A leg measurement early warning system based on trend analysis

Abstract

The application relates to the technical field of medical treatment, in particular to a leg measurement and early warning system based on trend analysis, a data acquisition module used for continuously collecting user leg girth data; a data preprocessing module used for filtering and removing abnormal values of the collected leg girth data; a change rate calculation module used for calculating daily change rates and weekly change rates of the leg girth; a risk judgment module used for comprehensively judging risks according to the daily change rates, the weekly change rates and change trends; and an early warning output module used for outputting corresponding early warning information according to risk levels; the application can realize precise positioning and measurement of patient legs, automatically collect data, automatically process and analyze the data, intelligently filter interference signals, quantitatively analyze patient edema change trends, realize multi-dimensional risk fusion judgment, and then automatically trigger a special device for risk early warning, so that the needs of early screening and real-time monitoring of lower limb deep vein thrombosis in clinical treatment can be met.

Description

Technical Field

[0001] This invention relates to the field of medical technology, specifically to a leg-detection early warning system based on trend analysis. Background Technology

[0002] Deep vein thrombosis (DVT) of the lower extremities is a common critical illness in clinical practice, frequently occurring in postoperative bed rest, trauma immobilization, elderly individuals with chronic diseases, and those who sit for long periods. Abnormal blood clotting within the deep veins causes lower extremity edema and pain. If the thrombus detaches, it can induce pulmonary embolism, seriously threatening the patient's life. Lower extremity edema is the most typical and obvious early sign of DVT formation. Monitoring changes in leg circumference and assessing the development trend of edema are core methods for early detection and intervention of thrombosis, and are also routine tasks in clinical nursing and intensive care monitoring.

[0003] Currently, clinical monitoring of lower extremity edema and the risk of venous thrombosis mainly relies on manual measurement of leg circumference using a soft measuring tape. The degree of edema is determined by recording daily values, comparing bilateral differences, and tracking cyclical changes. However, this traditional method has several inherent drawbacks: First, the measurement points rely on manual positioning by medical staff, which is prone to positional errors, resulting in poor data repeatability and meaningless comparisons between different measurements. Second, a single measurement only obtains instantaneous values, making it impossible to achieve continuous, timed data collection and capture the gradual trend of edema changes. Third, data requires manual analysis after recording, but data processing and analysis capabilities are inadequate, making it difficult to intuitively determine the representativeness of the data and accurately quantify key indicators such as daily and weekly changes in leg circumference. Fourth, early warning relies on the subjective judgment of medical staff, making it difficult to promptly identify slowly worsening, insidious edema and easily missing the early intervention window for thrombosis.

[0004] In summary, the current lack of a dedicated device capable of accurately locating measurement points, continuously collecting leg circumference data, intelligently filtering interference signals, quantitatively analyzing edema trends, and automatically triggering risk warnings makes it difficult to meet clinical needs for early screening and real-time monitoring of lower extremity deep vein thrombosis. Therefore, developing a lower extremity edema early warning device based on leg circumference trend analysis and possessing intelligent filtering and precise early warning functions has become an urgent technical problem to be solved in this field. Summary of the Invention

[0005] The purpose of this invention is to provide a leg-detection early warning system based on trend analysis to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: As an optional solution to the trend analysis-based leg measurement early warning system of the present invention, the trend analysis-based leg measurement early warning system includes a data acquisition module, a data preprocessing module, a rate of change calculation module, a trend analysis module, a risk judgment module, and an early warning output module. The data acquisition module is used to continuously collect users' leg circumference data; A data preprocessing module, connected to the data acquisition module, is used to filter and remove outliers from the acquired leg circumference data. The rate of change calculation module, connected to the data preprocessing module, is used to calculate the daily and weekly rate of change of leg circumference. The trend analysis module, connected to the data preprocessing module, is used to perform linear regression fitting on the preprocessed leg circumference data to obtain the leg circumference change trend. The risk assessment module is connected to the rate of change calculation module and the trend analysis module respectively, and is used to make a comprehensive risk assessment based on the daily rate of change, weekly rate of change and the trend of change. The early warning output module is connected to the risk determination module and is used to output corresponding early warning information according to the risk level.

[0007] As an optional solution of the trend analysis-based leg measurement early warning system described in this invention, the data acquisition module includes multiple sets of flexible sensors, a signal processing unit, and a body position detection unit. The flexible sensors are used to conform to the user's legs to collect leg circumference data; the signal processing unit is connected to the flexible sensors and is used to amplify and convert the sensor signals from analog to digital; the body position sensors are used to detect the user's body position information during measurement. Flexible sensors are distributed at the location of the thigh circumference 15cm above the patella and the calf circumference at the location of the maximum circumference of the calf, and data are collected continuously for 5-10 days.

[0008] As an optional solution to the trend analysis-based leg detection early warning system described in this invention, the data preprocessing module is specifically used for: A sliding window filtering algorithm is used to take the average of the most recent 5-10 measurements as valid data. Outliers were removed using the 3σ criterion. The mean μ and standard deviation σ of the historical data were calculated, and measurements that exceeded the range of [μ-3σ, μ+3σ] were removed. Postural compensation is performed based on body position information to correct leg circumference data under different body positions to a unified benchmark value.

[0009] As an optional solution to the trend analysis-based leg-detection early warning system described in this invention, the change rate calculation module includes the following calculation formulas: Daily relative change rate:

[0010] Daily absolute change: , Weekly relative rate of change:

[0011] Rate of change relative to baseline

[0012] in, : The circumference of a specific part of the thigh (calf) on that day; Leg circumference at the same location and time the previous day; Average leg circumference this week: Average leg circumference last week; Individual baseline values ​​(such as leg circumference at admission or in a healthy state); Absolute difference in leg circumference between both sides:

[0013] Two-sided relative difference rate:

[0014] in, The circumference of the left leg. The measurement is the circumference of the right leg.

[0015] As an optional solution to the leg detection early warning system based on trend analysis described in this invention, the trend analysis module is specifically used for: Linear regression was performed on the effective leg circumference data of the past 7 days, and the slope k of the fitted line was calculated. Calculate the exponential moving average of leg circumference: EMA(t) = α×Q+(1-α)×EMA(t-1) in, EMA(t): Exponential moving average on day t; α: Smoothing coefficient; Q: Measured leg circumference on day t; EMA(t-1): Exponential moving average on day t-1.

[0016] EMA daily change rate:

[0017] EMA cumulative increment:

[0018] in, Smooth hind leg circumference on the same day, : The smoothed value of the previous day.

[0019] As an optional solution to the trend analysis-based leg detection early warning system described in this invention, the risk determination module is specifically used for: A daily change rate > 2%, a weekly change rate > 5%, or a difference in leg circumference on one side > 2cm is considered high risk. When the daily change rate is between 1% and 2%, the weekly change rate is between 3% and 5%, or the difference in unilateral leg circumference is between 1 and 2 cm, it is judged as medium risk; When the daily change rate is <1%, the weekly change rate is <3%, and the difference in leg circumference on one side is <1cm, it is judged as low risk; When the slope k of the fitted line is greater than 0.1 cm / day, if the current risk level is medium risk, it will be upgraded to high risk; if the current risk level is low risk, it will be upgraded to medium risk.

[0020] As an optional solution to the trend analysis-based leg detection early warning system described in this invention, the early warning output module is specifically used for: No warning information is output when the risk is low; local pop-up window and indicator light prompts are output when the risk is medium; local alarm is output and warning information is pushed to remote terminals when the risk is high.

[0021] As an optional solution to the trend analysis-based leg circumference early warning system described in this invention, it further includes: a personalized baseline module for establishing a user's personal leg circumference baseline and calculating the rate of change based on the personal baseline; a data storage module for storing historical leg circumference data and early warning records; and a communication module for data interaction with a remote terminal.

[0022] Compared with the prior art, the beneficial effects of the present invention are: This invention enables precise positioning and measurement of a patient's legs, automatic data collection, and automatic data processing and analysis. It intelligently filters interference signals, quantitatively analyzes the trend of edema changes, and achieves a multi-dimensional integrated assessment of thrombosis risk, resulting in more accurate and reliable early warnings. This invention overcomes the limitations of traditional methods that rely solely on single instantaneous values ​​or human experience by jointly judging three key indicators: daily / weekly leg circumference change rate, edema change trend, and difference in leg circumference between the left and right sides. It can comprehensively identify the nature of lower extremity edema from three perspectives: "acute fluctuations, slow progression, and bilateral asymmetry," effectively distinguishing between physiological fluctuations and pathological edema, significantly improving the accuracy of early identification of deep vein thrombosis, and reducing the rate of missed and false diagnoses.

[0023] Through data preprocessing and exponential smoothing algorithms, the system has stronger anti-interference capabilities. It adopts sliding window filtering, 3σ outlier removal and body position compensation to eliminate random errors caused by measurement posture, skin tightness and body position changes. Combined with exponential moving average (EMA), it further weakens short-term noise, highlights the true edema trend, and enables the device to work stably in complex clinical environments. It avoids false alarms caused by single measurement deviations and improves the continuity and reliability of monitoring.

[0024] By combining trend analysis with slope determination, the risk of occult thrombosis can be identified in advance. By fitting the trend of leg circumference changes over 7 consecutive days through linear regression, it is possible to capture occult edema that is slow, continuous, and progressively worsening. Even if the daily change rate does not exceed the standard, the slope can be used to make an early warning, giving medical staff more time to intervene and effectively reducing the risk of thrombosis progression and serious complications such as pulmonary embolism.

[0025] The system adopts a personalized baseline and graded early warning mechanism to better meet actual clinical needs. It establishes a personal leg circumference baseline and avoids the incompatibility of general thresholds with patients of different body types and underlying diseases. At the same time, it sets up three levels of risk assessment and graded early warning output, which not only reduces unnecessary alarm interference, but also provides timely audible and visual alarms and remote push notifications in case of high risk. It is especially suitable for high-risk groups such as those who are bedridden after surgery, elderly people with chronic diseases, and those who are immobilized for a long time.

[0026] The data is storable and traceable, supporting long-term continuous monitoring and remote management systems. It has functions for storing historical data, recording change rates, and retaining early warning logs, enabling the retrospective analysis of leg circumference change trends and comparison of disease progression. With the communication module, data can be uploaded to remote terminals, facilitating remote monitoring by medical staff, home rehabilitation tracking, and batch screening, significantly improving nursing efficiency and reducing the workload of medical staff. Attached Figure Description

[0027] Figure 1 This is a flowchart of a leg-detection early warning system based on trend analysis. Detailed Implementation

[0028] Example 1: Please refer to Figure 1 The present invention provides a technical solution: A leg-measuring early warning system based on trend analysis includes a data acquisition module, a data preprocessing module, a rate of change calculation module, a trend analysis module, a risk assessment module, and an early warning output module. The data acquisition module is used to continuously collect users' leg circumference data; A data preprocessing module, connected to the data acquisition module, is used to filter and remove outliers from the acquired leg circumference data. The rate of change calculation module, connected to the data preprocessing module, is used to calculate the daily and weekly rate of change of leg circumference. The trend analysis module, connected to the data preprocessing module, is used to perform linear regression fitting on the preprocessed leg circumference data to obtain the leg circumference change trend. The risk assessment module is connected to the rate of change calculation module and the trend analysis module respectively, and is used to make a comprehensive risk assessment based on the daily rate of change, weekly rate of change and the trend of change. The early warning output module is connected to the risk determination module and is used to output corresponding early warning information according to the risk level.

[0029] It also includes a personalized baseline module for establishing a user's personal leg circumference baseline and calculating the rate of change based on the personal baseline; a data storage module for storing historical leg circumference data and early warning records; and a communication module for data interaction with remote terminals.

[0030] The system collects leg circumference data from users over a continuous 7-day period, calculates the daily average, and uses the multi-day average as the individual baseline value. This is used for subsequent rate of change calculation to avoid misjudgment of the general threshold due to individual body size differences; The data storage module stores daily raw leg circumference, preprocessed data, rate of change, EMA curve, risk level, and early warning records, supporting historical review and comparison; the communication module uses Bluetooth, WiFi, or wired network to achieve data interaction with hospital nursing systems, cloud platforms, and mobile terminals, supporting remote viewing and remote early warning.

[0031] The data acquisition module includes multiple sets of flexible sensors, a signal processing unit, and a body position detection unit. The flexible sensors are used to conform to the user's legs to collect leg circumference data. The signal processing unit is connected to the flexible sensors and is used to amplify and convert the sensor signals from analog to digital. The body position sensors are used to detect the user's body position information during measurement. It should be further explained that the flexible sensors are distributed at the position of the thigh circumference corresponding to 15cm above the patella and the position of the calf circumference corresponding to the maximum circumference of the calf. The flexible sensors fit the skin of the leg and collect the leg circumference signal in real time. The signal processing unit amplifies, filters and converts the analog signal output by the sensor into a digital leg circumference value. The body position detection unit uses a gyroscope sensor to identify the user's current body position, including lying, sitting, and standing positions, and synchronously sends the body position information to the data preprocessing module; The system continuously collects leg circumference data for 5 to 10 days, forming a leg circumference time series for the same time period each day, which is used for subsequent trend analysis and baseline establishment.

[0032] Example 2: This example is an improvement upon Example 1. Please refer to [link / reference]. Figure 1Specifically, the data preprocessing module performs noise reduction, outlier removal, and postural compensation on the collected raw leg circumference data. The steps are as follows: S1: Sliding window filtering: The sliding window filtering algorithm is used to take the average of the most recent 5-10 measurements as the effective data, thereby reducing the fluctuation of a single measurement. S2: 3σ criterion for outlier removal: Calculate the mean μ and standard deviation σ of historical leg circumference data. If a measurement value exceeds the interval [μ-3σ, μ+3σ], it is determined to be an outlier and removed, and the valid data from the previous time point is used to replace it. S3: Posture Compensation Correction: Based on the lying, sitting, and standing postures identified by the posture detection unit, a preset compensation coefficient is invoked. The leg circumference in the standing position is approximately 2% higher than that in the lying position. The leg circumference in the sitting position is approximately 1% higher than that in the lying position. The system uniformly corrects the leg circumference under different body positions to the supine baseline value to ensure data comparability.

[0033] The rate of change calculation module calculates the following indicators based on the preprocessed data: Daily relative change rate:

[0034] : The circumference of a specific part of the thigh (calf) on that day; The circumference of the leg at the same location and time the previous day is used to reflect changes in acute edema. A sudden increase suggests the risk of acute thrombosis.

[0035] Daily absolute change:

[0036] The unit is cm. In clinical practice, a difference in circumference of unilateral calf greater than 2 cm is considered abnormal.

[0037] Weekly relative rate of change:

[0038] Average leg circumference this week: Average leg circumference last week is used to assess progressive edema, chronic thrombosis, and abnormal cardiac and renal function.

[0039] Rate of change relative to baseline

[0040] It reflects the degree of change in current leg circumference relative to a healthy baseline and is suitable for long-term bed rest, postoperative rehabilitation, and home monitoring.

[0041] Absolute difference in leg circumference between both sides:

[0042] Two-sided relative difference rate:

[0043] in, The circumference of the left leg. The difference in circumference between the two legs is an important criterion for diagnosing deep vein thrombosis in the lower extremities.

[0044] Example 3: This example is an improvement on Example 1. Please refer to [link / reference]. Figure 1 Specifically, the trend analysis module is used to extract the edema development trend from continuous data, including linear regression analysis and exponential smoothing analysis (EMA). Linear regression was performed on the effective leg circumference data of the past 7 days, and the slope k of the fitted line was calculated. Calculate the exponential moving average of leg circumference: EMA(t) = α×Q+(1-α)×EMA(t-1) in, EMA(t): Exponential moving average on day t; α: Smoothing coefficient; Q: Measured leg circumference on day t; EMA(t-1): Exponential moving average on day t-1.

[0045] EMA daily change rate:

[0046] EMA cumulative increase over 3 consecutive days:

[0047] in, Smooth hind leg circumference on the same day, The previous day's smoothed value indicates a pathological edema trend when the EMA rises for three consecutive days and the cumulative increase is greater than 0.5 cm.

[0048] Example 4: This example is an improvement on Example 3. Please refer to [link / reference]. Figure 1 Specifically, the risk assessment module combines rate of change and trend indicators to conduct a three-level risk assessment: a. High risk: Satisfy any of the following conditions: Daily change rate > 2%; Weekly rate of change > 5%; The difference in leg circumference on one side is greater than 2cm.

[0049] b. Medium risk: Satisfy any of the following conditions: Daily change rate 1%–2%; Weekly change rate 3%–5%; The difference in leg circumference on one side is 1-2 cm.

[0050] c. Low risk: Simultaneously satisfy: Daily change rate <1%; Weekly rate of change <3%; The difference in leg circumference on one side is less than 1cm.

[0051] Trend escalation rule: If the linear regression slope k > 0.1 cm / day, it indicates that edema is continuously worsening. Currently classified as medium risk → upgraded to high risk; The current risk level is low; it has been upgraded to medium risk.

[0052] It should be added that the data acquisition module uses high-precision body temperature measurement to collect the temperature between the patient's legs and calculate the temperature difference. When a temperature difference exists, the trend escalation rule is to redefine the level of risk assessment based on the temperature difference and then upgrade the assessment accordingly. Absolute temperature difference between the two sides: in, Temperature of the skin on the left leg. The temperature of the skin on the right leg. The temperature difference between the two legs was considered, and a grading threshold was set. ΔT < 0.5℃: Normal 0.5℃ ≤ ΔT < 1.0℃: Medium risk ΔT ≥ 1.0℃: High risk When the absolute temperature difference between the two sides is determined to be normal, the original assessment level remains unchanged. When the absolute temperature difference between the two sides is determined to be medium risk, the original assessment level is upgraded by one level. When the absolute temperature difference between the two sides is determined to be high risk, it is directly upgraded to high risk, and the alarm information is directly output by the early warning output module. Furthermore, if the temperature difference continues to rise for three consecutive days, even if it does not exceed 1°C, the risk level can be raised by one level.

[0053] Example 5: This example is an improvement on Example 4. Please refer to [link / reference]. Figure 1 Specifically, the early warning output module executes a tiered early warning strategy based on the final risk level: No warning information is issued when the risk is low; When the risk level is medium, a local pop-up window and indicator light will be displayed. When a high risk occurs, a local alarm is output and a warning message is pushed to a remote terminal.

[0054] This invention performs sliding window preprocessing, daily / weekly change rate threshold determination, and linear regression trend fitting on the collected leg circumference data, balancing low computational cost and clinical applicability. It focuses on "personalized baseline calibration," "calculation of change rate for postural compensation," and "multi-dimensional risk fusion determination," which can effectively provide timely early warning based on the actual condition of the patient's lower limbs and meet the clinical needs for early screening and real-time monitoring of deep vein thrombosis in the lower limbs.

[0055] This article uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only for the purpose of helping to understand the method and core ideas of the present invention. The above descriptions are only preferred embodiments of the present invention. It should be noted that due to the limitations of textual expression, while there are objectively infinite specific structures, those skilled in the art can make several improvements, modifications, or changes without departing from the principles of the present invention, and can also combine the above technical features in an appropriate manner. These improvements, modifications, changes, or combinations, or the direct application of the inventive concept and technical solution to other situations without modification, should all be considered within the scope of protection of the present invention.

Claims

1. A leg-detection early warning system based on trend analysis, characterized in that: It includes a data acquisition module, a data preprocessing module, a rate of change calculation module, a trend analysis module, a risk assessment module, and an early warning output module; The data acquisition module is used to continuously collect users' leg circumference data; A data preprocessing module, connected to the data acquisition module, is used to filter and remove outliers from the acquired leg circumference data. The rate of change calculation module, connected to the data preprocessing module, is used to calculate the daily and weekly rate of change of leg circumference. The trend analysis module, connected to the data preprocessing module, is used to perform linear regression fitting on the preprocessed leg circumference data to obtain the leg circumference change trend. The risk assessment module is connected to the rate of change calculation module and the trend analysis module respectively, and is used to make a comprehensive risk assessment based on the daily rate of change, weekly rate of change and the trend of change. The early warning output module is connected to the risk determination module and is used to output corresponding early warning information according to the risk level.

2. The leg detection early warning system based on trend analysis according to claim 1, characterized in that: The data acquisition module includes multiple sets of flexible sensors, a signal processing unit, and a body position detection unit. The flexible sensors are used to collect leg circumference data by conforming to the user's legs. The signal processing unit is connected to the flexible sensors and is used to amplify and convert the sensor signals from analog to digital. The body position sensors are used to detect the user's body position information during measurement.

3. The leg detection early warning system based on trend analysis according to claim 1, characterized in that: The data preprocessing module is specifically used for: A sliding window filtering algorithm is used to take the average of the most recent 5-10 measurements as valid data. Outliers were removed using the 3σ criterion. The mean μ and standard deviation σ of the historical data were calculated, and measurements that exceeded the range of [μ-3σ, μ+3σ] were removed. Postural compensation is performed based on body position information to correct leg circumference data under different body positions to a unified benchmark value.

4. The leg detection early warning system based on trend analysis according to claim 1, characterized in that: The rate of change calculation module includes the following formulas: Daily relative change rate: ， Daily absolute change: ， Weekly relative rate of change: ， Rate of change relative to baseline ， in, : The circumference of a specific part of the thigh (calf) on that day; Leg circumference at the same location and time the previous day; Average leg circumference this week: Average leg circumference last week; Individual baseline values ​​(such as leg circumference at admission or in a healthy state); Absolute difference in leg circumference between both sides: ， Two-sided relative difference rate: ， in, The circumference of the left leg. The measurement is the circumference of the right leg.

5. The leg detection early warning system based on trend analysis according to claim 1, characterized in that: The trend analysis module is specifically used for: Linear regression was performed on the effective leg circumference data of the past 7 days, and the slope k of the fitted line was calculated. Calculate the exponential moving average of leg circumference: EMA(t) = α×Q+(1-α)×EMA(t-1), in, EMA(t): Exponential moving average on day t; α: Smoothing coefficient; Q: Measured leg circumference on day t; EMA(t-1): Exponential moving average on day t-1. The trend analysis module also includes the following formulas: EMA daily change rate: ， EMA cumulative increment: ， in, Smooth hind leg circumference on the same day, : The smoothed value of the previous day.

6. The leg detection early warning system based on trend analysis according to claim 1, characterized in that: The risk assessment module is specifically used for: A daily change rate > 2%, a weekly change rate > 5%, or a difference in leg circumference on one side > 2cm is considered high risk. When the daily change rate is between 1% and 2%, the weekly change rate is between 3% and 5%, or the difference in unilateral leg circumference is between 1 and 2 cm, it is judged as medium risk; When the daily change rate is <1%, the weekly change rate is <3%, and the difference in leg circumference on one side is <1cm, it is judged as low risk; When the slope k of the fitted line is greater than 0.1 cm / day, if the current risk level is medium risk, it will be upgraded to high risk; if the current risk level is low risk, it will be upgraded to medium risk.

7. The leg detection early warning system based on trend analysis according to claim 1, characterized in that: The early warning output module is specifically used for: No warning information is output when the risk is low; local pop-up window and indicator light prompts are output when the risk is medium; local alarm is output and warning information is pushed to remote terminals when the risk is high.

8. The leg detection early warning system based on trend analysis according to claim 1, characterized in that: It also includes a personalized baseline module, which is used to establish a user's personal leg circumference baseline and calculate the rate of change based on the personal baseline; The data storage module is used to store historical leg circumference data and early warning records; The communication module is used for data interaction with remote terminals.

Images