阅读说明：本技术 一种基于LoRa通信的智能输液系统 (Intelligent infusion system based on LoRa communication ) 是由 叶芝慧 张紫嫣 于 2019-06-29 设计创作，主要内容包括：本发明公开了一种基于LoRa通信的智能输液系统,包括输液监测单元、呼叫器、中继器、Lora通信模块和医护站上位机,所述输液监测单元、呼叫器通过Lora通信模块与中继器连接,所述医护站上位机通过网线与中继器连接。本发明通过LoRa无线通信技术,提高了信息传输的鲁棒性、安全性,通过设置称重校准模块,提高了数据测量的准确性。(The invention discloses an intelligent infusion system based on LoRa communication, which comprises an infusion monitoring unit, a pager, a repeater, a Lora communication module and a medical care station upper computer, wherein the infusion monitoring unit and the pager are connected with the repeater through the Lora communication module, and the medical care station upper computer is connected with the repeater through a network cable. According to the invention, the robustness and the safety of information transmission are improved through the LoRa wireless communication technology, and the accuracy of data measurement is improved through the weighing calibration module.)

1. The utility model provides an intelligent transfusion system based on loRa communication which characterized in that: including infusion monitoring unit, calling set, repeater, Lora communication module and the host computer of doctorsing and nurses, infusion monitoring unit, calling set pass through Lora communication module and are connected with the repeater, the host computer of doctorsing and nurses passes through the net twine and is connected with the repeater.

2. The intelligent infusion system based on LoRa communication of claim 1, wherein: the Lora communication module adopts a star network structure, and in the information transmission process, node data are enabled to access the Lora gateway in turn by adopting a polling mode.

3. The intelligent infusion system based on LoRa communication of claim 1, wherein: the infusion monitoring unit is provided with a weighing calibration module, and the compensation of zero temperature drift is carried out by adopting a wavelet neural network and an artificial fish swarm algorithm symmetric retransmission sensor.

4. The intelligent infusion system based on LoRa communication of claim 1, wherein: the Lora communication module is provided with a safety verification access module, and the safety verification access of the Lora communication module, the infusion monitoring unit and the caller is realized by utilizing the physical unclonable function property of the chip.

5. The intelligent infusion system based on LoRa communication of claim 1, wherein: infusion monitoring unit and calling set up near the sick bed, repeater and Lora communication module set up in the hospital corridor.

6. The intelligent infusion system based on LoRa communication of claim 1, wherein: each sickbed corresponds to one infusion monitor and one calling device, RFID tags are arranged on the sickbeds in an infusion area respectively, and each infusion monitor and each calling device are bound with the tags on the sickbeds through a radio frequency identification technology.

Technical Field

The invention relates to the field of medical information, in particular to an intelligent infusion system based on Lora communication.

Background

The intelligent transfusion system can automatically judge the residual transfusion dosage, estimate the residual time, judge the situations of needle blockage, needle leakage, empty bottle, drip stop and the like, transmit the transfusion state information of each bed to the upper computer of the medical care station in real time through the Internet of things, enable nurses to see the transfusion process of each bed in the ward and replace the liquid medicine in time in the medical care station, and is an important research direction of intelligent medical treatment. The existing intelligent transfusion system mostly adopts Wifi and ZigBee for wireless communication, the transmission distance is small, and the intelligent transfusion system is easily influenced by interference.

Disclosure of Invention

The invention aims to provide an intelligent infusion system based on Lora communication.

The technical solution for realizing the purpose of the invention is as follows: the utility model provides an intelligent transfusion system based on Lora communication, includes infusion monitoring unit, repeater, calling set, wireless communication module and the host computer of doctorsing and nurses a station, infusion monitoring unit, calling set pass through Lora communication module and are connected with the repeater, the host computer of doctorsing and nurses a station passes through the net twine and is connected with the repeater. The transfusion monitoring unit is responsible for monitoring the weight of the liquid medicine, the transfusion progress and other information and transmitting the information to the repeater through the LoRa wireless communication technology; the caller is used for sending out an emergency call signal and transmitting the emergency call signal to the repeater through the LoRa wireless communication technology; the repeater acquires infusion related information from the infusion monitoring unit, acquires a calling signal from the caller, and transmits the calling signal to the upper computer of the medical care station through a network cable; the upper computer of the medical care station provides power for the repeater and sends out infusion ending reminding or calling reminding according to the information obtained from the repeater.

Compared with the prior art, the invention has the advantages that: 1) the invention improves the robustness and safety of information transmission by the LoRa wireless communication technology; 2) according to the invention, the weighing calibration module is arranged, so that the accuracy of data measurement is improved.

Drawings

Fig. 1 is a block diagram of a smart infusion system based on LoRa communication.

Fig. 2 is a flow chart of polling the Lora gateway.

Fig. 3 is a schematic diagram of a secure access Lora gateway.

Fig. 4 is a schematic view of a site in accordance with an embodiment of the present invention.

FIG. 5 is a specific flowchart of the wavelet neural network and the artificial fish swarm algorithm.

Detailed Description

The invention is further illustrated by the following examples in conjunction with the accompanying drawings.

As shown in figure 1, intelligent infusion system based on Lora communication, including infusion monitoring unit, repeater, calling set, wireless communication module and the host computer of medical care station, infusion monitoring unit, calling set pass through Lora communication module and are connected with the repeater, the host computer of medical care station passes through the net twine and is connected with the repeater. The transfusion monitoring unit is responsible for monitoring the weight of the liquid medicine, the transfusion progress and other information and transmitting the information to the repeater through the LoRa wireless communication technology; the caller is used for sending out an emergency call signal and transmitting the emergency call signal to the repeater through the LoRa wireless communication technology; the repeater acquires infusion related information from the infusion monitoring unit, acquires a calling signal from the caller, and transmits the calling signal to the upper computer of the medical care station through a network cable; the upper computer of the medical care station provides power for the repeater and sends out infusion ending reminding or calling reminding according to the information obtained from the repeater.

In some embodiments, the Lora communication module employs a star network architecture. Because the Lora gateway can only perform information intersection with the same terminal node at a certain time, in the process of wireless transmission, the Lora communication module adopts a polling mode to enable node data to access the Lora gateway in turn according to a certain period, and the polling flow is shown in fig. 2.

Generally, an infusion monitoring unit measures the weight of a medical fluid through a weighing sensor, thereby determining the infusion progress. However, the weighing sensor is easily affected by temperature to generate a zero drift phenomenon, so that the weighing information is inaccurate, and the alarm of the transfusion ending is advanced or lagged. In order to deal with the zero drift phenomenon generated by the weighing sensor under the influence of temperature, the invention arranges a weighing calibration module in the transfusion monitoring unit and adopts a wavelet neural network and an artificial fish swarm algorithm symmetrical retransmission sensor to compensate the zero temperature drift.

Before temperature compensation is carried out, firstly, voltage data output by weighing sensors under different conditions are collected through two modes of fixing the weight of liquid medicine, changing the temperature and fixing the temperature and changing the weight of the liquid medicine; then, learning the data sample by using a wavelet neural network, and gradually approximating the output of the neural network to an expected output value through training the wavelet neural network; because the wavelet neural network is easy to fall into local optimum, an artificial fish swarm algorithm is introduced to check the wavelet neural network so as to obtain a global optimum solution.

the specific process of the wavelet neural network and the artificial fish swarm algorithm is described as follows:

(1) Initializing parameters: for the number m of nodes of the input layer, the number n of nodes of the hidden layer, the number l of nodes of the output layer and the expansion factor a of the wavelet_jTranslation factor b_jAnd the connection weight v between the input layer and the hidden layer_jiThe connection weight w from the hidden layer to the output layer_kjGiving initial value, selecting hidden layer neuron functionAnd an error function E;

(2) Starting wavelet neural network search: inputting a learning sample (gravity weighed by a weighing sensor under different temperature conditions) and corresponding expected output (voltage output by the weighing sensor), calculating the output of an output layer by using a formula (1), and calculating an error by using a formula (2);

Wherein M is the number of modes of the input sample,AndAre respectively provided withJ (j ═ 1,2, …, N) th desired output and actual output for the ith (i ═ 1,2, …, M) th mode;

(3) Judging whether to end the neural network algorithm: when the error is smaller than a certain preset threshold value, stopping the learning of the neural network, switching to an artificial fish swarm algorithm, otherwise, inputting a next sample, returning to the step 2, and continuing the learning of the neural network;

(4) Fish shoal initialization: initializing N artificial fishes by utilizing the output result of the wavelet neural network output layer, wherein the output result of the wavelet neural network is the state of the ith artificial fish, the error function in the step 2 is the adaptive value function of the artificial fish, the artificial fish schools are gathered at the same position, and the optimal solution on the bulletin board is the output result of the wavelet neural network;

(5) And (3) carrying out verification of the optimal solution: the method comprises the following steps that a fish school executes foraging behavior, a place with higher food concentration (namely a place with smaller adaptive value and smaller error) in a Visual field range (Visual) is searched according to a certain Step length (Step), if the position is not searched, the Step is randomly advanced one Step, the adaptive value of the artificial fish is changed, the optimal solution on a bulletin board is not updated, and when iteration is carried out for a certain number of times, the optimal solution of the bulletin board is not changed all the time, the output result of the wavelet neural network can be judged to be the global optimal result; if the fish school finds a place with higher food concentration in the foraging process, the optimal solution on the bulletin board is changed, and the result is fed back to the wavelet neural network to continue training, and the process is circulated until a global optimal solution is found.

In some embodiments, the Lora communication module is provided with a security verification access module, and the security verification access between the Lora communication module and the infusion monitoring unit and between the Lora communication module and the caller is implemented by using the property of a physical unclonable function of a chip, and a flow of the security verification access is shown in fig. 3, and specifically includes the following contents:

Because chip manufacturing processes in each infusion monitoring module and the calling device are different, the Lora gateway sends excitation signals to different infusion monitoring units and different calling devices respectively to obtain different response signals, and the responses are unique and can not be duplicated and can be used as the ID (identity) of each infusion monitoring unit or calling device, namely an identity verification code; the Lora gateway puts the identity verification codes into a storage module for ID access verification when the infusion monitor and the caller poll.

As shown in fig. 4, in some embodiments, the infusion monitoring unit and the beeper are disposed near the patient bed, and the repeater and the Lora communication module are disposed in the hospital corridor. Because the sickbeds are close to the infusion monitoring unit and the calling device, each sickbed is respectively bound with one infusion monitoring unit and one calling device by adopting a radio frequency identification technology. RFID labels are respectively arranged on sickbeds in infusion areas, each infusion monitor and the calling device are bound with the label on the sickbed, and after the labels are bound, the host of the medical care station can distinguish information sent by infusion monitoring units and calling devices from different beds.

To verify the validity of the inventive scheme, the following simulation experiment was performed.

lora communication test

In this embodiment, the Lora communication module is a wireless serial port module (UART) based on an SX1278 radio frequency chip manufactured by SEMTECH corporation, which is a product E32-433T20DC manufactured by eggbaud corporation, and the module uses TTL level. The default settings of the E32-433T20DC product are kept unchanged, namely the air transmission rate is 2.4kbps, the transmitting power is 20dBm, and the plane through-wall test and the cross-floor test are carried out in the environment with the presence of the interference of walls, electronic equipment, optical equipment and the like and including interference of 4G, WiFi, microwaves, millimeter waves and the like.

(1) plane through-wall testing

first, a transmission/reception test of data packets was performed in different rooms on the same floor, and the test results are shown in table 1 when 150000 transmission packets were transmitted. It can be seen that when the transmission path is as long as 60 meters (six walls apart), the packet loss rate is only 1.187%, which is higher than the communication quality in Wifi, ZigBee and other manners.

TABLE 1 test result table for plane wall penetration

Number of wall surfaces	Number of transmission packets	Number of lost packets	packet loss rate
				1(10m)	150000	416	0.277％
2(20m)	150000	425	0.283％
				3(30m)	150000	478	0.319％
4(40m)	150000	579	0.386％
				5(50m)	150000	1216	0.811％
6(60m)	150000	1780	1.187％

(2) Cross-floor testing

And respectively sending data packets to other floors by taking the third floor as a reference, wherein the test results are shown in a table 2. It can be seen that when the signal is transmitted to the sixth floor, the packet loss rate is only 1.271%, which is higher than the communication quality in Wifi, ZigBee and other manners.

TABLE 2 Cross-storey test result table

Partition floor	Number of transmission packets	Number of lost packets	packet loss rate
				1 (to four storied building)	120000	355	0.296％
2 (to five storied building)	120000	390	0.325％
				3 (to six storied building)	120000	1525	1.271％

Second, weighing calibration test

In the weighing calibration module, a plurality of weighing sensors can be selected, different weighing sensors have different zero temperature drift rules due to differences in manufacturing processes, and therefore temperature compensation is performed according to the rules of the weighing sensors. Here, taking a PT14 type pressure sensor as an example, its output voltage data in various cases was measured, as shown in table 3.

TABLE 3 MEASUREMENT DATA TABLE FOR PT14 TYPE PRESSURE SENSOR

t(℃)	P(KPa)	100	101	102	103	104	105	106	107	108
											23.4	UP(mV)	21.6	43.0	61.8	81.0	99.1	118.9	139.2	153.4	183.5
30	UP(mV)	20.0	39.6	58.4	78.9	98.7	120.2	137.0	150.2	179.9
											37	UP(mV)	18.8	39.4	58.9	77.9	95.4	113.9	133.3	152.8	173.9
44	UP(mV)	5.5	27.9	45.6	63.1	83.0	101.4	120.0	139.5	157.7
											54	UP(mV)	7.9	27.2	46.4	63.5	84.5	102.4	120.0	139.4	158.3
60	UP(mV)	4.6	21.2	37.6	54.1	73.1	95.4	103.8	126.9	148.7
											70	UP(mV)	4.9	18.5	31.3	49.6	67.4	86.9	96.8	118.9	137.3

In Table 3, t is the ambient temperature at which the pressure sensor is located, P is the pressure carried by the pressure sensor, U_PIs the output voltage of the pressure sensor.

After the data acquisition is completed, the wavelet neural network and the artificial fish swarm algorithm are executed according to the graph 5. The parameter initialization comprises the following steps: setting 9 input layer nodes of the wavelet neural network to be 9 according to 9 values (100-108) of the pressure P, determining the number of modes of input samples to be 7 according to 7 values of the temperature t, and setting 1 output node to be used for outputting results; according to the experience of the existing research, the number of nodes of the hidden layer is set to be 16, the maximum training times of the network is set to be 500, the learning rate is 0.01, the momentum factor is 0.95, and the sum of squares of errors of training targets is 0.0001; the hidden layer neuron function adopts a Morlet wavelet function, and the function expression is as follows:

in addition, the size N of the fish school is set to 20, the Visual field Visual is set to 0.7, the Step size Step is set to 0.5, and the maximum number of iterations is 400. The above parameters are not intended to limit the present invention, and modifications and optimization may be made in the embodiments according to the circumstances.

Secondly, training a wavelet neural network, taking the gravity and the output voltage of the weighing sensor weighed under different temperature conditions as learning samples, calculating an output result by using a formula (1), and calculating an error by using a formula (2); and switching to an artificial fish school algorithm until the calculated error is smaller than a set threshold value, and taking the output result of the wavelet neural network as the initial state of the fish school.

Verifying whether the output is globally optimal by having the artificial fish swarm perform foraging behavior. Setting the current state of the ith artificial fish as X_iCorresponding adaptation value is Y_i(calculated by equation (2)), randomly selecting a state X within the Visual range thereof_vIf the corresponding adaptive value Y is as in formula (4)_v<Y_iIf the best solution on the bulletin board is updated to X_vFeeding back to the wavelet neural network to enable the wavelet neural network to continue training so as to improve the result; otherwise, the artificial fish advances one step at random and changes to the state X_inexte.g., equation (5), without the bulletin board having to be updated.

X_v＝X_i+Rand()*Visual (4)

X_inext＝X_i+Rand()*Step (5)

And repeating the steps until the wavelet neural network outputs the optimal result, namely the temperature compensation achieves the optimal effect under the specified error.