Intelligent wound monitoring dressing system based on multi-parameter biosensing

Abstract

The invention discloses an intelligent wound monitoring dressing system based on multi-parameter biosensing, and belongs to the technical field of medical instruments. The method comprises the steps of determining a wound supervision range of an intelligent medical dressing, and obtaining a dynamic parameter Di of a wound in a supervision period based on the wound supervision range in combination with an intelligent wound supervision dressing system constructed by a multi-modal data processing platform; full-process intelligent optimization from data collection to clinical intervention is achieved, real-time monitoring ensures that wound microenvironment changes are captured in time, multi-parameter collaborative diagnosis provides a scientific basis for risk assessment, normal logs can be automatically generated, invalid intervention is reduced, and the risk assessment efficiency is improved. According to the invention, the system is simple in structure and convenient to operate, and can trigger targeted nursing actions through abnormal signals and multi-stage alarms to construct a monitoring-analysis-intervention closed-loop mode, so that the workload of medical staff is reduced, and wound recovery is effectively promoted by accurately regulating and controlling the wound healing environment.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, specifically to an intelligent wound monitoring dressing system based on multi-parameter biosensing. Background Technology

[0002] Traditional medical dressings only have the basic functions of physically isolating the wound and absorbing exudate. They cannot detect changes in key biochemical and physical indicators of the wound microenvironment in real time (such as exudate volume, pH value, temperature, etc.). During the nursing process, medical staff need to frequently remove the dressing for manual inspection. This not only interferes with the stable healing environment of the wound, but also significantly increases the risk of secondary infection.

[0003] Existing technologies have many limitations: traditional medical dressings only have physical isolation and exudate absorption functions, and cannot sense changes in the biochemical and physical indicators of the wound microenvironment in real time. Nursing staff need to frequently remove the dressing for inspection, which not only interferes with the wound healing environment but also significantly increases the risk of infection. At the same time, existing electronic dressings mostly focus on monitoring a single parameter and lack the ability to analyze multiple indicators in a coordinated manner. The rigid circuit design results in poor dressing fit, affecting patient comfort. The power supply scheme has problems such as large size or insufficient practicality, and the early warning mechanism mostly relies on single threshold judgment, which is prone to false alarms or missed alarms, making it difficult to meet the dynamic monitoring needs of complex wounds.

[0004] To address the aforementioned technical shortcomings, a solution is proposed. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent wound monitoring dressing system based on multi-parameter biosensing to solve the problems mentioned above.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent wound monitoring dressing system based on multi-parameter biosensing, comprising determining the wound monitoring range of the intelligent medical dressing, and constructing an intelligent wound monitoring dressing system in conjunction with a multimodal data processing platform, as detailed below:

[0007] Multi-level joint data acquisition module: Based on the wound monitoring scope, it obtains the dynamic parameter Di of the wound within the monitoring period, as well as a historical database constructed from previously collected recorded data and pre-stored thresholds;

[0008] The shallow risk data assessment module acquires integrated pH and temperature / humidity values. It retrieves historical data to compare and analyze the integrated pH values, generating a pH anomaly signal RYz and a pH gradient ΔpH / Δt. The integrated temperature / humidity values ​​are analyzed in conjunction with the wound shape and data collection location to obtain local temperature and humidity differences between the wound center and edges. These differences are then compared with historical data to generate a temperature / humidity anomaly signal WSYz and a temperature change ΔWt.

[0009] The deep risk trend judgment module obtains the integrated exudate value, performs sensitive data extraction on the integrated exudate value to obtain the amount of protease-active exudate, generates an exudate risk signal by comparing with historical data, constructs a curve model by combining with the historical database, and obtains the rate of change of exudate volume ΔV.

[0010] Re-inspection decision control module: Receives multiple sets of signals, extracts pH gradient ΔpH / Δt, temperature change ΔWt, and exudate volume change rate ΔV from the multiple sets of signals, and generates multi-level alarm signals by combining them with a three-dimensional decision model, triggering the display of pre-stored dressing change reminders or correcting the accuracy of data collection.

[0011] Furthermore, the data acquisition process of the multi-level joint data acquisition module for the wound monitoring range is as follows:

[0012] The system obtains the operation instructions of the intelligent wound monitoring dressing system, constructs a monitoring cycle based on the time of instruction execution to the current time, and divides the monitoring cycle into several sets of cycle nodes with equal timestamp intervals. The integrated values ​​of pH, temperature and humidity, and exudate collected from the wound within the cycle nodes are marked as the original set of dynamic parameters. The original set of dynamic parameters is denoised to remove jump values ​​caused by poor sensor contact, and the different parameter data are unified with timestamps. Combined with the collection coordinates and time, spatiotemporal marking is performed to obtain the dynamic parameter Di.

[0013] Based on the timeline of the regulatory cycle, the obtained dynamic parameters Di are stored in batches in the multimodal data processing platform to build a temporary cache library. The multimodal data processing platform internally stores historical databases constructed from previously collected data records and pre-stored thresholds.

[0014] Furthermore, the analysis process of the shallow risk data assessment module for the integrated pH value in the dynamic parameter Di is as follows:

[0015] The potential data of the iridium oxide thin film three-electrode system was measured in the range of -0.2V to +0.6V, and was generated every N minutes. The potential data was marked as pH detection values. The pH detection values ​​of three consecutive times were integrated and the average value was marked as the integrated pH value. Historical pH data of the wound in the previous 24 hours were retrieved from the historical database, as well as preset historical normal pH data and preset historical abnormal pH data of the same type of wound in the same recovery period. The baseline range was established by the timestamp trend of the time period nodes. The historical pH data, preset historical normal pH data, and preset historical abnormal pH data were sequentially input into the baseline range to construct a normal range with the upper and lower limits of the preset historical normal pH data, a trend comparison range with the upper and lower limits of the historical pH data, and an abnormal range between the upper and lower limits of the preset historical abnormal pH data and the upper and lower limits of the historical normal pH data.

[0016] Furthermore, when the pH detection value is brought into the baseline range, if the pH detection value is completely within the normal range and within the trend comparison range along the time stamp of the periodic node, a normal pH log Ri is generated; if the pH detection value is completely within the normal range along the time stamp of the periodic node but has fluctuations exceeding the upper and lower limits of the trend comparison range, a pH abnormal log RYi is generated; if the pH detection value enters the abnormal range along the time stamp of the periodic node, a pH abnormal signal RYz is generated, and the pH detection value along the time stamp of the periodic node is marked as the pH gradient ΔpH / Δt.

[0017] Furthermore, the analysis process of the shallow risk data assessment module for the integrated temperature and humidity values ​​in the dynamic parameter Di is as follows:

[0018] The temperature / humidity values ​​updated every N seconds by the NTC thermistor array are obtained and marked as integrated temperature and humidity values. The wound shape is divided into a central region and an edge region. Based on this, the corresponding values ​​are obtained from the region where the NTC thermistor array is located and averaged. The difference between the average values ​​of the central region and the edge region is extracted and marked as local difference values.

[0019] Furthermore, the system retrieves pre-stored standard temperature and humidity thresholds for the same type of wound within the same time period from the historical database, as well as integrated temperature and humidity values ​​recorded N hours ago for the current wound. The system extracts the local differences between the pre-stored temperature and humidity thresholds and the integrated temperature and humidity values, marking these differences as the difference coefficients. The system then compares the local differences in the integrated temperature and humidity values ​​with the pre-stored temperature and humidity thresholds: if the local difference is close to the pre-stored temperature and humidity threshold and the difference is less than the difference coefficient, the current integrated temperature and humidity value of the wound is considered to be within the standard range, and a temperature and humidity log (WSi) is generated; if the local difference is close to or far from the pre-stored temperature and humidity threshold and the difference is greater than the difference coefficient, the current integrated temperature and humidity value of the wound is considered to be outside the standard range, and a temperature and humidity anomaly signal (WSYz) is generated. The process of comparing the local differences in the integrated temperature and humidity values ​​with the pre-stored temperature and humidity thresholds is marked as temperature change (ΔWt).

[0020] Furthermore, the analysis process of the deep risk trend judgment module on the integrated value of permeation in the dynamic parameter Di is as follows:

[0021] The real-time monitoring capacitance value of the serpentine conductive fiber network, recorded every N minutes, is obtained and marked as percolation integrated data. Type-sensitive data is extracted from the percolation integrated data to obtain the percolation amount of protease activity and other comprehensive percolation amounts. Historical percolation data for the past N days and pre-stored percolation difference thresholds are retrieved from the historical database. The old enzyme activity value and comprehensive value of each day from the far date to the recent date are extracted from the historical percolation data. A floating curve is constructed by continuously comparing the old enzyme activity value and comprehensive value of the extracted days. The curve model is constructed with the timestamp of the period node as the X-axis and the unit amount of percolation integrated data as the Y-axis.

[0022] Furthermore, the amount of protease-active exudate and other comprehensive exudate are compared with the old enzyme activity value and the comprehensive value: if the amount of protease-active exudate is close to the old enzyme activity value, the amount of other comprehensive exudate is close to the comprehensive value, and the difference between the two groups is less than the exudate difference threshold, then the wound recovery status is judged to be poor, and an exudate risk signal is generated; if the amount of protease-active exudate is far from the old enzyme activity value, the amount of other comprehensive exudate is far from the comprehensive value, and the difference between the two groups is greater than the exudate difference threshold, then the wound recovery status is judged to be good, and an exudate log record SYi is generated.

[0023] The protease activity exudate volume and other comprehensive exudate volumes are plotted in the curve model by connecting the second points according to the periodic node arrangement. When the curves of protease activity exudate volume, other comprehensive exudate volumes and old enzyme activity values ​​and comprehensive values ​​continue to supplement the end of the curve, an abnormal exudate state log is generated. When the curves of protease activity exudate volume, other comprehensive exudate volumes and old enzyme activity values ​​and comprehensive values ​​continue to supplement the end of the curve, a normal exudate state log is generated. The floating curve change marker is constructed by comparing the protease activity exudate volume and other comprehensive exudate volume in the integrated exudate data with the data in the historical database as the exudate volume change rate ΔV.

[0024] Furthermore, the re-inspection decision control module processes the received signals as follows:

[0025] After obtaining the pH anomaly signal RYz, the temperature and humidity anomaly signal WSYz, and the seepage risk signal, the three-dimensional decision model pre-stored in the historical database is retrieved. The pH gradient ΔpH / Δt, temperature change ΔWt, and seepage rate change ΔV from the original parameters of the three sets of signals are extracted and input into the three-dimensional decision model. Based on the differences that exist during the comparison of the three-dimensional decision model, multi-level alarm signals are triggered.

[0026] The beneficial effects of this invention are:

[0027] 1. The present invention precisely captures the dynamic parameters of the wound microenvironment through a multi-level combined data acquisition module. Relying on an iridium oxide thin film three-electrode system, an NTC thermistor array, and a serpentine conductive fiber network, it can obtain the dynamic parameter Di in real time, and ensure the data accuracy through noise reduction processing and spatio-temporal marking. Compared with the traditional dressing that needs to be frequently uncovered for inspection, it can continuously monitor without interfering with the wound healing environment, significantly reducing the risk of secondary infection. The standardized acquisition and storage of the dynamic parameter Di provide a high-quality data basis for subsequent risk assessment, enabling real-time and precise monitoring, reducing the infection risk, and ensuring data reliability.

[0028] 2. The present invention adopts an analysis logic that combines shallow risk assessment and deep trend judgment. Through multi-dimensional linkage analysis of the pH gradient △pH / △t, temperature change △Wt, and exudate volume change rate △V, it constructs a three-dimensional decision model to achieve multi-level alarms. By comparing with thresholds, it can accurately distinguish the normal healing and infection risk states, avoiding false alarms and missed alarms caused by single-threshold judgment. Medical staff can quickly respond based on the clear alarm levels. The first-level alarm strengthens monitoring, and the third-level alarm initiates emergency dressing change, greatly improving the clinical decision-making efficiency, achieving multi-parameter collaborative diagnosis, and enhancing the early warning accuracy and clinical decision-making efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0030] Figure 1 is the system flow block diagram of the present invention;

[0031] Figure 2 is the schematic diagram of the intelligent silicone dynamic antibacterial dressing of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0033] Example 1: Please refer to Figure 1 - Figure 2As shown, this embodiment is an intelligent wound monitoring dressing system based on multi-parameter biosensing, including determining the wound monitoring range of the intelligent medical dressing, and constructing the intelligent wound monitoring dressing system in conjunction with a multi-modal data processing platform, as detailed below:

[0034] Multi-level joint data acquisition module: Based on the wound monitoring range, it obtains the dynamic parameter Di of the wound within the monitoring period, as well as a historical database constructed from previously collected recorded data and pre-stored thresholds; the data acquisition process of the multi-level joint data acquisition module for the wound monitoring range is as follows:

[0035] After human-computer interaction, the intelligent wound monitoring dressing system generates an operation command in the multimodal data processing platform and sends it to the multi-level joint data acquisition module. The multi-level joint data acquisition module constructs a monitoring cycle based on the time of the command execution to the current time, and performs time stamp equal division planning on the monitoring cycle to obtain several sets of cycle nodes with equal timestamp intervals. The integrated values ​​of pH, temperature and humidity, and exudate collected in the cycle nodes are marked as the original set of dynamic parameters.

[0036] The original set of dynamic parameters is denoised to remove jump values ​​caused by poor sensor contact. Different parameter data are unified with timestamps and spatiotemporally marked by combining the acquisition coordinates and time to obtain the dynamic parameter Di.

[0037] Based on the timeline of the regulatory cycle, the obtained dynamic parameters Di are stored in batches in the multimodal data processing platform to build a temporary cache library. The multimodal data processing platform internally stores historical databases constructed from previously collected data records and pre-stored thresholds.

[0038] The shallow risk data assessment module obtains integrated pH and integrated temperature and humidity values, retrieves historical data to compare and analyze the integrated pH values, and generates a pH anomaly signal RYz and a pH gradient ΔpH / Δt. The analysis process of the integrated pH value in the dynamic parameter Di by the shallow risk data assessment module is as follows:

[0039] pH integrated numerical data source: Detection of the iridium oxide thin-film three-electrode system within the range of -0.2V to +0.6V, with potential data generated every N minutes, where N represents a positive natural number, which can be 5 or 10. The iridium oxide thin-film three-electrode system is located in... Figure 2In the middle dressing contact layer, a three-electrode system based on an iridium oxide thin film electrode and a reference electrode is used to detect the redox potential in the range of -0.2V to +0.6V by cyclic voltammetry, covering a pH range of 5.0-8.5, while the traditional colorimetric method is only 6.0-7.5. It integrates a pH sensor and antibacterial hydrogel, and marks the potential data as the pH detection value. It integrates three consecutive pH detection values, calculates the average value and marks it as the integrated pH value. It retrieves historical pH data from the database of the wound in the previous 24 hours, as well as preset historical normal pH data and preset historical abnormal pH data of the same type of wound in the same recovery period. The preset historical normal pH data can take values ​​in the range of 5.5-6.5, and the preset historical abnormal pH data can take values ​​in the range of >7.3 or <5.0. The baseline range is established by the timestamp trend of the time period nodes.

[0040] Substitute historical pH data, preset historical normal pH data, and preset historical abnormal pH data into the baseline range in sequence to construct a normal range with the upper and lower limits of the preset historical normal pH data, a trend comparison range with the upper and lower limits of the historical pH data, the trend comparison range can be taken as ±0.3 of the pH fluctuation range in the previous 24 hours of the wound, and an abnormal range between the upper and lower limits of the preset historical abnormal pH data and the upper and lower limits of the historical normal pH data.

[0041] When the pH test value is brought into the baseline range, if the pH test value is completely within the normal range and within the trend comparison range along the time stamp of the cycle node, a pH normal log Ri is generated. The pH normal log Ri is sent to the multimodal data processing platform and the re-examination decision control module through the shallow risk data assessment module. The multimodal data processing platform will store it in the historical database and mark it as normal fluctuation. The intelligent wound monitoring dressing system and the nursing station terminal receive a summary report every 2 hours without additional intervention. If the pH normal log Ri is generated for 12 consecutive hours, the intelligent wound monitoring dressing system will automatically extend the data upload interval to 30 minutes to reduce energy consumption.

[0042] If the pH test value is entirely within the normal range along the time stamp of the cycle node but fluctuates beyond the upper and lower limits of the trend comparison range, a pH anomaly log RYi is generated. The pH anomaly log RYi is sent to the multimodal data processing platform and the re-examination decision control module through the shallow risk data assessment module. Immediately, a prompt box pops up on the intelligent wound monitoring dressing system and nursing station terminal, displaying the text message "pH value trend fluctuation abnormal (not exceeding the safe range)" and attaching the integrated pH value curves of the most recent 3 times. Medical staff need to check the wound baseline comparison map through the APP within 2 hours to confirm whether the fluctuation is caused by the patient's diet or medication. If the fluctuation continues, the pH test frequency is increased to once every 5 minutes.

[0043] If the pH test value enters the abnormal range along the time stamp of the cycle node, a pH abnormal signal RYz is generated. The pH abnormal signal RYz is sent to the multimodal data processing platform and the re-examination decision control module through the shallow risk data assessment module. It immediately sends an audible and visual alarm to the nursing station through the BLE module. The intelligent wound monitoring dressing system and the nursing station terminal display text information or audio in the style of "pH value enters the abnormal range, infection risk increases".

[0044] Medical staff must arrive within 30 minutes to examine the wound, remove the dressing to check for redness and swelling, and simultaneously use a portable pH meter to verify the result. If an abnormality is confirmed, local antibacterial treatment will be initiated, such as applying a silver ion hydrogel. The pH measurement value will be marked as the pH gradient ΔpH / Δt along the time stamp of the periodic node.

[0045] The shallow risk data assessment module performs a linked analysis of the integrated temperature and humidity values ​​based on the wound shape and the collection location to obtain the local temperature and humidity differences between the wound center and the edge. It then compares these values ​​with historical data to generate anomaly signals WSYz and temperature change ΔWt. The analysis process of the integrated temperature and humidity values ​​in the dynamic parameter Di by the shallow risk data assessment module is as follows:

[0046] The integrated temperature and humidity data source is: Temperature / humidity values ​​updated every N seconds by an NTC thermistor array are obtained and marked as integrated temperature and humidity values, for example... Figure 2 As shown, its dressing middle layer integrates an NTC thermistor array. The temperature value of the NTC thermistor array is updated every 5 seconds, and the humidity value comes from the humidity sensing unit of the same array.

[0047] The average values ​​were taken from the central and peripheral regions of the wound. The central region was within 3 cm in diameter, and the peripheral region was 3-5 cm away from the central region. The integrated temperature and humidity values ​​were obtained. The Kalman filter algorithm was used to eliminate the interference of ambient temperature on pH measurement. The central and peripheral regions were divided according to the shape of the wound. The corresponding values ​​were obtained from the coordinates of the NTC thermistor array and averaged. The difference between the average values ​​of the central and peripheral regions was extracted and marked as the local difference value.

[0048] The system retrieves pre-stored standard temperature and humidity thresholds for similar wounds within the same time period from the historical database, as well as integrated old values ​​of temperature and humidity recorded N hours ago for the current wound. The local differences between the pre-stored temperature and humidity thresholds and the integrated old values ​​are extracted and marked as difference coefficients. These local differences are then compared with the pre-stored temperature and humidity thresholds. It should be noted that the pre-stored temperature and humidity thresholds are: center-to-edge temperature difference ≤ 1℃, humidity difference ≤ 20%; the difference coefficient is 1.2 times the temperature and humidity difference of the current wound 8 hours ago. For example, if the temperature difference 8 hours ago was 0.5℃, the difference coefficient would be 0.6℃.

[0049] When the local difference value is close to the pre-stored temperature and humidity threshold, and the difference between the two is less than the difference coefficient, it is determined that the current integrated temperature and humidity value of the wound is within the standard range, and a temperature and humidity log (WSi) is generated. The temperature and humidity log (WSi) is sent to the multimodal data processing platform and the re-examination decision control module through the shallow risk data assessment module. The intelligent wound monitoring dressing system automatically archives the temperature and humidity log (WSi) received by the multimodal data processing platform and generates a temperature and humidity stability report, which is summarized weekly in the department's quality management database as a basis for wound care compliance. If there are three consecutive days of temperature and humidity log (WSi), it indicates that the dressing change cycle can be extended from the usual 5 days to 7 days.

[0050] When the local difference value is close to or far from the pre-stored temperature and humidity threshold, and the difference between the two is greater than the difference coefficient, it is determined that the current integrated temperature and humidity value of the wound is outside the standard range, and a temperature and humidity abnormality signal WSYz is generated. The temperature and humidity abnormality signal WSYz is sent to the multimodal data processing platform and the re-examination decision control module through the shallow risk data assessment module. The intelligent wound monitoring dressing system and nursing station terminal display the text information in the style of "center-edge temperature and humidity difference exceeds the standard", with an attached regional temperature distribution heat map.

[0051] Medical staff need to check if the dressing is loose. If the edge is raised by more than 2mm, it should be reapplied. If it is well attached, check if the ambient temperature is too high. If the ward is not air-conditioned in summer, adjust the ward temperature and humidity to 24-26℃ and 50-60% if necessary.

[0052] The process of comparing the local differences in the integrated temperature and humidity values ​​with the pre-stored temperature and humidity thresholds is denoted as temperature change ΔWt.

[0053] Example 2:

[0054] The deep risk trend judgment module obtains the integrated exudate value, performs sensitive data extraction on the integrated exudate value to obtain the amount of protease-active exudate, generates an exudate risk signal by comparing it with historical data, constructs a curve model based on the historical database, and obtains the rate of change of exudate volume ΔV. The analysis process of the integrated exudate value in the dynamic parameter Di by the deep risk trend judgment module is as follows:

[0055] Permeation integration numerical data acquisition: Record the real-time monitoring capacitance value of the serpentine conductive fiber network every N minutes, where N can be 5. Figure 2 As shown, a conductive fiber network is arranged in a serpentine pattern on the side of the middle layer of the dressing. The material is a silver nanowire composite. The amount of exudate is calculated by measuring the change in capacitance between adjacent electrodes, with a sensitivity of 0.1 ml / cm². This data is marked as integrated exudate data. Sensitive data of the type is extracted from the integrated exudate data. After calculating the amount of exudate by capacitance change, the protease activity and the total amount of exudate are extracted. The protease activity, such as MMP-9 concentration, has a sensitivity of 0.1 ml / cm². The protease activity exudate and other total amounts of exudate are obtained. Historical exudate data for the past 3 days and pre-stored exudate difference thresholds are retrieved from the historical database. It should be noted that the pre-stored exudate difference thresholds are: protease activity difference ≤ 15 ng / mL and total exudate difference ≤ 0.5 ml / cm².

[0056] The floating curve is constructed by continuously comparing the old enzyme activity value and the comprehensive value of the single day from the far date to the recent date in the historical data of exudate, as well as the old enzyme activity value and the comprehensive value of the extracted days. The curve model is constructed with the timestamp of the period node as the X-axis and the unit quantity of the integrated exudate data as the Y-axis.

[0057] Compare the protease activity exudate volume and other comprehensive exudate volumes with the old enzyme activity volume and comprehensive volume values:

[0058] If the amount of exudate with protease activity is close to the old enzyme activity value, and the other comprehensive exudate amounts are close to the comprehensive values, and the difference between the two groups is less than the exudate difference threshold, then the wound recovery status is judged to be poor, and an exudate risk signal is generated. The exudate risk signal is sent to the multimodal data processing platform and the re-examination decision control module through the deep risk trend judgment module. The multimodal data processing platform immediately sends the exudate risk signal to the nursing station and the wristband of the on-duty nurse, displaying the text information in the style of "abnormal exudate amount and protease activity, possibly indicating active inflammation".

[0059] The text message includes operating steps, including:

[0060] Step 1: Record the color of the exudate when changing the dressing. If it is yellow-green, it indicates an infection.

[0061] Step two: Adjust the release rate of the antibacterial module, such as increasing the sustained-release dose of the microcapsule by 20%;

[0062] Step 3: Monitor the change in seepage volume every hour until the signal is cleared;

[0063] If the amount of exudate with protease activity is far from the old enzyme activity value, and the other comprehensive exudate amounts are far from the comprehensive values, and the difference between the two ratios is greater than the exudate difference threshold, then the wound recovery status is judged to be good, and an exudate log record SYi is generated. The exudate log record SYi is sent to the multimodal data processing platform and the re-examination decision control module through the deep risk trend judgment module. The exudate log record SYi is stored in the historical database and marked as "good recovery". The intelligent wound monitoring dressing system automatically generates exudate integrated data summary information and pushes it to the patient's family APP every day, prompting the text information in the style of "wound exudate reduced, healing normal". If it is SYi for 3 consecutive days, the exudate detection interval is automatically extended to 20 minutes.

[0064] The protease activity exudate volume and other comprehensive exudate volumes were plotted into the curve model by connecting the second points according to the arrangement of the period nodes.

[0065] When the curves for protease activity exudate volume, other comprehensive exudate volume, old enzyme activity value, and comprehensive value supplementation show an upward trend at the end, an exudate abnormality log is generated. The exudate abnormality log is sent to the multimodal data processing platform and the re-examination decision control module through the deep risk trend judgment module. After triggering, the nursing station terminal displays the text "Explicit upward trend in exudate volume" and marks the protease activity value.

[0066] Healthcare workers must check the absorbency of the dressing within 1 hour. If the middle layer of absorbent foam is close to saturation and the weight increases by more than 5g, the dressing should be replaced and an exudate sample should be taken for testing to detect whether it contains bacterial metabolites. It should be noted that the criteria for judging an upward trend is an increase in exudate volume of more than 0.2ml / cm² for three consecutive cycle nodes.

[0067] When the curves for protease activity exudate volume, other comprehensive exudate volume, and old enzyme activity value and comprehensive value continue to supplement show a downward trend at the end, a normal state log for exudate is generated. The normal state log for exudate is sent to the multimodal data processing platform through the deep risk trend judgment module for archiving in the historical database.

[0068] The change marker of floating curves was constructed by comparing the protease activity exudate volume and other comprehensive exudate volumes in the exudate integrated data with the data in the historical database, and the change rate of exudate volume ΔV was used as the marker.

[0069] Example 3:

[0070] The re-inspection decision control module receives multiple sets of signals, extracts the pH gradient ΔpH / Δt, temperature change ΔWt, and exudate volume change rate ΔV from these signals, and generates multi-level alarm signals based on a three-dimensional decision model. This triggers pre-stored dressing change reminders or corrects data acquisition accuracy logic. The re-inspection decision control module processes the received signals as follows:

[0071] After obtaining the pH anomaly signal RYz, the temperature and humidity anomaly signal WSYz, and the seepage risk signal, the three-dimensional decision model pre-stored in the historical database is retrieved, and the pH gradient ΔpH / Δt, temperature change ΔWt, and seepage rate of change ΔV from the original parameters of the three sets of signals are extracted and input into the three-dimensional decision model:

[0072] (ΔV>2ml / h)∩(ΔpH / Δt>0.15 / h)∩(ΔWT>1.5℃)

[0073] Wherein, ΔV represents the rate of change of exudate volume; 2 ml / h represents the threshold for comparing the rate of change of exudate volume (ΔV), which is the critical value for distinguishing between normal exudate and abnormal inflammatory exudate. In the early postoperative period of a normal wound, the amount of exudate will gradually decrease. For example, in Example 1, the baseline ΔV after surgery is about 1.2 ml / h. If the rate of change of exudate volume ΔV continues to exceed 2 ml / h, it suggests that there may be active bleeding, poor drainage, or increased inflammatory exudate due to infection. This threshold is set based on clinical data. Below this value, it is usually normal exudate during the tissue repair period. Above this value, the risk of poor wound healing should be noted. ΔpH / Δt represents the pH gradient, and 0.15 / h represents the threshold for comparing the pH gradient (ΔpH / Δt). 0.15 / h is the critical value for judging the rate of abnormal pH change. The pH of the normal wound microenvironment is stable at 5.5-6.5. If the pH value increases or decreases by more than 0.15 per hour, it suggests an imbalance in the wound microenvironment. For example, during infection, bacterial metabolism produces ammonia, which leads to a rapid increase in pH. Example 1 clearly states that "pH A sustained pH value > 7.3 for 2 hours and a ΔpH / Δt ratio > 0.15 / h serve as an infection warning condition, confirming the clinical value of this threshold. Compared to a single pH threshold, the ΔpH / Δt threshold better reflects dynamic trends and avoids misjudgments caused by individual baseline differences. ΔWT represents temperature change; 1.5℃ represents the temperature change (ΔT) comparison threshold. 1.5℃ is the temperature critical value that distinguishes between physiological inflammation and pathological infection. In the early stages of normal wound healing, there may be a slight temperature increase (ΔT < 1℃). If it exceeds 1.5℃, it is usually related to local infection or aggravated inflammation. Increased tissue metabolism and vasodilation lead to increased heat production. This threshold is set in conjunction with the normal human body temperature fluctuation range of 36.5-37.5℃, which avoids overreacting to mild inflammation and can promptly identify significant temperature increases caused by infection.

[0074] When a single parameter is abnormal, such as the exudate volume change rate ΔV exceeding the standard, a level one alarm signal is generated. The level one alarm signal is sent to the multimodal data processing platform through the re-examination decision control module. The nursing station terminal displays the text message "Single indicator abnormal, closely monitor" and automatically increases the detection frequency of this parameter to once every 3 minutes. The APP pushes the text message "Slight fluctuation in wound indicators, no need to change dressing for now" to the smart devices of the personnel associated with the patient.

[0075] When two parameters are abnormal, for example, only the rate of change of exudate volume ΔV, pH gradient ΔpH / Δt, and temperature change ΔWt exceed the standard, a level 3 alarm signal is generated. The level 3 alarm signal is sent to the multimodal data processing platform through the re-inspection decision control module to initiate an emergency response.

[0076] Step 1: Send an alert via SMS to the department head;

[0077] Step 2: Automatically generate a dressing change list, including breathable and waterproof membrane and antibacterial hydrogel;

[0078] Step 3, data acquisition accuracy correction: Control the NTC thermistor array to recalibrate, remove jump values ​​and verify a second time. If it is still abnormal, prompt the message "The sensor may be contaminated and the dressing needs to be replaced";

[0079] When a single signal occurs repeatedly, such as three abnormal temperature and humidity signals WSYz within one hour but no other abnormalities, the system initiates a self-correction process:

[0080] Step 1: Pause the upload of this parameter data, and restart the sensor after it has been powered off for 5 seconds;

[0081] Step 2: Call the normal parameter range of wounds of the same location and type in the historical database to correct the deviation of the current data;

[0082] Step 3: If the problem persists after correction, the terminal will display a message in the style of "It is recommended to check the sensor fit (poor contact may be caused by excessive leakage)" to guide medical staff to press the edge of the dressing again to ensure that the sensor unit is in contact with the skin.

[0083] Combining Examples 1, 2, and 3, intelligent optimization of the entire process from data collection to clinical intervention was achieved. Real-time monitoring ensures that changes in the wound microenvironment are captured in a timely manner, and multi-parameter collaborative diagnosis provides a scientific basis for risk assessment. The combination of these two aspects enables coordinated operation, which can automatically generate routine logs to reduce ineffective interventions, and trigger targeted nursing actions through abnormal signals and multi-level alarms, thus constructing a closed-loop model of monitoring-analysis-intervention. This not only reduces the workload of medical staff, but also effectively promotes wound recovery by precisely regulating the wound healing environment, filling the gap in the functional integration and clinical applicability of domestically produced high-end intelligent dressings.

[0084] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined in the claims, all of which should fall within the protection scope of the present invention.

[0085] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0086] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A smart wound monitoring dressing system based on multi-parameter biosensing, characterized in that, This includes defining the wound monitoring scope of intelligent medical dressings and constructing an intelligent wound monitoring dressing system using a multimodal data processing platform, as detailed below: Multi-level joint data acquisition module: Based on the wound monitoring scope, it obtains the dynamic parameter Di of the wound within the monitoring period, as well as a historical database constructed from previously collected recorded data and pre-stored thresholds; The shallow risk data assessment module obtains integrated pH and integrated temperature and humidity values, retrieves historical data to compare and analyze the integrated pH values, and generates the pH anomaly signal RYz and pH gradient ΔpH / Δt. The integrated temperature and humidity values ​​are analyzed in conjunction with the wound shape and the data collection location to obtain the local temperature and humidity differences between the wound center and the edge. These differences are then compared with historical data to generate anomaly signals WSYz and temperature change ΔWt. The deep risk trend judgment module obtains the integrated exudate value, performs sensitive data extraction on the integrated exudate value to obtain the amount of protease-active exudate, generates an exudate risk signal by comparing with historical data, constructs a curve model by combining with the historical database, and obtains the rate of change of exudate volume ΔV. Re-inspection decision control module: Receives multiple sets of signals, extracts pH gradient ΔpH / Δt, temperature change ΔWt, and exudate volume change rate ΔV from the multiple sets of signals, and generates multi-level alarm signals by combining them with a three-dimensional decision model, triggering the display of pre-stored dressing change reminders or correcting the accuracy of data collection.

2. The intelligent wound monitoring dressing system based on multi-parameter biosensing according to claim 1, characterized in that, The data acquisition process for the wound monitoring area by the multi-level joint data acquisition module is as follows: The system obtains the operation instructions of the intelligent wound monitoring dressing system, constructs a monitoring cycle based on the time of instruction execution to the current time, and performs time stamp equal division of the monitoring cycle to obtain several sets of cycle nodes with equal timestamp intervals. The integrated values ​​of pH, temperature and humidity, and exudate collected from the wound within the cycle nodes are marked as the original set of dynamic parameters. The original set of dynamic parameters is denoised to remove jump values ​​caused by poor sensor contact, and the different parameter data are unified with timestamps. Combined with the collection coordinates and time, spatiotemporal marking is performed to obtain the dynamic parameter Di. Based on the timeline of the regulatory cycle, the obtained dynamic parameters Di are stored in batches in the multimodal data processing platform to build a temporary cache library. The multimodal data processing platform internally stores historical databases constructed from previously collected data records and pre-stored thresholds.

3. The intelligent wound monitoring dressing system based on multi-parameter biosensing according to claim 1, characterized in that, The analysis process of the shallow risk data assessment module for the integrated pH value in the dynamic parameter Di is as follows: The potential data of the iridium oxide thin film three-electrode system was measured in the range of -0.2V to +0.6V, and was generated every N minutes. The potential data was marked as pH detection values. The pH detection values ​​of three consecutive times were integrated and the average value was marked as the integrated pH value. Historical pH data of the wound in the previous 24 hours were retrieved from the historical database, as well as preset historical normal pH data and preset historical abnormal pH data of the same type of wound in the same recovery period. The baseline range was established by the timestamp trend of the time period nodes. The historical pH data, preset historical normal pH data, and preset historical abnormal pH data were sequentially input into the baseline range to construct a normal range with the upper and lower limits of the preset historical normal pH data, a trend comparison range with the upper and lower limits of the historical pH data, and an abnormal range between the upper and lower limits of the preset historical abnormal pH data and the upper and lower limits of the historical normal pH data.

4. The intelligent wound monitoring dressing system based on multi-parameter biosensing according to claim 3, characterized in that, When the pH detection value is brought into the baseline range, if the pH detection value is completely within the normal range and within the trend comparison range along the time stamp of the periodic node, a normal pH log Ri is generated; if the pH detection value is completely within the normal range along the time stamp of the periodic node but has fluctuations exceeding the upper and lower limits of the trend comparison range, a pH abnormal log RYi is generated; if the pH detection value enters the abnormal range along the time stamp of the periodic node, a pH abnormal signal RYz is generated, and the pH detection value along the time stamp of the periodic node is marked as the pH gradient ΔpH / Δt.

5. The intelligent wound monitoring dressing system based on multi-parameter biosensing according to claim 1, characterized in that, The analysis process of the shallow risk data assessment module for the integrated temperature and humidity values ​​in the dynamic parameter Di is as follows: The temperature / humidity values ​​updated every N seconds by the NTC thermistor array are obtained and marked as integrated temperature and humidity values. The wound shape is divided into a central region and an edge region. Based on this, the corresponding values ​​are obtained from the region where the NTC thermistor array is located and averaged. The difference between the average values ​​of the central region and the edge region is extracted and marked as local difference values.

6. The intelligent wound monitoring dressing system based on multi-parameter biosensing according to claim 5, characterized in that, Retrieve the pre-stored standard temperature and humidity thresholds for the same type of wound within the same time period from the historical database, as well as the integrated old values ​​of temperature and humidity recorded N hours ago for the current wound in the historical database. Extract the difference between the local difference values ​​of the pre-stored temperature and humidity thresholds and the integrated old values ​​of temperature and humidity, and mark them as difference coefficients. Compare the local difference values ​​of the integrated temperature and humidity values ​​with the pre-stored temperature and humidity thresholds: when the local difference value is close to the pre-stored temperature and humidity threshold, and the difference between the two is less than the difference coefficient, it is determined that the current integrated temperature and humidity value of the wound is within the standard range, and a temperature and humidity log WSi is generated. When the local difference value is close to or far from the pre-stored temperature and humidity threshold, and the difference between the two is greater than the difference coefficient, it is determined that the current integrated temperature and humidity value of the wound is outside the standard range, and a temperature and humidity abnormality signal WSYz is generated. The process of comparing the local difference value of the integrated temperature and humidity value with the pre-stored temperature and humidity threshold is marked as temperature change △Wt.

7. The intelligent wound monitoring dressing system based on multi-parameter biosensing according to claim 1, characterized in that, The analysis process of the deep risk trend judgment module on the integrated value of permeation in the dynamic parameter Di is as follows: The real-time monitoring capacitance value of the serpentine conductive fiber network, recorded every N minutes, is obtained and marked as percolation integrated data. Type-sensitive data is extracted from the percolation integrated data to obtain the percolation amount of protease activity and other comprehensive percolation amounts. Historical percolation data for the past N days and pre-stored percolation difference thresholds are retrieved from the historical database. The old enzyme activity value and comprehensive value of each day from the far date to the recent date are extracted from the historical percolation data. A floating curve is constructed by continuously comparing the old enzyme activity value and comprehensive value of the extracted days. The curve model is constructed with the timestamp of the period node as the X-axis and the unit amount of percolation integrated data as the Y-axis.

8. The intelligent wound monitoring dressing system based on multi-parameter biosensing according to claim 7, characterized in that, The amount of exudate with protease activity and other comprehensive exudates are compared with the old enzyme activity value and the comprehensive value. If the amount of exudate with protease activity is close to the old enzyme activity value, the amount of other comprehensive exudates is close to the comprehensive value, and the difference between the two ratios is less than the exudate difference threshold, then the wound recovery status is judged to be poor, and an exudate risk signal is generated. If the amount of exudate with protease activity is far from the old enzyme activity value, the amount of other comprehensive exudates is far from the comprehensive value, and the difference between the two ratios is greater than the exudate difference threshold, then the wound recovery status is judged to be good, and an exudate log record SYi is generated. The protease activity exudate volume and other comprehensive exudate volumes are plotted in the curve model by connecting the second points according to the periodic node arrangement. When the end of the curve supplemented by the protease activity exudate volume, other comprehensive exudate volumes and the old enzyme activity value and comprehensive value shows an upward trend, an abnormal exudate state log is generated. When the end of the curve supplemented by the protease activity exudate volume, other comprehensive exudate volumes and the old enzyme activity value and comprehensive value shows a downward trend, a normal exudate state log is generated. The floating curve change marker is constructed by comparing the protease activity exudate volume and other comprehensive exudate volume in the integrated exudate data with the data in the historical database as the exudate volume change rate ΔV.

9. The intelligent wound monitoring dressing system based on multi-parameter biosensing according to claim 1, characterized in that, The re-inspection decision control module processes the received signals as follows: After obtaining the pH anomaly signal RYz, the temperature and humidity anomaly signal WSYz, and the seepage risk signal, the three-dimensional decision model pre-stored in the historical database is retrieved. The pH gradient ΔpH / Δt, temperature change ΔWt, and seepage rate change ΔV from the original parameters of the three sets of signals are extracted and input into the three-dimensional decision model. Based on the differences that exist during the comparison of the three-dimensional decision model, multi-level alarm signals are triggered.

Images