International Journal of Engineering Trends and Technology (IJETT) – Volume23 Number 5- May 2015
Overview on Configurable and Mixed ECG Signals
Monitoring System
V. H. Satonkar #1, D. S. Shilwant *2, S. P. Kharde #3
#
ME Student, Electronics and Telecommunication Engineering ,Dr. Babasaheb Ambedkar Marathwada University,
Aurangabad (M.S.), India.
Abstract— Electrocardiogram (ECG) analysis is used in the
medical area and Monitoring electrocardiogram or analysis
provides idea about the disease to the cardiologists. Now a day’s
complex signal processing and low power required, and to
achieve this goal different researchers used different devices like
SoC, ZigBee, PSoC, HRV ASIC chip, sensors, Microcontroller
unit, and algorithms for analysis like continuous wavelet
transform .This paper gives an overview on configurable and
mixed ECG signals monitoring system and literature survey
provides an idea about the existing system.
Keywords— Electrocardiogram, cardiologists, Soc, Zigbee,
Figure 1 Elements of ECG signal.
Microcontroller, PSoC.
Here Segments are nothing but isoelectric lines
periods between different waves, where Waves are
Now a day heart related problems are increased deviations of the signals from the base line (PQRST)
and hence it is important to diagnosis the cardiac and Intervals are the period consists of Wave and
cycle which is generated by the polarization and Segment.
depolarization
of
cardiac
tissue
and
II. LITERATURE SURVEY
Electrocardiogram (ECG) picks up electrical
impulses and translates into a waveform. These
DeChazal, P., O'Dwyer, M., Reilly, R.B. [1],
waves are useful for diagnostic purposes on human provides system for the automatic processing of the
hearts for the diagnosis of heart abnormalities. But electrocardiogram (ECG) and classification
at the time of electrical impulse generated by methods of heartbeats. This system or method
cardiac cycle some motion artefacts gets added and works on one of the five beat classes recommended
poor or configurable and mixed ECG signals get by ANSI/AAMI EC57:1998 standard. These heart
generated, which causes poor result, which cause beats are normal beat, VEB i.e. ventricular ectopic
wrong clinical diagnosis. Before classification of beat, SVEB i.e. supraventricular ectopic beat, or
ECG signals it is necessary to provide pure ECG may be fusion of a normal and a VEB, or unknown
signal to the ECG monitoring system. This beat type. Heart beat data was obtained from MITcondition generates higher level challenges to the BIH arrhythmia database; the 44 non pacemaker
system developer such as removing motion recordings were taken. Database was divided into
artefacts and detection of accurate features. Fig. 1 two datasets which contains 50 000 beats from 22
show one cardiac cycle and different elements of recordings. From these two datasets first dataset
ECG signals
was used to select a classifier configuration and
second dataset was used to provide an independent
performance
assessment
of
the
selected
configuration. Dataset were compared by using
twelve configurations processing feature sets
derived from two ECG, these feature sets were
based on ECG morphology, heartbeat intervals, and
I. INTRODUCTION
ISSN: 2231-5381
http://www.ijettjournal.org
Page 230
International Journal of Engineering Trends and Technology (IJETT) – Volume23 Number 5- May 2015
RR-intervals. This system provides a sensitivity of complexity microcontroller was estimated to
75.9% 77.7%, a positive productivity of 38.5% and consume32.5 V. all these features makes it portable
81.9, and a false positive rate of 4.7% and 1.2% for HRV monitoring system.
the SVEB class and the VEB class respectively.
Hulzink, J. et al. [5], presents a voltage-scalable
Yu Hen Hu et al. [2], present a MOE approach i.e. digital signal processing system. This system is
mixture-of-experts approach to demonstrate the designed for the use in a wireless sensor node
feasibility of having a patient-adaptable ECG beat (WSN) for ambulatory monitoring of biomedical
classification algorithm. This mixture-of-experts signals. The aim of this system is ambulatory
was based on a SOM/LVQ-based approach and monitoring, power consumption, which directly
used to develop customized electrocardiogram translates to the WSN battery lifetime and size,
classifier, which is useful to improve the must be kept as low as possible. The proposed
performance of ECG processing. By using a MOE processing platform is an event-driven system and
classifier large ECG database of many patients was architecture uses effective system partitioning to
taken which is then tested with MIT/BIH enable duty cycling, single instruction multiple data
arrhythmia database and it is observed that the (SIMD) instructions, power gating, voltage scaling,
significant performance enhancement using this multiple clock domains, multiple voltage domains,
approach.
and extensive clock gating. For heart-beat detection
Hyejungkim, et al. [3], presents a mixed signal continuous wavelet transform (CWT) shows that
ECG SoC monitoring system. This system contains the platform preserves the sensitivity and positive
integrated analog as a front end and DSP as a back productivity of the algorithm, also achieves the
end. For impedance measurement of ECG signal lowest energy/sample for Electrocardiogram (ECG).
and band power extraction AFE supports Naveen Verma et al. [6], presents a low-power SoC
concurrent 3-channel ECG monitoring. For motion which is useful for performs EEG acquisition and
artifact, r-peak, HRV analysis and classification of feature extraction. Feature extraction is required for
arrhythmia SIMD processor provides additional continuous detection of seizure onset in epilepsy
algorithm as advanced functionality. Without patients. The SoC corresponds to one EEG channel
affecting the information content of the input and up to 18 channels may be worn to detect
signals the adaptive sampling ADC provides the seizures as part of a chronic treatment system. SoC
equivalent data rate of the ADC output and SoC system contains instrumentation-amplifier, which is
provides best power consumption i.e. 31.1µW from used for chopper-stabilization in a topology that
1.2 V supply.
achieves high input-impedance and rejects large
Massagram, W. et al. [4], present application- electrode-offsets and operates at 1 V, ADC employs
specific integrated circuit i.e. ASIC and this system power-gating for low energy-per-conversion while
is designed for digital heart rate variability (HRV) using static-biasing for comparator precision and
parameter monitoring and assessment. System digital processor that streams features-vectors to a
ASIC chip was used to measures beat-to-beat (RR) central device where machine-learning classifier is
intervals. System stores RR interval HRV used for seizure detection. EEG feature extraction
parameters into its internal memory in real time. processor contains low-power hardware whose
These HRV parameter then tested with wide range parameters are determined through validation via
of short-term and long-term ECG signals obtained patient data. The sensing and local processing
from Physionet, provides MITBIH datasets. System system power by 14x by reducing the rate of
detects R peaks with millisecond accuracy, and wireless EEG data transmission. System derived
stores up to 2 min of continuous RR interval data. Feature vectors at a rate of 0.5 Hz, and SoC
The ASIC chip was fabricated in a 0.5 m operates from a 1 V supply and system consuming
complementary
metal-oxide
semiconductor 9 J per feature vector.
technology on a 33 mm2 die area, with a measured
Uttam. U. Deshpande and V. R. Kulkarni [7],
dynamic power consumption of 10 W and measured presents system which is design for a low power
leakage current of 2.62 nA. The HRV monitoring sensor node which acquires ECG signal, process
system consists HRV ASIC chip and a low and transmits it over wireless medium. In this
ISSN: 2231-5381
http://www.ijettjournal.org
Page 231
International Journal of Engineering Trends and Technology (IJETT) – Volume23 Number 5- May 2015
system PSoC was used i.e. Programmable System types of LEDs and photodiode packed in Velcro
on Chip processor performs rapid, complex signal strip are used for facing to a patient’s fingertip for
processing and required low power, also PSoC’s pulse oximetry and heartbeat. Three colours LED
capabilities extend its use in designing intelligent with LDR are used for pH level of a patient.
wireless sensor node. This ECG monitoring system Microcontroller unit is used for interfacing with
detects R peaks at regular intervals, calculate heart wireless module, processing all biomedical sensor
rate and classify the signals. By using simple data sending to base PC. PSoC circuits represent a
threshold technique monitoring system provides new concept in embedded systems design that
high accuracy, low error rate and good noise replaces multiple traditional MCU-based system
immunity. System CY3271sensor node enables components with one, low cost single-chip
transmitter only when critical heart rate is observed programmable device.
instead of sending data continuously to the base
III. CONCLUSIONS
station and also provides reduction in power
The Electrocardiogram (ECG) analysis is used in
consumption.
Marco Altini et al. [8] present an ECG patch the medical area and analysis provides idea about
aiming at long term patient monitoring system. This the disease to the cardiologists. In some last year’s
system contains recently standardized Bluetooth many researchers are done different research in this
Low Energy (BLE) technology combined with a area and this paper gives an overview on
customized ultra-low-power ECG System on Chip configurable and mixed ECG signals monitoring
(ECG SoC) and including Digital Signal Processing system and literature survey provides an idea about
(DSP). This design provides ultra-low power the existing system.
systems, which is able to continuously monitor
REFERENCES
patients and also performing on board signal [1] De Chazal, P., O'Dwyer, M., Reilly, R.B.,” Automatic classification of
heartbeats using ECG morphology and heartbeat interval features”,
processing. System done board signal processing
Biomedical Engineering, IEEE Transactions on (Volume:51 , Issue:
such as filtering, data compression, beat detection
7 ),2004.
and motion artifact. At the time of computing beat [2] Yu Hen Hu,SurekhaPalreddy, and Willis J. Tompkins,” A PatientAdaptable ECG Beat Classifier Using a Mixture of Experts Approach”,
detection and transmitting heart rate remotely via
IEEE Transactions on Biomedical Engineering, Vol. 44, No. 9, Sept
1997.
BLE, the ECG SoC and BLE leads to a total current
[3]
Hyejungkim, et al.,” A Configurable and Low-Power Mixed Signal
consumption of only 500μA at 3.7V. This feature
SoC for Portable ECG Monitoring Applications”,IEEE Transaction on
Biomedical Circuits and System, Vol.8, April 2014.
allows system up to one month lifetime with a
[4]
Massagram,
W. ,
Hafner,
N. ; Mingqi
Chen ; Macchiarulo,
400mAh Li-Po battery only.
L. ; Lubecke, V.M. ; Boric-Lubecke, O.. “Digital Heart-Rate
Variability Parameter Monitoring and Assessment ASIC,” Biomedical
D.J.R.Kiran Kumar and Nalini Kotnana [9],
Circuits and Systems, IEEE Transactions on , Volume:4 , Issue:
presents a portable real-time wireless health
1 ,2010.
monitoring system. Monitoring system is [5] Hulzink, J. , Konijnenburg, M., Ashouei, M., Breeschoten, A., Berset,
T., Huisken, J., Stuyt, J., de Groot, H., Barat, F., David, J., Van
implemented using Programmable System on Chip
Ginderdeuren, J.,”AnUltra Low Energy Biomedical Signal Processing
System Operating at Near-Threshold”, Biomedical Circuits and
(PSoC). This system is useful for monitoring of
Systems, IEEE Transactions on ,Volume:5 , Issue: 6 ,2011.
patients’ temperature, heart rate and oxygen [6] Naveen Verma, Ali Shoeb, Jose Bohorquez, Joel Dawson, John Guttag,
and Anantha P. Chandrakasan,” Micro-Power EEG Acquisition SoC
saturation in bloody, pH level of blood and ECG.
With Integrated Feature Extraction Processor for a Chronic Seizure
Here low cost, low power consumption and flexible
Detection System”, IEEE Journal of Solid-State Circuits, Vol. 45, No.
4, April 2010.
network topology ZigBee wireless module is used
[7]
Uttam. U. Deshpande, V. R. Kulkarni,” Wireless ECG monitoring
to sense the remote patient data. To sense the
system with remote data logging using PSoC and CyFi”, IJAREEIE,
Vol. 2, Issue 6, June 2013.
remote patient data here low cost, low power
[8]
Marco Altini, Salvatore Polito, Julien Penders, Hyejung Kim, Nick
consumption and flexible network topology ZigBee
Van Helleputte, Sunyoung Kim, Firat Yazicioglu,” An ECG Patch
Combining a Customized Ultra-Low-Power ECG SoC with Bluetooth
wireless module is used. Proposed systems sensor
Low Energy for Long Term Ambulatory Monitoring”, Wireless Health,
unit consists of temperature sensor, two types of
AMC, 2011.
LEDs and photodiode packed in Velcro strip, three [9] D.J.R.Kiran Kumar and Nalini Kotnana, “Design and Implementation
of Portable Health Monitoring system using PSoC Mixed Signal Array
colours LED with LDR, ECG, Microcontroller unit,
chip”, International Journal of Recent Technology and Engineering
(IJRTE),
Volume-1,
Issue-3,
August
2012.
PSoC circuits built by Cypress Microsystems. Two
ISSN: 2231-5381
http://www.ijettjournal.org
Page 232