An IoT project for Vital Signs Monitoring

Nowadays, the Internet of Things (IoT) projects are very popular and they are developed for numerous fields. In order to detect various medical problems on time, it is required to monitor the subjects either human or non-human. This could be used on regular or specific activities, like sport or work. It is necessary to determine the factors that could lead to medical problems. Another important aspect is to quantify the factors, to monitor them, to collect data and to make the proper interpretation. This could be achieved using dedicated sensors, controlled by an application embedded on a development board. When a dangerous value is reached, the system has to inform the subject (if human) or someone else (if non-human). This paper presents an Arduino based IoT project used for monitoring the vital signs for human and non-human and the results based on its usage. The paper details the hardware and software components of this project.


Introduction
Vital signs are very important in detecting and monitoring various medical problems from early stages because they show the quality of the basic functions of the body, indicating the general status of a person's physical health.The acceptable values may vary depending on multiple factors, like the person's sex, age, weight, medical background, life style, geographical location, and so on.The medical theory is consistently pointing out four primary vital signs: body temperature, pulse, blood pressure and respiratory rate.Vitals can be very useful for prevention, while the pain must be considered a sign of an already existing illness.The body temperature and pulse are showing a general view over a person's health condition, while the respiratory rate and blood pressure are highly specialized indicating potential respiratory dysfunctions or heart failure risks [1].This is the main reason why the most important vital signs are considered to be the body temperature (BT), and the pulse, also called heart rate (HR), so the continuous monitoring appears to be a must.Arduino-based development platforms are very popular for the development of such systems due their low price and capabilities [2], being easily programmed and integrated [3].One important aspect for this solution is the dimension, having in mind that such a device could be used by humans or non-humans.The paper is structured as follows.The sections Human Vitals and Vital Signs for Non-Human presents the main factors that can be monitored in order to detect signs of illness or infection for human, respectively nonhuman.The section Hardware Design describes the proposed Arduino-based solution for monitoring the vital signs on human or non-human.Software Design deals with the software component of the proposed system.The results are presented in Findings and Results section.The paper ends with conclusion and future work.

Human Vitals
The normal body temperature recorded at skin level is about 36.5⁰C for adults (and 37.0⁰C for babies and children).A value between 36.5⁰C and 37.5⁰C is considered to be normal for adults.Fever (or hyperthermia) occurs when the value is higher than 37.5⁰C and it usually indicates infection or illness.The severe fever is called hyperpyrexia and it appears when the body temperature is over 40.0⁰C.When the body becomes too hot, the following two critical conditions may occurheat exhaustion and heatstroke, as described below [4]:  heat exhaustionbecause of the heat, the body starts losing salts and water, leading to degradation of the person's physical condition, like feeling dizzy, faint or sick, weakness, sweating, thirst, muscle cramps, etc. Severe symptoms may include seizures or the loss of consciousness. heatstrokethe body is not able anymore to cool down by itself so the temperature is going higher until a dangerous level is touched.A not spotted heat exhaustion may lead to heatstroke that can be life threatening.A drop below 36.5⁰C of body temperature is known as hypothermia (the reverse of hyperthermia), usually being life threatening so it should be addresses as a real medical emergency.Cold environments are usually producing hypothermia, like falling into cold water, having no enough heat in the house or staying outside for a long time in cold conditions without wearing proper warm clothes.
The second important vital sign is the pulse, also known as HR (heart rate), indicating the heart beat rate while pumping blood and usually measured in BPM (beats per minute).The normal adult resting pulse is between 60 and 100 beats per minute, while the athletes may have a resting rate between 40 and 60 beats per minute.The pulse is affected by the activity a person is doing, higher when doing exercises and lower when deep resting or sleeping.The maximum pulse rate a person may achieve without being in danger to lose consciousness can be simply computed by using the following formula: So, for example, the HR of a 40 years old adult doing exercises should be under 180 beats per minute, going over is a clear sign of a dangerous situation that must be avoided by taking a few minutes of rest.Exercises are having great benefits over the body, leading to a healthier and happier life.
While it is very important to constantly monitor the vital signs, only a few people are doing this daily, even if it is very simple to quickly check the body temperature and the pulse.

Vital Signs for Non-Human
We used statistical analysis to check if the human most important vitals are also vital when discussing about non-humans.For this reason, we used various data sets concerning dogs, cats and other common animals.For example, based on data from Table 1, it is analyzed a data set containing details about horse colic (severe abdominal discomfort that must be treated as emergency because of the high rate of mortality) [7].

0.004
These results lead us to choose to monitor the pulse and the temperature for both human and non-human.

Hardware Design
There are several books and papers dealing with Arduino projects, like [10] and [11].The projects can be developed for almost anything in real life, even for health monitoring.We can easily notice the human vital signs seem equally important for non-humans as well, so the continuous monitoring becomes a very wise decision that allows to early notice any health issue signs.In this respect, we propose a system that is depicted in Figure 2. The proposed system is based on an Arduino board and it gathers data from pulse sensor (heart rate), temperature sensor, accelerometer, GPS and, in case of emergency, it sends SMS using a GSM module.Also, the system shows the collected data on a display and it uploads data to a server using the cellular line.The board and the components are powered by an accumulator in order to assure portability and autonomy.

Fig. 2. The proposed system components
On the market, we can find devices being able to track such signs, like [5], or proposals like [6], but we were thinking about the opportunity to design an inexpensive mobile device to be used for continuous heart rate and body temperature monitoring for humans and non-humans, too.

Fig. 3. The Arduino-based assembled device
Figure 3 represents the setup we did by using the components that are briefly presented below.The IR Temperature sensor, TMP007, is an integrated microelectromechanical system (MEMS) thermopile sensor that contactless measures the temperature [9].The sensor can read temperature between -40°C to +125°C and it has a supply voltage between 2.5V and 5.5V.The HR sensor, XD-58C, works at a sample rate of 500Hz and it is powered at 5V.The accelerometer is CJMCU-116, based on MPU-6500 Integrated Gravity 6-axis Gyro Acceleration Module with SPI/I2C interfaces.The maximum power supply is 3.3 V.It provides data for three axes.The GPS, GSM and Bluetooth board is based on SIM808 integrated circuit.The GSM module is used for sending data over a GPRS connection.The location coordinates are provided by the GPS module and are collected by the application in order to track the subject.The supply voltage is between 5 and 18V.
The OLED display has 0.96 inches and it uses SPI/I2C.The display has a resolution of 128 x 64 pixels and it uses tow colors: yellow and blue.The development board is compatible with Arduino Nano (ATmega328p and CH340).The board works at 16MHz and is powered at 5V.It has 14 I/O pins (6 PWM and 8 ADC).All components are connected using a breadboard.The breadboard can be easily identified in Figure 3 as it connects the Arduino board and the other modules.The Arduino-based board is collecting data about the heart rate, body temperature and location (one reading per second) and is uploading these details in the cloud by using the GPRS connection at a 3-minute interval (less than 500 entries during a day).This time interval could be customized according to the needs, but for the regular monitoring we consider a 3-minutes timer offers a good balance between the quantity of data and the vitals monitoring benefits.The system is using the OLED display to show the current readings for the pulse and temperature together with the GPS and GPRS status.

Software Design
The collected data are processed by a module developed using the dedicated Arduino environment and third-party libraries when necessary.Once the setup is finished, the data are uploaded into the cloud once at every 3 minutes in a dedicated so-called channel.The Great Wearable channel settings page is presented in Figure 4.He holds a PhD diploma in Economic Cybernetics and Statistics, having the title Models and techniques to perform the economic information audit.He is co-author of books and articles in information audit and ICT fields.Also, he has published articles in proceedings of national and international conferences, symposiums, workshops in the fields of data quality, software quality, information audit and juridical aspects in ICT field.He is evaluator of ANEVAR association.

1 DOI: 10
.12948/issn14531305/21.2.2017.03 important factors explaining the health condition for the horse, as is depicted in Figure1.The pulse clearly indicates a sign of illness, most of the horses with higher pulse values finally died due to the colic.

Fig. 1
Fig. 1 Ranking results and the BoxPlot chart

Table 1 .
Factors that influences non-human health conditions

higher values may occur due to infection 0.967 By
applying ranking, we can easily see that the Pulse and Body Temperature are

Table 2 .
Results of Principal Component Analysis