Hemodialysis patient internal fistula pressing hemostasis self-adaptive regulation and control system fused with NB-IoT technology

Abstract

The invention relates to the technical field of intelligent nursing, in particular to a hemodialysis patient internal fistula pressing hemostasis self-adaptive regulation and control system fused with the NB-IoT technology, and the system comprises a sensor module which collects internal fistula puncture point pressure, local temperature and patient body movement data; the processing module is used for outputting a pressure regulation and control instruction through a self-adaptive control algorithm based on the internal fistula elasticity modulus, the blood coagulation function and the historical hemostasis threshold value of the patient after preprocessing the data, and has a fuzzy PID strategy and an abnormal value elimination function; the execution module adjusts the pressing pressure according to the instruction, and a feedback sensor is arranged in the execution module to form closed-loop control; the NB-IoT communication module encrypts and uploads data to a cloud platform through AES-128, and supports remote viewing and regulation; the power module is designed with low power consumption, and the endurance is longer than or equal to 7 days. Through real-time monitoring, self-adaptive control and remote interactive closed loop, the hemostasis success rate is increased to be larger than or equal to 92%, the subcutaneous hematoma occurrence rate is reduced to be smaller than or equal to 3%, the safety of a patient and the service life of internal fistula are guaranteed, and diagnosis and treatment convenience is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent nursing, in particular to a blood dialysis patient internal fistula pressing hemostasis self-adaptive control system fusing NB-IoT technology. BACKGROUND

[0002] As the main treatment method for patients with end-stage renal disease, one of the core links of hemodialysis is to establish and maintain effective vascular access. Arteriovenous fistula has become the first choice for vascular access type due to its low infection rate and long service life. However, postoperative hemostasis management of internal fistula is directly related to the maintenance of access function and patient safety: if the pressing pressure is insufficient, it is easy to cause subcutaneous hematoma, pseudoaneurysm and other complications; if the pressure is too large, it may cause internal fistula occlusion and loss of dialysis function. According to statistics, internal fistula related complications account for 23%-35% of hemodialysis patient hospitalization reasons, of which improper hemostasis accounts for as high as 41%.

[0003] The existing hemostasis technology has significant limitations: traditional manual pressing relies on the experience of medical staff, the pressure control accuracy is low (±10-15mmHg), and the bleeding situation cannot be continuously monitored; mechanical pressing device realizes pressure quantification, but adopts open loop control mode, without considering individual differences (such as coagulation function, blood vessel elasticity) and dynamic physiological changes of patients, resulting in a failure rate of hemostasis of 12%-18%. In addition, the existing equipment lacks remote monitoring capability, and medical staff need to frequently patrol on site, increasing the labor cost and management difficulty.

[0004] With the development of Internet of Things technology, intelligentization of medical equipment has become a trend. Narrowband Internet of Things (NB-IoT) provides an ideal solution for remote medical monitoring due to its low power consumption, wide coverage and high penetration. However, existing NB-IoT medical devices mainly focus on data acquisition and transmission, lack edge computing capability, and cannot realize real-time decision and adaptive control.

[0005] The existing hemostasis of hemodialysis patient internal fistula has the following defects:

[0006] 1. Pressure regulation has no individual adaptability, and hemostasis effect is unstable

[0007] The existing technology adopts a unified pressure standard or experience setting, without considering individual differences of internal fistula elasticity, coagulation function and previous hemostasis threshold of patients. Fixed pressure equipment only outputs single gear pressure, and manual pressing relies on medical experience, which is easy to cause hematoma due to insufficient pressure or excessive pressure, resulting in low success rate of hemostasis and many complications.

[0008] 2. Monitoring dimension is single, and it is difficult to identify overall risk

[0009] The prior art only monitors the pressing pressure, does not monitor the local temperature of the puncture point (cannot prewarn the inflammatory reaction) and the body movement of the patient (cannot identify the pressure deviation caused by the limb movement), and is easy to cause the hemostasis failure due to the omission of the key risk factors.

[0010] 3. The data processing and regulation logic is simple and lacks precision

[0011] The pressure data lacks effective pretreatment (is easy to be interfered by noise), is mostly open-loop control (cannot dynamically correct the pressure deviation), and the scheme is not optimized in combination with the historical hemostasis data of the patient, so that the regulation response is lagged and the precision is low.

[0012] 4. No remote interaction and whole-cycle data management capability

[0013] The device locally and independently operates, cannot remotely transmit the hemostasis data, and the medical staff is difficult to intervene in the out-of-hospital abnormality in real time; the whole-cycle hemostasis data cannot be stored, and it is difficult to trace the rule and develop the subsequent individualized scheme.

[0014] 5. Poor adaptation of the endurance and practicability

[0015] The endurance capability of the power module is weak, and frequent charging is needed; the execution mechanism is slow in response, and cannot be quickly adjusted when the pressure deviation occurs, so that the patient is inconvenient for long-term use and efficient hemostasis in the clinic. SUMMARY

[0016] The purpose of the present application is to provide a hemodialysis patient internal fistula pressing hemostasis self-adaptive regulation system fusing NB-IoT technology, so as to solve the problems of poor adaptability of the traditional hemodialysis patient internal fistula hemostasis pressure, single monitoring dimension and lack of remote management in the background art.

[0017] In order to achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0018] The hemodialysis patient internal fistula pressing hemostasis self-adaptive regulation system fusing NB-IoT technology comprises:

[0019] A sensor module is used for collecting the pressure data, local temperature data and body movement data of the puncture point of the patient internal fistula in real time;

[0020] A processing module is electrically connected with the sensor module, is used for pretreating the pressure data, temperature data and body movement data, and outputs a pressure regulation instruction based on an adaptive control algorithm;

[0021] An execution module is electrically connected with the processing module, is used for adjusting the pressing pressure on the puncture point according to the pressure regulation instruction;

[0022] An NB-IoT communication module, electrically connected with the processing module, configured to upload the pressure data, temperature data, body movement data and system state information to a cloud platform, and receive control instructions sent by the cloud platform or a user terminal;

[0023] A power module configured to supply power to the sensor module, the processing module, the execution module and the NB-IoT communication module;

[0024] The adaptive control algorithm is based on individual physiological parameters of the patient's internal fistula, including internal fistula elastic modulus E, coagulation function index C and historical hemostasis pressure threshold P hist , dynamically adjusts the target pressure P target , and specifically satisfies:

[0025] ;

[0026] In the formula, k is the current time, and a and β are weight coefficients, 0

[0027] As a preferred embodiment, the sensor module comprises:

[0028] A pressure sensor, which is a MEMS piezoresistive sensor, has a range of 0-100 mmHg and an accuracy of ±0.5 mmHg, and is arranged at the contact interface between the hemostasis pressing device and the patient's skin;

[0029] A temperature sensor, which is an infrared non-contact sensor, has a range of 25-45℃ and an accuracy of ±0.1℃, and is used to monitor the skin temperature around the puncture point to assist in determining whether there is an inflammatory reaction;

[0030] A three-axis acceleration sensor, which has a range of ±16g and a sampling frequency of 100Hz, is used to detect the amplitude and frequency of the patient's body movement, and when the amplitude of the body movement exceeds a threshold , , a pressure compensation mechanism is triggered.

[0031] As a preferred embodiment, the processing module comprises a microprocessor MCU, a storage unit and a signal conditioning circuit; the storage unit stores patient electronic medical record data and parameters of the adaptive control algorithm, the patient electronic medical record data includes internal fistula position, previous hemostasis failure record and prothrombin time, and the parameters of the adaptive control algorithm include a, β and P hist ; the signal conditioning circuit is used to filter and analog-to-digital convert the original sensor signals, the filter is a 5th order Butterworth low-pass filter with a cutoff frequency of 10Hz, and the ADC resolution of the analog-to-digital conversion is 16 bits.

[0032] Preferably, the actuation module is an electric actuator or a miniature air pump, with an output pressure range of 20-80 mmHg and a response time of ≤0.5s; the actuation module has a built-in pressure feedback sensor for real-time transmission of the actual output pressure. The process module then forms a closed-loop control.

[0033] Preferably, the NB-IoT communication module supports the LTE Cat-NB1 protocol, operates in the Band 5 / 8 / 20 frequency band, and has a transmission rate of 500bps-25kbps. It uses the AES-128 encryption algorithm to encrypt transmitted data, ensuring patient privacy and data security. The cloud platform is used to store the patient's full-cycle hemostasis data, including timestamps, pressure curves, temperature changes, and body movement records. It also supports doctors to remotely view and adjust control parameters through user terminals, including mobile apps or computers.

[0034] Preferably, the adaptive control algorithm further includes a fuzzy PID control strategy, used to adjust the current pressure error based on the fuzzy PID control strategy. and error change rate / Δ Dynamically adjust the proportional coefficient K of the PID controller p Integral coefficient K i Differential coefficient K d Δ =1s; The fuzzy PID control rule is as follows:

[0035] If | |>5mmHg and| If |>2mmHg / s, then increase K p And decrease K i To quickly suppress errors, increment ΔK p =0.5, decrease ΔK i =0.1;

[0036] If | |≤2mmHg and| If |≤1mmHg / s, then decrease K p And increase K i To avoid overshoot, reduce ΔK. p =0.3, increment ΔK i =0.05;

[0037] Controller output satisfy:

[0038] .

[0039] As preferred, the system further comprises a data preprocessing unit for removing outliers from the raw pressure data collected by the sensor; the outlier removal employs a statistical method based on interquartile range IQR, and the specific steps are as follows:

[0040] S1: calculating the first quartile Q1, the third quartile Q3 and IQR = Q3-Q1 of the pressure data of the last 10 sampling points;

[0041] S2: defining the lower threshold L = Q1-1.5*IQR and the upper threshold U = Q3+1.5*IQR;

[0042] S3: if the pressure data P(k) of a sampling point is less than L or greater than U, it is determined as an outlier, and the previous valid data is used to replace it by using the forward filling method.

[0043] As preferred, the user terminal comprises:

[0044] a data display unit for displaying the pressure-time curve, the temperature-time curve and the body movement amplitude-time curve in the form of a line graph in real time;

[0045] a warning unit for triggering an audible and visual alarm when any of the following conditions is detected:

[0046] (1) the actual pressure P(k) deviates from the target pressure P(k) by more than ±5 mmHg for more than 10 seconds; act target (2) the local temperature T(k) is greater than 38℃ for more than 5 minutes;

[0047] (3) the body movement amplitude A(k) is greater than 10℃, accompanied by a sudden drop in pressure, ΔP(k) <-10 mmHg / s.

[0048]

[0049] As preferred, the power module comprises a rechargeable lithium battery with a capacity of ≥2000 mAh, a wireless charging coil supporting a charging power of 5W according to the Qi protocol, and a low-power consumption management chip; the low-power consumption management chip is used to control the time-sharing wake-up of each module, the sampling frequency of the sensor module is 1 Hz, the sleep-wake cycle of the processing module is 10s, the NB-IoT communication module uploads data 3 times a day, the overall power consumption of the system is ≤10 mW, and the endurance time is ≥7 days.

[0050] As preferred, the system improves the hemostasis success rate from 78% of the traditional fixed pressure device to ≥92% through a closed-loop mechanism of real-time monitoring-adaptive control-remote interaction, and reduces the incidence of subcutaneous hematoma caused by excessive compression from 15% to ≤3%.

[0051] Compared with the prior art, the present application has the following advantages: ​​

[0052] I. Hemostatic effect is greatly improved, and the incidence of complications is significantly reduced

[0053] Through the core mechanism of real-time monitoring-adaptive regulation-closed-loop feedback, the clinical hemostatic outcome is directly improved:

[0054] Compared with traditional fixed pressure hemostatic equipment, the success rate of hemostasis is increased from 78% to ≥92%, effectively avoiding the problem of hemostatic failure such as bleeding at the puncture point caused by insufficient pressure;

[0055] Relying on the target pressure mode of dynamic adjustment, rather than fixed pressure, the incidence of subcutaneous hematoma caused by excessive compression is reduced from 15% to ≤3%, reducing the risk of patient pain and internal fistula damage;

[0056] The response time of the execution module is ≤0.5s, and the built-in pressure feedback sensor forms a closed-loop control, which can quickly correct the pressure deviation and avoid the influence of pressure fluctuation on the hemostatic effect.

[0057] II. Precise adaptation to individual differences of patients, more targeted regulation

[0058] The system abandons the one-size-fits-all fixed pressure scheme and realizes individualized regulation based on the patient's own characteristics and actual situation:

[0059] The adaptive control algorithm takes the patient's individual physiological parameters as the core, including the internal fistula elastic modulus, coagulation function indicators, and historical hemostatic pressure threshold, and dynamically calculates the target pressure through the weight coefficient trained by logistic regression, adapting to the internal fistula conditions and coagulation speed of different patients;

[0060] The three-axis acceleration sensor has a range of ±16g and a sampling frequency of 100Hz, which can detect the patient's body movement amplitude in real time. When the amplitude exceeds the threshold 5, the pressure compensation mechanism is triggered immediately to avoid insufficient compression pressure caused by limb movement;

[0061] The data preprocessing unit ensures the accuracy of sensor data such as pressure and temperature through quartile range outlier rejection, 16-bit ADC analog-to-digital conversion, and 5th order Butterworth low-pass filtering, providing reliable data support for individualized regulation.

[0062] III. Support remote interaction and whole-cycle data management, optimize diagnosis and treatment process

[0063] Relying on NB-IoT technology to build an interconnected system of patients-equipment-cloud platform-medical staff, breaking the time and space constraints:

[0064] Data remote uploading and storage: The NB-IoT module supports LTE Cat-NB1 protocol, working frequency band Band5 / 8 / 20, and uploads data such as pressure curve, temperature change, body movement record, and system state to the cloud platform through AES-128 encryption algorithm, realizing the retention of patient's whole-cycle hemostasis data, and facilitating medical staff to trace the hemostasis process and analyze the rules;

[0065] Remote regulation and real-time viewing: Doctors can remotely view the patient's hemostasis state through mobile phone APP or computer terminal, and can also adjust the control parameters, without the need for the patient to make a site re-visit to complete the intervention;

[0066] Patient terminal visual management: The user terminal displays the pressure-time, temperature-time, and body movement amplitude-time curves in the form of a line chart in real time, and patients and their families can intuitively master the hemostasis progress, improving the treatment participation.

[0067] Four, multi-dimensional security protection, considering data privacy and patient safety

[0068] The system builds a perfect security system from three aspects of data security, abnormal early warning, and device reliability:

[0069] Data privacy protection: AES-128 encryption algorithm is used in the transmission process to prevent the leakage of patient electronic medical record data (including internal fistula location, prothrombin time, and past hemostasis failure record) and hemostasis data, which meets the medical data privacy specification;

[0070] Multi-scenario abnormal early warning: When the actual pressure deviates from the target pressure ±5mmHg and lasts for 10 seconds, the local temperature is greater than 38℃ and lasts for 5 minutes, the body movement amplitude exceeds the standard and is accompanied by a sudden drop in pressure (ΔP < -10mmHg / s), etc., the system triggers an audible and light alarm to remind medical staff or patients to handle it in time, avoiding delay in intervention;

[0071] Device operation reliability: The pressure sensor accuracy is ±0.5mmHg, and the temperature sensor accuracy is ±0.1℃. The execution module realizes closed-loop pressure control through pressure feedback, and the processing module uses signal conditioning circuit to ensure data processing anti-interference ability, ensuring long-term stable operation of the device and reducing safety risks caused by device errors.

[0072] Five, low-power design adapts to long-term use, improving patient convenience

[0073] The system optimizes the design in terms of power consumption control and use convenience for the long-term use of patients:

[0074] Long endurance: adopt rechargeable lithium battery with capacity ≥2000mAh, match low-power management chip to control each module to wake up in time, sensor module sampling frequency 1Hz, processing module sleep-wake cycle 10s, NB-IoT communication module upload data 3 times a day, overall power consumption ≤10mW, single charge endurance ≥7 days, no need for patients to charge frequently;

[0075] Convenient charging method: support Qi protocol charging power 5W wireless charging coil, patients can charge without plugging in the cable, more convenient operation, especially suitable for patients with difficulty in moving;

[0076] Offline data storage: processing module storage unit can store patient electronic medical record data and adaptive control algorithm parameters, even if there is no network for a short time, the device can normally complete hemostasis regulation, and after network recovery, data is supplemented through NB-IoT, which does not affect the continuity of use. BRIEF DESCRIPTION OF DRAWINGS

[0077] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, and together with the embodiments of the present application, make a further detailed explanation, but do not constitute a limitation of the present application.

[0078] Figure 1 The blood dialysis patient internal fistula pressing hemostasis adaptive control system of the present application is composed of a block diagram. DETAILED DESCRIPTION

[0079] The technical solutions in the embodiments of the present application will be described below in conjunction with the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0080] Embodiment one: conventional clinical outpatient scene adaptive system

[0081] This embodiment is designed for the conventional diagnosis and treatment scene of hemodialysis clinic, suitable for most patients who need regular dialysis and moderate activity, and the core goal is to balance the hemostasis effect and the convenience of use.

[0082] 1. Sensor module

[0083] Pressure sensor: MEMS piezoresistive sensor, range covers 0-100mmHg, accuracy control in ±0.5mmHg, directly embedded in the contact interface of the hemostasis pressing device and the patient's skin, real-time capture of puncture point pressing pressure data.

[0084] Temperature sensor: Infrared non-contact sensor is selected, measurement range 25-45℃, accuracy ±0.1℃, align the skin area within 3cm around the puncture point, and assist in judging whether there is an inflammatory reaction through temperature change.

[0085] Triaxial acceleration sensor: range set to ±16g, sampling frequency 100Hz, used to monitor the amplitude and frequency of patient's arm movement, when the body movement amplitude exceeds 5 (standard threshold), the pressure compensation mechanism is automatically triggered to avoid insufficient pressing pressure caused by limb movement.

[0086] 2. Processing module

[0087] The core is a high-performance microprocessor (MCU) with a storage unit and a signal conditioning circuit. The storage unit pre-stores patient electronic medical record data (including internal fistula specific location, past hemostasis failure record, prothrombin time) and adaptive control algorithm parameters. The signal conditioning circuit first filters the original sensor signal through a 5th order Butterworth low pass filter (cutoff frequency 10Hz), and then completes analog-to-digital conversion through a 16-bit resolution ADC to ensure data accuracy.

[0088] Built-in data preprocessing function, using statistical method based on interquartile range (IQR) to remove pressure data outliers: take the first quartile (Q1), third quartile (Q3) and IQR of the last 10 sampling points, define the lower limit L=Q1-1.5×IQR and the upper limit U=Q3+1.5×IQR, if a sampling point data exceeds the range, use the previous valid data to replace it.

[0089] 3. Execution module

[0090] An electric push rod is used as the execution component, with an output pressure range of 20-80mmHg and a response time of ≤0.5s, which can quickly adjust the pressing force according to the control instructions. The push rod has a built-in pressure feedback sensor that transmits the actual output pressure back to the processing module in real time, forming a closed-loop control to avoid pressure deviation.

[0091] 4. NB-IoT communication module and cloud platform

[0092] Supports LTE Cat-NB1 protocol, working frequency band covers Band5 / 8 / 20, transmission rate 500bps-25kbps, uses AES-128 encryption algorithm to protect transmission data and prevent patient privacy leakage.

[0093] The cloud platform stores patient hemostasis data throughout the cycle (including timestamp, pressure change curve, temperature fluctuation record, body movement event), and outpatient doctors can access the data through the background to trace the hemostasis process.

[0094] 5. Power module

[0095] Equipped with 2000mAh rechargeable lithium battery, with wireless charging coil supporting Qi protocol (5W power), it is convenient for patients to supplement power when waiting in outpatient; Low-power management chip controls the time-sharing wake-up of each module: sensor module sampling frequency 1Hz, processing module sleep-wake cycle 10s, NB-IoT communication module uploads data 3 times a day, overall power consumption ≤10mW, single charge endurance ≥7 days.

[0096] 6. User terminal and early warning function

[0097] The user terminal is a mobile phone APP, which displays the pressure-time, temperature-time, and body movement amplitude-time curves in the form of a broken line graph in real time, making it convenient for patients and medical staff to intuitively view the status.

[0098] Sound and light alarms are triggered when the following conditions occur: actual pressure deviates from target pressure ±5mmHg and lasts more than 10 seconds, local temperature >38℃ and lasts more than 5 minutes, body movement amplitude >10 with sudden pressure drop, and timely remind medical staff to intervene.

[0099] 7. System effect

[0100] Through the closed-loop mechanism of real-time monitoring, self-adaptive regulation, and remote interaction, the hemostatic success rate is increased from 78% of traditional fixed pressure equipment to ≥92%, and the incidence of subcutaneous hematoma caused by excessive compression is reduced from 15% to ≤3%.

[0101] Example Two: Low-power scene adaptation system for community dialysis center

[0102] This example is designed for community dialysis center patients (mostly home and center dialysis combination, need to carry equipment for a long time), the core goal is to prolong the endurance time, reduce the charging frequency, and at the same time guarantee the stability of remote data transmission.

[0103] 1. Sensor module

[0104] The pressure sensor and temperature sensor model and precision are the same as in Example One, only the sensor sampling strategy is fine-tuned: the pressure sensor samples once every 1.2 seconds, and the temperature sensor samples once every 3 seconds, reducing power consumption without affecting data effectiveness.

[0105] The three-axis acceleration sensor maintains a ±16g range and a 100Hz sampling frequency to ensure body movement monitoring sensitivity and avoid missing body movement events due to extended sampling intervals.

[0106] 2. Processing module

[0107] Low-power MCU is selected, and the storage unit is expanded to support 6-month patient electronic medical record data storage, which is convenient for community doctors to track patient hemostasis rules for a long time; the signal conditioning circuit retains 5-order Butterworth low-pass filter and 16-bit ADC to ensure that data processing accuracy is not affected by power optimization.

[0108] The data preprocessing logic is the same as in Example One, and the outlier rejection algorithm remains the same to ensure data reliability.

[0109] 3. Execution module

[0110] A miniature air pump is used instead of an electric push rod, and the output pressure range is still 20-80 mmHg, with a response time ≤0.5s, and a built-in pressure feedback sensor for closed-loop control; the miniature air pump has lower power consumption when starting and stopping, and is suitable for long-lasting needs.

[0111] 4. NB-IoT communication module and cloud platform

[0112] The communication protocol, working frequency band, and encryption method are the same as in Example One, and the data upload strategy is adjusted: only 1 hour after dialysis, at night, and in the morning, each upload 1 time of data, reducing unnecessary transmission power consumption; the cloud platform adds a community doctor exclusive view, highlighting the patient's hemostasis data trend in the past week, making it easy for doctors to quickly assess.

[0113] 5. Power module

[0114] Upgraded to a 2500mAh rechargeable lithium battery, and the wireless charging coil still supports the Qi protocol 5W power; the low-power management chip optimizes the wake-up cycle: the sensor module sampling frequency is 0.8Hz, the processing module sleep-wake cycle is 15s, and the NB-IoT communication module uploads data 2 times a day, with a total power consumption ≤8mW, and a single charge endurance ≥10 days.

[0115] 6. User terminal and early warning function

[0116] The user terminal includes a mobile phone APP and a community center computer terminal, and the computer terminal supports batch viewing of multiple patient data, making it easy for community medical staff to manage; the APP adds a battery life reminder function, which pushes a charging prompt when the battery is below 20%.

[0117] The early warning trigger condition is the same as in Example One, and the alarm volume and light intensity can be adjusted to adapt to different environments at home and in the community.

[0118] 7. System effect

[0119] The hemostasis success rate is ≥92%, the subcutaneous hematoma incidence is ≤3%, the endurance time meets the needs of patients for 3-5 days without charging, and the charging burden of community patients is reduced.

[0120] Example Three: System Adapted to High Mobility Patients

[0121] This embodiment is designed for dialysis patients who need to go out frequently (such as commuting, short trips), and the core goal is to enhance the anti-interference ability and body motion adaptability of the device to ensure stable hemostasis effect during movement.

[0122] 1. Sensor module

[0123] Pressure sensor: Based on the MEMS piezoresistive sensor, a shockproof shell is added to avoid pressure data fluctuations caused by bumps during movement, with a range of 0-100 mmHg and an accuracy of ±0.5 mmHg.

[0124] Temperature sensor: Optimize the detection angle and reduce the monitoring range to within 2 cm around the puncture point to reduce the interference of external environmental temperature (such as wind blowing, clothing covering) on the measurement results, with unchanged range and accuracy.

[0125] Three-axis acceleration sensor: Improve data processing priority, trigger pressure compensation mechanism immediately when monitoring body motion amplitude exceeds 5, compensation response speed is 20% faster than regular scene; At the same time, increase the moving mode, when the body motion frequency is continuous 30 seconds > 5 times / minute, automatically increase the sampling frequency to 200Hz, more accurately capture the limb activity state.

[0126] 2. Processing module

[0127] Use industrial-grade MCU with stronger anti-interference ability, signal conditioning circuit adds electromagnetic shielding layer to avoid electromagnetic interference from mobile phones, elevators and other devices during travel; Storage unit supports offline storage of 30 days of hemostasis data to prevent data loss due to poor network during travel.

[0128] Data preprocessing algorithm adds fast outlier judgment function, when the data of a sampling point deviates from the previous 3 valid data by more than 10 mmHg, it is immediately judged as an outlier and replaced, reducing errors caused by sudden interference during movement.

[0129] 3. Execution module

[0130] Choose high-stability electric push rod, output pressure error controlled within ±0.3 mmHg, higher accuracy than regular scene; Push rod has built-in buffer structure to reduce the impact of vibration on pressing pressure during movement, response time still maintains ≤0.5s, ensuring the timeliness of pressure adjustment.

[0131] 4. NB-IoT communication module and cloud platform

[0132] Support multi-band adaptive switching (Band 5 / 8 / 20 priority, while compatible with Band 1 / 3), improve network coverage in different areas (such as remote areas, high-rise buildings); transmission rate remains 500bps-25kbps, uses breakpoint resume technology, automatically supplements offline data after network recovery.

[0133] Cloud platform adds mobile trajectory association function, which can lightly associate patient hemostasis data with geographic location (requires patient authorization) to help doctors analyze the impact of different environments (such as in the car, outdoors) on hemostasis effect.

[0134] 5. Power module

[0135] Use 2200mAh high-rate rechargeable lithium battery, support fast charging (charge to 50% in 15 minutes), adapt to the need of patients to quickly charge when going out; wireless charging coil increases waterproof design to avoid accidental water damage when going out.

[0136] Low-power management chip uses dynamic wake-up strategy: when the patient is stationary, the sensor sampling frequency is 1Hz and the processing module sleeps for 10s; when the patient moves, automatically adjust the sensor sampling frequency to 2Hz and the processing module sleeps for 5s, balance power consumption and data timeliness, overall power consumption ≤9mW, single charge endurance ≥8 days.

[0137] 6. User terminal and early warning function

[0138] User terminal APP adds emergency contact function, automatically sends alarm information and patient's approximate location (requires authorization) to pre-set contacts (such as family members, attending doctors) when triggering an alarm; APP interface is simplified into two core blocks: status overview and alarm center, making it easy for patients to quickly check on the go.

[0139] Early warning adds low battery emergency alarm, when the battery is lower than 15%, in addition to sending reminders, automatically reduce the power consumption of unnecessary functions (such as historical data query), to prioritize hemostasis control and alarm functions.

[0140] 7. System effect

[0141] Hemostasis success rate ≥92%, subcutaneous hematoma incidence ≤3%, in the scenario of patients moving 5-10km per day, the device stability is consistent with the conventional scenario, meeting the needs of high-mobility patients going out.

[0142] This hemodialysis patient internal fistula compression hemostasis adaptive control system based on NB-IoT technology focuses on the clinical core needs of hemodialysis patient internal fistula hemostasis, combines NB-IoT communication, adaptive control, high-precision sensing and other technologies, and forms significant advantages in hemostasis effect, personalized adaptation, remote management, safety, and ease of use, as follows:

[0143] I. Hemostatic effect is greatly improved, and the incidence of complications is significantly reduced

[0144] The system directly improves the clinical hemostatic outcome through the core mechanism of real-time monitoring-adaptive regulation-closed-loop feedback:

[0145] Compared with traditional fixed pressure hemostatic equipment, the success rate of hemostasis is increased from 78% to ≥92%, effectively avoiding the problem of hemostatic failure such as bleeding at the puncture point caused by insufficient pressure;

[0146] Relying on the target pressure mode of dynamic adjustment, rather than fixed pressure, the incidence of subcutaneous hematoma caused by excessive compression is reduced from 15% to ≤3%, reducing the risk of patient pain and internal fistula damage;

[0147] The response time of the execution module is ≤0.5s, and the built-in pressure feedback sensor forms a closed-loop control, which can quickly correct the pressure deviation and avoid the influence of pressure fluctuation on the hemostatic effect.

[0148] II. Precise adaptation to individual differences of patients, more targeted regulation

[0149] The system abandons the one-size-fits-all fixed pressure scheme and realizes individualized regulation based on the patient's own characteristics and actual situation:

[0150] The adaptive control algorithm takes the patient's individual physiological parameters as the core, including the internal fistula elastic modulus, coagulation function indicators, and historical hemostatic pressure threshold, and dynamically calculates the target pressure through the weight coefficient trained by logistic regression, adapting to the internal fistula conditions and coagulation speed of different patients;

[0151] The three-axis acceleration sensor has a range of ±16g and a sampling frequency of 100Hz, which can detect the patient's body movement amplitude in real time. When the amplitude exceeds the threshold 5, the pressure compensation mechanism is triggered immediately to avoid insufficient compression pressure caused by limb movement;

[0152] The data preprocessing unit ensures the accuracy of sensor data such as pressure and temperature through quartile range outlier rejection, 16-bit ADC analog-to-digital conversion, and 5th order Butterworth low-pass filtering, providing reliable data support for individualized regulation.

[0153] III. Support remote interaction and whole-cycle data management, optimize diagnosis and treatment process

[0154] Relying on NB-IoT technology to build an interconnected system of patients-equipment-cloud platform-medical staff, breaking the time and space constraints:

[0155] Data remote uploading and storage: The NB-IoT module supports LTE Cat-NB1 protocol, working frequency bands Band5 / 8 / 20, and uploads data such as pressure curve, temperature change, body movement record, system state to the cloud platform through AES-128 encryption algorithm, realizing the retention of patient's whole cycle hemostasis data, and facilitating medical staff to trace the hemostasis process and analyze the rules;

[0156] Remote control and real-time viewing: Doctors can remotely view the patient's hemostasis state through mobile phone APP or computer terminal, and can also adjust the control parameters, without the need for patients to make on-site re-examination to complete the intervention;

[0157] Patient terminal visual management: The user terminal displays the pressure-time, temperature-time, and body movement amplitude-time curves in the form of a line chart in real time, and patients and their families can intuitively understand the hemostasis progress and improve the treatment participation.

[0158] Four, multi-dimensional security protection, considering data privacy and patient safety

[0159] The system builds a perfect security system from three aspects of data security, abnormal early warning, and device reliability:

[0160] Data privacy protection: AES-128 encryption algorithm is used in the transmission process to prevent the leakage of patient electronic medical record data (including internal fistula location, prothrombin time, and past hemostasis failure record) and hemostasis data, which meets the medical data privacy specification;

[0161] Multi-scenario abnormal early warning: When the actual pressure deviates from the target pressure ±5mmHg and lasts for 10 seconds, the local temperature is greater than 38℃ and lasts for 5 minutes, the body movement amplitude exceeds the standard and is accompanied by a sudden drop in pressure (ΔP < -10mmHg / s), etc., the system triggers an audible and light alarm to remind medical staff or patients to handle it in time and avoid delay intervention;

[0162] Device operation reliability: The pressure sensor accuracy is ±0.5mmHg, and the temperature sensor accuracy is ±0.1℃. The execution module realizes closed-loop pressure control through pressure feedback, and the processing module uses signal conditioning circuit to ensure data processing anti-interference ability, ensuring long-term stable operation of the device and reducing safety risks caused by device errors.

[0163] Five, low-power design adapts to long-term use, improving patient convenience

[0164] The system optimizes the design in terms of power consumption control and use convenience for the long-term use of patients:

[0165] Long endurance: adopt rechargeable lithium battery with capacity ≥2000mAh, cooperate with low power management chip to control each module to wake up in time, sensor module sampling frequency 1Hz, processing module sleep-wake cycle 10s, NB-IoT communication module upload data 3 times a day, overall power consumption ≤10mW, single charge endurance ≥7 days, no need for patients to charge frequently;

[0166] Convenient charging method: support Qi protocol charging power 5W wireless charging coil, patients do not need to plug in the cable to charge, operation is more convenient, especially suitable for patients with difficulty in moving;

[0167] Offline data storage: processing module storage unit can store patient electronic medical record data and adaptive control algorithm parameters, even if there is no network for a short time, the device can also complete the hemostasis control normally, after the network is restored, the data is supplemented through NB-IoT, which does not affect the continuity of use.

[0168] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only preferred examples of the present application and are not intended to limit the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. An adaptive control system for pressure-based hemostasis in arteriovenous fistulas of hemodialysis patients integrating NB-IoT technology, characterized in that, include: The sensor module is used to collect pressure data, local temperature data, and patient movement data at the puncture site of the arteriovenous fistula in real time. The processing module, electrically connected to the sensor module, is used to preprocess the pressure data, temperature data, and body motion data, and output pressure regulation commands based on an adaptive control algorithm. An execution module, electrically connected to the processing module, is used to adjust the pressure applied to the puncture point according to the pressure control command. The NB-IoT communication module is electrically connected to the processing module and is used to upload the pressure data, temperature data, body movement data and system status information to the cloud platform, and to receive control commands sent by the cloud platform or user terminal. A power supply module is used to supply power to the sensor module, processing module, execution module and NB-IoT communication module; The adaptive control algorithm is based on the individual physiological parameters of the patient's arteriovenous fistula, including the fistula's elastic modulus E, coagulation function index C, and historical hemostatic pressure threshold P. hist Dynamically adjust the target pressure P target Specifically, it satisfies: ； In the formula, k is the current time, α and β are weight coefficients, 0 < α, β < 1, and α + β < 1, which are obtained by training a logistic regression model based on the patient's historical hemostasis effect data.

2. The adaptive control system for arteriovenous fistula pressure hemostasis in hemodialysis patients integrating NB-IoT technology according to claim 1, characterized in that, The sensor module includes: The pressure sensor, a MEMS piezoresistive sensor, has a range of 0-100 mmHg and an accuracy of ±0.5 mmHg. It is located at the interface between the hemostatic compression device and the patient's skin. The temperature sensor, an infrared non-contact sensor, has a range of 25-45℃ and an accuracy of ±0.1℃. It is used to monitor the skin temperature around the puncture point and help determine whether there is an inflammatory reaction. A triaxial accelerometer with a range of ±16g and a sampling frequency of 100Hz is used to detect the amplitude and frequency of patient body movements. When the amplitude of body movements exceeds a threshold... hour, This triggers the stress compensation mechanism.

3. The adaptive control system for arteriovenous fistula pressure hemostasis in hemodialysis patients integrating NB-IoT technology according to claim 1, characterized in that, The processing module includes a microprocessor (MCU), a storage unit, and a signal conditioning circuit. The storage unit stores patient electronic medical record data and parameters of the adaptive control algorithm. The patient electronic medical record data includes the location of the arteriovenous fistula, previous hemostasis failure records, and prothrombin time. The parameters of the adaptive control algorithm include α, β, and P. hist The signal conditioning circuit is used to filter and perform analog-to-digital conversion on the original sensor signal. The filtering adopts a 5th-order Butterworth low-pass filter with a cutoff frequency of 10Hz, and the ADC resolution of the analog-to-digital conversion is 16 bits.

4. The adaptive control system for arteriovenous fistula pressure hemostasis in hemodialysis patients integrating NB-IoT technology according to claim 1, characterized in that, The actuator module is an electric actuator or a miniature air pump, with an output pressure range of 20-80 mmHg and a response time of ≤0.5s; the actuator module has a built-in pressure feedback sensor for real-time transmission of the actual output pressure. The process module then forms a closed-loop control.

5. The adaptive control system for pressure-based hemostasis of arteriovenous fistula in hemodialysis patients integrating NB-IoT technology according to claim 1, characterized in that, The NB-IoT communication module supports the LTE Cat-NB1 protocol, operates in Band 5 / 8 / 20, and has a transmission rate of 500bps-25kbps. It uses AES-128 encryption to encrypt transmitted data, ensuring patient privacy and data security. The cloud platform stores the patient's full-cycle hemostasis data, including timestamps, pressure curves, temperature changes, and body movement records. It also allows doctors to remotely view and adjust control parameters via user terminals, including mobile apps or computers.

6. The adaptive control system for arteriovenous fistula pressure hemostasis in hemodialysis patients integrating NB-IoT technology according to claim 1, characterized in that, The adaptive control algorithm further includes a fuzzy PID control strategy, used to adjust the current pressure error. and error change rate Dynamically adjust the proportional coefficient K of the PID controller p Integral coefficient K i Differential coefficient K d Δ =1s; The fuzzy PID control rule is as follows: If | |>5mmHg and| If |>2mmHg / s, then increase K p And decrease K i To quickly suppress errors, increment ΔK p =0.5, decrease ΔK i =0.1; If | |≤2mmHg and| If |≤1mmHg / s, then decrease K p And increase K i To avoid overshoot, reduce ΔK. p =0.3, increment ΔK i =0.05; Controller output satisfy: 。 7. The adaptive control system for arteriovenous fistula pressure hemostasis in hemodialysis patients integrating NB-IoT technology according to claim 1, characterized in that, It also includes a data preprocessing unit for outlier removal from the raw pressure data collected by the sensor; the outlier removal employs a statistical method based on interquartile range (IQR), and the specific steps are as follows: S1: Calculate the first quartile Q1, the third quartile Q3, and IQR = Q3 - Q1 of the pressure data from the most recent 10 sampling points; S2: Define the lower threshold L = Q1 - 1.5 × IQR, and the upper threshold U = Q3 + 1.5 × IQR; S3: If the pressure data P(k) at a certain sampling point is less than L or greater than U, it is determined to be an outlier and the forward filling method is used to replace it with the valid data from the previous moment.

8. The adaptive control system for arteriovenous fistula pressure hemostasis in hemodialysis patients integrating NB-IoT technology according to claim 1, characterized in that, The user terminal includes: The data display unit is used to display pressure-time curves, temperature-time curves, and body motion amplitude-time curves in real time in the form of line graphs; The warning unit will trigger an audible and visual alarm when it detects any of the following conditions: (1) Actual pressure P act (k) Deviation from target pressure P target (k) Exceeding ±5 mmHg for more than 10 seconds; (2) The local temperature T(k) > 38℃ and lasts for more than 5 minutes; (3) The amplitude of body movement A(k) > 10℃, accompanied by a sudden drop in pressure, ΔP(k) < -10mmHg / s.

9. The adaptive control system for arteriovenous fistula pressure hemostasis in hemodialysis patients integrating NB-IoT technology according to claim 1, characterized in that, The power module includes a rechargeable lithium battery with a capacity of ≥2000mAh, a wireless charging coil supporting Qi protocol charging power of 5W, and a low-power management chip. The low-power management chip is used to control the time-sharing wake-up of each module, the sampling frequency of the sensor module is 1Hz, the sleep-wake cycle of the processing module is 10s, and the NB-IoT communication module uploads data 3 times a day, so that the overall power consumption of the system is ≤10mW and the battery life is ≥7 days.

10. The adaptive control system for arteriovenous fistula pressure hemostasis in hemodialysis patients integrating NB-IoT technology according to claims 1-9, characterized in that, The system improves the hemostasis success rate from 78% of traditional fixed pressure devices to ≥92% through a closed-loop mechanism of real-time monitoring, adaptive control, and remote interaction, and reduces the incidence of subcutaneous hematoma caused by excessive pressure from 15% to ≤3%.

Images