scholarly journals An IoT and Fog Computing-Based Monitoring System for Cardiovascular Patients with Automatic ECG Classification Using Deep Neural Networks

Sensors ◽  
2020 ◽  
Vol 20 (24) ◽  
pp. 7353
Author(s):  
Jaime A. Rincon ◽  
Solanye Guerra-Ojeda ◽  
Carlos Carrascosa ◽  
Vicente Julian

Telemedicine and all types of monitoring systems have proven to be a useful and low-cost tool with a high level of applicability in cardiology. The objective of this work is to present an IoT-based monitoring system for cardiovascular patients. The system sends the ECG signal to a Fog layer service by using the LoRa communication protocol. Also, it includes an AI algorithm based on deep learning for the detection of Atrial Fibrillation and other heart rhythms. The automatic detection of arrhythmias can be complementary to the diagnosis made by the physician, achieving a better clinical vision that improves therapeutic decision making. The performance of the proposed system is evaluated on a dataset of 8.528 short single-lead ECG records using two merge MobileNet networks that classify data with an accuracy of 90% for atrial fibrillation.

2020 ◽  
Vol 7 (1) ◽  
pp. 16
Author(s):  
Nuzhat Ahmed ◽  
Yong Zhu

Atrial fibrillation, often called AF is considered to be the most common type of cardiac arrhythmia, which is a major healthcare challenge. Early detection of AF and the appropriate treatment is crucial if the symptoms seem to be consistent and persistent. This research work focused on the development of a heart monitoring system which could be considered as a feasible solution in early detection of potential AF in real time. The objective was to bridge the gap in the market for a low-cost, at home use, noninvasive heart health monitoring system specifically designed to periodically monitor heart health in subjects with AF disorder concerns. The main characteristic of AF disorder is the considerably higher heartbeat and the varying period between observed R waves in electrocardiogram (ECG) signals. This proposed research was conducted to develop a low cost and easy to use device that measures and analyzes the heartbeat variations, varying time period between successive R peaks of the ECG signal and compares the result with the normal heart rate and RR intervals. Upon exceeding the threshold values, this device creates an alert to notify about the possible AF detection. The prototype for this research consisted of a Bitalino ECG sensor and electrodes, an Arduino microcontroller, and a simple circuit. The data was acquired and analyzed using the Arduino software in real time. The prototype was used to analyze healthy ECG data and using the MIT-BIH database the real AF patient data was analyzed, and reasonable threshold values were found, which yielded a reasonable success rate of AF detection.


In the present work, we have designed a health monitoring system based on Node MCU to monitor temperature, heart rate and oxygen saturation level (SpO2) signals, sensed by respective sensors. The necessary signal conditioning circuits have been designed in our laboratory using off-the shelf electronic components. A Data acquisition system has been designed using ESP 32 Node MCU. The designed system is a low-cost alternative to the commercially available USB controller based health monitoring systems. Firmware has been developed and deployed into the Node MCU using arduino IDE. The acquired data has been displayed on OLED display. The result shows maximum errors in the measured parameters within 2%. The designed system helps to achieve portability, high functionality and low cost which makes it an easy accessible tool for public, hospital, sports healthcare and other medical purposes.


2021 ◽  
Vol 2042 (1) ◽  
pp. 012042
Author(s):  
Olivier Steiger ◽  
Reto Marek

Abstract Monitoring systems are essential for the energy-efficient and comfortable operation of buildings. However, today's monitoring solutions are relatively expensive in terms of purchase, installation, and maintenance. At the same time, there is a need for low-cost monitoring systems, especially for smaller buildings. To address this need, a novel do-it-yourself, low-cost building monitoring system based on open technologies has been developed. The system is intended to be assembled and put into operation by laymen in accordance with given instructions. Accordingly, all work stages must be simple and obvious. This paper describes the low-cost monitoring system and its prototype implementation.


Author(s):  
Patrik Šarga ◽  
Patrik Strnisko

The presented paper describes the creation of a “low-cost” monitoring security system based on the IoT platform connecting to the cloud. We focused on heat exchanger station, which is used in a block of flats. A simulation solution was devel-oped together with 3D visualization and a practical test. Final monitoring system informed the operator about the current state of the heat exchanger station and operator can intervene in time so that heat exchanger station is not suddenly damaged or cause some damage. Such monitoring system will find application in practice, but also in the teaching process, as the preparation of graduates for the modern monitoring systems, which are increasingly used in practice, will be improved. The presented paper is a scientific and methodological publication.


2011 ◽  
Vol 128-129 ◽  
pp. 74-78
Author(s):  
Kai Xu ◽  
Ying Hai Shao ◽  
Gang Wang

The proposed system is designed to monitor patients with atrial fibrillation (AF) in family. This system mainly consists of wireless sensor networks (WSNs), which contains several mobile sensor nodes and coordinator for acquisition of bio-signals, and an embedded computer (EC) for signal processing. The WSNs are responsible to acquire and transmit Electrocardiogram (ECG). The EC is to extract the AF signal using nonlinear blind source extraction (BSE) algorithm. The extracted AF signal is then utilized to intelligently judge whether or not AF is on, based on which the system will send alert information to related doctors via Ethernet. In the meantime, the extracted AF signal is displayed on liquid crystal display (LCD), and then is also sent to relate doctors. The system aims to be low-cost, low-power consumption, small size and long-distance (up to thirty meters) transmission, can be further integrated into other healthcare monitoring system, and is expected to have great potential in family monitoring.


Smart Cities ◽  
2020 ◽  
Vol 3 (1) ◽  
pp. 138-156 ◽  
Author(s):  
Spiridon Vergis ◽  
Vasileios Komianos ◽  
Georgios Tsoumanis ◽  
Athanasios Tsipis ◽  
Konstantinos Oikonomou

With the rapid increase of vehicles in use worldwide, the need for efficient traffic monitoring systems has arisen. This work proposes a low-cost vehicular traffic monitoring system using IoT devices and fog computing. The system is based on a three-tiered architecture which is composed of (i) the mobile tracking system that records the positions of the vehicles using GPS technologies; (ii) the information gathering system which gathers all the data collected by the mobile tracking system; and (iii) the fog devices that process the data collected and extract the information needed. The system is tested in the town of Corfu during a period of increased tourism when the traffic is considered to be relatively dense. The mobile tracking system devices are placed on taxis and with the help of professional taxi drivers the accuracy of the data collected is evaluated. The system is able to record the movement of the vehicles accurately using its own independent data. The results can be remotely accessed by utilizing fog and cloud computing infrastructure established to process the data and upload it on a server. The system is used to give a better understanding of the speed variance in the center of the town during different dates and hours. In conclusion the system presented in this study can be utilized to monitor the traffic and provide vital information about its behavior in relation to time.


10.6036/10075 ◽  
2021 ◽  
Vol DYNA-ACELERADO (0) ◽  
pp. [ 7 pp.]-[ 7 pp.]
Author(s):  
JOAO PEDRO NIEVES DA COSTA ◽  
PAULO AVILA ◽  
JOAO BASTOS ◽  
LUIS PINTO FERREIRA

The industry 4.0 revolution provides the machines with a sensory and communicational capacity, which allows them to monitor and collect large amounts of information. This kind of data have an impact on planning, maintenance, and management of production, enabling real time reaction, efficiency increase, and the development of predictive and process improvement models. The most recent machines are prepared to communicate with the existing monitoring systems, however, many (around 60%) do not. The objective of this work is to present the proposal of a system for remote monitoring of equipment in real time that meets the requirements of low cost, simplicity, and flexibility. The system monitors the equipment in a simple and agile way, regardless of its sophistication, installation constraints and company resources. A prototype of a system was developed and tested both laboratory conditions and a productive environment. The proposed architecture of the system comprises of a sensor that transmits the machine’s signal wirelessly to a gateway which is responsible of collecting all surrounding signals and send it to the cloud. During the testing and assessment of the tools, the results validated the developed prototype. As main result, the proposed solution offers to the industrial market a new low-cost monitoring system based in mature and tested technology laid upon flexible and scalable solutions. Industry 4.0, Machine Monitoring, Beacon, Bluetooth BLE, Remote Monitoring, Low Cost, SME’s, b-Remote


2021 ◽  
Vol 11 (16) ◽  
pp. 7313
Author(s):  
Seung Soo Kwak ◽  
Yun Chan Im ◽  
Yong Sin Kim

As smart grids develop rapidly, low-cost monitoring systems for pole-mounted transformers increase in demand. Even though battery-powered wireless monitoring systems appear to provide optimal solutions, they consume large amounts of energy for continuous sampling and data transmission. Operation and maintenance costs then increase owing to reduced battery lifetime and battery replacement. To overcome this problem, this paper presents an event-driven battery-powered wireless monitoring system that monitors abnormalities of a transformer and transmits data only if an abnormality occurs. When the proposed event controller detects an abnormality, it enables a root mean square (RMS) converter and a peak detector for sampling and transmitting the maximum RMS value of the abnormal signal and then falls into sleep mode until the next event to save energy. Simulation and experimental results show that the proposed system enhances battery lifetime by up to two orders of magnitude compared to a conventional battery-powered wireless monitoring system.


2012 ◽  
Vol 2012 ◽  
pp. 1-6 ◽  
Author(s):  
Peter Haemers ◽  
Piet Claus ◽  
Rik Willems

Atrial fibrillation (AF) is the most common cardiac arrhythmia and imposes a huge clinical and economic burden. AF is correlated with an increased morbidity and mortality, mainly due to stroke and heart failure. Cardiovascular imaging modalities, including echocardiography, computed tomography (CT), and cardiovascular magnetic resonance (CMR), play a central role in the workup and treatment of AF. One of the major advantages of CMR is the high contrast to noise ratio combined with good spatial and temporal resolution, without any radiation burden. This allows a detailed assessment of the structure and function of the left atrium (LA). Of particular interest is the ability to visualize the extent of LA wall injury. We provide a focused review of the value of CMR in identifying the underlying pathophysiological mechanisms of AF, its role in stroke prevention and in the guidance of radiofrequency catheter ablation. CMR is a promising technique that could add valuable information for therapeutic decision making in specific subpopulations with AF.


Sign in / Sign up

Export Citation Format

Share Document