Intelligent limb thrombus monitoring system

Abstract

The invention discloses an intelligent limb thrombus monitoring system, which relates to the technical field of intelligent medical treatment and comprises a limb data acquisition terminal, a wireless transmission module, an intelligent analysis early warning unit and a visual display terminal, the limb data acquisition terminal is used for synchronously acquiring limb multi-dimensional physiological parameters, and the limb multi-dimensional physiological parameters comprise limb circumference / diameter change data, skin surface temperature data, local skin tension data and capillary blood oxygen saturation data. The intelligent limb thrombus monitoring system provided by the invention realizes synchronous real-time monitoring of key indexes such as limb circumference, temperature, tension, blood oxygen and the like on the premise of not influencing the activity of a patient through a miniaturized and multi-sensing integrated acquisition node, so that the objectivity and comprehensiveness of data acquisition are greatly improved; an intelligent analysis algorithm adopted by the system performs cooperative judgment based on individualized basic data and a multi-parameter time sequence trend.

Description

Technical Field

[0001] This invention relates to the field of smart medical technology, specifically to an intelligent monitoring system for limb thrombosis. Background Technology

[0002] With the increasing clinical application of interventional cardiovascular surgery, the incidence of postoperative complications such as subcutaneous hemorrhage, swelling, and thrombosis in the puncture-side limb is also rising. If these complications are not detected and managed promptly, they can negatively impact the patient's postoperative recovery and may even lead to more serious clinical problems. Currently, clinical monitoring of these complications mainly relies on medical staff visually observing changes in the limb's appearance and measuring the limb's circumference with a tape measure; this is a crucial step in ensuring postoperative patient recovery.

[0003] Existing manual monitoring methods have certain limitations in clinical applications. The results of manual observation and measurement are easily influenced by subjective factors; different medical personnel may have different judgment standards and measurement accuracy, making it difficult to accurately capture subtle physiological changes in the limbs. Manual monitoring is an intermittent operation, unable to achieve 24-hour continuous monitoring, easily missing subtle early signs of complications, leading to a lag in the detection of abnormalities. Furthermore, the data generated by manual monitoring lacks systematic recording and management, making effective retrospective analysis difficult and failing to provide sufficient data support for research on the development patterns of complications and clinical treatment decisions. In addition, relying solely on limb circumference as a single indicator lacks the synergistic analysis of multi-dimensional physiological parameters, leaving room for improvement in the accuracy of early identification of subcutaneous hematomas and thrombosis. To address these issues, we propose an intelligent limb thrombosis monitoring system. Summary of the Invention

[0004] To address the aforementioned technical problems, this paper provides an intelligent monitoring system for limb thrombosis. This solution overcomes the significant limitations of existing manual monitoring methods, which are susceptible to subjective influences, struggle to capture subtle physiological changes in the limbs, are intermittent in operation, easily miss early signs of complications, lack systematic data recording, and cannot achieve effective retrospective analysis. Furthermore, relying solely on a single circumference indicator lacks multi-dimensional parameter coordination, resulting in insufficient accuracy in early identification of complications.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: An intelligent limb thrombosis monitoring system, comprising: Limb data acquisition terminal, wireless transmission module, intelligent analysis and early warning unit, and visualization display terminal; The limb data acquisition terminal is used to synchronously collect multi-dimensional physiological parameters of the limb, including: limb circumference / diameter change data, skin surface temperature data, local skin tension data, and capillary blood oxygen saturation data. The wireless transmission module is used to transmit the collected physiological data to the analysis and early warning unit and the display terminal in real time. The intelligent analysis and early warning unit is used to process the multi-dimensional physiological parameters of the limbs collected in real time, and to trigger an audible and visual early warning when the data is abnormal. The visualization display terminal is used to display physiological parameter values, change curves and early warning information in real time, and has the ability to store and trace historical data.

[0006] Preferably, the method for synchronously collecting multi-dimensional physiological parameters of the limbs includes: At least two data acquisition nodes are attached to the target limb of the patient at different anatomical locations; Configure each data acquisition node to enter a scanning collaborative working mode, and cyclically collect data from the limb monitoring area at set time intervals; Each acquisition node synchronously acquires data on changes in limb circumference / diameter, skin surface temperature, local skin tension, and capillary blood oxygen saturation at its location.

[0007] Preferably, the method for real-time transmission of collected physiological data by the wireless transmission module includes: The limb data acquisition terminal packages the collected multi-dimensional physiological parameter data through its integrated wireless communication unit; Based on the distance between the data receiving terminal and the data acquisition terminal and the network environment, the communication link is automatically selected or switched between Bluetooth and Wi-Fi protocols. Data packets are simultaneously sent to the intelligent analysis and early warning unit and the visualization display terminal via the selected communication link.

[0008] Preferably, the method for processing the real-time acquired multi-dimensional physiological parameters of the limbs includes: The intelligent analysis and early warning unit receives real-time physiological parameter data uploaded via a wireless transmission module; Based on pre-set individualized baseline data, the relative rate of change of each physiological parameter relative to its baseline value is calculated in real time. Perform time-series analysis on the relative rates of change of each parameter and plot dynamic trend curves; Monitor the trend curve to identify whether a single physiological parameter shows a continuous abnormal trend.

[0009] Preferably, the method for acquiring and setting the preset preoperative individualized basic data includes: Before monitoring begins, baseline values ​​of various physiological parameters of the patient's target limb are obtained through a limb data acquisition terminal or manual measurement. The basic value is entered into the intelligent analysis and early warning unit as the benchmark for subsequent calculation of the relative rate of change; Based on clinical treatment guidelines and individual patient differences, personalized warning threshold ranges for various physiological parameters are set on the basis of baseline values.

[0010] Preferably, the method for triggering an audible and visual warning when data is abnormal includes: When a single physiological parameter is found to show a continuous abnormal trend, it is further determined whether at least one pre-set associated physiological parameter also shows an abnormal trend. If the condition of continuous abnormality of a single parameter and synchronous abnormality of at least one associated parameter is met, the warning level is determined based on the combination of abnormal parameters and the slope of the changing trend. Based on the determined warning level, a corresponding audible and visual warning signal is generated and displayed on a visual display terminal.

[0011] Preferably, the method for classifying the early warning level based on the slope of the abnormal trend includes: Calculate the instantaneous slope or recent average slope of the abnormal parameter change trend curve at the warning trigger time point; Multiple slope threshold ranges are set, each corresponding to a different clinical risk level; The calculated slope is compared with the threshold range to determine the specific risk level to which this warning belongs.

[0012] Preferably, the method for the visualization display terminal to display physiological parameter values, change curves, and early warning information in real time includes: Receive processed data and warning instructions from the intelligent analysis and early warning unit; The real-time physiological parameter values ​​of each acquisition node are dynamically displayed in partitions on the display interface; Simultaneously plot and update the dynamic trend curves of various physiological parameters; When an early warning command is received, an early warning window will pop up in a prominent position on the interface, displaying the abnormal parameter type, occurrence time, corresponding data collection node location, and early warning level.

[0013] Preferably, the method for the visualization display terminal to realize historical data storage and retrieval includes: The system continuously stores full-cycle monitoring data, including all physiological parameters and their trend information, uploaded by the intelligent analysis and early warning unit, in chronological order. Provides a data query interface that allows users to specify target time periods and target physiological parameters; Based on the query command, retrieve and redraw the change curve of the target parameter within the specified time period from the stored data to complete the data backtracking.

[0014] Preferably, the overall workflow of the system also includes initialization and configuration steps: After the data acquisition nodes are attached, the system is started and the device performs a self-test and wireless pairing. On the visual display terminal, complete the binding of patient information, the entry of preoperative basic data, and the configuration of personalized early warning thresholds; After confirming that the configuration is correct, start the real-time monitoring and data analysis functions of all data acquisition nodes and intelligent analysis and early warning units.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: The intelligent limb thrombosis monitoring system proposed in this invention, through miniaturized, multi-sensor integrated acquisition nodes, enables simultaneous real-time monitoring of key indicators such as limb circumference, temperature, tension, and blood oxygenation without affecting patient activity. This significantly improves the objectivity and comprehensiveness of data acquisition. The intelligent analysis algorithm employed by the system makes collaborative judgments based on individualized basic data and multi-parameter time-series trends, significantly reducing the risk of false alarms and making the early warning mechanism more aligned with the clinical pathological development pattern. This helps to identify signs of subcutaneous hematoma and thrombosis formation at an early stage. Abnormal situations can be promptly alerted to medical staff through audible and visual warnings, buying valuable time for clinical intervention. The system has complete monitoring data storage and retrospective functions, providing reliable data support for postoperative recovery assessment, complication research, and treatment decisions. The application of this system not only improves monitoring efficiency and accuracy but also reduces the workload of medical staff, demonstrating good clinical applicability and promotional value. Attached Figure Description

[0016] Figure 1 This is a system module diagram of the present invention. Detailed Implementation

[0017] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0018] Reference Figure 1 As shown, an intelligent limb thrombosis monitoring system includes: a limb data acquisition terminal, a wireless transmission module, an intelligent analysis and early warning unit, and a visualization display terminal; The limb data acquisition terminal is used to synchronously collect multi-dimensional physiological parameters of the limb, including: limb circumference / diameter change data, skin surface temperature data, local skin tension data, and capillary blood oxygen saturation data. The method for synchronously collecting multi-dimensional physiological parameters of the limb includes: attaching at least two acquisition nodes to the target limb of the patient at different anatomical locations; configuring each acquisition node to enter a scanning collaborative working mode, and cyclically collecting data from the limb monitoring area at set time intervals; and synchronously acquiring limb circumference / diameter change data, skin surface temperature data, local skin tension data, and capillary blood oxygen saturation data at the location of each acquisition node.

[0019] The scanning collaborative working mode employs a master-slave node collaborative architecture for synchronization. During system initialization, the node with the strongest signal strength is automatically identified as the active node, with the rest being slave nodes. The active node has a built-in high-precision real-time clock module, and clock calibration of all nodes is achieved through wireless synchronization signals, with timing control accuracy down to the millisecond level. During operation, the active node sends acquisition trigger commands at set time intervals. Upon receiving the commands, the slave nodes synchronously start the acquisition process. After each node completes acquisition, it sends a data ready signal back to the active node. The active node confirms that all nodes' data is ready and then triggers the transmission process uniformly, avoiding signal interference caused by multiple nodes acquiring or transmitting data simultaneously. The time interval is set based on clinical monitoring needs. The default time interval is 1 minute during the postoperative acute risk period, and automatically adjusts to 5 minutes after 24 hours of continuous monitoring without abnormalities. Manual adjustment by medical staff via a visual terminal is also supported, with an adjustment range of 1-30 minutes. The adjustment is based on the severity of the patient's condition, the type of surgery, and clinical treatment guidelines. Shorter interval monitoring is used for high-risk patients, while longer interval monitoring is used for patients in the stable phase to reduce power consumption.

[0020] The wireless transmission module is used to transmit the collected physiological data to the analysis and early warning unit and the display terminal in real time. The method for real-time transmission of collected physiological data by the wireless transmission module includes: the limb data acquisition terminal packages the collected multi-dimensional physiological parameter data through the wireless communication unit integrated therein; automatically selects or switches the communication link between Bluetooth protocol and Wi-Fi protocol according to the distance between the data receiving terminal and the acquisition terminal and the network environment; and synchronously sends the data packet to the intelligent analysis and early warning unit and the visualization display terminal through the selected communication link.

[0021] The decision-making algorithm for automatically selecting or switching communication links employs a multi-parameter quantitative evaluation model. First, the signal strength detection module obtains an estimated distance between the receiving terminal and the acquisition node. The Bluetooth protocol's adaptation distance is ≤10 meters, and the Wi-Fi protocol's adaptation distance is >10 meters. Simultaneously, the network environment quality is quantitatively evaluated using two core indicators: Received Signal Strength Indicator (RSSI) and packet loss rate. An RSSI ≥ -60dB and a packet loss rate <1% is considered a high-quality network environment; an RSSI between -80dB and -60dB and a packet loss rate between 1% and 5% is considered a general network environment; and an RSSI < -80dB or a packet loss rate >5% is considered a poor network environment. The decision logic is as follows: when the distance is ≤10 meters, Bluetooth transmission is prioritized; when the distance is >10 meters, Wi-Fi is selected if it is a high-quality or general network environment; otherwise, Bluetooth is switched to expand the signal search range if it is a poor network environment. When RSSI changes suddenly or the packet loss rate exceeds 5% for three consecutive sampling periods during communication, link switching is automatically triggered. During the switching process, a caching mechanism is used to temporarily store data to avoid data loss.

[0022] Data packaging employs a custom frame format protocol. The data packet structure includes a header, data segments, and a checksum. The header occupies 8 bytes and contains the acquisition node ID (2 bytes), timestamp (4 bytes), and data type identifier (2 bytes). The data segments are arranged in the order of limb circumference, temperature, tension, and blood oxygen saturation, with each parameter occupying 4 bytes and stored using 16-bit floating-point type to balance precision and storage capacity. The checksum is generated using the CRC-16 algorithm, occupying 2 bytes, and is used for data integrity verification at the receiving end. Data compression uses the LZ77 algorithm, with a compression ratio controlled between 1.5:1 and 2:1 to reduce transmission bandwidth usage. Data encryption uses the AES-256 encryption algorithm to encrypt the entire data packet. The key is automatically generated during system initialization and synchronized to each terminal to ensure data transmission security.

[0023] The data synchronization and reliable transmission confirmation and retransmission mechanism adopts a dual-terminal response mode. After the acquisition terminal sends a data packet, the intelligent analysis and early warning unit and the visualization display terminal must return an acknowledgment (ACK) signal within 1 second. The acquisition terminal only determines that the transmission is successful after receiving ACK signals from both terminals. If no ACK signal is received from either terminal, or if a negative acknowledgment (NACK) signal is received, the retransmission process is automatically initiated. The retransmission intervals are 100ms, 200ms, and 500ms respectively, with a maximum of 3 retransmissions. After 3 failed retransmissions, a transmission failure is recorded and a fault alarm signal is sent. At the same time, buffered data is prioritized for transmission after the link is restored.

[0024] The intelligent analysis and early warning unit is used to process the multi-dimensional physiological parameters of the limbs collected in real time, and to trigger an audible and visual early warning when the data is abnormal. The method for processing real-time collected multi-dimensional physiological parameters of the limbs includes: the intelligent analysis and early warning unit receiving real-time physiological parameter data uploaded via a wireless transmission module; calculating the relative rate of change of each physiological parameter relative to its baseline value based on preset preoperative individualized baseline data; performing time-series analysis on the relative rate of change of each parameter and plotting a dynamic trend curve; monitoring the trend curve and identifying whether a single physiological parameter exhibits a continuous abnormal trend.

[0025] The specific mathematical formula for the relative rate of change is: in This represents the real-time relative rate of change of a certain physiological parameter. This is the real-time collected value of this parameter. This is the preoperative individualized baseline value for this parameter. In the formula, The instantaneous data collected synchronously by the acquisition nodes is filtered and processed to remove noise interference. The baseline values ​​set preoperatively serve as a reference standard for evaluating parameter changes, and the relative rate of change is also considered. The positive and negative values ​​represent the increase or decrease of the parameter relative to the base value, respectively, and the magnitude of the absolute value reflects the magnitude of the change.

[0026] Preoperative individualized baseline data measurement must follow standardized procedures. Limb circumference is measured using a medical soft measuring tape, perpendicular to the limb's longitudinal axis, with a tightness allowing a finger's width to fit between the tape and the limb. Each measurement point is repeated three times, and the average value is used as the baseline. Skin surface temperature is measured using a high-precision infrared thermometer, with the measurement distance controlled within 1 cm, avoiding hair-covered areas. Each measurement point is measured twice, with a 30-second interval, and the average value is used. Local skin tension is measured using a portable skin tension meter, with the probe perpendicularly placed on the skin surface, maintaining constant pressure for one second before reading the value. This is repeated three times, and outliers are removed before the average value is used. Capillary oxygen saturation is measured using a portable pulse oximeter, with the probe worn on the fingertip or toe of the target limb, and the value recorded after stabilizing for three seconds. Five consecutive measurements are taken, and the average value is used. All manual measurements must be completed within 24 hours before surgery. During measurement, the patient should remain supine with the limb in a naturally relaxed state, avoiding pressure, activity, and drastic changes in ambient temperature that could affect the measurement results.

[0027] Personalized early warning threshold ranges are set using a multiple linear regression model, dynamically adjusted based on clinical treatment guidelines and individual patient differences. Key influencing factors include patient age, underlying diseases (diabetes, hypertension, history of thrombosis, etc.), type of surgery (major orthopedic surgery, abdominal surgery, vascular surgery, etc.), and basic limb condition (degree of edema, skin condition, etc.). The model expression is: in For personalized early warning thresholds, The clinical standard threshold, Adjustment coefficient for individuals, Age is an influencing factor. Basic disease influencing factors, As a factor influencing surgical type, These are the weighting coefficients for each factor. For example, diabetic patients have poor vascular elasticity, and temperature is related to the blood oxygen saturation threshold. The coefficient was adjusted to 0.9, which is the threshold for circumference change in patients with a history of thrombosis. The coefficient was adjusted to 0.8; for patients undergoing major orthopedic surgery. For patients with higher factor weights than those undergoing abdominal surgery, the threshold range is appropriately narrowed. The threshold range is divided into an upper threshold and a lower threshold; an early warning assessment is triggered when the threshold exceeds the range and persistent abnormal conditions are met.

[0028] In time series analysis, the Kalman filter algorithm is used to remove data noise when plotting dynamic trend curves. During the filtering process, state equations and observation equations are set, and the optimal estimate is obtained through iterative calculation to ensure the curve is smooth and accurately reflects the parameter change trend. For missing data within the sampling interval, a cubic spline interpolation algorithm is used to supplement the missing data. The interpolation function is: in The interpolation interval index is... , , , These are the interpolation coefficients, which are solved using boundary and continuity conditions to ensure the smoothness and continuity of the interpolated curve. The horizontal axis of the trend curve represents the time axis, and the vertical axis represents the relative rate of change. The data points are connected sequentially according to the sampling time, and the curve segment corresponding to the latest collected data is updated in real time.

[0029] The criteria for identifying a persistent abnormal trend are: the relative rate of change of a single physiological parameter. If the data exceeds the personalized early warning threshold range for three consecutive sampling periods, or exceeds the threshold range for four out of five consecutive sampling periods, and the direction of change is consistent (continuously increasing or continuously decreasing), a buffer judgment mechanism is set up. If the data exceeds the threshold in a single sampling period but returns to the threshold in the next sampling period, it is judged as an occasional fluctuation and is not considered a continuous abnormal trend. If the data exceeds the threshold by more than 50% of the upper or lower limit of the threshold, the anomaly assessment is directly triggered regardless of the number of consecutive periods.

[0030] The method for acquiring and setting the pre-set individualized basic data before surgery includes: before the monitoring begins, acquiring the basic values ​​of various physiological parameters of the patient's target limb through a limb data acquisition terminal or manual measurement; entering the basic values ​​into the intelligent analysis and early warning unit as the benchmark for subsequent calculation of relative change rate; and setting personalized early warning threshold ranges for various physiological parameters based on the benchmark values ​​according to clinical diagnosis and treatment guidelines and individual patient differences.

[0031] The method for triggering audible and visual warnings when data is abnormal includes: when a single physiological parameter is identified to show a continuous abnormal trend, further determining whether at least one pre-set associated physiological parameter shows a synchronous abnormal trend; if the condition of continuous abnormality of the single parameter and synchronous abnormality of at least one associated parameter is met, then the warning level is determined according to the combination of abnormal parameters and the slope of the change trend; according to the determined warning level, a corresponding audible and visual warning signal is generated and prompted through a visual display terminal.

[0032] The associated physiological parameters are set based on the physiological and pathological mechanisms of thrombosis and their clinical relevance. Specific associated combinations are as follows: limb circumference is associated with local skin tension; when venous return is obstructed due to thrombosis, limb swelling will simultaneously increase circumference and skin tension. Skin surface temperature is associated with capillary oxygen saturation; insufficient blood perfusion after local thrombosis leads to decreased skin temperature and a decrease in capillary oxygen saturation. Limb circumference is also associated with skin temperature; severe swelling may be accompanied by abnormal temperature due to impaired local blood circulation. The associated parameters for each parameter are pre-stored in the system and can be updated according to clinical research progress and treatment needs.

[0033] The decision tree model process from anomaly identification to early warning signal generation is as follows: First, identify a continuous abnormal trend of a single parameter and label the abnormal parameter type; Second, retrieve the associated parameters corresponding to the abnormal parameter and determine whether at least one associated parameter simultaneously exhibits a continuous abnormal trend; Third, if no associated parameter is synchronously abnormal, record the abnormal event and continue monitoring without triggering an audible and visual warning; if associated parameters are synchronously abnormal, extract the combination of abnormal parameters and the slope of the trend curve of each parameter; Fourth, determine the warning level based on the slope calculation results; Fifth, generate the corresponding audible and visual warning signal based on the warning level and simultaneously send it to the visualization display terminal and the local speaker and indicator light of the intelligent analysis and early warning unit; Sixth, after receiving confirmation from medical staff, stop the audible and visual warning, and continuously monitor and record parameter changes; if there is no confirmation within 10 minutes, automatically upgrade the warning level by one level and resend the warning signal.

[0034] The method for classifying the warning level based on the slope of the abnormal trend includes: calculating the instantaneous slope or recent average slope of the abnormal parameter change trend curve at the warning trigger time; setting multiple slope threshold intervals, each corresponding to a different clinical risk level; and comparing the calculated slope with the threshold intervals to determine the specific risk level to which this warning belongs.

[0035] The instantaneous slope is calculated using the five-point difference method, taking the relative rate of change data of the warning trigger time point and two sampling points before and after it. The calculation formula is as follows: in The instantaneous slope , To trigger the early warning, the relative rate of change of the first two sampling points, , The relative rate of change between the two sampling points after the warning is triggered. The sampling time interval is specified. The recent average slope is calculated using a sliding window method with a window length of 10 sampling periods. The calculation formula is as follows: in The recent average slope The relative rate of change at the time of the warning trigger. The relative rate of change 10 sampling periods ago. The sampling time interval is specified. When the parameter change trend is stable, the instantaneous slope is used for evaluation; when the parameter change fluctuates, the recent average slope is used for evaluation to ensure the accuracy of the classification.

[0036] The slope threshold range and corresponding clinical risk level are set as follows: Low-risk warning (Level I): or An absolute value < 0.2% / min corresponds to the early stage of thrombosis, with a low clinical risk, requiring enhanced monitoring; Intermediate risk warning (Level II): 0.2% / min ≤ or An absolute value <0.5% / min corresponds to the progression stage of thrombosis, indicating significant venous return obstruction, requiring timely intervention; High-risk warning (Level III): or An absolute value ≥0.5% / min indicates an acute phase of thrombosis or a high risk of thrombus detachment, posing a high clinical risk and requiring emergency treatment. The baseline values ​​of the slope thresholds corresponding to different combinations of abnormal parameters can be adjusted appropriately. For example, when circumference and tension are simultaneously abnormal, the baseline threshold value can be reduced by 20% to improve warning sensitivity.

[0037] The visualization display terminal is used to display physiological parameter values, change curves and early warning information in real time, and has the ability to store and trace historical data.

[0038] The method for the visualization display terminal to display physiological parameter values, change curves and early warning information in real time includes: receiving data and early warning instructions processed by the intelligent analysis and early warning unit; dynamically displaying the real-time physiological parameter values ​​of each acquisition node in sections on the display interface; synchronously drawing and updating the dynamic change trend curves of each physiological parameter; and when an early warning instruction is received, popping up an early warning window in a prominent position on the interface to display the abnormal parameter type, occurrence time, corresponding acquisition node location and early warning level.

[0039] The visual display terminal interface adopts a segmented layout design with an overall resolution of 1920×1080. The upper left half of the interface is the parameter value display area, occupying 40% of the interface width. It is displayed according to the data acquisition nodes, with each node corresponding to a display module. The module sequentially displays the real-time values ​​and relative rates of change of limb circumference, temperature, tension, and blood oxygen saturation. The numerical font size is 16pt, black under normal conditions, and red when exceeding the threshold. The numerical refresh rate is consistent with the sampling period. The upper right part is the trend curve display area, occupying 60% of the interface width. It supports the overlay display of multiple curves. Each curve corresponds to a physiological parameter and is distinguished by different colors (circumference blue, temperature red, tension green, blood oxygen purple). The curve line width is 2px. A parameter switching button is set at the top of the interface, which can select single-parameter or multi-parameter curve display. In the curve coordinate system, the horizontal axis is the time axis, which defaults to displaying data for the past 1 hour. It can be adjusted to the past 24 hours or the past 6 hours using the scaling control. The vertical axis is the relative rate of change. The threshold range is -50% to 50%, marked by dashed lines indicating the upper and lower limits. The lower half of the interface is an information display area, occupying 20% ​​of the interface height. The left side displays basic patient information, and the right side displays the system status (monitoring, standby, fault) and the connection status of the data acquisition nodes.

[0040] The warning window uses a pop-up design, positioned in the center of the interface without interrupting the main interface data display. The window size is 600px × 400px. The top displays the warning level indicator (low risk blue, medium risk yellow, high risk red) and the warning title. The middle section sequentially displays the abnormal parameter type, occurrence time (accurate to the second), corresponding data collection node location (labeled with the anatomical site), abnormal parameter value, and relative rate of change. At the bottom, there are two interactive buttons: "Confirm" and "View Details". Clicking the "Confirm" button closes the pop-up window, while the warning icon remains in the upper right corner of the main interface until the parameters return to normal. Clicking the "View Details" button switches the pop-up window to the detailed page, displaying the trend curves, slope values, and warning classification criteria for the abnormal parameters and related parameters.

[0041] The method for the visualization display terminal to realize historical data storage and backtracking includes: continuously storing full-cycle monitoring data uploaded by the intelligent analysis and early warning unit in chronological order, containing all physiological parameters and their trend information; providing a data query interface, allowing users to specify target time periods and target physiological parameters; and retrieving and redrawing the change curve of the target parameters within the specified time period from the stored data according to the query instructions, thereby completing data backtracking.

[0042] Historical data storage employs a hierarchical structure. The first layer categorizes data by patient ID, the second by monitoring date, and the third by sampling time. Each sampling time node stores corresponding data records, including the original parameter values ​​and relative rates of change for each collection point. The system displays filtered values, curve slopes, and system status information. Data is stored in JSON format for easy parsing and retrieval. The storage medium is a 1TB SSD built into the terminal, with support for cloud backup at an hourly rate. Cloud storage lasts 90 days, while local storage lasts 180 days. Historical data exceeding the storage period is automatically compressed and archived, and archived files can be exported to external storage devices.

[0043] The data query interface adopts a graphical interactive design. The left side of the interface features a query criteria area with a calendar control and parameter drop-down menus. The calendar control supports selecting start and end times, with time precision adjustable to the minute level. The parameter drop-down menu allows users to query single or multiple parameters, and also provides a filtering option for data collection nodes. The query process is as follows: the user selects the target time period, target parameters, and data collection nodes, clicks the "Query" button, and the system backend retrieves the corresponding data from the storage medium, decompresses and converts the data format, and re-applies the plotting algorithm to draw the trend curve. The curve display interface supports zooming and panning operations, allowing users to view specific values ​​at any time point. It also provides a data export function, supporting export to Excel or PDF formats. The exported file includes a table of parameter values ​​and a trend curve graph.

[0044] The overall workflow of the system also includes initialization and configuration steps: after the acquisition nodes are attached, the system is started and the device self-test and wireless pairing are performed; on the visual display terminal, patient information binding, preoperative basic data entry and personalized early warning threshold configuration are completed; after confirming that the configuration is correct, the real-time monitoring and data analysis functions of all acquisition nodes and intelligent analysis and early warning units are started.

[0045] The equipment self-test includes sensor calibration, battery level detection, wireless communication link detection, and data transmission path detection. Sensor calibration is automatically completed by the calibration module built into the acquisition node, comparing the preset standard value with the acquired value, calculating the calibration coefficient, and automatically correcting. Battery level detection reads the battery level of each acquisition node and terminal device, prompting charging when the battery level is below 10%. Wireless communication link detection tests the connection stability between the acquisition node and the two terminals, detecting RSSI value and packet loss rate. Data transmission path detection verifies the integrity of data transmission from the acquisition node to the analysis unit and then to the display terminal by sending test data packets. After the self-test is completed, a self-test report is generated, displaying the test results of each item. If any faults are found, the fault type and troubleshooting suggestions are marked. Only after the self-test is passed can the next step be performed.

[0046] The wireless pairing process uses the Bluetooth BLE protocol. After system startup, the data acquisition node automatically enters pairing mode, and its indicator light flashes yellow. The visual display terminal scans for nearby Bluetooth devices and lists the IDs of pairable data acquisition nodes. The user selects the corresponding node to complete the pairing. During the pairing process, a dynamic encryption algorithm is used to generate a pairing code, which is only valid in the current session. After successful pairing, the data acquisition node's indicator light turns solid green, and the terminal interface displays the node's connection status as "Paired." If pairing fails, the system automatically retryes, with a maximum of 5 retries. After 5 failed attempts, the system prompts the user to check the node's location or restart the device.

[0047] The necessary fields for patient information binding include: patient name, gender, age, hospital number, bed number, surgery name, surgery time, target limb for monitoring, history of underlying diseases, and history of thrombosis. The information entry interface provides a template import function, which can directly import patient information from the hospital's HIS system to avoid duplicate entry. After the entry is completed, a unique monitoring file is automatically generated, and the file number is consistent with the patient's hospital number, which facilitates data association and traceability.

[0048] The personalized early warning threshold configuration interface employs both slider and numerical input methods. The interface is categorized by physiological parameters, with each parameter corresponding to a slider control. The slider range is ±50% of the baseline value. Dragging the slider adjusts the upper and lower limits of the threshold. Manual input of specific values ​​is also supported; the system automatically prompts and restricts saving when the input value exceeds the range. The interface provides a clinical standard threshold reference button; clicking it displays the corresponding clinical standard value for medical staff to refer to when setting settings. After configuration, the correspondence between the threshold range and the baseline value can be previewed. Once confirmed, the configuration is saved, and the configured parameters are automatically synchronized to the intelligent analysis and early warning unit.

[0049] After activating the real-time monitoring function, the status of each module in the system is displayed synchronously via indicator lights and interface prompts. A solid green light on the data acquisition node indicates normal operation, a flashing yellow light indicates data transmission, and a solid red light indicates a fault. A solid green light on the intelligent analysis and early warning unit indicator indicates data analysis is in progress; when an early warning is triggered, the corresponding indicator light flashes, and an audio-visual alert is emitted simultaneously. The top of the visual display terminal interface displays a "Monitoring" status indicator, updating the working status of each module in real time. Users must click the "Start Monitoring" button on the interface to confirm activation. After activation, the system enters continuous monitoring mode and cannot be interrupted arbitrarily. Interruption of monitoring requires the medical staff's employee ID and password to ensure monitoring continuity and security.

[0050] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. An intelligent monitoring system for limb thrombosis, characterized in that, include: Limb data acquisition terminal, wireless transmission module, intelligent analysis and early warning unit, and visualization display terminal; The limb data acquisition terminal is used to synchronously collect multi-dimensional physiological parameters of the limb, including: limb circumference / diameter change data, skin surface temperature data, local skin tension data, and capillary blood oxygen saturation data. The wireless transmission module is used to transmit the collected physiological data to the analysis and early warning unit and the display terminal in real time. The intelligent analysis and early warning unit is used to process the multi-dimensional physiological parameters of the limbs collected in real time, and to trigger an audible and visual early warning when the data is abnormal. The visualization display terminal is used to display physiological parameter values, change curves and early warning information in real time, and has the ability to store and trace historical data.

2. The intelligent limb thrombosis monitoring system according to claim 1, characterized in that, The method for synchronously collecting multi-dimensional physiological parameters of the limbs includes: At least two data acquisition nodes are attached to the target limb of the patient at different anatomical locations; Configure each data acquisition node to enter a scanning collaborative working mode, and cyclically collect data from the limb monitoring area at set time intervals; Each acquisition node synchronously acquires data on changes in limb circumference / diameter, skin surface temperature, local skin tension, and capillary blood oxygen saturation at its location.

3. The intelligent limb thrombosis monitoring system according to claim 2, characterized in that, The method for real-time transmission of collected physiological data by the wireless transmission module includes: The limb data acquisition terminal packages the collected multi-dimensional physiological parameter data through its integrated wireless communication unit; Based on the distance between the data receiving terminal and the data acquisition terminal and the network environment, the communication link is automatically selected or switched between Bluetooth and Wi-Fi protocols. Data packets are simultaneously sent to the intelligent analysis and early warning unit and the visualization display terminal via the selected communication link.

4. The intelligent limb thrombosis monitoring system according to claim 3, characterized in that, The method for processing the real-time acquired multi-dimensional physiological parameters of the limbs includes: The intelligent analysis and early warning unit receives real-time physiological parameter data uploaded via a wireless transmission module; Based on pre-set individualized baseline data, the relative rate of change of each physiological parameter relative to its baseline value is calculated in real time. Perform time-series analysis on the relative rates of change of each parameter and plot dynamic trend curves; Monitor the trend curve to identify whether a single physiological parameter shows a continuous abnormal trend.

5. The intelligent limb thrombosis monitoring system according to claim 4, characterized in that, The methods for acquiring and setting the preset preoperative individualized basic data include: Before monitoring begins, baseline values ​​of various physiological parameters of the patient's target limb are obtained through a limb data acquisition terminal or manual measurement. The basic value is entered into the intelligent analysis and early warning unit as the benchmark for subsequent calculation of the relative rate of change; Based on clinical treatment guidelines and individual patient differences, personalized warning threshold ranges for various physiological parameters are set on the basis of baseline values.

6. The intelligent limb thrombosis monitoring system according to claim 5, characterized in that, The method for triggering audible and visual warnings when data is abnormal includes: When a single physiological parameter is found to show a continuous abnormal trend, it is further determined whether at least one pre-set associated physiological parameter also shows an abnormal trend. If the condition of continuous abnormality of a single parameter and synchronous abnormality of at least one associated parameter is met, the warning level is determined based on the combination of abnormal parameters and the slope of the changing trend. Based on the determined warning level, a corresponding audible and visual warning signal is generated and displayed on a visual display terminal.

7. The intelligent limb thrombosis monitoring system according to claim 6, characterized in that, The method for classifying early warning levels based on the slope of abnormal trends includes: Calculate the instantaneous slope or recent average slope of the abnormal parameter change trend curve at the warning trigger time point; Multiple slope threshold ranges are set, each corresponding to a different clinical risk level; The calculated slope is compared with the threshold range to determine the specific risk level to which this warning belongs.

8. The intelligent limb thrombosis monitoring system according to claim 7, characterized in that, The method for the visualization display terminal to display physiological parameter values, change curves, and early warning information in real time includes: Receive processed data and warning instructions from the intelligent analysis and early warning unit; The real-time physiological parameter values ​​of each acquisition node are dynamically displayed in partitions on the display interface; Simultaneously plot and update the dynamic trend curves of various physiological parameters; When an early warning command is received, an early warning window will pop up in a prominent position on the interface, displaying the abnormal parameter type, occurrence time, corresponding data collection node location, and early warning level.

9. The intelligent limb thrombosis monitoring system according to claim 8, characterized in that, The method for the visualization display terminal to realize historical data storage and retrieval includes: The system continuously stores full-cycle monitoring data, including all physiological parameters and their trend information, uploaded by the intelligent analysis and early warning unit, in chronological order. Provides a data query interface that allows users to specify target time periods and target physiological parameters; Based on the query command, retrieve and redraw the change curve of the target parameter within the specified time period from the stored data to complete the data backtracking.

10. The intelligent limb thrombosis monitoring system according to claim 9, characterized in that, The overall system workflow also includes initialization and configuration steps: After the data acquisition nodes are attached, the system is started and the device performs a self-test and wireless pairing. On the visual display terminal, complete the binding of patient information, the entry of preoperative basic data, and the configuration of personalized early warning thresholds; After confirming that the configuration is correct, start the real-time monitoring and data analysis functions of all data acquisition nodes and intelligent analysis and early warning units.

Images