Effective analysis and accurate detection of common diseases in ECG
signals and classifications and monitoring through cloud computing
technology
DHANANJAYA V, GEETHA M
Department of Computer Science and Engineering
IMPACT College of Engineering and Applied Sciences
Sahakarnagar, Bangalore
Karnataka, INDIA
Abstract:
In the modern age of technology and advancements, the world is shrinking and accessible to
everyone, in all ways including the medical field at the outset to save the lives. As the various experts are
spread over the globe with their expertise, even the human dis-orders are also growing equally with the
same speed. Hence it requires an assimilation of various sensors, embedded systems and some regularized
protocol for hassle free interaction over the finger tips. In the proposed work we are attempting to interface
few biosensors to capture ECG signals and to perform parametric estimation on breath-rate, heart beat
rate, systolic pulse and other required parameters to classify the signal into any disease oriented or normal
human being and uploaded the results along with type of disease. From the cloud, the concern medical
experts can access and can treat the patient for further diagnosis. In this paper we are considering
Epilepsy, heart beat rate, systolic pulse compared with normal condition of the human being. To classify
the recorded ECG signals to discriminate among above condition we make use of Artificial Neural
Network model and the entire processing of bio-medical signals are done on Matlab Platform. The
processing of database is selected from the standard universally available database MIMIC II. The Matlab
processed signals are again processed over a common protocol to communicate to the expert for exact
diagnosis of the patient conditions over the captured bio-signals, the entire system we make use of
modified LED algorithm and we term it as Health-Raid algorithm.
Key-Words:
-
ANN, Cloud Computing, ECG, Heart Monitoring System, MLED and RTP-RTCP
.
Received: May 23, 2021. Revised: March 13, 2022. Accepted: April 12, 2022. Published: May 7, 2022.
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2022.19.13
Dhananjaya V, Geetha M
E-ISSN: 2224-2902
107
Volume 19, 2022
1 Introduction
Electroencephalographic recording system serves to
the examination of a brain activity. Depending on a
human brain activity, several types of
electroencephalographic (ECG) waves can be
distinguished. Deviations of the normal ECG waves
corresponding to alpha or beta activity, i.e. sharp
spikes, can refer to pathologic patterns
accompanying neurotically illnesses. Based on a
detection of pathologic patterns recurring with
period of one or a few seconds in ECG signal,
electroencephalographic recording system has
been used to diagnose diseases such as epilepsy or
polio. Because of the important role of the ECG
examination in neurological diseases diagnostics,
there have been several attempts to simulate a real
ECG signal by artificial models. Modelling of the
ECG signals is a very wide academic area divided
in several branches. A brief overview of possible
branches of ECG modelling is introduced
.
2 Related Work
In the recent century it has witnessed the utilization
of advancements in the domain of information and
communication technology [15]. The information
communication and technology has a major impact
including medical clinical field. To diagnosis heart
related issues based on ECG, the ECG record
obtained is diagnosed on the input data received for
further follow up and consultation with the
utilization of technology advancements and tis
trends. The important organ of cardiovascular
system of any human body is the heart. The cardiac
tissues that relaxes and contracts in a particular
repetitive frequency which is responsible for the
circulation of blood over the body. With the usage
of the combination of the ECG & ECG signals we
are proposing to identify the disorders of the human
body i.e., cardiac disorder, paralysis, internal joint
disorders and it is compared with the normal human
being through the predefined protocol for
transmission in real time with suitable media for
data exchange [16].
In [2] syncretizes the principal idea of the Sea
Computing model in Internet of Things and the
steering convention of the remote sensor arrange,
and proposes another directing convention CASCR
(Context-Awareness in Sea Computing Routing
Protocol) for Internet of Things, in view of setting
mindfulness which has a place with the key
advancements of Internet of Things. Moreover, the
paper depicts the subtleties on the convention in the
work process, information structure and
quantitative calculation etc [14].
At long last, the recreation is given to investigate the
work execution of the convention CASCR. Hypothetical
examination and analysis check that CASCR
has higher vitality proficient and longer lifetime than the
congeneric conventions. The paper improves the
hypothetical establishment and makes some commitment
for remote sensor system traveling to Internet of Things
in this examination stage [17].
In [3] investigated IoT from various points and here we
outline the exercises learned by this audit. To start with,
from the market openings points of view, venture on this
new innovation is sane for associations looking for
market intensity. From the design perspective, the
layered structure of IoT frameworks is embraced well by
IoT systems and research endeavours [4]. In any case,
the quantity of layers and their degrees are characterized
distinctively relying upon the fundamental frameworks
and innovations. As adaptability and interoperability
have an extraordinary significance in IoT applications,
expanding the design without hardly lifting a finger these
issues. The five-layer engineering [4, 5] present such a
model. The intense increment in urbanization in the
course of recent years requires practical, proficient, and
savvy answers for transportation, governance,
environment, personal satisfaction, etc. The Internet of
Things offers many complex and omnipresent
applications for keen urban communities [13]. The
vitality request of IoT applications is expanded, while
IoT gadgets keep on developing in the two numbers and
prerequisites. Along these lines, brilliant city
arrangements must be able to productively use vitality
and handle the related difficulties. Vitality the executives
is considered as a key worldview for the
acknowledgment of complex vitality frameworks in
shrewd urban communities [6]. In this article, we present
a short outline of vitality the executives and difficulties
in brilliant urban areas. We at that point give a bringing
together structure to vitality productive enhancement and
planning of IoT-based shrewd urban communities. We
likewise examine the vitality gathering in shrewd urban
areas, which is a promising answer for broadening the
lifetime of low-control gadgets and its related difficulties
[12]. We detail two contextual analyses. The first targets
vitality effective planning for brilliant homes, and the
second covers remote power exchange for IoT gadgets in
shrewd urban areas. Reproduction results for the
contextual investigations exhibit the enormous effect of
vitality proficient planning advancement and remote
power exchange on the exhibition of IoT in shrewd
urban areas [7].
Constant mixed media applications are frequently sent to
give basic data in circumstances, for example, news
inclusion of an occasion or an episode or an rescue
vehicle racing to give crisis care. Portable cell inclusion
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2022.19.13
Dhananjaya V, Geetha M
E-ISSN: 2224-2902
108
Volume 19, 2022
and execution of a supplier system differs both
spatially and transiently and moves the capacity of
the system to help continuous system access and
application progression [20]. We recommend that
between suppliers handoffs could reduce this issue,
given the ongoing multiplication of Dual SIM Dual
Active (DSDA) gadgets what's more, the likelihood
of Multi SIM Multi Active gadgets by arranging
workstations with numerous remote broadband
associations. We propose an application QoS
mindful portability the board approach. The product
execution is named as SeaMoX and dependent on
SeaMo+, a previous usage. We utilize live video
spilling for instance application and exhibit the
effect of system choice and handoff by looking at
the playback at the recipient, which uses a versatile
jitter cushion algorithm [21].
3 System Architecture
In the proposed system it can be broadly classified
into six different steps namely initially data
acquiring through the bio-sensor network for the
analysis of various parameters such as Heart rate
(HR), Respiration rate (RESP), Systolic blood
pressure (ABPsys), PLETH)Oxygen saturation
(SpO2) and common diseases. Which are fed as
input for the ECG signal construction and after
drawing Raw ECG signals it is segmented into five
different types of ECG waves which are frequency
dependent signals for analysis. The overall
representation of the proposed system is as shown
in Fig.1.
ECG signal: ECG signals are the powerful tool that
has the capability to reflect all the activities of the
brain from all the various parts of the brain viz.,
Central Nervous System (CNS) which comprised of
Spinal cord and Peripheral Nervous System (PNS).
ECG signals are essential enough to understand the
dynamism of brain and also to monitor the various
physiological states with its diagnosis of
neurological analysis. These ECG signals are the
respondent signals recorded with 10000–100000
neurons parallel and gathered with various bio-
sensors attached to human body in part of the scalp
region. These signals not only limited to study of
neurons but also it is highly subjective to detect
Epilepsy, tumorous cell behaviours and compared
with the normal recording of ECG [8, 9].
Majorly ECG systems in these days, are digital are
amplified and is digitized via an available analogy-
to digital converter, after being passed through an
anti-aliasing filter. Analogy-to digital sampling
typically occurs at 256-512 Hz in clinical scalp
ECG; sampling rates of up to 20 kHz are used in
some research applications [18]. In these Scalp ECG
signals are recorded using various biosensors which are
similar to older electrodes to capture raw ECG signal for
processing.
The ECG signals are comprised of five different types of
signals which are classified based on the frequency range
with the RAW data analysis ranging from 0.5 - 45 Hz
and it is digitally enhanced with 8 or 9 bit representation
to get 256 Hz or 512 Hz [23].
ECG Signal: Electrocardiography (ECG) is the process
of recording the electrical activity of the heart over a
period of time using electrodes placed over the skin.
These electrodes detect the tiny electrical changes on the
skin that arise from the heart
muscle's electrophysiological pattern
of depolarizing and repolarising during each heartbeat. It
is very commonly performed to detect any cardiac
problems [10].
Database:
In this paper standard clinical database from the MIMIC
is an openly available dataset developed by the MIT Lab
for Computational Physiology, comprising de-identified
health data associated with ~40,000 critical care patients.
Test images: convert to packets and transmit through our
network simulation. The Properties of image having
360*360mm dimension. Test images having same file
type PNG File with the file size of 320kB [22]. The ECG
and ECG signals are acquired from standard database of
MIMIC-II and through pre-processing of the signals are
converted into packest based header format shown in
Fig.3. In this work, totally four different multimedia
metadata’s are considered such as ECG, EEG, Audio and
Image data’s and all datasets are converted into packets,
these packets has session ID, sun-SID, sub-sub-SID,
flow ID, sub-sub-FID, node counter and timing setup for
counter of packets shown in Fig.3.
With the acquired database from the ECG different
sensors from the human body, the complete set of data
received from the sensor will form the wireless body
sensor network ready for the data transmission using the
required protocols i.e., RTP & RTCP for error free
communication, then as defined by the essential software
for analytics and data processing.
Thinkspeak is considered with the peculiar password in
which only expert doctor can view from the remote place
for the consultation and advice by receiving the
cumulative data from RTP & RTCP through e-health
monitoring system, in the proposed system in the RTP &
RTCP protocol data compression and decompression is
achieved using wavelet based technique with
consideration of symlet and debauche method.
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2022.19.13
Dhananjaya V, Geetha M
E-ISSN: 2224-2902
109
Volume 19, 2022
Fig.1. Proposed block diagram of Data Packets
transmission and storage in Cloud
3.1 Holomorphic Cryptography algorithm
ECG signal is input to the algebraically
holomorphic which are converted into integers for
the given plaintext and any
key integer .
Inputs: Plaintext and Secrete Key: P1 and SK1
Outputs: Integer Value: Iv
X2 = [P1, SK1] mod
X0= P1. (QQ’+PQ1) mod X
= QQ’+PP1Q1-T1 PQ0 for any integer values
of T1
=
Iv = [outputs] =
X0 for the integer
K2
=Y+Q
+PQ for the given integer P and Q.
The output of the CT is a nearly to numerous of
integer P.
In a traditional 12-lead ECG, ten cathodes are put on
the patient's appendages and on the outside of the
chest. The general extent of the heart's electrical
potential is then estimated from twelve unique points
("leads") and is recorded over some stretch of time
(normally ten seconds) [11]. Right now, by and
large greatness and bearing of the heart's electrical
depolarization is caught at every minute all through
the cardiovascular cycle. The chart of voltage
versus time delivered by this non-invasive clinical
technique is an electrocardiogram. There are three
primary parts to an ECG: the P wave, which speaks to
the depolarization of the atria; the QRS complex, which
speaks to the depolarization of the ventricles; and the T
wave, which speaks to the repolarization of the ventricles
[19].
Fig 2: Proposed Automated diagnosis system using
advanced protocol and adaptive compressive sensing
technique with cloud storage.
By the analysis of individual ECG signals it
considered for the analysis of wavelet based approach
in which CWT and DWT are used in combination as
mother wavelet due to its advantage of variable
window size with wider response for the low
frequency and sharp for the higher frequency signals
of ECG. These mother wavelets are derived from the
8 sensor
data
acquisition
Wireless
body
network
Transmitter
(RTP &
RTCP)
Thinkspeak
Receiver
(RTP & RTCP)
E-health
Monitorin
Health
adviser
(Doctor)
Prescription
(Advice)
Data analysis
and diagnosis
process
ECG signal database from MIMIC-II
Filtration and normalization for removal artifacts
Analysis of signal and detection of decreases
Calculation of epilepsy, Heart Rate, Sio2, and six
common ECG diseases based on input signals
Selection if best mother wavelet (dbi (i=1 to
45)
Compression of data using db1 (best
wavelet)
Data transmission using WBSN – MLED
Calculation of energy and power for MLED
protocol
Decompression using best mother wavelet
(db1protocol)
Calculation of CR, PRD and SNR
protocol
Calculation of energy and power for MLED
protocol
Fuzzy Interface system (FIS) with
membership and ANN classification
Heart
Rate
Spo2
Epilepsy
Normal
Sinus
rhyth
Atrial
fibrillation
Sinus
tachycar
Junction
al
Sinus
bradycar
Ventricular
fibrillation
Atrial flutter
Ventricular
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2022.19.13
Dhananjaya V, Geetha M
E-ISSN: 2224-2902
110
Volume 19, 2022
various functions through the dilation and
translation which are well suited for analysis for
sudden and transient signal variation. These
wavelet based approach is opted for the reason of
its capability of irregular impulse signal arrival at
various interval of time.
In the mother wavelet based approach g(n) shown
in Fig.1. Acts as high pass filter and h(n) will
serve as low pass filter in DWT analysis [24].
In CWT analysis it is analysed as scaling and
translation parameter for signal representation.
With the feature extraction from the wavelet
based analysis it is an essential parameter for the
neural network based classifier for identification
of normal or any disorder oriented ECG signals
viz., Epilepsy, tumour or any other kind of
disorders of the patient under consideration.
These neural network based approach has been
provided with parameters such as homogeneity,
energy, correlation and contrast at the input
layers. All the parameters under consideration is
treated with predefined weights Wi and applied
with transformations for the frequency and time
domain. The decision network is provided with
the activation parameter such as ’ for
classification of ECG signal as shown in Fig.2.
Table 1: Various values of ECG signals, Wavelets
and co-efficient.
The PRD analysis is done based on two different
wavelets namely Symlet wavelet and debauche
wavelet with the available coefficient in the range of 1
to 20.
The wavelet with the co-efficient 5, 10, 15 and 20 are
tabulated in the table 2 and comparative graph is
plotted as shown in fig 10, for the ECG signal with
Tumour detection and Epilepsy disorder it is noted that
PRD value should be as minimum as possible almost
located around the axis would be good for accurate
identification of disorder when compared with the
normal condition. The various values of PRD is
compared and analysed as shown in table1. Different
signal analysis and to increase the accuracy of disorder
detection.
The entire proposed algorithm is implemented based
on modified LED (Local Emergency Detection)
protocol as disused below
Fig.3. Proposed Automated diagnosis system
using advanced protocol and adaptive compressive
sensing technique with cloud storage.
3.2 FORMATION OF THE CLUSTERS FOR
WSN
SYMLET
Wavelet for
TUMOUR
ECG Signal
Co-
efficient
GAMMA
BETA
ALPHA
THETA
DELTA
PRD
5
26
13
13
6
1
-
1.55
10
53
28
15
5
1
1.33
15
57
14
7
7
1
0.69
20
29.00
29
14
4
1
1.36
DEBAUCHE
Wavelet for
TUMOUR
ECG Signal
5
47
24
12
6
1
0.09
10
53
14
7
3
1
-
0.88
15
30
29
14
9
1
0.83
20
36
36
14
4
1
0.48
SYMLET
Wavelet for
EPILEPSY
ECG Signal
5
16
8
8
4
1
-0.9
10
33
18
9
3
1
0.92
15
35
9
4
2
1
0.54
20
18
18
9
5
1
0.96
DEBAUCHE
Wavelet for
EPILEPSY
ECG Signal
5
29
15
7
4
1
0.13
10
33
9
4
2
1
-
0.38
15
18
18
8
6
1
0.34
20
23
23
9
3
1
0.09
Neural
Network
Se
g
Se
gm
Se
gm
Se
gm
Seg
me
Wavelet based
Dimensional
Modified LED
protocol for
Diagnostic
Analysis
Doctor / Expert
Consultation
ECG
signal
Bio-Sensor’s
network
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2022.19.13
Dhananjaya V, Geetha M
E-ISSN: 2224-2902
111
Volume 19, 2022
In this proposed work, four sensors are placed on
the human body, one on head to acquired ECG
signals, second on the heart to acquired ECG
signals, third on the mouth to record the audio
signals and fourth one database fundus images.
These datasets are processed for cluster
formulation to find the cluster head and base
station of each signal as shown in Fig.4. Let that
A is the cluster for CL1, CL2, CL3,……. CLA
and their base stations are b1, b2, b3,…… bA.
The equation (1) states that k-means objective
function with base station bi, this can be update
the CLi using equation (2).
----(1)
------(2)
For determination of distance between sensors
and base stations is measured by using Euclidean
distance and it is given by equation (3)
---------------------------(3)
Fig.4. Proposed flow diagram of network creation
and session ID and data storage in Cloud
Fig.5. Simulation output towards the parameters of
modified LED protocol.
The Fig.5. Shows the transmission of packets and
uploading into cloud through MLED protocol and
measurements of delay, energy consumed per packet
and number packets lose during the process. The
results shows that, delay and losses in the proposed
work are optimized as compared to existing LEACH
protocol. Every packet data is compressed using
adaptive compressive sensing technique which uses
the best mother wavelet selection i.e Daubechies
approach with lessor PRD. The vanishing moment (P)
of Daubechies is various from 1 to 45, out of which,
one of the P will be selected based on lessor value of
PRD. The lessor PRD values gives the more
compression as shown in Fig. 8 [26, 27,29] and same
packet is transmitted through RTP-RTCP protocol and
its flow chart is depicted in the Fig.4.
3.3 RESULTS AND DISCUSSION
The complete design as per Fig.2. has been developed
in MATLAB 2017a for more than 1000 sample of
ECG of both normal and abnormalities and sample
taken from MIMIC-II standard from Physionet.org.
Each and every sample is tested with proposed
algorithm for effective feature extraction and
classifications, finally the results produced by
proposed system are compared with standard results
given in the physionet.org and it is found that 98% of
sample are matched with standard results.
The identified abnormalities are successfully uploaded
into cloud by making use of MLED which includes
RTP-RTCP protocols with lesser energy consumption
and higher throughput. All ECG signals along their
abnormalities are stored in cloud for remotely
monitoring by medical experts, the abnormalities
calculations are measured by using heart rate and the
detailed discussion as follows.
The six common diseases in ECG are measured from
heart rate which has been calculated based on R-R
Deployment of the
network nodes
Deployment of
the Sensors nodes
If new
network
has
creation
Exchange ID of the
each node
Authentication and
session communication
establishment
Agree
transmiss
ion
protocol
and
Transmit the
packet into cloud
Is it
packet
session
ids?
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2022.19.13
Dhananjaya V, Geetha M
E-ISSN: 2224-2902
112
Volume 19, 2022
interval in seconds or minutes
i.e
The bradycardia disease will occurs in ECG
signals when heart rate is less than 60bpm, the R
wave in ECG signal is called Ventricular
depolarization and it plays major role to detect
bradycardia disease, the time interval between
one R to another R has measured and as shown in
Fig.6(a).
Fig.6 (a) Detection of R-R wave peaks
Fig.6.(b).Empty GUI of proposed diagnosis tool
for effective classification of different diseases.
The complete design is developed in MATLAB
2017a in the form of GUI as shown in Fig.6 (b).
in which, there are different options for selection
of ECG from data base, analysis of different diseases
and their characteristics. The Fully Automated Online
Artifact Removal (FAOAR) algorithm has been used
to remove or reduce the artifacts which is influenced in
ECG signal during the acquisition from neurological
operation. FAOAR is able to remove the maximum
blinks and artifacts in ECG signals as shown in Fig.7.
Fig.7. Simulated results of pre-processing for artifacts
removal
Fig.8. Simulated results of Data compression sensing
techniques using Best Mother Wavelet
As observed in Fig.8, which depicts the representation
of normal condition of ECG signal with mother
wavelet which is majorly used for compression of data
during transmission. It is noted that the density of ECG
signals are normal no abnormalities can be noted
majorly information is found in the main lobe of the
mother wavelet when compared with the other two
disorders as shown in Fig.7, which is used to represent
Epilepsy and tumour condition respectively. It can be
observed clearly that the density of the ECG signal is
high for the Epilipsey disorder than the tumour signal
under consideration. The mother wavelet output is the
normalized output of the main lobe the variability is
also depleted for the epilepsy whereas the tumour
condition retain the similar value of normalization.
The advantage of our proposed algorithm and system
ie., modified LED algorithm is disorders can be
identified noted by observing the ANN as shown in
Fig.9, 10, 11,12,13 & 14.
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2022.19.13
Dhananjaya V, Geetha M
E-ISSN: 2224-2902
113
Volume 19, 2022
Fig.9. Normal ECG signal and its heart rate, R-R
interval and QRS interval
Fig.10. Detection of Sinus tachycardia disease
and its heart rate, R-R interval and QRS interval.
Fig.11. Severe case detection for Ventricular
fibrillation disease and values of heart rate, R-R
interval and QRS interval are more the limit.
The diagnosis of different diseases are analysed
through the GUI and their results and
classifications are shown in Fig.9 to Fig.14
Fig.12. Moderate case detection in atrial
fibrillation disease and values of heart rate, R-R
interval and QRS interval are moderate values.
Fig.13. Mild case detection in sinus bradycardia
disease and values of heart rate, R-R interval and QRS
interval are moderate values
Fig.14. Moderate case detection in atrial flutter disease
and values of heart rate, R-R interval and QRS interval
are more than limit.
The atrial flutter disease occurs when heart rate is
between 250 to 350 bpm which is too fast that the P
wave are unable to identifiable.
In this paper, we have demonstrated to use multimedia
data of the patient data which is captured using
wireless capsule endoscopy images are processed
using protocols RTP and RTCP protocol. As our
application in the consideration is the data processing
using the wavelet and uploading onto cloud storage
Fig.15. Interfacing of multimedia datasets to cloud
through RTP-RTCP protocol by using thingSpeak.com
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2022.19.13
Dhananjaya V, Geetha M
E-ISSN: 2224-2902
114
Volume 19, 2022
Table 2: Comparative statement between previous
work and proposed work
The proposed work i.e RTP and RTCP protocol
for transmission of wireless capsule endoscopy
images, ECG, EEG and audio signals which are
called as multimedia data are successfully
simulated in MATLAB 2017a and tested for
WCE images and real time biomedical signals,
from the obtained results, it is concluded that
there is improvement for packet loss is 18.42%,
improvement in throughput is 15%, energy
efficiency is 3.02% and end to end delay is 4% as
shown in Table.1
Fig 16: Comparative graph of various ECG
signals and its frequency
Fig.17: various PRD values for different wavelets and
co-efficient
4 Conclusion
Patient monitoring and exact diagnosis of the disorder
is becoming an challenging requirement for the
existing healthcare system, meanwhile communication
of the patient information through an dependable
channel is more required without loss of any kind of
information and data which are essential for patient
analysis. Hence to meet the requirement of the today’s
world we have proposed a new system which is the
combination of the kind of wireless body sensor
network is termed as Health RAD system which
incorporates the modified LED protocol by calculating
the PRD values, thus provides the exact analysis of the
disorder. In this context we have discussed three
different conditions namely Normal ECG signal, ECG
with Epilepsy and EEG signal with tumour, with the
dependence of kind of ECG signal further it has
analysed keenly with the consideration if all the five
different types of ECG i.e., alpha, beta, gamma, delta
and theta signals, it is noticed that delta waves have
observed common frequency. For all the conditions,
the data before transmission are compressed using
mother wavelet technique for the better analysis of
compress data with essential PRD values.
We have examined with two different kind of wavelets
i.e., Symlet and Debauche, it is also pointed and
studied the effect of variability of coefficients in the
range of 5 to 20 in the interval of 5 units and the best
coefficient is 15 for symlet in tumour detection, 5 for
debauche wavelet. For epilepsy EEG signal coefficient
15 sample well in terms of symlet and coefficient 20
meets good for the debauche wavelet.
0
20
40
60
Frequency in Hz
ECG Waves
Comparitive Graph of ECG
Co-efficient
GAMMA
BETA
ALPHA
THETA
DELTA
-2
-1
0
1
2
normalized values of
PRD
PRD values
PRD
Parameter
Previous
work
Proposed work
Packet loss
37.42%
19%
Throughput
230Mb
1.0744e+03
End to End
delay
200ms
196ms
Delay Jitter
0.02s
0.0981s
Energy
Efficiency
18.35%
21.33
Average
energy
dissipation for
all rounds
NA
0.02004Jouls
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2022.19.13
Dhananjaya V, Geetha M
E-ISSN: 2224-2902
115
Volume 19, 2022
For data transmission using packet switching
network is employed in terms of nodes, packets
sent, packet dropped, propagation delay with its
energy it is also observed that proposed method is
best among the existing method and the
comparative graph are plotted. The interesting
factor of our consideration is the heart rate and
Spo2 are constant in all the conditions under
consideration.
5. Future Enhancement
Monitoring and accurate diagnosis of the disorder
can be improved by adding video feature to the
proposed work to meet the challenging
requirement for the existing healthcare system.
References:
[1]. C. Gentry, “A fully homomorphic encryption
scheme”, Ph.D Thesis, Stanford University,
2009
[2]. C. Gentry, “Fully homomorphic encryption
using ideal lattices”, In STOC, pp 169-178,
ACM, 2009.
[3]. N. P. Smart, F. Vercauteren, “Fully
homomorphic encryption with relatively small
key and ciphertext sizes” In Public Key
Cryptography - PKC'10, Vol. 6056 of LNCS,
pp. 420-443,Springer, 2010
[4]. M. V. Dijk, C. Gentry, S. Halevi, V.
Vaikuntanathan, “Fully homomorphic
encryption over the integers”, Proceedings of
Eurocrypt,
[5]. Vol. 6110 of LNCS, pp. 24-43, Springer, 2010.
[6]. C. Gentry, “Computing arbitrary functions of
encrypted data”, Communications of the ACM,
53(3), pp.97-105, 2010.
[7]. D. Stehlé, R. Steinfeld, “Faster fully
homomorphic encryption. ASIACRYPT’2010,
Vol. 6477 of LNCS, pp.377-394, Springer, 2010
[8]. N. Ogura, G. Yamamoto, T. Kobayashi, S.
Uchiyama, “An improvement of key generation
algorithm for Gentry’s homomorphic
[9]. encryption scheme”, Advances in Information
and Computer Security - IWSEC 2010, Vol.
6434 of LNCS, pp. 70–83, Springer, 2010.
[10]. J. S. Coron, A. Mandal, D. Naccache and M.
Tibouchi, “Fully homomorphic encryption over the
integers with shorter public keys”, CRYPTO 2011, P.
Rogaway (Ed.), Vol. 6841 of LNCS, pp. 487-504,
Springer, 2011.
[11]. Z. Brakerski, V. Vaikuntanathan, “Efficient fully
homomorphic encryption from (standard) LWE”,
Electronic Colloquium on Computational Complexity
(ECCC) 18: 109, 2011.
[12]. Z. Brakerski, V. Vaikuntanathan, “Fully
homomorphic encryption from ring-LWE and
security for key dependent messages. CRYPTO 2011,
pp.505-524.
[13]. Z. Brakerski, C Gentry, V. Vaikuntanathan, “Fully
homomorphic encryption without bootstrapping”,
Electronic Colloquium on Computational Complexity
(ECCC) 18: 111, 2011.
[14]. P. Scholl, N.P. Smart, “Improved key generation for
Gentry’s fully homomorphic encryption Scheme”,
Cryptology ePrint Archive: Report 2011/471,
http://eprint.iacr.org/2011/471
[15]. Y Govinda Ramaiah.et.al, "Efficient Public key
Homomorphic Encryption Over Integer Plaintexts",
978-1-4673-2588-2/12, 2012 IEEE.
[16]. Naveen Ghorpade.et.al, "Towards Achieving
Efficient and Secure way to Share the Data", 2017
IEEE 7th International Advance Computing
Conference, 978-1-5090-1560-3/17, 2017 IEEE, DOI
10.1109/IACC.2017.10.
[17]. Hongchao Zhou.et.al, "Efficient Homomorphic
Encryption on Integer Vectors and Its Applications",
This work was supported in part by Draper
Laboratory through the UR&D Program and by
AFOSR under Grant No. FA9550-11-1-0183.
[18]. Tianying Xie.et.al, "Efficient Integer Vector
Homomorphic Encryption Using Deep Learning for
Neural Networks",Springer Nature Switzerland AG
2018, L. Cheng et al. (Eds.): ICONIP 2018, LNCS
11301, pp. 83–95, 2018. https://doi.org/10.1007/978-
3-030-04167-0_8.
[19]. Pramod Kumar Siddharth,et.al, "A Homomorphic
Encryption Scheme Over Integers Based on
Carmichael’s Theorem",2016 International
Conference on Electrical, Electronics,
Communication, Computer and Optimization
Techniques (ICEECCOT),978-1-5090-4697-
3/16,2016 IEEE.
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2022.19.13
Dhananjaya V, Geetha M
E-ISSN: 2224-2902
116
Volume 19, 2022
[20]. Pan Yang.et.al, "An Efficient Secret Key
Homomorphic Encryption Used in Image
Processing Service",Hindawi Security and
Communication Networks Volume 2017,
Article ID 7695751, 11 pages
https://doi.org/10.1155/2017/7695751.
[21]. Hao Chen.et.al, "Logistic regression over
encrypted data from fully homomorphic
encryption",BMC Medical Genomics 2018,
11(Suppl 4):81, 10.1186/s12920-018-0397-z.
[22]. XunWang, et.al, "A More Efficient Fully
Homomorphic Encryption Scheme Based on
GSW and DM Schemes", Hindawi Security and
Communication Networks Volume 2018,
Article ID 8706940, 14 pages
https://doi.org/10.1155/2018/8706940.
[23]. Roger A. Hallman.et.al, "Homomorphic
Encryption for Secure Computation on Big
Data",In Proceedings of the 3rd International
Conference on Internet of Things, Big Data and
Security (IoTBDS 2018), pages 340-347, ISBN:
978-989-758-296-7, Copyright © 2018 by
SCITEPRESS – Science and Technology
Publications.
[24]. Jheng-Hao.et.al, "Low-Complexity VLSI
Design of Large Integer Multipliers for Fully
Homomorphic Encryption", IEEE
TRANSACTIONS ON VERY LARGE SCALE
INTEGRATION (VLSI) SYSTEMS, VOL. 26,
NO. 9, SEPTEMBER 2018, 1063-8210,2018
IEEE.
[25]. Zekeriya Erkin.et.al, "Generating Private
Recommendations Efficiently Using
Homomorphic Encryption and Data Packing",
IEEE TRANSACTIONS ON INFORMATION
FORENSICS AND SECURITY, VOL. 7, NO.
3, JUNE 2012, 1556-6013, 2012 IEEE.
[26]. Mohanad Dawoud.et.al, "Privacy-Preserving
Search in Data Clouds Using Normalized
Homomorphic Encryption",Euro-Par 2014
Workshops, Part II, LNCS 8806, pp. 62–72,
2014.Springer International Publishing
Switzerland 2014.
[27]. G. Dinesh Kumar.et.al, An Efficient
Watermarking Technique for Biometric
Images", Procedia Computer Science 115
(2017) 423–430, 1877-0509, 2017 Published by
Elsevier B.V.
[28]. Dhananjaya, Balasubramani R, "Accelerating
Information Security in Cloud Computing using a
Novel Holomorphic Scheme", International Journal
of Innovative Technology and Exploring Engineering
(IJITEE) ISSN: 2278-3075, Volume-9 Issue-1,
November 2019.
[29]. Veer Amol Motinath.et.al, "A Novel ECG Data
Compression Algorithm using Best Mother Wavelet
Selection", 2016 Intl. Conference on Advances in
Computing, Communications and Informatics
(ICACCI), Sept. 21-24, 2016, Jaipur, India.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the Creative
Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en_US
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2022.19.13
Dhananjaya V, Geetha M
E-ISSN: 2224-2902
117
Volume 19, 2022
