1 Introduction

Monitoring the vital data of individuals provides precautions against complications that may develop in the near future. Hence, possible casualties or serious damages to the human body can be prevented. Such monitoring becomes even more important in pandemic situations. The Covid-19 pandemic causes severe respiratory infections in humans, ranging from influenza to Middle East Respiratory Syndrome (MERS) and Severe Acute Respiratory Syndrome (SARS) [1]. Silent hypoxia is defined as a decrease/discrepancy in O2 levels without difficulty breathing in Covid-19 patients [2]. By the time Covid-19 patients realize that they are struggling to breathe, their condition has already significantly progressed to moderate to severe levels of pneumonia [3]. Detecting this silent form of hypoxia in Covid-19 patients before they start to experience difficulty breathing is critical to prevent pneumonia from progressing to a dangerous level [4]. For this reasons, it is crucial to have reliable non-contact thermometers and pulse oximeters during pandemic periods.

In this study, a wearable device consisting of a wristband and a finger-clip and a mobile app that can work in compatible with this device have been developed to monitor vital data such as body temperature, pulse and SpO2 values of individuals. The wristband and the finger-clip include various medical sensors, an MCU and a bluetooth module. The communication between the wristband and the smart mobile device is performed wirelessly via bluetooth. The smart mobile device is used as a recording, control and display device through the developed mobile app. Thus, the vital values of the individuals can be measured instantaneously by the patient himself/herself or his/her relatives (e.g. in the next room) with the help of the smart mobile device. In fact, like a holter device, vital values can be recorded at certain times of the day and at certain intervals. When the patient’s vital values reach a predefined risk level, the patient or his/her relatives can be warned instantly with a notification. The developed device is both a measuring device and a holter for body temperature, pulse and SpO2 values. This feature can be used by both the patient and his/her relatives. To the best of the authors’ knowledge, there has not been any previous research on a mobile-compatible device like the one in this study, which has a holter feature. Moreover, the highlights of our work can be summarized with the following points:

Silent hypoxia mitigation:

Our device and app can reduce the risks of silent hypoxia by monitoring people’s vital signs and enabling early intervention, especially for cases such as asymptomatic Covid-19.

Comprehensive vitals monitoring:

The device measures body temperature, pulse and SpO2, allowing real-time or, unlike other devices, holter-like recordings for detailed analysis.

Real-time notifications:

The device-users and their relatives receive instant notifications when vital values reach predefined risk levels, ensuring timely medical attention.

Accuracy and validation:

The device showed high accuracy with minimal error values: 1.8% SpO2, 3.5% pulse, 0.6% temperature.

Innovative mobile holter capability:

The mobile-compatible device with a holter feature, enabling continuous vital sign monitoring for enhanced patient care and data tracking.

2 Related work

In the literature, various studies have been conducted on analyzing and monitoring the condition patients [5,6,7]. Ganesh et al. [8, 9] developed an IoT-based, portable, cost-effective device that can measure heart rate and SpO2 level. Their device offers the opportunity to monitor the measurement results with an OLED (Organic Light-Emitting Diode) display. At the same time, the measurement results can be monitored in real time from an HTML (Hypertext Markup Language) page instead of a mobile application. Although their device is portable, it is not suitable for wearable use. In the study by Dhadge and Tilekar [10], a wearable system was developed to track the disease process of Covid-19 patients. Apart from pulse, body temperature and oxygen saturation level measurements, this system can also detect the patient’s anxiety. It does this by monitoring the patient’s hand movements. The vital values measured by the device are notified to the users through the blinking of an LED (Light-Emitting Diode) integrated into the system and a bluetooth-based messaging protocol.

Santos et al. [11] also introduced a wearable system to monitor Covid-19 infected patients in isolation services. Their system includes market products such as a wearable chest strap and a finger-worn pulse oximeter to measure the heart rate of patients and transmit this value via bluetooth. With this system, it is aimed to accelerate the outpatient treatment process. In the study by Khan et al. [12], a system was developed for monitoring Covid-19 disease. In this study, pulse rate, body temperature and blood oxygen saturation level can be measured using an Arduino. The measurement results can be monitored via an LCD. These results can be sent to a mobile app via bluetooth, but their device has low wearability.

Al Bassam et al. [13] reported an IoT-based Covid-19 tracking device. Their work can be analyzed in three layers. The first layer includes wearable technology. In this layer, location is determined using GPS (Global Positioning System), body temperature, pulse rate and blood oxygen saturation level (SpO2) are measured. The second layer, called the cloud layer, stores the patient’s symptoms, emergency contact information and location data along with the measurement results from the microcontroller (MCU). This layer provides secure access to the patient record. In the last layer, an interface is developed for real-time data retrieval. This layer includes both web and mobile software interfaces. The interfaces allow access to measurement results by sending e-mail/SMS alerts for emergencies.

Petrovic and Kocic [14] realized a computer vision process using an infrared sensor controlled by an Arduino and a thermal camera connected to a Raspberry Pi. The aim of this study was to check whether individuals are wearing masks or not, their body temperature and social distance in a non-contact way in public areas. The data received from the Raspberry Pi is transmitted to the main server and then sent to a security guard’s phone via mobile software. Thus, the security guard will prevent a potential Covid-19 case and mask wearing, body temperature and social distancing controls can be carried out quickly and effectively in public areas.

In addition to devices developed in the laboratory, there are also commercial products developed with the support of leading companies in the commercial market. ThermO2 by 221e [15], Fitbit by Google [16] and CheckmePod by Wearpulse [17] are some of them. With these commercial devices, body temperature and SpO2 values can be measured. However, while continuous measurements can be performed with Fitbit, ThermoO2 and CheckmePod are not suitable for continuous measurements like a holter device. Furthermore, these three devices provide automatic measurements, which means they are not directly programmable by their users. They literally do not offer full user programmability.

3 Materials and methods

The block diagram of the developed system is illustrated in Fig. 1. The person wearing the wristband can monitor his/her own vitals as shown in Fig. 1a or, the vitals can be monitored through the bluetooth module by relatives or paramedics who are not in the same room with him/her as in Fig. 1b.

Fig. 1
figure 1

Working diagram of the proposed system in (a) the same and (b) different indoor environments

Full size image

Arduino Nano [18, 19], was preferred as the control unit to read, process and send sensor data to the smart mobile device via bluetooth. Sparkfun Pulse Oximeter [20] was used to measure SpO2 and pulse data and DS18B20 temperature sensor was used to detect body temperature. A Bluetooth module (HM-10 BLE 4.0) was employed to transfer sensor data [21]. A smart mobile device with Android operating system was preferred for monitoring the sensor data. With the developed mobile app, measurement settings can be adjusted via this device. The proposed system works according to the flow diagram in Fig. 2.

Fig. 2
figure 2

The system workflow

Full size image

The proposed system begins by configuring the MAX30101 pulse oximeter component, temperature sensor and bluetooth module in the device prototype via a controller. If the bluetooth connection between the device and the mobile app is established, the instant or holter mode measurement settings are set via the mobile app. The measurement settings are saved in the EEPROM. If instant measurement mode is enabled, temperature, SpO2 and pulse values are measured at that moment and sent to the mobile app. Otherwise, i.e. in holter mode, temperature, SpO2 and pulse measurements are carried out periodically considering settings in EEPROM. These settings are values such as time interval and duration previously recorded by the user. Periodically measured values are instantly sent to the mobile app. If there is no bluetooth connection, it is temporarily saved in EEPROM to be sent later. As soon as the bluetooth connection is re-established, these recorded values are sent to the mobile app. Thus, it is ensured that all periodically measured values are completely sent to the mobile app.

4 Production and housing

The wristband and finger holder boxes (housing), CAD (Computer-Aided Design) models of which are shown in Fig. 3, were manufactured from PLA (Polylactic Acid) filament material using a 3D printer. The electronic components of the developed device were placed in the boxes. The black band/strap in Fig. 4 was attached to the wristband box so that the device could be worn on the wrist.

Fig. 3
figure 3

Housing designs for (a) wrist and (b) finger

Full size image
Fig. 4
figure 4

The wristband and finger-clip prototype

Full size image

The control unit, bluetooth module, charging unit, temperature sensor and power supply were placed inside the wristband box in Fig. 3a. A 3.7 V 1300 mAh 1S PX1300XP-40C model Li-Po battery was utilized as the power supply. With this battery, 5 V and 3.3 V output was obtained through a boost converter and the system (control card, bluetooth module, pulse oximeter, temperature sensor) was supplied with this output. There is a pulse oximeter inside the finger-clip in Fig. 3b. The weight of the device, which has been manufactured and assembled, is 123.6 g with the finger-clip and wristband. Figure 4 shows the assembled housings of the wristband and the finger-clip obtained using Computer Aided Design and Manufacturing (CAD-CAM) programs and the electronic components placed inside them.

5 Mobile application

A mobile app that can run on Android operating systems has been developed to control the device prototype via smart mobile phones and to monitor vitals such as body temperature, pulse and SpO2 from the device. The mobile app software was written in Visual Studio Code environment and tested on a smart phone with Android 7.0 Nougat operating system, 3 GB RAM and Bluetooth 4.1. The developed mobile app includes Main, Instant Measurement, Holter Measurement, Recorded Measurements, Send to Expert, Mode Selection and Settings windows.

5.1 Main and instant measurement windows

The main window contains buttons to access the following windows: Instant Measurement, Recorded Measurements, Send to Expert and Settings. The main window is shown in Fig. 5a.

Fig. 5
figure 5

The developed mobile app interface and its windows. (a) Main menu and (b) Instant measurement window

Full size image

The Instant Measurement window is prepared to measure the instant body temperature, pulse and SpO2 values of the person. After the measured values are sent to a smart mobile phone with Android operating system, they can be monitored in the Instant Measurement window in Fig. 5b.

5.2 Recorded measurements window

The Recorded Measurements window is used to display the recorded/logged data in the form of a list or graph. Data selection (editing) can be made on the saved data for submission to the query result based on date and time. Thus, detailed analysis can be performed on the recorded data. Figure 6a shows the recorded measurements in holter mode and Fig. 6b displays different numbers of recorded data in holter mode.

Fig. 6
figure 6

The mobile app interfaces and its windows. (a) Recorded measurements in holter mode and (b) Data recorded in various numbers in holter mode

Full size image

5.3 Send to expert window

The Send to Expert window allows the recorded data to be sent to medical healthcare professionals or other relatives of the person worn the wristband. The data can be sent as an e-mail. Figure 7a shows the Send to Expert window. Figure 7b illustrates an e-mail sent through the Send to Expert window.

Fig. 7
figure 7

The developed mobile app interfaces and its windows. (a) Send to Expert Window and (b) the e-mail sent through the Send to Expert Window

Full size image

5.4 Settings window

The settings window was used for scanning bluetooth devices and establishing a connection with the developed wristband and finger-clip. The selection of the operating mode (instant or holter measurement), determining the settings related to the selected mode (measurement frequency and risk levels) and saving these settings to the EEPROM in the MCU are also executed in this window. Figure 8a shows the mode selection and Fig. 8b shows the settings window.

Fig. 8
figure 8

Developed mobile app interfaces and its windows. (a) Mode selection, (b) Settings window where risk level and measurement range are set

Full size image

The developed device can measure a person’s vitals in two different modes: 1) instant measurement mode and 2) holter mode. In the first mode, a single measurement is carried out and the measurement results are saved to the database at the same time. In the second mode, measurements can be taken automatically at set intervals over a pre-set period of time and the measurement results are regularly recorded in the database. In both instant measurement and holter mode, when a person’s vitals reach a predefined risk level, He/She or his/her relatives are immediately alerted with a notification. In holter mode, the bluetooth connection is checked periodically to ensure the continuity of the connection. Thus, the measurement values taken from the wristband are precisely recorded in the database.

6 Results

The developed device and mobile app were evaluated through various tests. First, the prototype device was compared with commercial products. Then, the capabilities of the device were tested in the field in the accompaniment of an expert healthcare professional. In addition, a battery endurance test was conducted.

6.1 Comparison test

The developed device was compared with Armoline Al-651 model pulse oximeter for SpO2 and pulse values and Loobex infrared thermometer for temperature values. The measurements were recorded on eight people, six males and two females, under the same conditions. The measurement results are listed in Table 1. The percentage errors were calculated from these measurements and the mean square errors (RMS %) was found for each vital data.

Table 1 Comparing the prototype with a commercial product among different genders and age groups 
Full size table

6.2 Connectivity-mailing and the field testing

The prototype has been tested in home and hospital environments. In Fig. 9, in a hospital room under the guidance of an expert healthcare professional, the vital data of the user wearing the device is monitored through a mobile app by the healthcare professional. The device was tested in different modes (instant and holter mode) in the company of a healthcare professional and received positive feedbacks from him.

Fig. 9
figure 9

Testing of the prototype in a hospital with an expert healthcare professional

Full size image

6.3 Endurance and long-term measurement test

The developed device was tested under two different power supplies. The first test is the battery endurance test. In this test, the power supply of the device was provided by a 1300 mAh Li-Po battery and the device was configured to operate in holter mode for 12 hours at 45-minute measurement intervals in the settings window of the developed mobile app. The same configuration settings were also adjusted when the device was powered by an external power supply via a cable. The wired supply from an external power source constituted the second test. The measurement results for both tests are given in Table 2 and Table 3 respectively.

Table 2 Long-term measurement testing in holter mode supported by internal battery
Full size table
Table 3 Long-term measurement test in holter mode for up to 12 hours with external voltage supply
Full size table

The device provides up to approx. 7 hours of use with internal battery support. To test measurements over 7 hours, the device was powered by an external voltage source. In this long-term test, the device performed successful measurements for 12 hours at the specified measurement intervals. These measurements are listed in Table 3.

7 Discussion

The developed device was compared with competitive works in the literature as well as with the commercial products. The comparison list is provided in Table 4. Compared to other studies, the proposed work stands out with the ability to perform continuous and configurable/ programmable holter mode measurements. Thanks to the mobile application developed specifically for the proposed work, the measurement results can be easily accessed. Moreover, the proposed device prototype offers direct user programmability, high wearability, and portability. In future work, it is planned to increase the battery capacity of the proposed device and thus extend its operating endurance. Moreover, wireless communication technologies [22,23,24] such as LORA, NbIOT or WiFi [25] can be used to extend its connectivity. Currently, the mobile app developed for the device runs on the Android operating systems. Improvements are ongoing for this mobile app to work on the IOS operating system as well.

Table 4 Comparison with works in the literature and the commercial market
Full size table

With the works in Table 4 and commercial products, individuals’ vital signs can be monitored instantaneously, continuously and with additional measurement options such as holter mode. Vital signs are displayed to users through various screen technologies or mobile applications. The most of the devices here are wearable, portable and low-weight, providing convenience to their users. The device prototype proposed in this paper is distinguished from the others by its ability to be directly programmable by the user and continuous measurement in holter mode. Additionally, the laboratory-developed devices in Table 4 are more cost-effective compared to commercial products. This is because the devices developed in the laboratory environment were produced by avoiding commercial concerns. However, when all of the works and products in Table 4 are reviewed, it is obvious that monitoring the vital signs of individuals is a critically important research topic.

8 Conclusion

In this paper, a research on the measurement and monitoring of vital data, the importance of which has been greatly felt with the recent Covid-19 pandemic, has been carried out. In this research, a wearable device prototype consisting of a wristband and a finger-clip to measure and monitor pulse, SpO2 and temperature data of individuals and a mobile app compatible with this prototype were developed. The prototype is portable and wearable and can work as a holter device. The designed mobile app offers rich features such as instant-holter mode measurement options, sending measured data via e-mail, graphical display of measured data, setting risk levels and notifications based on these levels. The device and mobile app have been evaluated in various tests such as benchmarking, field and endurance. The device has been compared with commercial products and as a result of this comparison, RMSEs of 1.87%, 3.53% and 0.60% were calculated for SpO2, pulse and temperature measurements, respectively. The device offers up to 7 hours of use with the internal battery. The device has also been tested for up to 12 hours with external battery.