Intelligent infusion monitor with complication early warning function and use method thereof

Abstract

The invention relates to an intelligent infusion monitor with a complication early warning function and a using method thereof.The intelligent infusion monitor comprises a monitor body, the monitor body is composed of a front shell, a rear cover plate, an infusion tube drip cup sleeve and multiple functional modules, the middle of the front portion of the front shell is sunken to be used for containing an infusion tube, and a tube clamp is installed above the sunken portion; the rear portion of the front shell and the infusion tube drip cup sleeve are used for containing the multiple functional modules and the control panel, the rear cover plate covers the rear portion of the front shell, the infusion tube drip cup sleeve is used for containing the dripping speed detection module, the multiple functional modules comprise an infusion tube flow control module, a power module, a liquid shortage detection module and the dripping speed detection module, and the modules are all controlled by the control panel. The control panel is provided with a 4G / 5G Internet of Things chip, a capacitance acquisition chip, an integrated sensor chip (altitude, atmospheric pressure and temperature), a switch button and an indicating lamp. The device is simple and reasonable in overall structure, efficient and stable, the working intensity of medical workers can be greatly reduced, and infusion complications are reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of infusion monitors, and relates to an intelligent infusion monitor with a complication early warning function and a method of using the same. Background Art

[0002] Infusion monitors are medical devices that monitor the drip rate and remaining liquid volume of infusions. Existing infusion monitors have the following problems:

[0003] 1. Low data collection accuracy

[0004] 1. The accuracy of infusion drip rate monitoring is insufficient. Single infrared drip rate detection is easily affected by factors such as ambient light, liquid type, and infusion tube material;

[0005] 2. Liquid shortage detection uses capacitor electrodes and infrared acquisition methods, which are easily affected by factors such as ambient temperature, liquid type, and infusion tube material.

[0006] 2. The effect of infrared drip rate alone on the properties of some infused solutions

[0007] The infrared light used for infusion drip rate detection is generally near-infrared light, but near-infrared light will produce different degrees of chemical reactions on the following drugs: a. Doxorubicin (DOX) nano-drug delivery system, which triggers the rapid and ultrasensitive release of doxorubicin after 5 seconds of near-infrared light (NIR) irradiation; b. Gefitinib prodrug system, which specifically activates drug release through glutathione (GSH) under near-infrared light irradiation; c. Quinolones; d. Hormonal drugs;

[0008] 3. Lack of personalization and dynamism in infusion rate adjustment

[0009] The infusion rate for different populations and health conditions can only be indicated by alarms. Patients need to control the infusion rate according to the medical staff, or the medical staff need to make special inspections and interventions. This conventional method not only makes patients unable to rest assured, but also increases the workload of medical staff.

[0010] 4. Inconvenience in inspecting the infusion process

[0011] When medical staff conduct inspections on infusions, they need to observe them closely, and even some infusion monitors may block the infusion drip pot, which increases the workload of medical staff.

[0012] 5. It is difficult to provide early warning of infusion complications (such as drug extravasation, phlebitis, etc.)

[0013] In summary, current infusion monitors only monitor the infusion process and cannot provide early warning of potential complications during the infusion process. However, infusion monitoring has two most important functional goals: 1. Smoothly monitor the infusion process to confirm that the drug solution has successfully entered the vein and completed the treatment process; 2. Provide early warning and preventive intervention for adverse complications that may occur during the infusion process.

[0014] Therefore, an intelligent infusion monitor with complication warning function is designed to overcome the above problems. Summary of the Invention

[0015] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide an intelligent infusion monitor with complication warning function and its use method, which has a simple and reasonable structure, is efficient and stable, and greatly reduces the workload of medical staff.

[0016] The present invention is achieved through the following technical solution: an intelligent infusion monitor with a complication warning function, which includes a monitor body, which is composed of a front shell and a rear cover plate and multiple functional modules arranged in the front shell. The front middle depression of the front shell is used to place the infusion tube, and a tube clamp is installed above the depression to clamp the infusion tube. An infusion tube drip pot cover is provided below the front shell, and the infusion tube drip pot cover is used to cover the infusion tube drip pot. Multiple functional modules and a control board are placed in the front shell and the infusion tube drip pot cover, and the rear cover plate covers the rear of the front shell. The multiple functional modules include an infusion tube flow control module, a power module, a liquid shortage detection module, and a drip rate detection module. The modules are all controlled by the control board, and the control board is also equipped with a 4G / 5G Internet of Things chip, a capacitor acquisition chip, an integrated sensor chip, a switch button, and an indicator light device. The integrated sensor chip is used to collect altitude data, atmospheric pressure data, and temperature data.

[0017] Preferably: a liquid shortage detection module is installed in the middle depression of the front shell, and a power supply module and an infusion tube flow control module are installed on both sides of the liquid shortage detection module respectively. Mounting columns are provided below the power supply module and above the infusion tube flow control module. A fixing hole is also provided on the control panel at a position relative to the mounting column, which is fixed to the rear of the power supply module, the infusion tube flow control module and the liquid shortage detection module through the fixing hole. A button rod is also provided below the power supply module, which extends to the button of the front shell in front and to the switch button on the control panel in the rear. The dripping rate detection module is installed on the outside of the infusion tube drip pot cover to detect the dripping situation in the infusion tube drip pot.

[0018] Preferably: the control board is equipped with a main control chip, a capacitance acquisition chip, an integrated sensor chip, a switch button, an indicator light device, a sound device, a 4G / 5G Internet of Things chip, and a battery management chip, wherein the battery management chip is connected to the battery and the charging port, the 4G / 5G Internet of Things chip is connected to an external device, the indicator light device is connected to the IO port of the main chip installed on the control board, the sound device is connected to the IO port of the main chip on the control board, the capacitance acquisition chip is connected to the IO port of the main chip, and the ADC sampling function integrated in the capacitance acquisition chip is connected to the liquid shortage detection module and the drip rate detection module, converting the capacitance change into a digital signal, and then transmitting it to the main control chip for data analysis. The main control chip simultaneously obtains the integrated sensor chip for analysis and calculation, and finally obtains the compensated parameter value, and connects the obtained data to the external device through the 4G / 5G Internet of Things chip for analysis. The main control chip is also connected to the infusion tube flow control module, and the infusion tube flow control module can be put into operation in time by detecting and analyzing the infusion situation. Preferably, the liquid deficiency detection module is composed of two arc-shaped electrode pieces, which are respectively installed on the left and right sides to form a semicircular channel. The infusion tube is confined between the two solitary electrode pieces by the tube clamp above. The two arc-shaped electrode pieces are connected to the capacitance acquisition chip through cables below to collect data and transmit it to the main chip. The capacitance acquisition chip has an integrated ADC sampling circuit.

[0019] Preferably, the infusion tube flow control module is composed of a stepper motor, a motor bracket, a power gear, and a lock tongue rod. The stepper motor is installed on the motor bracket, the power gear is installed above the motor bracket, the power gear is connected to the transmission shaft of the stepper motor, and the lock tongue rod is provided with a gear edge on the side and a lock tongue at the end. The gear edge is connected to the power gear, and the lock tongue is driven forward by the motor to clamp the infusion tube.

[0020] Preferably, the dripping speed detection module is composed of a dripping speed capacitor detection module and a dripping speed infrared detection module, wherein the dripping speed capacitor detection module is composed of capacitor electrodes arranged on opposite sides, and the dripping speed infrared detection module is composed of infrared radiating tubes arranged on opposite sides, and the capacitor electrodes arranged on opposite sides are installed above the PCB substrates arranged on the left and right sides. A through hole for installing the infrared radiating tube is opened on the electrode piece for installing the infrared radiating tube, one end of the infrared radiating tube is fixed to the rear of the PCB substrate through the electrode piece, and the other end is inserted into the through hole opened on the infusion tube drip pot cover and fixed. The PCB substrate is suspended outside the infusion tube drip pot cover, and the upper part is inserted into the bottom of the front shell and fixed. The infrared radiating tube is composed of an infrared transmitter and an infrared receiver, which are arranged on opposite sides. One infrared receiver is provided and at least two infrared transmitters are provided.

[0021] A method for using an intelligent infusion monitor with a complication warning function, the method comprising an installation method, a disassembly method, and a warning method, wherein:

[0022] Installation method: pinch the upper end of the drip pot of the infusion set with one hand, hold the monitor with the other hand, and pull the monitor down so that the drip pot cover inside the monitor fits the upper end of the drip pot of the infusion set. With a little force, the pipe is clamped into the pipe clamp at the upper end of the product. Press and hold the switch button for 3 seconds. At this time, the control panel turns on the liquid shortage detection module, drip rate capacitance detection module, and drip rate infrared detection module to monitor the infusion process. When the monitor detects that the infusion is completed, it triggers the infusion tube flow control module to automatically complete the pipe locking. All collected data and abnormalities can be fed back to the medical staff or patient display terminal;

[0023] Disassembly method: Short press the monitor switch or control it through the mobile phone to unlock the monitor. Pinch the upper end of the drip pot of the infusion device with one hand and gently pull out the infusion tube to make it free from the tube clamp. Hold the monitor with the other hand and pull up the monitor to fully expose the drip pot of the infusion tube. Then you can remove the monitor.

[0024] The early warning method is specifically:

[0025] 1) Data acquisition: The aforementioned device is used to monitor the droplets in the drip pot using infrared and capacitive detection. The infrared sampling frequency is ≥100 Hz, capturing the time intervals and morphological changes of the droplets.

[0026] 2) Feature extraction algorithm:

[0027] Time Series Analysis

[0028] a. Calculate the time difference (Δt) between adjacent droplets after moving average filtering;

[0029] b. Abnormal determination: Δt>1.5 times the baseline value for more than 5 cycles;

[0030] 3) Frequency domain feature extraction:

[0031] Perform fast Fourier transform (FFT) on the Δt sequence to identify the characteristic frequency components of 0.1-0.5 Hz;

[0032] 4) Decision-making algorithm:

[0033] a. Support vector machine (SVM) classification;

[0034] b. Input features: Δt variance, main frequency component, pressure gradient;

[0035] c. Training data: Hundreds of normal / needle removal samples collected clinically to improve the accuracy of early warning;

[0036] 5) Dynamic threshold warning:

[0037] a.Baseline adaptive adjustment formula:

[0038] b.Threshold ≥ 1.5σ

[0039] Where σ is the average drip rate in the previous 3 minutes, and an alarm is triggered when the threshold is exceeded;

[0040] 6) Complication trend prediction model:

[0041] Use the autoregressive integrated moving average model time series to predict the drip rate in the next 3 minutes, and issue an early warning when the predicted value falls below the safety threshold;

[0042] Through time series analysis and feature extraction, we fit the vital sign parameters associated with intravenous treatment complications and the training data of drip rate changes during the occurrence of hundreds of complications to provide early warning of abnormal situations.

[0043] Feature extraction:

[0044] Feature vector = [heart rate change rate, blood oxygen fluctuation amplitude, drip rate change rate, ...];

[0045] Early warning model:

[0046] P(anomaly) = sigmoid(W·eigenvector+b);

[0047] Among them, W is the weight matrix, b is the bias term, and sigmoid is the activation function.

[0048] As an example, the specific working method of the dripping speed detection module is as follows:

[0049] 1) Infrared detection: 940nm infrared emitting and receiving tubes are installed on both sides of the drip pot of the infusion tube. The detection is achieved by using the characteristic of liquid droplets blocking infrared light. It is suitable for most conventional liquid medicines.

[0050] 2) Capacitance detection: Counting by collecting regular changes in capacitance as the droplet falls;

[0051] 3) Combine the data collected by the integrated sensor chip to develop an accurate algorithm to improve the accuracy of drip rate detection;

[0052] The specific algorithm is:

[0053] Dripping rate = α * infrared detection result + (1-α) * capacitance detection result;

[0054] Among them, α is the weight coefficient. When α is 0, the result of capacitance detection is used. When α>0, the dripping rate is calculated by combining infrared and capacitance, achieving the purpose of infrared and capacitance dual detection.

[0055] In the process of obtaining the capacitance test results, the reference capacitance value is calibrated under standard conditions (such as 25°C, 1013hPa, and 0m above sea level). For other areas with significant changes in temperature, air pressure, and altitude, and for ambient temperatures, a compensation coefficient should be added when recording the capacitance change rate. The specific formula for the compensation coefficient is:

[0056] C eompensated =C raw *(1+K T *ΔT+Kp*ΔP+K A *ΔA)*(1+Δ∈ r / ∈ r0 ))

[0057] K T ,K P ,K A : Temperature, air pressure, and altitude compensation coefficients are determined through calibration experiments.

[0058] ΔT=T urent -T caibration

[0059] ΔP=P eurrent -P ealibration

[0060] ΔA=A current -A ealibration

[0061] Where C is the capacitance; K T is the temperature; K P is the air pressure; K A is the altitude; Δ∈ r , is the change of dielectric constant of the liquid.

[0062] Preferably, the liquid shortage detection module adopts a non-contact detection technology, which realizes liquid level detection by monitoring the changes in the electric field around the infusion tube. The specific method is as follows: when there is liquid in the infusion tube, the liquid acts as a dielectric to change the surrounding electric field distribution, causing the capacitance value of the detection electrode to increase; when the liquid is emptied, the capacitance value returns to its initial state, and the data is collected to the capacitance acquisition chip, and the capacitance change is converted into a digital signal output, and the liquid shortage alarm is realized through the linear change curve rule and threshold judgment.

[0063] Preferably, the liquid shortage detection module further includes a liquid shortage detection compensation algorithm. Since the capacitance is affected by the temperature and the medicine in the dropper, a compensation algorithm is added to the liquid shortage detection module. The specific compensation algorithm is:

[0064] C 补偿 =C 测量 ●(1+ΔT / T0)●(1+Δ∈ r / ∈ r0)

[0065] Among them, C 补偿 is the capacitance value after compensation, △T is the temperature change, Δ∈ r , is the change of dielectric constant of the liquid.

[0066] Compared with the prior art, the present invention has the following beneficial effects:

[0067] 1. The present invention adopts a dual droplet recognition solution innovation: it uses infrared light and capacitance detection fusion technology to improve the accuracy of drip rate detection. At the same time, when the drug properties of the patient's infused liquid change under near-infrared light irradiation, the infrared drip rate detection is turned off and switched to capacitance drip rate monitoring mode only. Although the drip rate recognition accuracy is reduced (capacitance detection recognition accuracy ≥ 90%), the infusion drip rate monitoring can still be kept in a controllable state.

[0068] 2. The liquid shortage detection chip of the present invention adopts MC11S and non-contact detection technology. By monitoring the linear change of capacitance value when liquid shortage occurs in the infusion pipeline, and combining the capacitance change compensation caused by temperature change, it can more accurately improve the accuracy of liquid shortage and dripping rate judgment, and reduce detection errors caused by bubbles or pipeline material.

[0069] 3. The present invention is designed with a micro DC stepper motor to drive the lock tongue lever to squeeze and change the flow diameter of the infusion tube, dynamically adjusting the infusion rate to ensure infusion safety and effectiveness. The innovative value of this function lies in: it can safely infuse according to standards when the patient is not feeling or resting, and it also reduces the difficulty for medical staff to accurately control the drip rate;

[0070] 4. The indicator light of the present invention can enable medical staff to observe the indicator light from a distance and multiple angles without disturbing the patient, and quickly identify the infusion status and abnormal type, which reduces the workload of medical staff and improves work efficiency.

[0071] 5. Through a large number of complications associated with changes in infusion drip rate, an infusion database is constructed to form an early warning algorithm for infusion complications, thereby reducing medical accidents caused by infusion complications.

[0072] In summary, the infusion monitor designed by the present invention can well realize the practical medical scientific and technological value and improve the core competitiveness in the market. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 It is a schematic diagram of the overall structure of the present invention (front).

[0074] Figure 2 It is a schematic diagram of the overall structure of the present invention (back).

[0075] Figure 3 The internal structure of the present invention is shown in FIG. Figure 1 (back).

[0076] Figure 4 The internal structure of the present invention is shown in FIG. Figure 2 (back).

[0077] Figure 5 It is a structural diagram of the control board in the present invention.

[0078] Figure 6 It is a structural diagram of the liquid shortage detection module in the present invention.

[0079] Figure 7 It is a structural diagram of the infusion tube flow control module in the present invention.

[0080] Figure 8 It is a structural diagram of the dripping rate detection module in the present invention. DETAILED DESCRIPTION

[0081] In order to enable those skilled in the art to more clearly understand the objectives, technical solutions and advantages of the present invention, the present invention is further described below with reference to the accompanying drawings and embodiments.

[0082] In the description of the present invention, it should be understood that the orientations or positional relationships indicated by terms such as “upper”, “lower”, “left”, “right”, “inside”, “outside”, “horizontal” and “vertical” are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention, rather than indicating or implying that the device or component referred to must have a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0083] The present invention will be described in detail below with reference to the accompanying drawings: Figure 1-3 As shown, an intelligent infusion monitor with a complication warning function includes a monitor body, which is composed of a front shell 1 and a rear cover 2 and a variety of functional modules arranged in the front shell 1. The front middle depression of the front shell 1 is used to place the infusion tube 3, and a tube clamp 4 is installed above the depression to clamp the infusion tube 3. An infusion tube drip pot cover 6 is provided below the front shell 1, and the infusion tube drip pot cover 6 is used to cover the infusion tube drip pot. Various functional modules and a control board 5 are placed in the front shell 1 and the infusion tube drip pot cover 6. The rear cover 2 covers the rear of the front shell 1. The various functional modules include an infusion tube flow control module 7, a power module 8, a liquid shortage detection module 9, and a drip rate detection module 10. The modules are all controlled by the control board 5, and a 4G / 5G Internet of Things chip 15, a capacitance acquisition chip 11, an integrated sensor chip 12, a switch button 13, and an indicator light device 14 are also installed on the control board 5. The integrated sensor chip 12 is used to collect altitude data, atmospheric pressure data, and temperature data.

[0084] A liquid shortage detection module 9 is installed in the middle recess of the front shell 1, and a power supply module 8 and an infusion tube flow control module 7 are installed on both sides of the liquid shortage detection module 9. Installation columns are provided below the power supply module 8 and above the infusion tube flow control module 9. A fixing hole is also provided on the control panel 5 at a position relative to the installation column, and is fixed to the rear of the power supply module 8, the infusion tube flow control module 7 and the liquid shortage detection module 9 through the fixing hole. A button rod 16 is also provided below the power supply module 8, and the button rod 16 extends to the button 17 of the front shell 1 in front and extends to the switch button 13 on the control panel 5 in the rear. The dripping rate detection module 10 is installed on the outside of the infusion tube drip pot sleeve 6 for detecting the dripping situation in the infusion tube drip pot 25.

[0085] like Figure 4-5 As shown, the control board 5 is equipped with a main control chip 19, a capacitance acquisition chip 11, an altitude atmospheric pressure, an integrated sensor chip 12, a switch button 13, an indicator light device 14, a sound device 24, a 4G / 5G Internet of Things chip 15, and a battery management chip 27, wherein the battery management chip 27 is connected to the battery 22 and the charging port 23, the 4G / 5G Internet of Things chip 15 is connected to an external device, the indicator light device 14 is connected to the main chip IO port installed on the control board 5, the sound device 24 is connected to the main chip IO port on the control board, the capacitance acquisition chip 11 is connected to the main chip IO port, and the capacitance The ADC sampling function 26 integrated in the acquisition chip 11 is connected to the liquid shortage detection module 9 and the drip rate detection module 10, converts the capacitance change into a digital signal, and then transmits it to the main control chip 19 for data analysis. The main control chip 19 simultaneously obtains the integrated sensor chip 12 for analysis and calculation, and finally obtains the compensated parameter value. The obtained data is connected to the external device through the 4G / 5G Internet of Things chip 15 for analysis. The main control chip 19 is also connected to the infusion tube flow control module 7. By detecting and analyzing the infusion situation, the infusion tube flow control module 7 can be put into operation in time.

[0086] like Figure 6 As shown, the liquid shortage detection module 9 is composed of two arc-shaped electrode pieces 28, which are respectively installed on the left and right sides to form a semicircular channel. The infusion tube 3 is confined between the two solitary electrode pieces 28 through the tube clamp 4 above. The two arc-shaped electrode pieces 28 are connected to the capacitor acquisition chip 11 through cables below to collect data and transmit it to the main chip. The capacitor acquisition chip 11 has an integrated ADC sampling circuit.

[0087] like Figure 7As shown, the infusion tube flow control module 7 is composed of a stepper motor 29, a motor bracket 30, a power gear 31, and a lock tongue rod 32. The stepper motor 29 is installed on the motor bracket 30, and the power gear 31 is installed above the motor bracket 30. The power gear 31 is connected to the transmission shaft 33 of the stepper motor 29. The lock tongue rod 32 has a gear edge 34 on the side and a lock tongue 35 at the end. The gear edge 34 is connected to the power gear 31, and the lock tongue 35 is driven forward by the motor to clamp the infusion tube 3.

[0088] like Figure 8 As shown, the dripping speed detection module 10 is composed of a dripping speed capacitor detection module and a dripping speed infrared detection module, wherein the dripping speed capacitor detection module is composed of capacitor electrodes 36 arranged on opposite sides, and the dripping speed infrared detection module is composed of infrared radiating tubes arranged oppositely, and the capacitor electrodes 36 arranged on opposite sides are installed above the PCB substrates 37 arranged on the left and right sides. A through hole for installing the infrared radiating tube is opened on the electrode piece for installing the infrared radiating tube. One end of the infrared radiating tube passes through the electrode piece and is fixed behind the PCB substrate 37, and the other end is inserted into the through hole opened on the infusion tube drip pot cover 6 and is fixed. The PCB substrate 37 is suspended on the outside of the infusion tube drip pot cover 6 and is inserted into the bottom of the front shell 1 and fixed. The infrared radiating tube is composed of an infrared emitter 39 and an infrared receiver 38, which are arranged on opposite sides. One infrared receiver 38 is provided and at least two infrared emitters 39 are provided.

[0089] A method for using an intelligent infusion monitor with a complication warning function, the method comprising an installation method, a disassembly method, and a warning method, wherein:

[0090] Installation method: pinch the upper end of the drip pot of the infusion set with one hand, hold the monitor with the other hand, and pull the monitor down so that the drip pot cover inside the monitor fits the upper end of the drip pot of the infusion set. With a little force, the pipe is clamped into the pipe clamp at the upper end of the product. Press and hold the switch for 3 seconds. At this time, the control panel 5 turns on the liquid shortage detection module 9, the dripping rate capacitance detection module, and the dripping rate infrared detection module to monitor the infusion process. When the monitor detects that the infusion is completed, it triggers the infusion tube flow control module to automatically complete the pipe locking. All collected data and abnormalities can be fed back to the medical staff or patient display terminal;

[0091] Disassembly method: Short press the monitor switch or control it through the mobile phone to unlock the monitor. Pinch the upper end of the drip pot of the infusion device with one hand and gently pull out the infusion tube to make it free from the tube clamp. Hold the monitor with the other hand and pull up the monitor to fully expose the drip pot of the infusion tube. Then you can remove the monitor.

[0092] The early warning method is specifically:

[0093] 1) Data acquisition: The aforementioned device is used to monitor the droplets in the drip pot using infrared and capacitive detection. The infrared sampling frequency is ≥100 Hz, capturing the time intervals and morphological changes of the droplets.

[0094] 2) Feature extraction algorithm:

[0095] Time Series Analysis

[0096] a. Calculate the time difference (Δt) between adjacent droplets after moving average filtering;

[0097] b. Abnormal determination: Δt>1.5 times the baseline value for more than 5 cycles;

[0098] 3) Frequency domain feature extraction:

[0099] Perform fast Fourier transform (FFT) on the Δt sequence to identify the characteristic frequency components of 0.1-0.5Hz

[0100] 4) Decision-making algorithm:

[0101] a. Support vector machine (SVM) classification;

[0102] b. Input features: Δt variance, main frequency component, pressure gradient;

[0103] c. Training data: Hundreds of normal / needle removal samples collected clinically to improve the accuracy of early warning

[0104] 5) Dynamic threshold warning:

[0105] a.Baseline adaptive adjustment formula:

[0106] b.Threshold ≥ 1.5σ

[0107] Where σ is the average drip rate in the previous 3 minutes, and an alarm is triggered when the threshold is exceeded;

[0108] 6) Complication trend prediction model:

[0109] Use the autoregressive integrated moving average model time series to predict the drip rate in the next 3 minutes, and issue an early warning when the predicted value falls below the safety threshold;

[0110] Through time series analysis and feature extraction, we fitted the vital sign parameters associated with intravenous therapy complications, hundreds of complication cases,

[0111] The training data of drip rate change rules during the production process can provide early warning of abnormal situations;

[0112] Feature extraction:

[0113] Feature vector = [heart rate change rate, blood oxygen fluctuation amplitude, drip rate change rate, ...];

[0114] Early warning model:

[0115] P(anomaly) = sigmoid(W·eigenvector+b);

[0116] Among them, W is the weight matrix, b is the bias term, and sigmoid is the activation function.

[0117] The specific working method of the dripping speed detection module is:

[0118] 1) Infrared detection: 940nm infrared emitting tubes and infrared receiving tubes are installed on both sides of the drip pot of the infusion tube. The detection is achieved by using the characteristic of liquid droplets blocking infrared light. It is suitable for most conventional liquid medicines.

[0119] 2) Capacitance detection: Counting drops by capturing regular changes in capacitance as they fall. Specifically, this method uses multi-band capacitance measurement (1kHz-1MHz) to measure the capacitance of falling drops. Infusion information is sent via the HIS system. Capacitance values for different drops are collected and aggregated to build a library of liquid dielectric fingerprints. Intelligent algorithms are then used to gradually optimize the accuracy of capacitance drop rate detection.

[0120] 3) Combine the data collected by the integrated sensor chip to develop an accurate algorithm to improve the accuracy of drip rate detection;

[0121] Temperature integrated sensor acquisition includes:

[0122] Auxiliary sensor, model: ICP-20100

[0123] The capacitance detection method based on dynamic compensation of dielectric constant eliminates the influence of liquid type (such as different conductivity) on the detection results. Temperature sensor: (compensates for changes in dielectric constant of medium)

[0124] Atmospheric pressure sensor: (compensates for changes in liquid density / viscosity)

[0125] Altitude sensor: (indirectly compensates for the effect of air pressure on droplet formation)

[0126] The specific algorithm is:

[0127] Dripping rate = α * infrared detection result + (1-α) * capacitance detection result;

[0128] Among them, α is the weight coefficient. When α is 0, the result of capacitance detection is used. When α>0, the dripping rate is calculated by combining infrared and capacitance, achieving the purpose of infrared and capacitance dual detection.

[0129] In the process of obtaining the capacitance test results, if the reference capacitance value is calibrated under a standard environment, a compensation coefficient should be added when recording the capacitance change rate due to changes in temperature, air pressure, and altitude in special regions. The specific formula for the compensation coefficient is:

[0130] C eompensated =C raw *(1+K T *ΔT+Kp*ΔP+K A *ΔA)*(1+Δ∈ r / ∈ r0 ))

[0131] K T ,K P ,K A : Temperature, air pressure, and altitude compensation coefficients are determined through calibration experiments.

[0132] ΔT=T urent -T caibration

[0133] ΔP=P eurrent -P ealibration

[0134] ΔA=A current -A ealibration

[0135] Where C is the capacitance; K T is the temperature; K P is the air pressure; K A is the altitude; Δ∈ r , is the change of dielectric constant of the liquid.

[0136] Environmental capacitance calibration process:

[0137] a. Calibrate the reference capacitance value under standard environment (such as 25℃, 1013hPa, altitude 0m);

[0138] b. By changing the temperature / air pressure / altitude, record the capacitance change rate and fit the compensation coefficient;

[0139] The liquid shortage detection module uses non-contact detection technology, which realizes liquid level detection by monitoring the changes in the electric field around the infusion tube. The specific method is as follows: when there is liquid in the infusion tube, the liquid acts as a dielectric to change the surrounding electric field distribution, causing the capacitance value of the detection electrode to increase; when the liquid is emptied, the capacitance value returns to its initial state, and the data is collected to the ADC module, which converts the capacitance change into a digital signal output, and realizes the liquid shortage alarm through the linear change curve rule and threshold judgment.

[0140] The liquid shortage detection module also includes a liquid shortage detection compensation algorithm. Since the capacitance is affected by the temperature and the medicine in the dropper, a compensation algorithm is added to the liquid shortage detection module. The specific compensation algorithm is:

[0141] C 补偿 =C 测量 ●(1+ΔT / T0)●(1+Δ∈ r / ∈ r0 )

[0142] Among them, C 补偿 is the capacitance value after compensation, △T is the temperature change, Δ∈ r , is the change of dielectric constant of the liquid.

[0143] The liquid shortage detection module uses MC11S and non-contact detection technology to detect liquid level by monitoring the changes in the electric field around the infusion tube. Its working principle is as follows:

[0144] Capacitance change mechanism: When there is liquid in the infusion tube, the liquid acts as a dielectric and changes the surrounding electric field distribution, causing the capacitance value of the MC11S detection electrode to increase; when the liquid is drained, the capacitance value returns to its initial state.

[0145] Signal conversion: MC11S has an internal integrated ADC module that converts capacitance changes into digital signal output, and implements liquid shortage alarm through linear change curve rules and threshold judgment.

[0146] Innovative advantages of this design: Compared with traditional photoelectric or gravity sensors, capacitive detection has the following advantages: a. Non-contact: No direct contact with the infusion tube is required, avoiding the risk of contamination. b. Strong anti-interference ability: Not affected by liquid color or lighting conditions. c. Low power consumption: The operating current of the MC11S chip is only 0.5mA, which is suitable for battery-powered scenarios and reduces the impact of the monitor's weight on the infusion tube falling off.

[0147] The design features of the present invention are as follows:

[0148] drip rate control

[0149] The product is designed with a DC stepper motor, which is driven by the electric pulse control of the stepper motor to move the lock tongue rod. Its movement accuracy can reach 0.05 mm in length. The flow diameter of the infusion tube is changed by squeezing, thereby changing the infusion drip rate, and the lock tongue rod is automatically fine-tuned by comparing the data collected from the drip rate. This control method is based on the patient's age, health status and vital signs data, and refers to medical standard requirements.

[0150] Indicator device display

[0151] The indicator light device consists of a medical inspection drip rate indicator light and an infusion abnormality indicator light, among which:

[0152] a. Medical inspection drip rate indicator: An RGB flashing indicator light is set on the top of the product. When the infusion is normal, it will show green light, and its flashing frequency is exactly the same as the drip rate. When the infusion is abnormal, it will flash yellow. When the infusion is completed or a complication is warned, it will show red. This makes it easy for medical staff to inspect the infusion status from a distance, improves the work efficiency of medical staff, and reduces the workload. It also makes it easier for accompanying personnel to detect infusion abnormalities. And since it is set on the top, it will not cause light interference to the rest of the patient and surrounding personnel.

[0153] Infusion abnormality indicator light: There is an indicator light at the bottom of the product. When the dripping rate is abnormal, another light will remind the patient of the abnormal event through flashing reminders and colors (red light, yellow light). The indicator light is always off when the dripping rate is normal to ensure that it will not affect the patient. The indicator light is designed at the bottom to ensure that the patient and the caregiver can clearly know when it flashes.

[0154] Complications warning algorithm:

[0155] Fluid mechanics principles of abnormal drip rate caused by needle removal

[0156] Pressure gradient mutation: When the needle is partially or completely separated from the blood vessel, the resistance of the infusion system suddenly changes from the intravascular pressure (about 5-15 mmHg) to the interstitial pressure (negative pressure or close to atmospheric pressure). At this time, the liquid flow resistance increases by 3-5 times, which manifests as a sudden drop in drip rate (such as from 60 drops / min to less than 10 drops / min) or intermittent stagnation.

[0157] Liquid column oscillation effect

[0158] When the liquid seeps into the tissue space after the needle is removed, the liquid column in the infusion line will form periodic oscillations:

[0159] a. Seepage stage: the liquid column pressure decreases and the dripping rate slows down.

[0160] b. Retraction phase: The elastic blood vessel wall retracts to generate reverse pressure, briefly restoring the drip rate. This oscillation presents a periodic fluctuation of 0.1-0.5 Hz.

[0161] The present invention uses ARIMA time series to predict the drip rate in the next 3 minutes, and issues an early warning when the predicted value is lower than a safety threshold (e.g. <15 drops / min).

[0162] 1) The intelligent infusion monitor with complication warning function designed by the present invention can effectively solve the insufficient accuracy of existing infusion drip rate detection and liquid shortage detection; the infrared light used for drip rate detection in the intelligent infusion monitor with complication warning function can change the medicinal properties of some liquid medicines; the infusion speed can be well personalized and dynamically adjusted automatically; during the infusion process, when the monitor detects that the infusion is completed, the infusion tube flow control module is triggered to automatically complete the tube locking. All collected data and abnormalities can be fed back to the medical staff or patient display terminal. This design does not require medical staff to check frequently, greatly reducing workload.

[0163] The specific embodiments described herein are merely illustrative of the principles and effects of the present invention and are not intended to limit the present invention. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical concepts disclosed herein are intended to be covered by the claims of the present invention.

Claims

1. An intelligent infusion monitor with complication warning function, comprising a monitor body, the body comprising a front shell, a rear cover, and multiple functional modules disposed within the front shell, characterized in that: The middle depression in the front of the front shell is used to place the infusion tube, and a tube clamp is installed above the depression to clamp the infusion tube. An infusion tube drip pot cover is provided below the front shell, and the infusion tube drip pot cover is used to cover the infusion tube drip pot. Various functional modules and a control board are placed in the front shell and the infusion tube drip pot cover, and the rear cover covers the rear of the front shell. The various functional modules include an infusion tube flow control module, a power module, a liquid shortage detection module, and a dripping rate detection module. The modules are all controlled by the control board, and a 4G / 5G Internet of Things chip, a capacitor acquisition chip, an integrated sensor chip, a switch button, and an indicator light device are also installed on the control board. The integrated sensor chip is used to collect altitude data, atmospheric pressure data, and temperature data.

2. The intelligent infusion monitor with complication warning function according to claim 1, characterized in that: A liquid shortage detection module is installed in the middle depression of the front shell, and a power supply module and an infusion tube flow control module are installed on both sides of the liquid shortage detection module respectively. Mounting columns are provided below the power supply module and above the infusion tube flow control module. A fixing hole is also provided on the control panel at a position relative to the mounting column, which is fixed to the rear of the power supply module, the infusion tube flow control module and the liquid shortage detection module through the fixing hole. A button rod is also provided below the power supply module, which extends to the button of the front shell in front and to the switch button on the control panel in the rear. The dripping rate detection module is installed on the outside of the infusion tube drip pot cover for detecting the dripping situation in the infusion tube drip pot.

3. The intelligent infusion monitor with complication warning function according to claim 1, characterized in that: The control board is equipped with a main control chip, a capacitance acquisition chip, an integrated sensor chip, a switch button, an indicator light device, a sound device, a 4G / 5G Internet of Things chip, and a battery management chip, wherein the battery management chip is connected to the battery and the charging port, the 4G / 5G Internet of Things chip is connected to an external device, the indicator light device is connected to the IO port of the main chip installed on the control board, the sound device is connected to the IO port of the main chip on the control board, the capacitance acquisition chip is connected to the IO port of the main chip, and the ADC sampling function integrated in the capacitance acquisition chip is connected to the liquid shortage detection module and the drip rate detection module, converting the capacitance change into a digital signal, and then transmitting it to the main control chip for data analysis. The main control chip simultaneously obtains the integrated sensor chip for analysis and calculation, and finally obtains the compensated parameter value, and connects the obtained data to the external device through the 4G / 5G Internet of Things chip for analysis. The main control chip is also connected to the infusion tube flow control module, and the infusion tube flow control module can be put into operation in time by detecting and analyzing the infusion situation.

4. The intelligent infusion monitor with complication warning function according to claim 1, characterized in that: The liquid deficiency detection module consists of two arc-shaped pole pieces, which are installed on the left and right sides respectively to form a semicircular channel. The infusion tube is confined between the two isolated pole pieces through the tube clamp above. The bottom of the two arc-shaped pole pieces is connected to the capacitance acquisition chip through a cable to collect data and transmit it to the main chip. The capacitance acquisition chip has an integrated ADC sampling circuit.

5. The intelligent infusion monitor with complication warning function according to claim 1, characterized in that: The infusion tube flow control module consists of a stepper motor, a motor bracket, a power gear, and a lock tongue rod. The stepper motor is installed on the motor bracket, the power gear is installed above the motor bracket, the power gear is connected to the transmission shaft of the stepper motor, the lock tongue rod has a gear edge on the side, and the end is a lock tongue. The gear edge is connected to the power gear, and the lock tongue is driven forward by the motor to clamp the infusion tube.

6. The intelligent infusion monitor with complication warning function according to claim 1, characterized in that: The dripping speed detection module is composed of a dripping speed capacitor detection module and a dripping speed infrared detection module, wherein the dripping speed capacitor detection module is composed of capacitor electrodes arranged on opposite sides, and the dripping speed infrared detection module is composed of infrared radiating tubes arranged on opposite sides. The capacitor electrodes arranged on opposite sides are installed above the PCB substrates arranged on the left and right sides. A through hole for installing the infrared radiating tube is opened on the electrode piece for installing the infrared radiating tube. One end of the infrared radiating tube passes through the electrode piece and is fixed to the rear of the PCB substrate, and the other end is inserted into the through hole opened on the infusion tube drip pot cover and fixed. The PCB substrate is suspended outside the infusion tube drip pot cover and inserted into the bottom of the front shell and fixed. The infrared radiating tube is composed of an infrared transmitter and an infrared receiver, which are arranged on opposite sides. One infrared receiver is provided and at least two infrared transmitters are provided.

7. A method for using the intelligent infusion monitor with complication warning function according to any one of claims 1 to 6, characterized in that: The method includes an installation method, a disassembly method, and an early warning method, wherein: Installation method: pinch the upper end of the drip pot of the infusion set with one hand, hold the monitor with the other hand, and pull the monitor down so that the drip pot cover inside the monitor fits the upper end of the drip pot of the infusion set. With a little force, the pipe is clamped into the pipe clamp at the upper end of the product. Press and hold the switch button for 3 seconds. At this time, the control panel turns on the liquid shortage detection module, drip rate capacitance detection module, and drip rate infrared detection module to monitor the infusion process. When the monitor detects that the infusion is completed, it triggers the infusion tube flow control module to automatically complete the pipe locking. All collected data and abnormalities can be fed back to the medical staff or patient display terminal; Disassembly method: Short press the monitor switch or control it through the mobile phone to unlock the monitor. Pinch the upper end of the drip pot of the infusion device with one hand and gently pull out the infusion tube to make it free from the tube clamp. Hold the monitor with the other hand and pull up the monitor to fully expose the drip pot of the infusion tube. Then you can remove the monitor. The early warning method is specifically as follows: 1) Data acquisition: The aforementioned device is used to monitor the droplets in the drip pot using infrared and capacitive detection. The infrared sampling frequency is ≥100 Hz, capturing the time intervals and morphological changes of the droplets. 2) Feature extraction algorithm: Time Series Analysis a. Calculate the time difference (Δt) between adjacent droplets after moving average filtering; b. Abnormal determination: Δt>1.5 times the baseline value for more than 5 cycles; 3) Frequency domain feature extraction: Perform fast Fourier transform (FFT) on the Δt sequence to identify the characteristic frequency components of 0.1-0.5 Hz; 4) Decision-making algorithm: a. Support vector machine (SVM) classification; b. Input features: Δt variance, main frequency component, pressure gradient; c. Training data: Hundreds of normal / needle removal samples collected clinically to improve the accuracy of early warning; 5) Dynamic threshold warning: a. Baseline adaptive adjustment formula: b.Threshold ≥ 1.5σ Where σ is the average drip rate in the previous 3 minutes, and an alarm is triggered when the threshold is exceeded; 6) Complication trend prediction model: Use the autoregressive integrated moving average model time series to predict the drip rate in the next 3 minutes, and issue an early warning when the predicted value falls below the safety threshold; Through time series analysis and feature extraction, we fit the vital sign parameters associated with intravenous treatment complications and the training data of drip rate changes during the occurrence of hundreds of complications to provide early warning of abnormal situations. Feature extraction: Feature vector = [heart rate change rate, blood oxygen fluctuation amplitude, drip rate change rate, ...]; Early warning model: P(anomaly) = sigmoid(W·eigenvector+b); Among them, W is the weight matrix, b is the bias term, and sigmoid is the activation function.

8. The method for using the intelligent infusion monitor with complication warning function according to claim 7, characterized in that: The specific working method of the dripping speed detection module is: 1) Infrared detection: Infrared emitting and receiving tubes are installed on both sides of the drip pot of the infusion tube. The detection is achieved by using the characteristic of liquid droplets blocking infrared light. It is suitable for most conventional liquid medicines. 2) Capacitance detection: Counting by collecting regular changes in capacitance as the droplet falls; 3) Combine the data collected by the integrated sensor chip to develop an accurate algorithm to improve the accuracy of drip rate detection; The specific algorithm is: Dripping rate = α * infrared detection result + (1-α) * capacitance detection result; Among them, α is the weight coefficient. When α is 0, the result of capacitance detection is used. When α>0, the dripping rate is calculated by combining infrared and capacitance, achieving the purpose of infrared and capacitance dual detection. In the process of obtaining the capacitance test results, the reference capacitance value is calibrated under standard conditions (such as 25°C, 1013hPa, and 0m above sea level). For other areas with significant changes in temperature, air pressure, and altitude, a compensation coefficient should be added when recording the capacitance change rate, and a compensation method for changes in the dielectric constant of the liquid should be added. The specific formula for the compensation coefficient is: C eompensated =C raw *(1+K T *ΔT+Kp*ΔP+K A *ΔA)*(1+Δ∈ r / ∈ r0 )) K T ,K P ,K A : Temperature, air pressure, and altitude compensation coefficients are determined through calibration experiments. ΔT=T urent -T caibration ΔP=P eurrent -P ealibration ΔA=A current -IN ealibration Where C is the capacitance; K T is the temperature; K P is the air pressure; K A is the altitude; Δ∈ r , is the change of dielectric constant of the liquid.

9. The method for using the intelligent infusion monitor with complication warning function according to claim 7, characterized in that: The liquid shortage detection module adopts non-contact detection technology, which realizes liquid level detection by monitoring the changes in the electric field around the infusion tube. The specific method is as follows: when there is liquid in the infusion tube, the liquid acts as a dielectric to change the surrounding electric field distribution, causing the capacitance value of the detection electrode to increase; when the liquid is emptied, the capacitance value returns to its initial state, and the data is collected to the capacitance acquisition chip, and the capacitance change is converted into a digital signal output. The liquid shortage alarm is realized through the linear change curve rule and threshold judgment.

10. The method for using the intelligent infusion monitor with complication warning function according to claim 9, characterized in that: The liquid shortage detection module also includes a liquid shortage detection compensation algorithm. Since the capacitance is affected by the temperature and the medicine in the dropper, a compensation algorithm is added to the liquid shortage detection module. The specific compensation algorithm is: C 补偿 =C 测量 ●(1+ΔT / T0)●(1+Δ∈ r / ∈ r0 ) Among them, C 补偿 is the capacitance value after compensation, △T is the temperature change, Δ∈ r , is the change of dielectric constant of the liquid.

Images