>25 kPa) from causing tissue ischemia (inflation is automatically stopped when blood oxygen saturation is <95%), or excessive pressure (<8 kPa) from affecting wound closure. The module also supports adjusting the pressure buffer coefficient according to the patient's activity intensity (rest / mild activity / moderate activity) (the higher the activity intensity, the larger the buffer coefficient; for example, 10% redundancy is retained in the chamber inflation volume during moderate activity), realizing personalized, positional, and dynamic management of bandaging pressure.

[0026] In the present application, a pain-inflammation correlation analysis module is also included, which correlates the patient pain data (pain score 0-10, pain type: sharp pain / bloating pain / burning pain, pain duration) obtained by the user interaction module with the temperature fluctuation value, exudate volume, and refractive index deviation data of the wound state analysis module in multiple dimensions; a pain-inflammation mapping matrix is constructed: sharp pain + temperature fluctuation value > 1℃ + exudate refractive index deviation > 0.015, corresponding to "acute inflammation" (warning threshold pain score ≥ 4); bloating pain + exudate volume > 4mL / h + pressure gradient > 4kPa / cm, corresponding to "exudate compression pain" (warning threshold ≥ 3); burning pain + temperature > 38℃ + refractive index deviation > 0.02, corresponding to "infectious pain" (warning threshold ≥ 2); the module calculates the difference between the pain score and the corresponding threshold in real time, and if the difference is ≥ 2 and lasts for 1 hour, an important attention reminder is pushed to the medical care end (displaying the pain type and the associated inflammation indicators), and targeted pain relief measures are added in the intelligent nursing suggestion: acute inflammation recommends cold compress + non-steroidal anti-inflammatory drug gel (such as diclofenac diethylamine emulsion, with a coverage range covering the pain area), exudate compression pain suggests adjusting the pressure of the dressing + drainage operation, and infectious pain suggests antibiotic external use + timely medical treatment; the module also supports recording the pain relief effect (re-evaluation after 30 minutes of nursing), and if the score decreases by < 2, the nursing plan is automatically upgraded (such as increasing the frequency of drug use), improving the comprehensive evaluation capability of wound inflammation and patient comfort.

[0027] In the present application, a wound healing trend prediction module is also included, which predicts the wound state in the next 7 days based on the historical monitoring data (pressure mean value in the last 14 days, temperature standard deviation, exudate volume peak value, refractive index deviation mean value) in the data storage and update module using an improved long short-term memory (LSTM) algorithm; the daily feature parameters are extracted to construct a prediction vector: (t-day temperature mean, unit ℃), (t-day pressure standard deviation, unit kPa), (t-day exudate volume peak value, unit mL / h), Δn(t) (t-day refractive index deviation mean value), and the healing trend index H(t) is calculated by the formula where H(t) is the healing trend index at time t (a positive value indicates healing progress, a negative value indicates healing delay, and the larger the absolute value, the more obvious the trend), k is the trend coefficient (0.01d⁻ 1), the derivative term reflects the rate of change of the characteristic parameter with time (weight 3 is added to n(t) to highlight the infection-related indicators); According to H(t), the healing trend is divided into four levels: rapid healing (H(t)≥0.8), normal healing (0.3≤H(t)<0.8), slow healing (-0.3<H(t)<0.3), and delayed risk (H(t)≤-0.3); When H(t) is less than or equal to -0.3 for 3 consecutive days, the system automatically generates an expert consultation request (with the monitoring curve and trend analysis of the last 7 days) and pushes it to the department director terminal, and adds "strengthening nutritional support" (recommended high-protein diet, daily protein intake 1.5g / kg body weight) and "increasing monitoring frequency" (from 10Hz to 20Hz) measures in the intelligent nursing suggestion; The module updates the LSTM model weight based on 500+ new healing cases every month, and the prediction accuracy is improved to more than 85%, which helps medical staff to intervene in potential healing problems in advance.

[0028] In the present application, a sensor failure self-diagnosis module is also included, which monitors the optical signal intensity of the optical fiber sensing array module in real time (normal range -30dBm to -10dBm), data transmission rate (normal ≥10Hz) and signal-to-noise ratio (normal ≥20dB), and determines the failure type: signal intensity < -40dBm is "optical fiber breakage", data transmission rate < 5Hz is "connection loose", and signal-to-noise ratio < 15dB is "signal attenuation"; For distributed optical fiber, the fault point is located by optical time domain reflectometer (OTDR) (accuracy ±5mm), and the relative position of the fault point and the wound is marked (such as "2cm from the upper section of the incision"); If the fault point is located in the non-wound area (distance from the wound edge >3cm), automatically enable the redundant optical fiber channel (set 1 redundant connection point every 5cm, total 4-6), switching time ≤1s; For point-type optical fiber, if a single fiber fails, the monitoring data of the adjacent two fibers is calculated by linear interpolation (error ≤5%), such as middle point-type optical fiber failure, the pressure and temperature values of the upper and lower sections of the optical fiber are weighted and averaged (the closer the distance, the greater the weight); Fault information is synchronized to the medical staff terminal (displaying fault type, positioning result, and impact range), and a hierarchical maintenance guide is pushed: "temporary repair steps (use optical fiber fusion splicer, fusion loss ≤0.1dB) + contact engineer on-site time" for optical fiber breakage, and "replug optical fiber connector (force 5-10N) + test signal method" for connection loose; The module also supports remote operation and maintenance, and engineers can obtain fault data (optical signal waveform, failure time sequence) through the cloud to remotely guide medical staff to troubleshoot problems, and the fault repair rate is improved to 90%, reducing equipment downtime.

[0029] In the present application, a multi-patient centralized management module is also included, which adds two levels of management interfaces, ward level and department level, at the medical care end: the ward level interface displays the patient list by ward number (such as 301 ward, 302 ward), each column is labeled with patient name, surgery type, postoperative days, and current wound state (green normal, yellow warning, red emergency), and clicking on the patient name can view real-time monitoring data and nursing records; the department level interface calculates the incidence of wound complications (updated every week, such as 3.2% for coronary artery bypass surgery and 2.8% for heart valve replacement surgery) by surgery type (coronary artery bypass surgery, heart valve replacement surgery), generates trend charts (changes in the past 3 months), and marks high-incidence complication types (such as excessive exudation accounting for 60%); the module introduces AI-assisted triage function, sets triage priority based on patient wound state emergency level (red emergency > yellow warning > green normal), underlying disease risk (combined diabetes > hypertension > no underlying disease), and postoperative days (≤3 days > 4-7 days >>7 days), and automatically recommends the order of medical staff visits (such as a 301 ward patient with diabetes 2 days after surgery, with a wound infection risk, priority 1; a 302 ward patient without underlying disease 5 days after surgery, with mild inflammation, priority 3); when the same ward has 3 consecutive patients with the same abnormality (such as excessive exudation), the system automatically analyzes common factors (such as using the same batch of dressings, the same nursing group operation), and pushes out investigation suggestions (such as "check the absorption performance of the dressing, compare with other batch data"); the module also supports exporting statistical reports (Excel / PDF format), including patient numbers, complication categories, nursing measure execution rates, and other data, to assist department quality management and clinical research.

[0030] In the present application, a patient activity state correlation module is also included, which collects patient activity data through integrated three-axis acceleration sensor (sampling rate 50 Hz, range ±2g) and gyroscope (sampling rate 100 Hz, range ±2000° / s), and combines with wound pressure changes to subdivide the activity state into: static (acceleration <0.1g, angular velocity <10° / s, such as bed rest), light activity (0.1g≤acceleration <0.3g, angular velocity 10-50° / s, such as slow walking, sitting up), moderate activity (0.3g≤acceleration <0.5g, angular velocity 50-100° / s, such as going up and down stairs, slowly turning around), heavy activity (acceleration ≥0.5g, angular velocity ≥100° / s, such as rapid turning, bending to pick up objects); the module builds an "activity intensity-wound pressure" correlation model, and counts the pressure peak safety threshold under different activity states: static state ≤15kPa, light activity ≤20kPa, moderate activity ≤25kPa, heavy activity ≤30kPa; when the pressure peak under a certain activity state exceeds the threshold for 3 times in a row, the intelligent nursing suggestion module pushes an activity restriction reminder (such as "current rapid turning causes wound pressure 32kPa, it is suggested to slow down the turning speed to angular velocity <50° / s"), and adjusts the pressure buffering coefficient of the bandage (light activity coefficient 1.0, moderate 1.2, heavy 1.5, the larger the coefficient, the more redundant the chamber inflation volume); the module also supports associated rehabilitation exercise plans, and recommends appropriate activity intensity according to postoperative days (only static / light activity is allowed for 1-3 days after surgery, moderate activity can be added for 4-7 days, and heavy activity can be tried for >7 days); when the patient performs rehabilitation exercise, the pressure change is monitored in real time, and if the threshold is exceeded, the exercise is paused and a reminder is given, balancing rehabilitation progress and wound protection.

[0031] In the present application, the intelligent judgment module of disinfection opportunity is also included, which combines the temperature data, exudate volume data, pressure stability (pressure fluctuation within 5 minutes ≤2kPa is stable), patient activity state (rest / light activity), and environmental data (ward temperature 22-26℃, humidity 40-60%) of the wound state analysis module to construct a disinfection suitability evaluation system; the disinfection suitability index calculation rules are set: temperature ≤37℃ (20 points), exudate volume ≤2mL / h (25 points), pressure stability (20 points), patient rest (20 points), and environmental temperature and humidity meet the standards (15 points), with a total score of 100 points; index 80-100 points is "suitable for disinfection" (best disinfection effect, minimum wound irritation), 50-79 points is "more suitable" (disinfection time needs to be shortened, wound exposure time does not exceed 3 minutes), and <50 points is "not suitable" (easy to cause infection or delay wound healing); when the index ≥80 and the distance from the last disinfection is more than 12 hours, the user interaction module pushes the disinfection reminder (text+voice, with "suitable disinfection" label) to the patient end, and displays the disinfection operation guidance video (steps: clean hands→open disinfectant→wipe the 5cm area around the wound edge in a ring→dry→record time, with voice commentary for each step) through the AR function; if the index is 50-79 points and the distance from the last disinfection is more than 24 hours, the "more suitable disinfection" reminder is pushed, and it is suggested to complete it within 1 hour (avoiding the peak of patient activity); if the environmental humidity is >60%, the drying time after disinfection is automatically increased (from 5 minutes to 8 minutes); the module also records the suitability index and effect (change of exudate volume 2 hours after disinfection) of each disinfection, optimizes the subsequent reminder opportunity, and improves the disinfection effect and patient cooperation degree.

[0032] This invention also includes a clinical guideline dynamic adaptation module. This module constructs a fully automated mechanism for the entire process of "guideline capture-analysis-arbitration-update": Every quarter, it captures the latest guidelines (in both Chinese and English) for cardiovascular surgery wound care from authoritative sources such as UpToDate, the official website of the Chinese Nursing Association, and the *Journal of Cardiothoracic Surgery*. Key recommendations are extracted using Natural Language Processing (NLP) technology and structured according to categories such as "wound assessment-nursing measures-complication management-consumable selection" (e.g., "Postoperative sternal incision infection management: intravenous cephalosporins are the first choice, combined with local silver ion dressings, changed once daily"). For conflicting content between different guidelines (e.g., Guideline A recommends using alginate dressings for excessive exudate, while Guideline B recommends foam dressings), the module calls and stores the data. The system stores and updates local case efficacy data, calculates the effectiveness rates of two treatment options (alginate dressing 85%, foam dressing 82%), and prioritizes the option with the higher effectiveness rate. If the effectiveness rates are close (difference < 5%), arbitration is based on patient group characteristics (e.g., if 30% of local patients are diabetic, the option suitable for diabetic wounds is prioritized). The structured guideline content is automatically updated to the intelligent nursing suggestion module's solution generation logic (e.g., adding "minimally invasive small incision nursing process"), and guideline update notifications are pushed to medical staff (marked "new content" and "replaces old content"). The module also supports departmental customization adjustments; for example, cardiac surgery departments can manually add internal nursing guidelines (requiring department head approval). The system's nursing suggestions conform to international guidelines and are adapted to local clinical practice, enhancing professional authority and clinical applicability.

[0033] The following two examples further illustrate the specific implementation of this system: Example 1: Wound monitoring and nursing care for hospitalized patients undergoing coronary artery bypass grafting (with diabetes) I. Scenario and Basic Patient Information This embodiment was applied in the cardiac surgery ward of a tertiary hospital. The patient was a 65-year-old male who underwent coronary artery bypass grafting (CABG) with cardiopulmonary bypass for "triple-vessel coronary artery disease." On the second postoperative day, he had an 18cm midline sternal incision. He also had type 2 diabetes, a BMI of 25 (overweight), and no history of allergies. The ward was equipped with an electrocardiogram monitor (30cm from the wound), an infusion pump, and other equipment. The system needed to be resistant to electromagnetic interference. Postoperatively, the patient was primarily bedridden, occasionally needing to sit up to eat, and the system needed to be adapted to the impact of positional changes on the bandage pressure.

[0034] II. Detailed Implementation Process of Each Module 1. Deployment of fiber optic sensor array modules A customized approach was adopted to address the unique characteristics of the sternal incision used in coronary artery bypass grafting (CABG). Distributed optical fiber: polyimide encapsulated optical fiber is selected, and is fixed in double helix shape along the incision edge (pitch 2 cm), covering the full length of 18 cm of the incision. The two ends of the optical fiber are connected to a demodulator to monitor the circumferential pressure and surface temperature. Anti-electromagnetic interference test shows that under the working state of the 30 cm electrocardiograph, the signal deviation is ≤0.1 kPa / 0.1℃, and the shielding effectiveness is 42 dB. Point optical fiber: three point optical fibers are implanted into the subcutaneous tissue of the upper, middle and lower segments of the incision respectively at a depth of 0.8 cm, to monitor the interstitial fluid pressure (0-30 kPa) and the refractive index of exudate (1.33-1.38). During implantation, a minimally invasive needle is used, and postoperative X-ray examination shows no development interference, and the patient has no obvious foreign body sensation.

[0035] 2. Multi-parameter data acquisition and wound state analysis Data acquisition: a dual-channel optical fiber demodulator is configured, which is connected to a differential amplification circuit (gain 100 times, common mode rejection ratio 85 dB) and a 50 Hz notch filter to remove the interference of the electrocardiograph, and data is collected at a frequency of 10 Hz; at the same time, the patient's vital signs are obtained through medical-grade Bluetooth (BLE5.0): heart rate 75 times / min, systolic pressure 130 mmHg, diastolic pressure 80 mmHg, and basic information is input into the system.

[0036] Threshold adjustment and state determination: Individual threshold adjustment: due to the combination of diabetes, the wound state analysis module reduces the inflammation temperature threshold from the regular 37.5℃ to 37.2℃, and the exudate volume threshold from 5 mL / h to 3 mL / h; Feature extraction: data collected 2 hours after surgery: distributed optical fiber monitors surface temperature 37.3℃, circumferential pressure 15-18 kPa; point optical fiber monitors interstitial fluid pressure 12 kPa, exudate refractive index 1.338 (deviation 0.008); Model determination: input improved support vector machine model (training cases contain 2000 cases of diabetes coronary artery bypass patient data), output "mild inflammation" (temperature exceeds threshold 0.1℃ + exudate refractive index deviation 0.008, does not reach the risk standard of infection).

[0037] 3. Exudate analysis and dressing pressure adjustment Exudate volume calculation: the exudate component dynamic analysis module is started, the monitoring time period t1=0, t2=60 min is set, n(t) is collected in real time from 1.338 to 1.340, T(t) is stable at 37.3℃, and the formula is used to calculate: c=0.02 mL・℃⁻ 1 ・min⁻ 1 , n0=1.33, and the integral term =∫0 60 (1.338-1.33)×37.3dt+∫0 60(1.340-1.33) x 37.3 dt = (0.008 x 37.3 x 30) + (0.01 x 37.3 x 30) = 8.952 + 11.19 = 20.142, Q = 0.02 x 20.142 = 0.403 mL, which is lower than the 3 mL / h threshold for diabetic patients, and there is no need to upgrade the care plan.

[0038] Pressure self-adaptive adjustment: The pressure self-adaptive adjustment module receives data from 5 distributed monitoring points (the incision is divided into 5 segments, each segment li = 3.6 cm): Pi is 15, 17, 16, 18, and 17 kPa, respectively, according to the formula Calculation: = (15 x 3.6 + 17 x 3.6 + 16 x 3.6 + 18 x 3.6 + 17 x 3.6) / 18 = (15 + 17 + 16 + 18 + 17) x 3.6 / 18 = 83 x 0.2 = 16.6 kPa; the patient is currently in a supine position, and the target = 12 kPa, the module controls the synchronous deflation of the 8 chambers of the inflatable bandage (rate 5 kPa / min), and after 1.5 minutes drops to 12 kPa, and the pressure gradient stabilizes at 2 kPa / cm, without tissue ischemia (oxygen saturation maintained at 98%).

[0039] 4. Intelligent care and user interaction Care plan generation: The intelligent care suggestion module generates a plan for "mild inflammation": local cold compress (15 minutes each time, interval 4 hours, cold compress bag temperature 4-6°C, wrapped with sterile gauze to avoid direct contact with the wound), replace silver ion antibacterial dressing (48 hours once, model Kanglebao 3420), with operation diagram (cold compress range marked as 5 cm from the incision edge).

[0040] User interaction operation: Medical staff side: 12-inch touch screen displays temperature-pressure-liquid exudation curve (updated trend every hour), "mild inflammation" is marked with a yellow warning, and medical staff can view the data for the past 24 hours (temperature up to 37.3°C, cumulative liquid exudation 0.8 mL) after inputting their ID; Patient side: mobile APP pushes cold compress reminders (text + voice: "Please perform local cold compress in 10 minutes, lasting 15 minutes"), and the patient records the pain score (4 points, stinging) in the APP after completion, which is synchronized to the medical staff side.

[0041] 5. Data storage and healing trend prediction Data storage: monitor data (pressure, temperature, exudate volume) encrypted storage to the local server of the ward (2TB, RAID5 backup), key records (mild inflammation determination results, dressing change time) upload to Ali Cloud Medical Cloud, block chain node (hospital information department, equipment manufacturer, CDC supervision node) synchronization, tamper-proof.

[0042] Trend prediction: the wound healing trend prediction module extracts data 2 days after surgery: = 37.2℃, = 1.5kPa, = 0.4mL / h, Δn = 0.007, calculated according to the formula = 0.01d⁻ 1 , the derivative term reflects the parameter change rate, H(t) = 0.01x(0.1+0.5x0.2+2x0.1+3x0.002) = 0.01x0.406 = 0.00406, it is recommended to strengthen nutritional support.

[0043] III. Effect verification and table analysis Table 1: Comparison of traditional nursing and this system nursing for coronary artery bypass grafting (diabetes) patients Evaluation index Traditional nursing care Intelligent system of the present application Number of monitoring parameters 2 items (surface temperature, appearance of exudation) 6 items (including subcutaneous pressure, refractive index) Abnormality identification time 2 hours (regular ward rounds) Real time (response ≤10 ms) Adaptability of dressing pressure Fixed pressure (15 kPa, regardless of body position) Body position adaptation (12 kPa in supine position) Degree of individualization of nursing plan General plan (no diabetes adaptation) Diabetes-specific threshold + plan Incidence of postoperative complications in 7 days 8% (2 cases of aggravated inflammation / 25 cases) 2% (1 case of inflammation / 50 cases) In Table 1, traditional nursing only monitors surface temperature and exudate appearance, and checks every 2 hours, which is easy to miss early inflammation signals, and the pressure of the dressing is fixed. Diabetic patients have a high risk of inflammation due to unadjusted threshold; this system monitors 6 parameters in real time, and the diabetes-specific threshold allows for accurate identification of mild inflammation within 2 hours after surgery, and the body position adaptive pressure avoids tissue ischemia, and the incidence of complications is reduced by 6 percentage points. For example, the patient's exudate refractive index increased to 1.345 on the 3rd day after surgery, and the system identified and pushed "increase dressing change frequency" within 10ms, while traditional nursing needed to wait until the next day to find out, significantly improving the timeliness and safety of nursing.

[0044] Example 2: Wound monitoring and nursing for home rehabilitation patients after heart valve replacement surgery I. Scene and patient basic information This embodiment is applied to the patient's home environment. The patient is a 50-year-old female who underwent minimally invasive heart valve replacement surgery due to "aortic valve stenosis". She was discharged for home rehabilitation 10 days after surgery, had no small incision on the right chest wall, 8cm in length, no underlying diseases, BMI 22 (normal), and needed to perform mild activities (slow walking, simple housework) daily. There is no professional medical staff in the home environment, and the system needs to support remote monitoring and self-care guidance.

[0045] II. Detailed implementation process of each module 1. Fiber sensing array and portable acquisition deployment ​Optical fiber arrangement: For a small incision on the chest wall, a single helical distributed optical fiber (pitch 1.5 cm, covering an 8 cm incision) is used, and two point optical fibers are implanted subcutaneously 0.6 cm away from both ends of the incision (avoiding rib positions to reduce discomfort during activity), and the optical fibers are all waterproof packaged (to adapt to home bathing protection), and the patient can wear it by himself (with wearing instructions, steps: clean the wound → paste the optical fiber fixing paste → connect the portable demodulator).

[0046] Portable acquisition device: A handheld optical fiber demodulator (weight 300g, endurance 8 hours, USB charging supported) is configured, connected with the patient's mobile phone APP through Bluetooth 5.0, the acquisition frequency is 10Hz, the data storage capacity is 16GB (can store 24 hours of data offline), and the anti-interference test shows that in the home WiFi (2.4GHz) environment, the signal deviation is ≤0.2kPa / 0.2℃.

[0047] 2. Multi-parameter acquisition and activity state correlation Data acquisition: At 9 am on the 10th day after the operation, when the patient is walking slowly (light activity), the system collects data: surface temperature 36.8℃ (fluctuation value 0.3℃), circumferential pressure 11-13kPa (gradient 1.8kPa / cm), interstitial fluid pressure 9kPa, and exudate refractive index 1.335 (deviation 0.005); simultaneously, through the integrated three-axis acceleration sensor (sampling rate 50Hz), the activity state is identified: acceleration 0.35g, angular velocity 45° / s, and it is determined as "light activity".

[0048] Activity-pressure correlation: The patient activity state correlation module is started, the light activity pressure threshold is ≤20kPa, the real-time monitored pressure peak value is 13kPa (lower than the threshold value), and no intervention is needed; if the patient quickly turns around (acceleration 0.55g, angular velocity 110° / s, determined as "heavy activity"), the pressure peak value rises to 26kPa (exceeding the threshold value 25kPa), the APP immediately pushes a voice reminder: "the current activity intensity is too high, the wound pressure is 26kPa, please slow down", and at the same time, the buffering coefficient of the bandage is adjusted to 1.5 (10% redundancy of chamber inflation amount is reserved).

[0049] 3. Exudate analysis and disinfection opportunity determination Exudate volume calculation: The exudate component dynamic analysis module sets t1=0, t2=60min, n(t) is stable at 1.335, T(t)=36.8℃, and calculates according to the formula: integral term=∫0 60 (1.335-1.33)×36.8dt=0.005×36.8×60=11.04, Q=0.02×11.04≈0.221mL, which is lower than the threshold value of 2mL / h on the 10th day after the operation, and it is determined as normal.

[0050] Disinfection timing determination: The disinfection timing intelligent determination module collects data: temperature 36.8℃ (20 minutes), exudate volume 0.221 mL / h (25 minutes), pressure fluctuation 0.5 kPa (stable, 20 minutes), patient is still (20 minutes), environmental temperature 24℃, humidity 50% (meets the standard, 15 minutes), total index 95 points (suitable for disinfection); 13 hours since the last disinfection, APP pushes disinfection reminder (with AR guidance: scan the wound area to display the ring-shaped wiping range, voice explanation "wipe the wound edge 5 cm with 0.5% povidone-iodine, dry for 5 minutes"), the patient completes and uploads the disinfection record to the medical staff end.

[0051] 4. Remote interaction and fault self-diagnosis Remote monitoring: The medical staff end views the patient's real-time data through the cloud: temperature 36.8℃, exudate volume 0.221 mL, pressure 12 kPa, wound state "normal healing" (green label); the patient uploads a wound photo (APP automatically crops the wound area without redness), the medical staff labels "good healing, continue current care" in the system and synchronizes it to the patient's APP.

[0052] Fault simulation and processing: intentionally disconnecting one point optical fiber connection (simulating failure), the sensor failure self-diagnosis module identifies "signal strength -45 dBm (broken)" within 1 second, locates the fault point "1 cm away from the right end of the incision", as the fault point is located at the wound edge (distance from wound 0.5 cm < 3 cm), automatically enables adjacent distributed optical fiber data interpolation calculation (error 4%), APP pushes fault prompt: "point optical fiber is broken, backup data is enabled, no impact on monitoring, suggest replacing during recheck", no monitoring interruption.

[0053] 5. Data storage and guideline adaptation Data storage: home data is first stored in the mobile phone APP (offline for 24 hours), synchronized to the cloud after networking, and the "normal healing" determination result and disinfection record are stored in the blockchain; the local server (1 TB hard disk connected to the patient's home router) backs up data for nearly 3 months, with automatic incremental backup every morning at 2 am.

[0054] Guideline adaptation: the clinical guideline dynamic adaptation module is updated quarterly, the latest "Guidelines for Postoperative Care of Minimally Invasive Heart Valve Replacement" recommends "gradually increasing activity intensity to moderate 10-14 days after surgery", the system automatically updates the rehabilitation exercise plan: allows the patient to go up and down stairs 2 times a day (moderate activity, acceleration 0.4g), the monitoring pressure does not exceed 25 kPa, APP pushes the updated exercise suggestion.

[0055] III. Effect verification and table analysis Table 2: Comparison of traditional follow-up and system care for heart valve replacement patients at home Evaluation index Traditional telephone follow-up Intelligent system of the present application Monitoring continuity 2 days once (5 minutes each time) Real time (24 hours without interruption) Accuracy of nursing guidance Oral description (without operation demonstration) AR visual guidance (accuracy ±1 cm) Fault response capability Interrupted monitoring (wait for reexamination) 1-second switching of backup data Patient's self-care ability Low (dependent on telephone consultation) High (AR + voice guidance) Incidence of postoperative complications in 30 days 5% (1 case of infection / 20 cases) 0.5% (no infection / 200 cases) In Table 2, the conventional home care relies on a telephone follow-up once a day for 2 days, only oral inquiry of wound condition, no parameter monitoring, and thus when a fault occurs, it needs to wait for a return visit for processing, and self-care of the patient is prone to errors; the system can let the patient accurately complete disinfection, activity and other nursing through real-time monitoring and AR visual guidance, and when a fault occurs, it can switch to standby data in 1 second, there is no monitoring blank, and the incidence of postoperative complications is reduced to 0.5% in 30 days. For example, the patient mistakenly applies the dressing too tightly 15 days after the operation, the system real-time monitors that the pressure rises to 23 kPa (exceeding the threshold value of 20 kPa), and immediately reminds adjustment, and the conventional follow-up needs to wait until the next telephone call to find out, thereby avoiding delayed healing due to excessive pressure, and greatly improving the safety of postoperative rehabilitation.

[0056] Referring to Figure 2 The figure clearly shows the advantage of the system in the comprehensiveness of parameter monitoring - the conventional manual monitoring only relies on naked eyes and hand feeling, and only two subjective parameters of surface temperature and appearance of exudate can be obtained, and micro signals such as subcutaneous pressure and exudate composition cannot be captured, and early changes of inflammation are prone to be missed; although the ordinary electronic sensor increases the circumferential pressure and the amount of exudate, it still lacks core parameters such as subcutaneous interstitial fluid pressure and refractive index, and it is difficult to accurately determine the wound state; the system combines distributed and point optical fibers, realizes full coverage of six parameters, especially the parameters of subcutaneous pressure and exudate refractive index, and can early reflect tissue ischemia and abnormal exudate composition, thereby providing data support for accurate identification of mild inflammation and infection risk, for example, the protein concentration can be calculated through the exudate refractive index, and the trend of tissue damage can be found 24 hours in advance, intervention is 1-2 days earlier than the conventional monitoring, and the risk of complications is reduced.

[0057] Referring to Figure 3 The figure shows the rationality of the system in self-adaptive adjustment of bandaging pressure - the conventional bandaging adopts a fixed pressure of 15 kPa, without considering the influence of body position and activity on the pressure: when in a lying position, the pressure of 15 kPa is too high, and is prone to cause tissue ischemia below the sternum (blood oxygen saturation is reduced to 94%); when in a standing position, the pressure of 15 kPa is too low, and cannot guarantee wound closure, affecting healing; and when in activity, the pressure of 15 kPa is insufficient, and is prone to cause increased exudate due to traction. The system calculates the average pressure through a formula, and dynamically adjusts according to the body position and activity intensity: the pressure is reduced to 12 kPa in a lying position (to avoid ischemia), is increased to 18 kPa in a standing position (to guarantee closure), and is increased to 25 kPa in moderate activity (to resist traction), and the pressure gradient is always ≤3 kPa / cm, thereby ensuring that there is no excessive pressure locally. For example, when the patient eats in a sitting position, the system pressure is 15 kPa, which can guarantee the stability of the wound and does not affect the expansion of the thorax, and the conventional fixed pressure causes the patient to feel chest tightness when eating, thereby improving the postoperative comfort of the patient.

[0058] Referring to Figure 4The figure highlights the unique value of the system in healing trend prediction - traditional nursing can only judge healing by wound appearance 7-10 days after surgery, without early trend prediction, and if healing is delayed, the best intervention opportunity is often missed; the system calculates a healing trend index by formula, the index is -0.25 (slow healing) 1 day after surgery, indicating that nutrition support needs to be strengthened, the index is -0.05 (close to normal) 3 days after surgery, the index is 0.28 (close to normal healing) 5 days after surgery, and the healing progress is dynamically reflected. For example, the index is -0.18 2 days after surgery, the system pushes the suggestion of "increasing protein intake", and the index gradually rises after the patient executes, and the traditional nursing needs to wait until 7 days after surgery to observe that the wound swelling is reduced to confirm that the healing is good, which is 2-3 days later than the system to master the healing trend. For patients at high risk of healing delay (index ≤-0.3), the system can push a request for expert consultation 3 days in advance, which is 1-2 days earlier than traditional nursing intervention, and reduces the incidence of healing delay.

[0059] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can make equivalent replacement or change within the technical range disclosed by the present application according to the technical solution and inventive concept of the present application, which should be covered within the protection scope of the present application.

Claims

1. An optical fiber sensing postoperative cardiovascular wound intelligent monitoring and nursing system, characterized in that, The following modules are included: Optical fiber sensor array module: customized solutions for coronary artery bypass graft, heart valve replacement, and congenital heart disease repair; coronary artery bypass graft uses double helix distributed optical fiber + 3 point optical fiber, and heart valve replacement uses single helix distributed optical fiber + 2 point optical fiber; the sensors are all polyimide encapsulated; Multi-parameter data acquisition module: configure a dual-channel optical fiber demodulator, receive optical signals, and process them through differential amplification and 50Hz notch filtering; simultaneously obtain patient basic information and real-time vital signs through medical-grade Bluetooth; Wound state analysis module: build an individual difference and multi-parameter fusion judgment model, adjust the abnormal threshold according to the patient's BMI and underlying diseases, extract features such as pressure gradient, temperature fluctuation value, and exudate refractive index deviation; use an improved support vector machine algorithm, combined with clinical case training, to determine the wound state as normal healing, mild inflammation, excessive exudation, pressure abnormality, and infection risk; Intelligent nursing suggestion module: generate a three-dimensional nursing plan based on the wound state, remind of disinfection for normal healing, use cold compress and antibacterial dressing for mild inflammation, recommend high-absorbency dressing and drainage operation for excessive exudation, link to bandaging adjustment for pressure abnormalities, and trigger an emergency plan and prompt contact with medical staff for infection risk; User interaction module: the medical staff's touch screen displays real-time data, judgment results, and historical data, and supports threshold adjustment and custom solutions; The patient's APP has voice reminders, AR guidance, pain rating, and photo labeling functions, and supports offline data storage; Data storage and update module: uses a three-level storage architecture of local, cloud, and blockchain; daily automatic incremental backup, monthly model feature weight update based on new cases using gradient descent algorithm, and annual nursing plan library update combined with the latest literature.

2. The optical fiber sensing postoperative cardiovascular wound intelligent monitoring and nursing system according to claim 1, characterized in that, Also included is a liquid infiltration component dynamic analysis module, which, on the basis of the refractive index data acquired by the multi-parameter data acquisition module, in combination with the liquid infiltration temperature and the acquisition time, calculates the liquid infiltration amount Q in a unit time through a formula Q = c ∫(n(t) - n0) dt + c ∫T(t) dt, wherein Q is the liquid infiltration amount in the period from t1 to t2, c is the optical fiber sensing coefficient, t1 is the monitoring start time, t2 is the monitoring end time, n(t) is the liquid infiltration refractive index at time t, n0 is the normal interstitial fluid refractive index, T(t) is the wound surface temperature at time t, and the integral term reflects the cumulative effect of the refractive index deviation and the temperature in the monitoring period; the module adds an indirect analysis function of the liquid infiltration component, and calculates the protein concentration in the liquid infiltration through the correlation between the refractive index deviation and the temperature.

3. The optical fiber sensing postoperative cardiovascular wound intelligent monitoring and nursing system according to claim 1, characterized in that, Further comprising a pressure self-adaptive adjustment module, which receives the pressure gradient data of the wound state analysis module, combines the patient's body size parameters and real-time body position, and realizes dynamic pressure adjustment through inflatable bandaging belts; the module first calculates the average pressure of the wound area by the formula calculating the average pressure of the wound area wherein is the average pressure, L is the total length of the wound, n is the number of monitoring points of the distributed optical fiber, is the pressure value of the ith monitoring point, is the length of the wound segment corresponding to the ith monitoring point; the target pressure of the supine position is set to 12 kPa, the target pressure of the sitting position is set to 15 kPa, and the target pressure of the standing position is set to 18 kPa.

4. The optical fiber sensing postoperative cardiovascular wound intelligent monitoring and nursing system according to claim 1, characterized in that, The pain and inflammation correlation analysis module correlates the patient's pain data obtained by the user interaction module with the temperature fluctuation value, exudate volume, and refractive index deviation data obtained by the wound state analysis module; builds a pain and inflammation mapping matrix: sharp pain and temperature fluctuation value > 1℃ and exudate refractive index deviation > 0.015 correspond to "acute inflammation"; dull pain and exudate volume > 4mL / h and pressure gradient > 4kPa / cm correspond to "exudate compression pain"; burning pain and temperature > 38℃ and refractive index deviation > 0.02 correspond to "infectious pain"; the module calculates the difference between the pain score and the corresponding threshold in real time.

5. The optical fiber sensing postoperative cardiovascular wound intelligent monitoring and care system according to claim 1, wherein, A wound healing trend prediction module is also included, which predicts the wound state in the next 7 days based on historical monitoring data in the data storage and update module using an improved long short-term memory algorithm; daily feature parameters are extracted to construct a prediction vector: , , , Δn(t), the healing trend index H(t) is calculated by formula , where H(t) is the healing trend index at time t, k is the trend coefficient, and the derivative term reflects the rate of change of the feature parameter with time; according to H(t), the healing trend is divided into four levels: rapid healing, normal healing, slow healing, and delayed risk; when H(t) is ≤-0.3 for 3 consecutive days, the system automatically generates an expert consultation request and pushes it to the department director's end, and adds the measures of "strengthening nutritional support" and "increasing monitoring frequency" in the intelligent nursing suggestion; the module updates the LSTM model weight based on 500 and newly added healing cases every month to assist medical staff in intervening in potential healing problems in advance.

6. The optical fiber sensing postoperative cardiovascular wound intelligent monitoring and care system according to claim 1, wherein, It also includes a sensor failure self-diagnosis module that monitors the optical signal intensity and signal-to-noise ratio of the optical fiber sensing array module in real time, determines the failure type: signal intensity < -40 dBm is "optical fiber break", data transmission rate < 5 Hz is "connection loose", signal-to-noise ratio < 15 dB is "signal attenuation"; for distributed optical fibers, locate the fault point by optical time domain reflectometer and mark the relative position of the fault point and the wound; if the fault point is located in the non-wound area, automatically enable the redundant optical fiber channel; for point-type optical fibers, if a single fiber fails, use the monitoring data of the adjacent two fibers to calculate by linear interpolation, and if the middle point-type optical fiber fails, use the weighted average of the pressure and temperature values of the upper and lower optical fibers; the fault information is synchronized to the medical care end and the hierarchical maintenance guidance is pushed: the optical fiber break pushes the temporary repair steps and the contact engineer's on-site time, the connection loose pushes the optical fiber joint re-plugging and signal testing method; the module also supports remote operation and maintenance.

7. The optical fiber sensing postoperative cardiovascular wound intelligent monitoring and care system according to claim 1, wherein, It also includes a multi-patient centralized management module that adds two levels of management interfaces at the medical care end: ward level and department level. The ward level interface displays the patient list by room number, with each column labeled with the patient's name, surgery type, postoperative days, and current wound status. Clicking on the patient's name allows you to view real-time monitoring data and nursing records. The department level interface calculates the incidence of wound complications by surgery type, generates trend charts, and marks the high-incidence complication types. The module introduces AI-assisted triage function, sets the triage priority based on the patient's wound state emergency level, underlying disease risk, and postoperative days, and automatically recommends the doctor's visit order; when the same ward has 3 consecutive patients with the same abnormality, the system automatically analyzes the common factors and pushes the investigation suggestions; the module also supports exporting statistical reports to assist department quality management and clinical research.

8. The optical fiber sensing postoperative cardiovascular wound intelligent monitoring and care system according to claim 1, wherein, It also includes a patient activity state correlation module that collects patient activity data through integrated three-axis acceleration sensors and gyroscopes, and combines wound pressure changes to subdivide activity states into: static, light activity, moderate activity, and heavy activity. The module builds an activity intensity and wound pressure correlation model to calculate the pressure peak safety threshold for different activity states: static state ≤ 15 kPa, light activity ≤ 20 kPa, moderate activity ≤ 25 kPa, and heavy activity ≤ 30 kPa. When the pressure peak value in a certain activity state exceeds the threshold for 3 consecutive times, the intelligent nursing suggestion module pushes an activity restriction reminder and adjusts the pressure buffer coefficient of the bandage. The module also supports associating with rehabilitation exercise plans and recommending appropriate activity intensity based on postoperative days. When the patient performs rehabilitation exercises, the pressure changes are monitored in real time.

9. The optical fiber sensing postoperative cardiovascular wound intelligent monitoring and care system according to claim 1, wherein, It also includes a disinfection opportunity intelligent judgment module, which combines temperature data, exudate volume data, pressure stability, patient activity state, and environmental data of the wound state analysis module to construct a disinfection suitability evaluation system. The disinfection suitability index calculation rules are set as follows: temperature ≤ 37℃, exudate volume ≤ 2mL / h, pressure stability, patient stillness, and environmental temperature and humidity meet the standards, with a total score of 100 points. An index of 80-100 points is "suitable for disinfection", 50-79 points is "more suitable", and < 50 points is "not suitable". When the index is ≥ 80 and the time since the last disinfection is more than 12 hours, the user interaction module pushes a disinfection reminder to the patient end and displays a disinfection operation guidance video through the AR function. If the index is 50-79 points and the time since the last disinfection is more than 24 hours, a "more suitable disinfection" reminder is pushed. If the environmental humidity is > 60%, the drying time after disinfection is automatically increased. The module also records the suitability index and effect of each disinfection to improve the disinfection effect and patient cooperation.

10. The optical fiber sensing postoperative cardiovascular wound intelligent monitoring and care system according to claim 1, wherein, It also includes a clinical guideline dynamic adaptation module that builds a guideline grabbing, analysis, arbitration, and updating full-process automation mechanism. The latest postoperative wound care guidelines are grabbed from authoritative channels such as UpToDate, the Chinese Nursing Society website, and the Journal of Cardiothoracic Surgery every quarter, and key recommendations are extracted through natural language processing technology. The guidelines are structured according to wound assessment, nursing measures, complication handling, and consumable selection. For conflicting content in different guidelines, the module calls local case efficacy data in the data storage and update module to calculate the efficiency of the two schemes, and the scheme with higher efficiency is preferred. If the efficiency is close, the patient population characteristics are combined for arbitration. The structured guideline content is automatically updated to the scheme generation logic of the intelligent nursing suggestion module and a guideline update notification is pushed to the medical and nursing ends. The module also supports department customization, such as manually adding department internal nursing specifications in the cardiology department. The system nursing suggestions comply with international guidelines and are adapted to local clinical practice.

Images