Design and Implementation of a Multi-Parameter IOT Wearable Device for Remote Patient Monitoring

Design and Implementation of a Multi-Parameter IOT Wearable Device for Remote Patient Monitoring

Ibharunujele Solomon Obhenbhen (1)

Department of Electrical/Electronics Engineering Ambrose Alli Universitty Ekpoma, Nigeria

Okoduwa Osedebamhen Emmanuel (2)

Department of Electrical/Electronics Engineering Ambrose Alli Universitty Ekpoma, Nigeria

Esekhaigbe Emmanuel (3)

Department of Electrical/Electronics Engineering Ambrose Alli Universitty Ekpoma, Nigeria

Odia Hope Ehimare (4)

Department of Electrical/Electronics Engineering Ambrose Alli Universitty Ekpoma, Nigeria

LarryOboh Isaiap

Department of Electrical/Electronics Engineering Ambrose Alli Universitty Ekpoma, Nigeria

Abstract – This Study presents the design and implementation of a multi-parameter Internet of Things (IoT) wearable device for continuous remote patient monitoring. The system is developed to address the limitations of traditional episodic vital sign measurement by enabling real-time, non-invasive, and continuous monitoring of key physiological parameters, namely heart rate, blood oxygen saturation (SpO), and body temperature. The proposed wearable integrates low-power biomedical sensors, an embedded microcontroller unit, and wireless communication technology to acquire, process, and transmit physiological data to a mobile application and cloud-based platform for visualization and alerting. Emphasis is placed on power efficiency, signal reliability in ambulatory conditions, and user comfort within a wearable form factor. The system architecture follows an edge gatewaycloud model, allowing preliminary processing at the device level and remote access to health data for patients and caregivers. Prototype testing demonstrates the feasibility of accurate vital sign acquisition and reliable data transmission under controlled conditions. The project provides a practical foundation for remote patient monitoring solutions and highlights the potential of IoT-enabled wearables to support proactive healthcare delivery, chronic disease management, and improved accessibility to medical monitoring services.

Keywords – wearable; architecture; healthcare; visualization; real-time

1. INTRODUCTION

   The global healthcare paradigm is undergoing a profound shift from reactive, hospital-centric treatment to proactive, personalized, and decentralized care. This transformation is driven by a confluence of pressures: aging populations, rising

   prevalence of chronic diseases, escalating healthcare costs, and a growing emphasis on preventive medicine [1]. Central to this new paradigm is the ability to continuously monitor vital signs the fundamental physiological metrics that offer a window into an individuals health status. Traditional episodic measurements, conducted in clinical settings with bulky, stationary equipment, are inherently limited. They provide mere snapshots in time, often missing critical transient events such as paroxysmal arrhythmias, nocturnal hypertension, or intermittent hypoxia, which are pivotal for early diagnosis and management [2].

   There appears to be increasing numbers of research studies investigating wearable, IoT-enabled and portable multi- parameter health monitoring systems that allow continuous, real-time and remote physiological monitoring. Consistently, throughout all of the various studies, the principal goal is always to improve patient safety by allowing for earlier identification of clinical deterioration and support for telemedicine and home-based care delivery through the use of multiple biosensors and wireless communications technologies. The most common parameters measured include ECG, heart rate, SpO, blood pressure, respiratory activity, body temperature, movement, and in some cases specialized indicator(such as heart sounds or seizure related biomarkers) [3], [4]. foundational work such as [5] investigated wearable clothing-based systems for non-invasively measuring ECG, breathing patterns and movement. This was primarily directed towards developing comfortable garments for use within homecare settings. As new wearable technology began to emerge, researchers continued to explore alternative methods for transmitting physiological data remotely. Researchers in ref. [6], [7] explored the use of wireless sensor networks utilizing

   GSM and ZigBee for enabling the remote transmission of physiological data with low power consumption and mobility. More recently, numerous researchers have begun to integrate IoT platforms and mobile applications to enable real-time data collection, processing and visualization. Examples of this can be found in [8], [9] who created wearable Android/Wi-Fi enabled systems that allowed for simultaneous multi parameter data collection and remote monitoring. While these examples represent a vast improvement over previously developed systems, [10], [11] built upon previous systems by expanding them into larger scale IoT enabled healthcare systems that supported cloud connections and remote clinical access. Telemedicine focused systems such as [12], [13] build upon the previously mentioned systems by incorporating low latency communication and real-time interaction between patients and healthcare professionals; however, like previously discussed systems they rely on reliable internet infrastructure and face challenges related to scalability.

   While researchers continue to develop new wearable systems that allow for the measurement of physiological signals, researchers have also continued to miniaturize wearable systems and increase their ability to measure multiple physiological parameters. Researchers such as[14], [15] were among the first to attempt to recreate hospital grade monitoring within a wearable format by combining ECG, SpO, temperature and movement measurements with innovations such as cuff less BP measurement using pulse transit time. While these studies demonstrate the feasibility of recreating hospital grade monitoring in a wearable format, many of the newer wearable systems being developed today have shifted their focus from recreating hospital grade monitoring towards low cost, mobile enabled home healthcare monitoring. Systems such as [16], [17] fall under this category. In addition to shifting their focus away from recreating hospital grade monitoring, researchers have also increased their focus on application specific wearables. An example of an application specific wearable is the seizure detection band proposed by [18].

   In addition to increasing the number of wearable devices being used in homecare environments, researchers have also increased their efforts to develop wearable systems that are powered in a more sustainable manner than traditional battery powered systems. An example of this type of system is the solar powered modular wearable network proposed by [19]. In addition to providing a wearable network that supports self sustained distributed monitoring, researchers have also proposed wearable networks that reduce the amount of energy consumed during each sample period[8]. An example of a wearable network that utilizes this approach is the adaptive sampling wearable network proposed by [20].Through the utilization of neural networks to optimize the sampling process of the wearable network, researchers were able to significantly decrease the overall energy consumption of the system.

   Researchers have also increased their focus on developing wearable systems with intelligent architectures that utilize machine learning algorithms to assist in diagnosing conditions more accurately and decreasing false alarm rates. One example of an intelligentwearable architecture is the patient specific monitoring system proposed by[21]. This wearable system utilized an FPGA based processor to fuse physiological signals collected from multiple sensors into a single health index. This

   fusion process resulted in a wearable system that was much more robust than previously developed systems when subjected to noise or motion artifacts. Another example of an intelligent wearable architecture is the wearable system proposed by [22]. This wearable system utilized machine learning techniques to assess patient condition more accurately than previously developed systems.

   The final area of interest for researchers involved in developing wearable technology includes designing systems that are adaptable across multiple clinical domains [23]. Examples of wearable systems that are adaptable across multiple clinical domains include those developed for cardiopulmonary rehabilitation as proposed by [24] and for fetal maternal monitoring as proposed by [25].

   As stated above, despite the numerous advances in wearable technology described above, several consistent limitations exist in terms of the current state-of-the-art in wearable technology. A significant gap exists in the integration of clinical-grade sensing, robust data processing, low-power operation, and secure, reliable IoT transmission into a single, affordable, and user-centric wearable device.

   Therefore, there is a pressing and well-defined need for a dedicated IoT-based wearable vital sign monitoring system. This system must bridge the identified gaps by providing continuous, accurate, and comfortable monitoring outside clinical settings; enabling real-time data transmission and alerting to facilitate proactive care; reducing the burden on formal healthcare systems; and being engineered with a focus on clinical relevance, power efficiency, and data security to serve as a credible bridge between consumer technology and medical-grade applications. This study aims to design and implement a prototype that addresses this multifaceted problem.

2. METHODOLOGY

   This session presents the comprehensive materials and methodological approach employed in the design and implementation of an IoT-based wearable vital sign monitor. The system integrates multiple physiological sensors with an ESP32 microcontroller to acquire, process, and transmit vital signs data to a cloud backend via Wi-Fi. A mobile application interfaces with the backend to display real-time measurements, provide health interpretations, and generate alerts. Hardware components, circuit design, firmware development, backend architecture, mobile application design, and validation protocols.

   1. Hardware Components

      Component	Part Number	Specifications	Selection Justification
      Microcontroller	ESP32 (ESP- WROOM- 32)	Dual-core, 240 MHz, 520KB	Integrated Wi-Fi eliminates need for separate module; ample processing

      Table 1: Selected Hardware Components

      		SRAM, Wi- Fi/BLE	for signal processing; low cost
      PPG Sensor	MAX30102	Integrated LEDs, driver, ADC; I²C interface	Industry-standard for HR/SpO; proven accuracy; low power
      Reference Pulse Sensor	SEN11574	Analog output, 3-5V	Provides second pulse measurement for validation and redundancy
      Temperature Sensor	DS18B20	Digital, 1- Wire interface, ±0.5°C accuracy	Waterproof probe option; simple interface; widely available
      Battery Charger	TP4056	Linear charger, 1A, thermal regulation	Built-in charge indication; protection features; standard in portable projects
      Battery	18650 Li- ion	3.7V, 2200mAh, protected	High capacity for extended operation; readily available
      Voltage Regulator	AMS1117- 3.3	LDO, 800mA, 3.3V output	Simple, reliable, sufficient current for all components

      300 nanoseconds and estimated bus capacitance of 400 picofarads, the maximum resistance is:

      =

      = 300×109 = 750 (1)

      400×1012

      However, to limit current consumption, a higher value of 4.7 k is chosen, which still provides adequate rise time for 400 kHz operation.

      SEN11574 Connection

      The SEN11574 connects to an analog input pin as shown in Fig 3.

   2. Hardware Design and Circuit Implementation

      Power Supply Circuit

      The power system is designed to provide stable 3.3V to all components from a 3.7V lithium-ion battery while enabling charging via USB. Fig 1 illustrates the complete power supply schematic.

      Fig 1: Power Supply Circuit Diagram

   3. Sensor Interface Circuits

      MAX30102 Connection

      The MAX30102 connects to the ESP32’s I²C bus with appropriate pull-up resistors as shown in Fig 2.

      Fig 2: MAX30102 Interface Circuit

      Pull-up resistors:

      SDA 4.7k 3.3V SCL 4.7k 3.3V

      The I²C pull-up resistor value is selected based on bus capacitance and desired rise time. For a maximum rise time of

      Fig 3: SEN11574 Interface Circuit

      GPIO34 is conFigd as an ADC input with 12-bit resolution, providing readings from 0 to 4095 corresponding to 0 to 3.3 volts.

      DS18B20 Connection

      The DS18B20 connects via the 1-Wire protocol as shown in Fig 4.

      Fig 4: DS18B20 Interface Circuit

      Pull-up resistor:

      DATA 4.7k 3.3V

   4. Complete System Schematic

      Fig 5 presents the complete system schematic showing all component interconnections.

      Fig 5 Complete System Schematic

   5. Signal Processing Algorithms

      Photoplethysmography Principles

      Photoplethysmography measures light absorption changes in vascular tissue. The Beer-Lambert law describes the relationship between light absorption and substance concentration:

      The overall health score combines multiple parameters with weighted contributions:

      = × + 2 × 2 + ×

      + × (11) with weights:

      = 0.3, 2 = 0.4, = 0.2, = 0.1

      = log

      (0)

      = (2)

      Individual parameter scores are calculated as:

      where A is absorbance, I is incident light intensity, I is transmitted intensity, is molar absorptivity, c is concentration, and d is path length.

      For pulse oximetry, the ratio of ratios R is calculated from the AC and DC components at red and infrared wavelengths:

      100 60 100

      = { 70 50 < 60 or 100 < 120

      40 40 < 50 or 120 < 140

      10 otherwise

      100 2 95

      = { 70 92 2 < 95

      = /

      /

      (3)

      2

      40 88 2 < 92

      10 2 < 88

      Oxygen saturation is then derived through an empirical calibration:

      2 = × (4)

      where a and b are constants determined through clinical validation. For the MAX30102, the typical equation is:

      100 36.1 37.2

      = { 70 37.2 < 38.0

      40 38.0 < 39.0

      10 > 39.0 or < 35.0

      100 5

      2 = 45.060 × 2 + 30.354 × + 94.845

      Digital Filtering

      A high-pass filter removes DC drift and low-frequency motion

      = { 80 10

      60 15

      30 otherwise

      artifacts from the PPG signal. The first-order difference equation is:

      The final score is normalized if some parameters are

      unavailable:

      [] = [] [ 1] + × [ 1] (5) where is the filter coefficient determining cutoff frequency. For a cutoff of 0.5 Hz with sampling rate of 100 Hz:

      =

      × 100 (12)

      = 2/ = 2×0.5/100 = 0.969 (6)

      A moving average filter provides low-pass filtering to reduce high-frequency noise:

   6. Cloud Backend Design

   Database Schema

   The PostgreSQL database implements three primary tables as

   [] = 1 1 [ ]

   (7)

   shown in Fig 3.7.

   =0

   For M = 5, this provides a cutoff frequency of approximately:

   0.443× = 0.443×100 = 8.86 (8)

   5

   Peak Detection Algorithm

   Heart rate detection uses adaptive threshold peak finding. The threshold is set as a percentage of the moving maximum:

   = 0.6 × max ([ : ])

   where N is the analysis window length, typically 300 samples corresponding to 3 seconds. Minimum peak distance prevents false detections from noise:

   = 300(corresponding to 200 BPM maximum) Heart rate is calculated from the average inter-beat interval:

   = 60 BPM (9)

   where t is the time between successive peaks in seconds.

   Temperature Conversion

   The DS18B20 returns temperature as a 16-bit signed integer. For 12-bit resolution, temperature in degrees Celsius is:

   Fig 6: Database Entity Relationship Diagram

   API Endpoints

   The FastAPI backend exposes RESTful endpoints as summarized in Table 3.2.

   Endpoint	Metho d	Purpos e	Request Body	Respons e
   /api/vitals	POST	Receive device data	Vital sign JSON	Created record
   /api/vitals/{device_id}/late st	GET	Get latest with health score	–	Health score response

   Table 2: API Endpoint Summary

   (°) =

   16

   (10)

   Health Score Calculation

   /api/vitals/{device_id}/hist ory	GET	Get historic al data	hours paramet er	List of vital signs
   /api/vitals/{device_id}/aler ts	GET	Get recent alerts	hours paramet er	List of alerts

   Alert Generation Thresholds

   Alerts are generated based on clinically relevant thresholds as defined in Table 3.

   TablE3: Alert Thresholds

   Parameter	Warning Threshold	Critical Threshold	Direction
   Heart Rate	< 60 or > 120 BPM	< 50 or > 140 BPM	Both
   SpO	< 92%	< 88%	Low
   Temperature	> 38.0°C	> 39.0°C or < 35.0°C	Both

   Testing and Validation Protocols

   Accuracy Validation

   1

   Heart rate accuracy is assessed through comparison with reference measurements. For N paired measurements, the mean error is:

   Fig 7: Assemble prototype (wearable strap)

   =

   ( , ,) (13)

   =1

   The standard deviation of errors is:

   1

   =

   1

   =1

   (

   )2 (14)

   Bland-Altman analysis calculates limits of agreement:

   = ± 1.96 × (15)

   Power Consumption Testing

   Average current consumption is calculated from measurements over a complete operating cycle:

   = ×+×

   (16)

   Fig 8: Assemble prototype (ESP32 and LM259S-ADJ)

   Expected battery life is then:

   =

   (17)

   where C_battery is battery capacity in milliampere-hours.

3. RESULT AND DISCUSSION

   This session presents the implementation of the IoT wearable vital sign monitor, covering component selection, Veroboard assembly, enclosure fabrication, firmware development, sensor validation, signal transmission verification, and mobile application testing.

   1. Hardware Assembly

      Layout Planning and Soldering Completed Assembly

      Fig 9: Assemble prototype (Inside cover showing LCD)

      System Integration

      End-to-end latency of 528 ms from sensor to mobile display met the design target. Average current of 20.29 mA enabled 4.5-day battery life with a 2200 mAh cell. The mobile application successfully displayed all parameters, calculated health scores, and generated appropriate alerts.

      Limitations

      SpO accuracy degraded below 90% saturation, a known limitation of reflectance PPG. Wi-Fi connectivity required proximity to a router, limiting mobility. The Veroboard assembly, while functional, was bulkier than a custom PCB would allow.

      Fig 10: Assembled prototype (front view)

   2. Enclosure Design and Fabrication

      A 100×60×25mm plastic project box was modified with precision cutouts for the USB port, charger input, status LEDs, sensor window, and temperature probe, as shown in Plate 4.1. The Veroboard was mounted using plastic standoffs, with the battery secured by adhesives Plate 4.1.

      Plate 4.1. displayed the completed prototype with all components enclosed. The final device weighed 145g and featured:

      1. USB and charger port access via rectangular cutouts

      2. Four 3mm LED windows for status indication

      3. 15×10mm MAX30102 sensor window

      4. Sealed DS18B20 probe exit

   3. Discussion

   Component Selection

   The ESP32 provided adequate processing power and integrated Wi-Fi. The MAX30102 delivered reliable PPG signals for heart rate and SpO extraction. The SEN11574 served effectively as a validation reference. The DS18B20 exceeded accuracy expectations with ±0.3°C performance.

   Sensor Performance

   Heart rate accuracy averaged ±0.3 BPM compared to manual counting, meeting project objectives. SpO accuracy was within

   ±2% for readings above 90%, consistent with literature for reflectance oximetry. Temperature accuracy of ±0.14°C exceeded the ±0.5°C requirement.

   Transmission Reliability

   Wi-Fi achieved 98% success rate at 10 meters through walls. Average transmission time of 215 ms provided acceptable latency for non-critical monitoring. Packet size of ~100 bytes minimized transmission power.

4. CONCLUSION

This study successfully achieved its aim of designing and implementing an IoT-based wearable vital sign monitor. The system integrated MAX30102, SEN11574, and DS18B20 sensors with an ESP32 microcontroller, transmitting data via Wi-Fi to a FastAPI cloud backend on Koyeb, with a Flutter mobile application providing real-time display, health scoring, and alerts.

All project objectives were met. The Veroboard assembly produced a functional circuit with proper connections. The enclosure fabrication yielded a compact 145g wearable device. Firmware development enabled successful sensor operation and Wi-Fi transmission. Sensor validation achieved heart rate accuracy of ±0.3 BPM, SpO accuracy of ±1.4 percent,and temperature accuracy of ±0.14 degrees Celsius. The cloud backend was successfully deployed on Koyeb with PostgreSQL database. The mobile application displayed real-time data, calculated health scores, and generated appropriate alerts. System integration achieved end-to-end latency of 528 milliseconds and battery life of 4.5 days.

Key contributions include the successful integration of multiple sensor types into a single wearable system, implementation of dual-pulse validation for enhanced reliability, development of a complete IoT ecosystem from embedded firmware to mobile frontend, and achievement of medical-grade temperature accuracy. The project also demonstrated a functional prototype at an affordable cost suitable for resource-limited settings and provided comprehensive documentation for future student projects.

Under optimum conditions, the system exceeded all target specifications. Heart rate accuracy surpassed the target of ±3 BPM. SpO accuracy exceeded the ±2 percent target for readings above 90 percent. Temperature accuracy exceeded the

±0.5 degrees Celsius target. Wi-Fi success rate, transmission latency, end-to-end latency, battery life, and alert response time all exceeded their respective targets. The total system cost remained under the 50,000 budget.

Recommendations for Future Work

Hardware improvements should include transitioning to a custom PCB for miniaturization, implementing energy harvesting for extended battery life targeting 7 to 10 days, and incorporating additional sensors such as accelerometers for motion artifact detection, galvanic skin response for stress monitoring, and single-lead ECG for enhanced cardiac assessment. Adding BLE capability would enable smartphone gateway mode to reduce Wi- Fi power consumption, and LoRaWAN would support long-range applications in rural areas. An SD card slot would provide offline data logging backup.

REFERANCES

1. D. Suwankhong and P. Liamputtong, Aging and Health Promotion, in Handbook of Concepts in Health, Health Behavior and Environmental Health, Springer, 2025, pp. 117.

2. M. V Perez et al., Large-scale assessment of a smartwatch to identify atrial fibrillation, New England Journal of Medicine, vol. 381, no. 20,

   pp. 19091917, 2019.

3. R. S. Nair et al., Recent Trends and Opportunities of Remote Multi- Parameter PMS using IoT, in 2021 3rd East Indonesia Conference on Computer and Information Technology (EIConCIT), 2021, pp. 325329.

4. S. Ding and X. Wang, Medical remote monitoring of multiple physiological parameters based on wireless embedded internet, IEEE Access, vol. 8, pp. 7827978292, 2020.

5. E. Sardini, M. Serpelloni, and M. Ometto, Multi-parameters wireless shirt for physiological monitoring, in 2011 IEEE International Symposium on Medical Measurements and Applications, 2011, pp. 316 321.

6. Z. Rui, L. Qi, and L. Chuang, Design of wireless wearable multi parameter monitoring system based on GSM communication, in 2016 International Symposium on Computer, Consumer and Control (IS3C), 2016, pp. 714717.

7. G. Zhang, A wearable device for health management detection of multiple physiological parameters based on ZigBee wireless networks, Measurement, vol. 165, p. 108168, 2020.

8. L. Jinming, Z. Cheng, L. Yinlong, and W. Yihe, Multi-parameter cardiac remote monitoring system based on Android, J. Excellence Comput. Sci. Eng, vol. 2, no. 2, pp. 1824, 2016.

9. H. Li, G. Sun, Y. Li, and R. Yang, Wearable wireless physiological monitoring system based on multi-sensor, Electronics (Basel)., vol. 10, no. 9, p. 986, 2021.

10. H. Tao, Research on wearable multi-health parameter monitoring system based on Internet of Things, in Conference on Infrared, Millimeter, Terahertz Waves and Applications (IMT2022), 2023, pp. 951955.

11. V. N. J. Robert, P. Ragupathy, K. Chandraprabha, A. S. Nandhini, and

    M. Gnanasekaran, Multi-Parameter Smart Health Monitoring System using Internet of Things, in 2022 Second International Conference on Artificial Intelligence and Smart Energy (ICAIS), 2022, pp. 13261334.

12. C. Pintavirooj, T. Keatsamarn, and T. Treebupachatsakul, Multi- parameter vital sign telemedicine system using web socket for COVID- 19 pandemics, in Healthcare, 2021, p. 285.

13. Q. Wu, P. Tang, and M. Yang, Data processing platform design and algorithm research of wearable sports physiological parameters detection based on medical internet of things, Measurement, vol. 165,

    p. 108172, 2020.

14. Z. Xu et al., A wearable multi-parameter physiological system, in Ubiquitous Information Technologies and Applications: CUTE 2013, Springer, 2014, pp. 643648.

15. L. Fang et al., Multi-parameter health monitoring watch, in 2017 IEEE 19th International Conference on e-Health Networking, Applications and Services (Healthcom), 2017, pp. 16.

16. V. Randazzo, J. Ferretti, and E. Pasero, A wearable smart device to monitor multiple vital parametersVITAL ECG, Electronics (Basel)., vol. 9, no. 2, p. 300, 2020.

17. T. V. Nguyen, Q. V. Ngo, C. C. Vu, and others, Design and Construction of Multi-Parameter Health Testing and Monitoring Equipment, Journal of Technical Education Science, vol. 20, no. 04SI,

    pp. 7484, 2025.

18. H. K. Chattar et al., Multi-parameter wearable band for wireless data collection from people with epilepsy, in 2022 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), 2022, pp. 15.

19. B. Lin et al., A Solar-Powered Modular Wearable Sensor Network for Distributed Multi-Parameter Health Monitoring, Sens. Actuators A Phys., p. 117670, 2026.

20. J. Hu, J. Wang, and H. Xie, Wearable bracelets with variable sampling frequency for measuring multiple physiological parameter of human, Comput. Commun., vol. 161, pp. 257265, 2020.

21. H. Alemzadeh, Z. Jin, Z. Kalbarczyk, and R. K. Iyer, An embedded reconfigurable architecture for patient-specific multi-paramater medical monitoring, in 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2011, pp. 18961900.

22. A. Athira, T. D. Devika, K. R. Varsha, and others, Design and development of IOT based multi-parameter patient monitoring system, in 2020 6th International Conference on Advanced Computing and Communication Systems (ICACCS), 2020, pp. 862866.

23. K. Sruthi, E. V Kripesh, and K. A. Unnikrishna Menon, A survey of remote patient monitoring systems for the measurement of multiple physiological parameters, Health Technol. (Berl)., vol. 7, no. 2, pp. 153159, 2017.

24. S.-H. Li, B.-S. Lin, C.-A. Wang, C.-T. Yang, and B.-S. Lin, Design of wearable and wireless multi-parameter monitoring system for evaluating cardiopulmonary function, Med. Eng. Phys., vol. 47, no. 1, pp. 144 150, 2017.

25. A. Bin Queyam, R. K. Meena, S. K. Pahuja, and D. Singh, An IoT based multi-parameter data acquisition system for efficient bio-telemonitoring of pregnant women at home, in 2018 8th International Conference on Cloud Computing, Data Science & Engineering (Confluence), 2018, pp. 1415.