Limb restraint strap self-adaptive pressure adjusting device and method

Abstract

The invention relates to the technical field of medical instruments, in particular to a limb restraint strap self-adaptive pressure adjusting device and method.The limb restraint strap self-adaptive pressure adjusting device comprises a pressure sensing unit, a multi-stage pressure sensor detects the pressure value between a limb of a patient and a restraint strap and generates a pressure signal, and a pressure signal processing module filters the pressure signal and identifies the activity state of the patient; the pressure data processing module performs multi-window pressure analysis based on the processed pressure data, identifies a pressure change trend, predicts a pressure change and generates a prediction result, and the adaptive threshold adjustment module dynamically adjusts a pressure threshold based on patient activity intensity, constraint duration and muscle tension. And the decision control module generates a pressure adjustment control instruction based on the prediction result and the adjusted pressure threshold value, and the pressure adjustment execution module comprises an air bag, an inflation device and an exhaust device and achieves dynamic adjustment of the tightness of the restraint strap.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular to a limb restraint belt adaptive pressure regulating device and method, which are used for intelligent pressure management when limb restraints are implemented on patients in a medical environment. Background Art

[0002] Limb restraints are a commonly used intervention in clinical practice to protect patient safety, particularly for patients who are unconscious, agitated, or at risk of self-harm. Traditional limb restraints typically employ a fixed pressure design. Once installed, the pressure is difficult to dynamically adjust based on the patient's condition, which presents a series of problems. On the one hand, overly tight restraints can lead to localized blood circulation obstruction, skin damage, and even nerve damage. On the other hand, overly loose restraints may not achieve the desired protective effect, leaving the patient vulnerable to escape and posing a safety hazard.

[0003] Existing restraints rely primarily on regular inspections and manual adjustments by medical staff, which not only increases their workload but also hinders timely response to changes in patient status due to the time between inspections. While some restraints with simple pressure detection capabilities are available on the market, most offer only passive alarms and lack intelligent predictive analysis and automatic adjustment capabilities, failing to meet clinical demands for safe, comfortable, and effective limb restraints. Summary of the Invention

[0004] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a limb restraint belt adaptive pressure adjustment device and method, which can monitor the pressure changes between the patient's limbs and the restraint belt in real time, predict the pressure trend, and automatically adjust the tightness of the restraint belt, thereby minimizing the discomfort and potential harm to the patient while ensuring the restraint effect.

[0005] The present invention provides a limb restraint belt adaptive pressure regulating device, comprising:

[0006] a pressure sensing unit, comprising a multi-stage pressure sensor, wherein the multi-stage pressure sensor comprises a first pressure sensor, a second pressure sensor, and a third pressure sensor, for detecting a pressure value between the patient's limb and the restraint belt and generating a pressure signal;

[0007] a pressure signal processing module, electrically connected to the pressure sensing unit, configured to receive the pressure signal, filter the pressure signal, identify the patient's activity state, and generate processed pressure data;

[0008] a pressure change prediction module, electrically connected to the pressure signal processing module, for performing multi-window pressure analysis based on the processed pressure data, identifying pressure change trends, predicting future pressure changes, and generating prediction results;

[0009] an adaptive threshold adjustment module, electrically connected to the pressure signal processing module and the pressure change prediction module, configured to dynamically adjust the pressure threshold based on the patient's activity intensity, restraint duration, and muscle tension, and generate an adjusted pressure threshold;

[0010] a decision control module, electrically connected to the pressure change prediction module and the adaptive threshold adjustment module, and configured to generate a pressure regulation control instruction based on the prediction result and the adjusted pressure threshold;

[0011] The pressure regulation execution module is electrically connected to the decision control module and includes an airbag, an inflation device and an exhaust device. It is used to receive the pressure regulation control instruction and adjust the air pressure in the airbag to achieve dynamic adjustment of the tightness of the restraint belt.

[0012] Preferably, the operating range of the multi-stage pressure sensor is:

[0013] The first pressure sensor monitors pressure changes within a range from 0 to a first threshold;

[0014] The second pressure sensor monitors pressure changes within a range from a first threshold to a second threshold;

[0015] The third pressure sensor monitors for pressure changes exceeding a second threshold.

[0016] Preferably, the pressure signal processing module includes:

[0017] a signal preprocessing unit, configured to perform digital filtering and baseline drift correction on the pressure signal;

[0018] An outlier identification unit, used to identify and remove abnormal data points;

[0019] A signal segmentation unit, for segmenting a continuous signal into an active segment and a rest segment;

[0020] A feature extraction unit, used to extract time domain features of the pressure waveform;

[0021] An activity recognition unit is used to distinguish the patient's active activities from external interference based on the time domain features.

[0022] Preferably, the pressure change prediction module includes:

[0023] Short-window analysis unit, used to analyze pressure data within 3 to 5 seconds and identify immediate pressure change trends;

[0024] The middle window analysis unit is used to analyze pressure data within 30 to 60 seconds and identify activity rhythms and patterns;

[0025] Long-window analysis unit, used to analyze pressure data within 5 to 10 minutes and learn patient activity patterns;

[0026] a trend matching unit, which matches real-time data with historical patterns;

[0027] The prediction calculation unit is used to calculate the probability and time of the pressure exceeding the threshold in the next 5 to 10 seconds.

[0028] Preferably, the adaptive threshold adjustment module includes:

[0029] A basic threshold setting unit, used to set initial threshold parameters according to the patient's body shape and limb circumference;

[0030] an activity coefficient calculation unit, for calculating an activity intensity coefficient by analyzing the frequency and amplitude of pressure fluctuations;

[0031] a time factor calculation unit, configured to calculate a time compensation factor based on a constraint duration;

[0032] A muscle tension assessment unit, used to assess muscle tension through pressure waveform characteristics;

[0033] The comprehensive threshold calculation unit is used to dynamically calculate the trigger thresholds of sensors at all levels based on the above parameters.

[0034] Preferably, the decision control module includes:

[0035] Safety assurance decision-making unit, used to monitor the ultimate pressure value and the duration of high pressure, and immediately generate a decompression command once it approaches a dangerous value;

[0036] Comfort optimization decision unit, used to generate adjustment instructions based on pressure distribution uniformity and long-term static pressure state;

[0037] A restraint effect maintenance decision unit, used to generate basic restraint force adjustment instructions based on activity intensity trends and escape attempt frequencies;

[0038] A decision priority management unit is used to arbitrate the instructions generated by each decision unit in the order of safety, comfort, and restraint effect;

[0039] The execution instruction generation unit is used to convert the arbitration decision into specific control instructions.

[0040] Preferably, the pressure regulation execution module further includes:

[0041] Air pressure sensor, used to detect the air pressure in the airbag;

[0042] A booster pump for inflating the airbag;

[0043] Pressure regulating valve, used to accurately adjust the pressure of the airflow entering the airbag;

[0044] A pressure relief valve, used to control the airbag deflation rate;

[0045] The execution feedback unit is used to collect the airbag pressure change data and feed it back to the decision control module.

[0046] As an advantage, it also includes:

[0047] an alarm module, electrically connected to the decision control module, for issuing an audible and visual alarm signal when the pressure exceeds a safe range or a system failure occurs;

[0048] A data storage module, electrically connected to the pressure signal processing module, for recording historical pressure data and system operating status;

[0049] Communication interface module, used for exchanging data with external medical information systems.

[0050] As an advantage, it also includes:

[0051] An intermediate connecting device, used to connect the restraint belt and the airbag, the intermediate connecting device comprising a belt body, an adjustment mechanism, a connecting device and a fixing mechanism;

[0052] The airbag is arranged in the middle of the belt body;

[0053] The adjustment mechanism is connected to the restraint belt;

[0054] The fixing mechanism is connected to the airbag;

[0055] The connecting device is used to connect the belt body and the airbag.

[0056] The adaptive pressure adjustment method of a limb restraint belt comprises:

[0057] Acquisition step, collecting the pressure value between the patient's limb and the restraint belt by using a multi-stage pressure sensor and generating a pressure signal;

[0058] a pressure signal processing step of filtering the pressure signal, identifying the patient's activity state, and generating processed pressure data;

[0059] a pressure change prediction step, performing multi-window pressure analysis based on the processed pressure data, identifying pressure change trends, predicting future pressure changes, and generating prediction results;

[0060] an adaptive threshold adjustment step, dynamically adjusting the pressure threshold based on the patient's activity intensity, restraint duration, and muscle tension, and generating an adjusted pressure threshold;

[0061] a decision control step of generating a pressure regulation control instruction based on the prediction result and the adjusted pressure threshold;

[0062] The pressure adjustment execution step receives the pressure adjustment control instruction and adjusts the air pressure in the airbag to achieve dynamic adjustment of the tightness of the restraint belt.

[0063] The present invention has the following beneficial effects:

[0064] 1. Intelligent perception and prediction capabilities: Through a multi-level pressure sensing system and prediction algorithm, it can capture the patient's limb movement status in real time and predict pressure change trends, achieving early response and significantly improving restraint safety.

[0065] 2. Adaptive adjustment capability: Dynamically adjust the pressure threshold based on the patient's individual characteristics and activity status to achieve precise restraint management and avoid the risks of over-restraint or under-restraint.

[0066] 3. Improve patient comfort: Through real-time pressure monitoring and precise control, unnecessary restraint pressure is reduced, the incidence of restraint-related complications is lowered, and the patient comfort experience is improved.

[0067] 4. Reduce the burden on medical staff: The automated pressure regulation system reduces the need for manual intervention, optimizes the allocation of medical resources, and improves the efficiency of restraint management.

[0068] 5. Comprehensive safety assurance mechanism: Multiple safety monitoring and automatic intervention functions can promptly identify and respond to potential risks, enhancing overall safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 This is a system structure diagram of the limb restraint belt adaptive pressure adjustment device of the present invention.

[0070] Figure 2 Schematic diagram of the layout of the pressure sensing unit of the present invention.

[0071] Figure 3 Schematic diagram of the pressure signal processing flow of the present invention.

[0072] Figure 4 This is a schematic diagram of multi-window pressure data analysis of the present invention.

[0073] Figure 5 Schematic diagram of the adaptive threshold adjustment process of the present invention.

[0074] Figure 6 Schematic diagram of the hierarchical structure of the decision control module of the present invention.

[0075] Figure 7 It is a structural diagram of the pressure regulation execution module of the present invention.

[0076] Figure 8 It is a structural schematic diagram of the intermediate connecting device of the present invention.

[0077] Figure 9 Schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0078] Please refer to the attached Figure 1-9 The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, but the embodiments of the present invention are not limited thereto.

[0079] like Figure 1 As shown, the adaptive pressure adjustment device for a limb restraint belt provided by the present invention includes: a pressure sensing unit 1, a pressure signal processing module 2, a pressure change prediction module 3, an adaptive threshold adjustment module 4, a decision control module 5, a pressure adjustment execution module 6, an alarm module 7, a data storage module 8, a communication interface module 9 and an intermediate connection device 10.

[0080] like Figure 2 As shown, the pressure sensing unit 1 comprises a multi-stage pressure sensor, comprising a first pressure sensor 11, a second pressure sensor 12, and a third pressure sensor 13. These sensors are arranged in a three-point triangle at key locations where the restraint strap contacts the patient's limb. Preferably, these sensors are located on the inner, outer, and central sides of the restraint strap, ensuring omnidirectional sensing of pressure changes in different directions. This layout is particularly suitable for capturing the varying pressure distributions generated by a patient's limb movements in different directions, such as when attempting to rotate their wrist or raise their arm.

[0081] In a preferred embodiment of the present invention, the working range of the multi-level pressure sensor is divided as follows: the first pressure sensor 11 monitors pressure changes in the range of 0 to a first threshold value, and the first threshold value is preferably set to 2.0 kPa, which is mainly used to capture early signs of slight activity of the patient, such as finger micro-movement or slight wrist rotation; the second pressure sensor 12 monitors pressure changes in the range of the first threshold value to the second threshold value, and the second threshold value is preferably set to 4.0 kPa, which is mainly used to identify obvious limb activities of the patient, such as trying to lift an arm or leg; the third pressure sensor 13 monitors pressure changes exceeding the second threshold value, which is mainly used to detect violent struggles of the patient, such as forcefully pulling the restraint belt or rapidly moving the limbs.

[0082] Each sensor stage utilizes a thin-film piezoresistive design with built-in signal conditioning circuitry to eliminate the effects of temperature drift. Sensitivity parameters vary: the first pressure sensor 11 has an accuracy of ±0.05 kPa, suitable for detecting minute pressure changes; the second pressure sensor 12 has an accuracy of ±0.1 kPa, balancing sensitivity and stability; and the third pressure sensor 13 has an accuracy of ±0.2 kPa, ensuring reliable operation even with large pressure changes. The sensor data sampling frequency is set to 100 Hz, meaning data is collected every 10 milliseconds, enabling millisecond-level pressure change responses—sufficient to capture the first signs of a patient's limb movement.

[0083] like Figure 3 As shown, the pressure signal processing module 2 is electrically connected to the pressure sensing unit 1, and is used to receive the pressure signal, filter the pressure signal, identify the patient's activity state, and generate processed pressure data.

[0084] In one embodiment of the present invention, the pressure signal processing module 2 includes: a signal preprocessing unit 21 , an outlier identification unit 22 , a signal segmentation unit 23 , a feature extraction unit 24 and an activity identification unit 25 .

[0085] The signal preprocessing unit 21 performs digital low-pass filtering and baseline drift correction on the received pressure signal. The digital low-pass filter uses a Butterworth filter with a cutoff frequency set to 10Hz, which can effectively eliminate high-frequency interference such as environmental vibration. In a clinical environment, this filtering setting can filter out interference caused by equipment vibration in the ward or operation of nearby beds, while retaining the patient's true activity signal. Baseline drift correction tracks sensor drift through a 60-second long-term window sliding average method and compensates in real time to ensure signal stability. This is particularly important for long-term constraints, because the sensor may drift during continuous use, affecting monitoring accuracy.

[0086] The outlier identification unit 22 uses statistical methods to identify and remove anomalous data points. This unit uses the 3σ principle to define outliers: when a data point deviates from the mean by more than three standard deviations, it is identified as an outlier and removed, preventing the system from falsely triggering. For example, if a caregiver accidentally touches the restraint while examining a patient, this can cause a transient pressure spike. This unit can identify and remove this interference, preventing the system from mistakenly interpreting it as patient movement and decompressing the patient.

[0087] The signal segmentation unit 23 divides the continuous pressure signal into active and resting segments. The segmentation algorithm is based on a dual judgment based on the pressure change rate and duration: when the pressure change rate exceeds 0.2 kPa / s and the duration exceeds 0.5 seconds, it is identified as an active segment; otherwise, it is identified as a resting segment. This dual judgment mechanism is particularly suitable for distinguishing active patient movement from external interference, such as the difference between a brief medical procedure and a patient's sustained attempts to move a limb.

[0088] The feature extraction unit 24 extracts time-domain features from the segmented signal segments, including peak value, mean value, coefficient of variation, rise time, and duration. These features constitute the feature vector of the patient's activity pattern, providing a basis for subsequent analysis. For example, a patient's limb twitching due to pain typically manifests as a brief, large pressure peak, while persistent activity due to restlessness manifests as a long-lasting pressure fluctuation with a high coefficient of variation.

[0089] The activity recognition unit 25 distinguishes between the patient's active movements and external interference based on the extracted feature vectors. Active movements typically exhibit regular fluctuations, while external interference often exhibits sudden changes. Using a pattern matching algorithm, this unit achieves approximately 85% accuracy in activity / interference classification. For example, when a patient repeatedly attempts to adjust their limb position due to discomfort, this produces periodic pressure fluctuations; whereas, temporary manipulation of restraints by medical personnel often results in irregular pressure variations.

[0090] like Figure 4 As shown, the pressure change prediction module 3 is electrically connected to the pressure signal processing module 2, and is used to perform multi-window pressure analysis based on the processed pressure data, identify pressure change trends, predict future pressure changes, and generate prediction results.

[0091] In one embodiment of the present invention, the pressure change prediction module 3 includes: a short window analysis unit 31 , a medium window analysis unit 32 , a long window analysis unit 33 , a trend matching unit 34 and a prediction calculation unit 35 .

[0092] The short-window analysis unit 31 analyzes the pressure data within 3 to 5 seconds and identifies the immediate pressure change trend. This unit uses the linear regression method to calculate the short-term pressure slope. The slope formula is:

[0093] ,

[0094] in, is the short-time pressure slope, in kPa / s; is the time point, the unit is s; is the pressure value at the corresponding time point, in kPa; is the time average value, in seconds; is the average pressure value in kPa; The number of data points in a short window is generally 300-500 points (based on a 100Hz sampling frequency).

[0095] when When the pressure is greater than the threshold of 0.3kPa / s, it is determined to be a rapid pressure increase trend, which usually means that the patient is starting a significant activity. For example, when a bedridden patient suddenly tries to lift his restrained arm, Values ​​can quickly exceed thresholds, and the system can identify this trend before pressure reaches levels that could cause discomfort.

[0096] The middle window analysis unit 32 analyzes the pressure data within 30 to 60 seconds to identify the activity rhythm and pattern. This unit extracts the frequency characteristics of the pressure fluctuations through fast Fourier transform (FFT):

[0097] ,

[0098] in, It is a complex value representing the amplitude and phase of the frequency component. is the pressure data sequence, the unit is kPa; is the number of data points in the middle window, generally 3000-6000 points; is the exponential term of Fourier transform, where is an imaginary unit; Represents the index of the frequency component, ranging from 0 to N-1; Represents the index in the time series, ranging from 0 to N-1.

[0099] By analyzing the primary frequency components, the system can identify periodic characteristics of patient activity, such as the approximately 0.3Hz fluctuations caused by breathing and the approximately 0.05-0.1Hz fluctuations caused by turning over. This is particularly important for predicting repetitive activity. For example, if the system identifies that a patient attempts to move their restrained leg every 60-90 seconds, it can prepare appropriate decompression measures in advance to alleviate the discomfort caused by the patient's struggle.

[0100] The long-window analysis unit 33 analyzes pressure data over a 5-10 minute period to learn the patient's activity patterns. This unit uses statistical methods to construct a patient activity pattern library, recording typical pressure feature sequences under different time periods and conditions. This learning mechanism is particularly important under long-term constraints, as patients often develop specific activity patterns. For example, some patients may exhibit a gradually increasing frequency of restless activity before the completion of an infusion.

[0101] The trend matching unit 34 matches the real-time data with the historical patterns. The matching degree is calculated using the Dynamic Time Warping (DTW) algorithm, which can effectively handle the nonlinear alignment problem of time series:

[0102] ,

[0103] Where DTW(X,Y) is the dynamic time warping distance between two time series X and Y, indicating the similarity between the two series; X and Y are the real-time data series and the series in the pattern library, respectively, both of which are time series of pressure values; is the Euclidean distance between corresponding points, that is is the subsequence after removing the first element; It means taking the minimum value of three cases, corresponding to three possible matching paths.

[0104] The match threshold is set at 0.75, with values ​​exceeding this threshold considered a close match. For example, if the system detects that the current pressure change pattern is highly similar to the pattern of the patient's previous attempt to remove the restraints, it will take preemptive measures, such as appropriate decompression and notifying medical staff, to prevent the patient from excessive struggling and causing skin abrasions.

[0105] The prediction calculation unit 35 calculates the probability and estimated time of the pressure exceeding the threshold in the next 5 to 10 seconds based on the multi-window analysis and matching results. The prediction probability model combines the linear trend and pattern matching results:

[0106] ,

[0107] in, The probability that the pressure exceeds the threshold in the next 5 to 10 seconds, ranging from 0 to 1; is the probability of exceeding the threshold value calculated based on the linear trend, ranging from 0 to 1; is the probability of exceeding the threshold value calculated based on pattern matching, ranging from 0 to 1; and is the weight coefficient, the preferred values ​​are 0.6 and 0.4 respectively, and satisfy .

[0108] when When the pressure is greater than 0.85, the system initiates decompression 0.5 seconds earlier. This early response is crucial for protecting patient safety, especially for those who may injure themselves due to intense struggle. In practice, this mechanism can reduce peak pressure by approximately 25-30%, significantly improving patient comfort and reducing restraint-related complications.

[0109] like Figure 5As shown, the adaptive threshold adjustment module 4 is electrically connected to the pressure signal processing module 2 and the pressure change prediction module 3, and is used to dynamically adjust the pressure threshold based on the patient's activity intensity, restraint duration and muscle tension, and generate an adjusted pressure threshold.

[0110] In one embodiment of the present invention, the adaptive threshold adjustment module 4 includes: a basic threshold setting unit 41 , an activity coefficient calculation unit 42 , a time factor calculation unit 43 , a muscle tension evaluation unit 44 and a comprehensive threshold calculation unit 45 .

[0111] The basic threshold setting unit 41 sets the initial threshold parameters according to the patient's body shape and limb circumference. The basic threshold calculation formula is:

[0112] ,

[0113] in, is the basic pressure threshold, in kPa; is the limb circumference in cm; is the proportionality coefficient, the preferred value is 0.07 kPa / cm; As the basic constant, the preferred value is 1.5 kPa.

[0114] This formula, derived from clinical experience, ensures effective restraint without excessive pressure on the patient. For example, for a patient with a wrist circumference of 16 cm, the calculated baseline threshold is approximately 2.6 kPa; for an obese patient, the circumference may reach 22 cm, corresponding to a baseline threshold of approximately 3.0 kPa. This differentiated setting ensures that patients of different body types receive appropriate restraint pressure.

[0115] The activity coefficient calculation unit 42 calculates the activity intensity coefficient by analyzing the pressure fluctuation frequency and amplitude. The activity intensity coefficient calculation formula is:

[0116] ,

[0117] in, is the activity intensity coefficient, dimensionless, ranging from 0 to 1.0; is the calculation window, preferably 30s; is the pressure change rate, in kPa / s; is the integral variable, representing time, in units of s; is the current time point, in seconds; It represents the integral of the absolute value of the pressure change rate within the time window T, and its dimension is kPa; The reference pressure is preferably 3.0 kPa. The greater the activity intensity, the closer the coefficient value is to 1.0.

[0118] For example, for a patient who is resting quietly, The value is usually in the range of 0.1-0.2; while patients who are restless and frequently try to move, The value may reach 0.7-0.9. The system dynamically adjusts its response sensitivity to patient activity based on this coefficient, lowering the trigger threshold when the activity intensity is high and increasing the system's response speed.

[0119] The time factor calculation unit 43 calculates the time compensation factor based on the constraint duration. The time factor calculation formula is:

[0120] ,

[0121] in, is the time factor, dimensionless, with a maximum value of 0.8; The duration of the constraint, in minutes; is the reference time, preferably 120 min; min means taking the smaller value of the two parameters.

[0122] As the restraint time increases, this factor increases, and the system will reduce the restraint pressure accordingly, reducing the discomfort and risk of complications caused by long-term restraint. For example, in the first 30 minutes of restraint, The value is 0.25; after 2 hours of restraint, When the value reaches the upper limit of 0.8, the system will significantly reduce the restraint pressure, giving the patient more room to move and preventing pressure sores and circulatory disorders caused by long-term restraint.

[0123] The muscle tension evaluation unit 44 evaluates the degree of muscle tension based on the pressure waveform characteristics. The muscle tension coefficient is calculated based on the high-frequency components of the pressure waveform:

[0124] ,

[0125] in, is the muscle tension coefficient, dimensionless, ranging from 0 to 1.2; is the pressure signal spectrum, indicating the frequency is The amplitude of the components; The lower limit of the muscle tremor frequency band is set to 2 Hz; The upper limit of the muscle tremor frequency band is set at 8 Hz; Indicates the frequency range arrive The sum of the absolute values ​​of the spectral components between ; Indicates from 0 to The sum of the absolute values ​​of all spectral components between ; is the correction factor, preferably 1.5.

[0126] The greater the muscle tension, the closer the coefficient is to 1.2. In clinical applications, this parameter is very important for identifying patients with muscle tension caused by pain, fear or drug effects. For example, patients with postoperative pain often show higher muscle tension. The value may be in the range of 0.8-1.0; while in patients receiving sedation, The value is usually low, around 0.2-0.4. The system will adjust the restraint force according to muscle tension, appropriately reducing the restraint force for patients with high muscle tension to prevent discomfort and blood circulation problems caused by continuous muscle tension.

[0127] The integrated threshold calculation unit 45 integrates the above parameters to dynamically calculate the trigger thresholds of each sensor level. The dynamic threshold calculation formula is:

[0128] ,

[0129] in, is the final dynamic threshold, in kPa; is the basic threshold, preferably 3.5 kPa; is the activity intensity coefficient; is the time factor; is the muscle tension coefficient; 、 and is the weight coefficient, and the preferred values ​​are 0.6, 0.3 and 0.4 respectively.

[0130] The system automatically recalculates the threshold parameters every 30 seconds to ensure that the restraint force is always at the optimal state. For example, for an agitated patient who has just been restrained (A=0.8, =0.1, M=0.9), the calculated dynamic threshold is about 4.38kPa; and for a stable patient who has been restrained for 2 hours (A=0.2, =0.8, M=0.3), with a dynamic threshold of only 4.07 kPa. This dynamic adjustment ensures that patients in different states receive the most appropriate restraint pressure, ensuring safety while minimizing discomfort.

[0131] like Figure 6 As shown, the decision control module 5 is electrically connected to the pressure change prediction module 3 and the adaptive threshold adjustment module 4, and is used to generate a pressure regulation control instruction based on the prediction result and the adjusted pressure threshold.

[0132] In one embodiment of the present invention, the decision control module 5 includes: a safety assurance decision unit 51 , a comfort optimization decision unit 52 , a constraint effect maintenance decision unit 53 , a decision priority management unit 54 and an execution instruction generation unit 55 .

[0133] The safety assurance decision unit 51 monitors the ultimate pressure value and the duration of high pressure, and immediately generates a decompression instruction once it approaches the dangerous value. The ultimate safety pressure set by this unit is 6.0kPa, and the continuous high pressure safety limit is 10 minutes. When the pressure is detected to exceed 5.7kPa or is continuously higher than 4.5kPa for more than 8 minutes, an early warning decompression is triggered; when the pressure is detected to exceed the ultimate value or the duration exceeds the safety limit, emergency decompression is immediately performed. These threshold settings are based on clinical research data and can effectively prevent the risk of tissue damage caused by restraint pressure.

[0134] The comfort optimization decision unit 52 generates adjustment instructions based on the pressure distribution uniformity and the long-term static pressure state. The pressure distribution uniformity is measured by the standard deviation of the three sensor readings, and the standard deviation calculation formula is:

[0135] ,

[0136] in, is the standard deviation of pressure distribution, in kPa; For the The pressure reading of each sensor is in kPa; is the average of the pressure readings of the three sensors, in kPa; It represents the sum of the squares of the differences between the three sensor pressure readings and the average value.

[0137] The standard deviation threshold is set at 0.8 kPa. When the standard deviation exceeds the threshold, the airbag inflation state is adjusted to equalize pressure at all points. For prolonged periods of static pressure (over 30 minutes), the unit will make fine adjustments according to a pre-set schedule (reducing pressure by 10% every 30 minutes, then restoring pressure for approximately 5 seconds) to prevent the formation of localized pressure ulcers. For example, for a patient restrained at the right wrist, if the pressure at the inner sensor is significantly higher than at other locations, the system will appropriately reduce the overall pressure and adjust the airbag inflation method to improve pressure distribution.

[0138] The restraint effect maintenance decision unit 53 generates a basic restraint force adjustment instruction based on the activity intensity trend and the frequency of the breakaway attempt. The activity intensity increase index calculation formula is:

[0139] ,

[0140] in, Add an exponent to the activity, dimensionless; The current activity intensity, that is, the activity intensity coefficient in the last 5 minutes; The average activity intensity in the past 30 minutes. When it is greater than 0.3, it is judged that the activity intensity has increased significantly, and the system will appropriately increase the basic restraint force to maintain the restraint effect. For example, when the patient starts to frequently try to move from a quiet state, The value may rise rapidly from close to 0 to above 0.5. At this time, the system will appropriately increase the restraint force to ensure the restraint effect, and notify medical staff to evaluate the cause of the patient's condition change.

[0141] The decision priority management unit 54 arbitrates the instructions generated by each decision unit based on the priorities of safety, comfort, and restraint effectiveness. The priority order is: safety assurance (highest priority) > comfort optimization > restraint effectiveness maintenance. When different decision units generate conflicting instructions, the higher-priority instruction overrides the lower-priority instruction. For example, if the comfort optimization unit recommends reducing pressure, while the restraint effectiveness maintenance unit recommends increasing pressure, the system will prioritize the comfort optimization suggestion. However, if the safety assurance unit issues a decompression instruction, it will be executed immediately, regardless of the decisions of the other units.

[0142] The execution instruction generation unit 55 converts the arbitrated decision into a specific control instruction, including the inflation / deflation rate, target pressure value, and adjustment duration. The control instruction format is a triplet (P_target, R_adjust, T_duration), which represents the target pressure (kPa), the adjustment rate (kPa / s), and the adjustment duration (s), respectively. For example, the instruction (3.2, 0.5, 4) indicates adjusting to a target pressure of 3.2 kPa at a rate of 0.5 kPa / s over 4 seconds. This precise control ensures a smooth and controllable pressure adjustment process, avoiding discomfort to the patient caused by sudden pressure changes.

[0143] like Figure 7 As shown, the pressure regulation execution module 6 is electrically connected to the decision control module 5, and includes an airbag 61, an inflation device 62 and an exhaust device 63, which is used to receive pressure regulation control instructions and adjust the air pressure in the airbag to achieve dynamic adjustment of the tightness of the restraint belt.

[0144] In one embodiment of the present invention, the pressure regulation execution module 6 further includes: an air pressure sensor 64 , a booster pump 65 , a pressure regulating valve 66 , a pressure relief valve 67 and an execution feedback unit 68 .

[0145] Air pressure sensor 64 monitors the air pressure within airbag 61 with a sampling frequency of 50 Hz (sampling every 20 milliseconds) and an accuracy of ±0.1 kPa, ensuring accurate pressure control. In clinical applications, this high-precision monitoring ensures that restraint pressure remains within a safe and effective range. Crucially, it enables timely detection of abnormalities such as airbag leaks.

[0146] Booster pump 65 is a diaphragm pump driven by a miniature DC motor. It has a maximum flow rate of 1.5 L / min, a maximum pressure of 10 kPa, and consumes less than 0.5 W. It is used to inflate airbag 61. The pump's compact size (approximately 30 mm × 25 mm × 15 mm) makes it suitable for integration into portable restraint devices. Its low power consumption enables the device to operate for extended periods without frequent recharging.

[0147] Pressure regulating valve 66 is an electromagnetic proportional control valve with a response time of less than 100ms and an adjustment accuracy of ±0.05kPa. It is used to precisely adjust the pressure of the airflow entering the airbag 61. This precise control ensures smooth adjustment of the restraint pressure, avoiding the discomfort caused by sudden pressure changes, and is particularly suitable for sensitive patients.

[0148] Pressure relief valve 67 is a normally closed electromagnetic valve with an opening time of less than 50ms. It controls the deflation rate of airbag 61. The system has three deflation rate settings: fast (2.0kPa / s), medium (1.0kPa / s), and slow (0.5kPa / s), designed for emergency decompression, routine adjustments, and fine-tuning, respectively. For example, if the patient is detected struggling violently or the pressure exceeds the safety threshold, the system activates fast deflation mode to rapidly reduce the restraint pressure. For routine fine-tuning, slow mode is used to ensure a comfortable and smooth adjustment process.

[0149] The execution feedback unit 68 collects pressure change data from the airbag 61 and feeds it back to the decision-making control module 5, forming a closed-loop control system. This feedback data, including the actual pressure value, pressure change rate, and execution delay, is used for system self-calibration and performance optimization. This closed-loop design ensures that the system can accurately execute the intended adjustments and adapt to the effects of varying environmental conditions (such as temperature fluctuations) and changes in patient status.

[0150] In a further embodiment of the present invention, the device further includes an alarm module 7 , a data storage module 8 and a communication interface module 9 .

[0151] Alarm module 7 is electrically connected to decision-making and control module 5 and is used to issue audible and visual alarm signals when pressure exceeds the safe range or a system failure occurs. Alarm module 7 has three levels of alarm: level one (yellow indicator light, pressure approaches the warning value); level two (orange indicator light and intermittent beeping, pressure exceeds the warning value); and level three (flashing red indicator light and continuous beeping, system failure or pressure exceeding the limit). In clinical applications, this multi-level alarm mechanism allows medical staff to promptly understand the patient's restraint status and take appropriate measures based on the degree of urgency.

[0152] The data storage module 8 is electrically connected to the pressure signal processing module 2 and is used to record historical pressure data and system operating status. With a storage capacity of 32MB, it can record approximately one week of pressure data using a compression algorithm and supports data export and retrospective analysis. This historical data is valuable for assessing patient restraint conditions, optimizing restraint plans, and managing medical quality.

[0153] Communication interface module 9 uses the Bluetooth Low Energy (BLE) protocol for data exchange with external medical information systems. It boasts a transmission rate of 1 Mbps and a range of up to 10 meters, supporting remote monitoring and parameter configuration. This allows medical staff to monitor patient restraint status in real time via mobile devices without frequent visits to the ward. It also facilitates remote adjustment of system parameters, improving equipment maintenance efficiency.

[0154] like Figure 8 As shown, the intermediate connecting device 10 is used to connect the restraint belt and the airbag 61 , and includes a belt body 101 , an adjustment mechanism 102 , a connecting device 103 and a fixing mechanism 104 .

[0155] The belt body 101 is made of medical-grade silicone material with good biocompatibility and elasticity, and is provided with an airbag 61 in the middle. This material selection ensures that long-term restraint will not cause skin allergic reactions, while the soft texture reduces the discomfort of the restraint belt contacting the skin.

[0156] Adjustment mechanism 102 utilizes an adjustable buckle design, connected to the restraint belt, and can be initially adjusted based on the patient's limb circumference. This design allows the device to adapt to the needs of patients of different body types, from children to adults, from thin to obese patients, all of whom can obtain the appropriate initial restraint state.

[0157] The securing mechanism 104 utilizes a resilient slot design, connected to the airbag 61, to prevent displacement of the airbag 61 during use. This is crucial for ensuring the stability of long-term restraints, particularly for active patients, as it avoids uneven pressure distribution and inaccurate monitoring caused by airbag displacement.

[0158] The connecting device 103 uses a quick connector design to connect the belt body 101 and the airbag 61, which is convenient for disassembly and maintenance. This modular design greatly improves the practicality of the device, allowing medical staff to quickly replace the airbag or belt body as needed, and also facilitates the cleaning and disinfection process of the device.

[0159] like Figure 9 As shown, the present invention also provides a method for adaptively adjusting pressure of a limb restraint belt, comprising the following steps:

[0160] Acquisition Steps: Multi-stage pressure sensors collect pressure between the patient's limb and the restraints, generating a pressure signal. The system samples at a 100Hz frequency to capture even subtle pressure changes. In clinical applications, this high-frequency sampling is crucial for early detection of patient movement, especially for patients who require close monitoring due to risk of agitation.

[0161] Pressure signal processing: The pressure signal is filtered to identify the patient's activity status and generate processed pressure data. This step first performs digital low-pass filtering to remove interference, then identifies and removes outliers. The continuous signal is then segmented into active and resting segments, pressure waveform features are extracted, and finally, active patient activity is distinguished from external interference. This processing flow ensures that the system can accurately identify the patient's true activity status, regardless of environmental interference or medical procedures.

[0162] Pressure Change Prediction Step: Multi-window pressure analysis is performed based on the processed pressure data to identify pressure change trends, predict future pressure changes, and generate prediction results. This step utilizes a three-tiered analysis architecture: short window (3-5 seconds), medium window (30-60 seconds), and long window (5-10 minutes). This allows for a comprehensive understanding of both immediate trends and long-term patterns. Combined with a dynamic time warping algorithm for pattern matching, the system predicts pressure changes within the next 5-10 seconds. This predictive capability enables the system to proactively respond to impending intense patient activity, mitigating discomfort and potential injury caused by sudden struggles.

[0163] Adaptive Threshold Adjustment: Dynamically adjusts the pressure threshold based on the patient's activity intensity, restraint duration, and muscle tension, generating an adjusted pressure threshold. This step considers the patient's individual characteristics, current state, and restraint history to dynamically optimize the trigger threshold, ensuring the system responds promptly to abnormal situations while avoiding oversensitivity that can lead to frequent false triggering. This personalized setting ensures that the restraint solution is optimally tailored to the patient's specific needs, balancing safety and comfort.

[0164] The decision-making control step generates pressure adjustment control instructions based on the predicted results and the adjusted pressure threshold. This step utilizes a hierarchical decision-making architecture that prioritizes safety, followed by comfort optimization, and finally maintaining restraint effectiveness, ensuring the system makes the most appropriate adjustment decisions under all circumstances. This prioritization reflects a design philosophy that prioritizes patient safety while also balancing comfort and effectiveness.

[0165] The pressure adjustment execution step receives pressure adjustment control commands and adjusts the airbag pressure to achieve dynamic adjustment of the restraint belt tightness. This step precisely controls the operating states of the booster pump, pressure regulating valve, and pressure relief valve to achieve precise regulation of the airbag pressure. The air pressure sensor provides real-time feedback, forming a complete closed-loop control system. This precise execution ensures that system decisions are accurately translated into actual pressure adjustments, providing the appropriate restraint force for the patient.

[0166] The adaptive pressure-regulating device for limb restraints of this invention utilizes innovative technologies such as multi-stage pressure sensing, intelligent signal processing, trend prediction, and adaptive control to achieve intelligent pressure management during limb restraint. In clinical trials at the ICU of a tertiary hospital, this device demonstrated the following significant advantages over traditional restraint devices:

[0167] 1. Pressure Control Precision: Traditional systems have a pressure fluctuation range of ±30% and a response time of over 2 seconds. This new system controls pressure fluctuations to within ±5% and a response time of less than 0.3 seconds, significantly improving restraint accuracy. This means that patients will not experience discomfort from sudden tightening of the restraints, nor will they be exposed to safety hazards caused by excessive loosening of the restraints.

[0168] 2. Safety: Through its pressure prediction and proactive response mechanism, this invention increases the prevention rate of potential risk events by approximately 87%, significantly reducing the incidence of restraint-related complications. In a three-month comparative trial, the incidence of restraint-related skin injuries in the patient group using this device was 65% lower than in the control group, and no serious complications such as nerve damage and circulatory disturbances occurred.

[0169] 3. Clinical Value: This device reduces the incidence of restraint-related complications (such as skin damage and circulatory impairment) by approximately 65%, reduces the workload of medical staff by approximately 75% (primarily by reducing the number of restraint adjustments and inspections), improves patient comfort and compliance, and reduces struggling behavior by approximately 33%. Numerous medical staff have reported that this device significantly improves the quality and efficiency of restraint management, reduces patient complaints of discomfort caused by restraints, and reduces medical staff workload and the risk of restraint-related disputes.

[0170] The above embodiments are merely preferred embodiments of the present invention and are not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. An adaptive pressure regulating device for a limb restraint, characterized in that: include: a pressure sensing unit, comprising a multi-stage pressure sensor, wherein the multi-stage pressure sensor comprises a first pressure sensor, a second pressure sensor, and a third pressure sensor, for detecting a pressure value between the patient's limb and the restraint belt and generating a pressure signal; a pressure signal processing module, electrically connected to the pressure sensing unit, configured to receive the pressure signal, filter the pressure signal, identify the patient's activity state, and generate processed pressure data; a pressure change prediction module, electrically connected to the pressure signal processing module, for performing multi-window pressure analysis based on the processed pressure data, identifying pressure change trends, predicting future pressure changes, and generating prediction results; an adaptive threshold adjustment module, electrically connected to the pressure signal processing module and the pressure change prediction module, configured to dynamically adjust the pressure threshold based on the patient's activity intensity, restraint duration, and muscle tension, and generate an adjusted pressure threshold; a decision control module, electrically connected to the pressure change prediction module and the adaptive threshold adjustment module, and configured to generate a pressure regulation control instruction based on the prediction result and the adjusted pressure threshold; The pressure regulation execution module is electrically connected to the decision control module and includes an airbag, an inflation device and an exhaust device. It is used to receive the pressure regulation control instruction and adjust the air pressure in the airbag to achieve dynamic adjustment of the tightness of the restraint belt.

2. The limb restraint belt adaptive pressure regulating device according to claim 1, characterized in that: The operating range of the multi-stage pressure sensor is: The first pressure sensor monitors pressure changes within a range from 0 to a first threshold; The second pressure sensor monitors pressure changes within a range from a first threshold to a second threshold; The third pressure sensor monitors for pressure changes exceeding a second threshold.

3. The limb restraint belt adaptive pressure regulating device according to claim 1, characterized in that: The pressure signal processing module includes: a signal preprocessing unit, configured to perform digital filtering and baseline drift correction on the pressure signal; An outlier identification unit, used to identify and remove abnormal data points; A signal segmentation unit, for segmenting a continuous signal into an active segment and a rest segment; A feature extraction unit, used to extract time domain features of the pressure waveform; An activity recognition unit is used to distinguish the patient's active activities from external interference based on the time domain features.

4. The limb restraint belt adaptive pressure regulating device according to claim 1, characterized in that: The pressure change prediction module includes: Short-window analysis unit, used to analyze pressure data within 3 to 5 seconds and identify immediate pressure change trends; The middle window analysis unit is used to analyze pressure data within 30 to 60 seconds and identify activity rhythms and patterns; Long-window analysis unit, used to analyze pressure data within 5 to 10 minutes and learn patient activity patterns; a trend matching unit, which matches real-time data with historical patterns; The prediction calculation unit is used to calculate the probability and time of the pressure exceeding the threshold in the next 5 to 10 seconds.

5. The limb restraint belt adaptive pressure regulating device according to claim 1, characterized in that: The adaptive threshold adjustment module includes: A basic threshold setting unit, used to set initial threshold parameters according to the patient's body shape and limb circumference; an activity coefficient calculation unit, for calculating an activity intensity coefficient by analyzing the frequency and amplitude of pressure fluctuations; a time factor calculation unit, configured to calculate a time compensation factor based on a constraint duration; A muscle tension assessment unit, used to assess muscle tension through pressure waveform characteristics; The comprehensive threshold calculation unit is used to dynamically calculate the trigger thresholds of sensors at all levels based on the above parameters.

6. The limb restraint belt adaptive pressure regulating device according to claim 1, characterized in that: The decision control module includes: Safety assurance decision-making unit, used to monitor the ultimate pressure value and the duration of high pressure, and immediately generate a decompression command once it approaches a dangerous value; Comfort optimization decision unit, used to generate adjustment instructions based on pressure distribution uniformity and long-term static pressure state; A restraint effect maintenance decision unit, used to generate basic restraint force adjustment instructions based on activity intensity trends and escape attempt frequencies; A decision priority management unit is used to arbitrate the instructions generated by each decision unit in the order of safety, comfort, and restraint effect; The execution instruction generation unit is used to convert the arbitration decision into specific control instructions.

7. The limb restraint belt adaptive pressure regulating device according to claim 1, characterized in that: The pressure regulation execution module also includes: Air pressure sensor, used to detect the air pressure in the airbag; A booster pump for inflating the airbag; Pressure regulating valve, used to accurately adjust the pressure of the airflow entering the airbag; A pressure relief valve, used to control the airbag deflation rate; The execution feedback unit is used to collect the airbag pressure change data and feed it back to the decision control module.

8. The limb restraint belt adaptive pressure regulating device according to claim 1, characterized in that: Also includes: an alarm module, electrically connected to the decision control module, for issuing an audible and visual alarm signal when the pressure exceeds a safe range or a system failure occurs; A data storage module, electrically connected to the pressure signal processing module, for recording historical pressure data and system operating status; Communication interface module, used for exchanging data with external medical information systems.

9. The limb restraint belt adaptive pressure regulating device according to claim 1, characterized in that: Also includes: An intermediate connecting device, used to connect the restraint belt and the airbag, the intermediate connecting device comprising a belt body, an adjustment mechanism, a connecting device and a fixing mechanism; The airbag is arranged in the middle of the belt body; The adjustment mechanism is connected to the restraint belt; The fixing mechanism is connected to the airbag; The connecting device is used to connect the belt body and the airbag.

10. A method for adaptively regulating pressure of a limb restraint belt, using the adaptive pressure regulating device for a limb restraint belt according to any one of claims 1 to 9, characterized in that: include: Acquisition step, collecting the pressure value between the patient's limb and the restraint belt by using a multi-stage pressure sensor and generating a pressure signal; a pressure signal processing step of filtering the pressure signal, identifying the patient's activity state, and generating processed pressure data; a pressure change prediction step, performing multi-window pressure analysis based on the processed pressure data, identifying pressure change trends, predicting future pressure changes, and generating prediction results; an adaptive threshold adjustment step, dynamically adjusting the pressure threshold based on the patient's activity intensity, restraint duration, and muscle tension, and generating an adjusted pressure threshold; a decision control step of generating a pressure regulation control instruction based on the prediction result and the adjusted pressure threshold; The pressure adjustment execution step receives the pressure adjustment control instruction and adjusts the air pressure in the airbag to achieve dynamic adjustment of the tightness of the restraint belt.

Images