Professional Documents
Culture Documents
net/publication/346487387
CITATIONS READS
0 230
4 authors, including:
Ameer Morad
University of Baghdad
16 PUBLICATIONS 47 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Analysis and Evaluation of Trnasport Protocols in Mobile Ad-Hoc Networks (MANET) View project
All content following this page was uploaded by Ameer Morad on 30 November 2020.
Abstract
The seizure epilepsy is risky because it happens randomly and leads to death in some cases. The standard epileptic
seizures monitoring system involves video/EEG (electro-encephalography), which bothers the patient, as EEG
electrodes are attached to the patient’s head.
Seriously, helping or alerting the patient before the seizure is one of the issue that attracts the researchers and
designers attention. So that there are spectrums of portable seizure detection systems available in markets which are
based on non-EEG signal.
The aim of this article is to provide a literature survey for the latest articles that cover many issues in the field of
designing portable real-time seizure detection that includes the use of multiple body signals, new algorithm methods,
and detection devices that are commercially available.
As a result, the reviewing process shows that there are many research articles that have covered wearable seizure
detection systems that based on body signals. The more effective monitoring and detection seizure system is the system
that uses multi-body signals, is highly comfortable and has low power consumption.
2. Epilepsy obvious. They could fall and harm their head, fall
into moving machinery or lose control of a car.
Epilepsy can be defined as a brain disorder
identified via an enduring pre-disposition for the
generation of epileptic seizures and via the 4. Seizure Types
physiological, neuro-biological, cognitive, and
social results of this disorder. Defining epilepsy Until the past few years, seizures were
needs the occurrence of a minimum of one categorized based on the international
epileptic seizure [4]. classification of seizures, which has been
This condition is the 4th most common suggested by the “International League against
neurologic disorder affecting people of different Epilepsy (ILAE) [10]. In 2017, ILAE announced a
ages [5]. There are about 65 million people new classification of seizures according to the
affected all around the world, with a high and onset of the seizure (i.e. focal or generalized), the
dramatic impact not only on the patient’s quality state of consciousness at the time of seizure
of life, but also on the professional development (complete or impaired awareness), and on the
and social behavior; the health system budget is symptoms that occur at the time (motor or
highly affected as well [6]. otherwise) [11].
Epilepsy may have an effect on people of any • Focal onset seizures (or partial seizures).This
age but it is most widespread among children and type of are more widespread than generalized
the elderly. A few kinds of epilepsy are inherited seizures and happen in one or more certain
and are caused by genetic factors. Other probable locations in the brain. In some of the situations,
epilepsy causes involve brain injuries because of they could spread to wider areas in the brain.
oxygen deprivation at birth or head traumas. In They typically happen as a result of some injuries,
several of the situations, the reason behind but in the majority of the cases the specific origins
epilepsy is simply not known [7]. are not known (idiopathic).
• Generalized onset seizures.
This type usually occurs in each of the brain sides.
3. Seizure Those seizures are combined with loss of
consciousness. Many types of those seizures are
A seizure is an unusual behavior (with genetically based. People typically have normal
symptoms or signs) that results from unusual neurologic functions between episodes.
discharges of cortical neurons. This phenomenon, • Unknown onset seizures.
which is observable, is finite in time. Epilepsy is a In some of the cases, the seizure onset happens
definition of a chronic state which is defined by unobserved and it is difficult to determine if it
repeated occurrences of seizures. A syndrome is a started out as focal or generalized.
set of symptoms and signs occurring together,
however, unlike disease, have no single known
cause or pathology [8]. 5. Seizure Detection Using Wearable
Seizures occur, in the majority of the patients, Biosensors
suddenly with no external symptoms. The sudden
occurrences of those seizure pose a serious In a clinical setting, Electro-Encephalo-Graphy
concern and accounts for mortality in patients that (EEG) in combination with video monitoring is
have uncontrolled epilepsy. Thus, timely the optimal standard for detecting and diagnosing
predicting seizures provides a sufficient time for different neuro-logical cases including epilepsy.
reducing risks of seizures and thereby improve the The capability for the automatic detection of
quality of life for the affected [9]. seizures is crucial for different diagnostic, safety,
A seizure could seem to happen in a and treatment causes. For example, the diagnosis
spontaneous way, with no external cause, even of a patient that has epilepsy needs capturing ictal
though in a great deal of patients a cause like a episodes with the use multi-channel EEG. Which
flashing light, redundant sounds or stressful cases could be time consuming and costly at the same
are easily characterized as triggers of seizure. time, and requires either a clinic stay or wearing
While seizures often happen during sleep, a ambulatory devices for a number of days [12].
seizure during the waking time could be a cause With the continuous need from the patients to
for the patient to harm themselves when they have a normal life while being observed for their
suddenly lose consciousness. The danger that this physiological activities, wearable biomedical
poses to the epileptic patient and others is
2
Alla Fikrat Alwindawi Al-Khwarizmi Engineering Journal, Vol. 16, No. 2, P.P. 1- 13 (2020)
sensors are playing a significant part. To achieve because of their stigma because of social
this, new adhesive tape type wearable sensors stigmatization. Therefore, it is necessary to find a
enable sensor integrations [13]. way to detect cases of epilepsy before using
There are several devices on the market which modern technology [16].
are designed for the detection of convulsive
seizures with the use of extra-cerebral (non-EEG)
signals. The majority of those devices use motion 6. Technologies Review
sensors due to the fact that convulsive movement 6.1 Researches Based on ECG
is identical amongst patients, and therefore readily
identifiable. On the other hand, there are no any
Most of epilepsy patients severed a Sudden
wearable devices available for the detection of
Unexpected Death (SUD) due to neurogenic
complicated partial seizures (CPS), as they are
cardiac arrhythmias, for that one study suggest
non-convulsive [14].
using Electrocardiograms (ECG) to monitor
Designing and developing of wearable bio-
implantable loop recorder [17-18]. The method
sensor systems for monitoring health has captured
showed effective automatic detection of
a great deal of focus in the industry as well as in
bradycardia and tachycardia.
scientific association in the past few years.
Van Elmpt et al. (2006) [19] held an
Primarily motivated by the increase in health-care
investigation which developed an algorithm based
expenses and powered by recent improvements in
on the analysis of heart rate pattern by using ECG.
the technology of miniature bio-sensing devices.
This method can determine the heart rate patterns
Such as wireless communications, smart textiles,
for the automatic detection of seizures in epilepsy
and micro-electronics. The constant development
patients.
concerning wearable sensor-based schemes will
Malarvili et al. (2009) [20] applied a method
have a great potential in transforming the future of
that contains a sequence of processes to monitor
health-care by allowing ubiquitous monitoring of
the epilepsy patients and used a supervised
a patient's’ health and permitting proactive
statistical classifier for HRV classification. The
management for personal health. Those systems
suggested method showed the sensitivity of HRV
may include different kinds of small physiologic
to the changes in the cardio regulatory system of
sensors, transmission modules, and processing
the heart, which helped detecting seizures in
abilities, and therefore is capable of facilitating
patients with epilepsy.
wearable low-cost unobtrusive results for constant
Recent development in miniaturizing ultra-low
day-round and anyplace health, also monitoring
power elements offer more intelligent wearable
activity and mental status wearable-systems for
health monitors. The evolvement and evaluation
the purpose of monitoring health could include
of a wireless wearable electrocardiogram (ECG)
different kinds of little sensors, implantable or
monitor for the detection of epileptic seizures
wearable. Those bio-sensors can measure
from variations in the cardiac rhythm is described.
important physiologic parameters such as hearts
The ECG data is analyzed with the use of
rate, oxygen saturation, blood pressure, the
embedded algorithms: a robust beat-detection
temperature of body and skin, rate of respiration,
algorithm in combination with a real-time
electrocardiogram, and others. The measurements
epileptic seizure detector [21].
that were gained are being communicated by a
An investigation has been held to test the
wired or wireless linked to the centralized node,
ability of algorithm based on Electrocardiographic
for instance, a personal digital assistant that is
(EKG) - based seizure detection [22]. The results
abbreviated as (PDA) or a micro-controller board
of EKG obtained a good quality that reached
that can afterward display the associated data on
(91%), which was acceptable and promising using
the user interface or transfers aggregated
these for further analysis.
important signs to the health center. The previous
Behbahani et al. (2016) [23] also developed an
depicts that a wearable system could include
algorithm based on heart rate analysis (HRV)
many different components such as: wireless
analysis by predicting when the seizures may
communication modules and links, user interface,
occur according to dynamical variations of ECG
smart textiles, power supplies, sensors, wearable
during the pre-ictal period. The presented method
materials, software, control and processing units,
showed a potential for monitoring epileptic
decision-making algorithms and advanced data
patients and improving their life quality. The
extraction [15].
general efficiency of the algorithm is an important
In most countries in the Middle East, epileptic
step for clinical implementations.
patients are feared to be known by the community
3
Alla Fikrat Alwindawi Al-Khwarizmi Engineering Journal, Vol. 16, No. 2, P.P. 1- 13 (2020)
The study of Shamim et al. (2016) [24] showed armband via a portable unit. The study used an
that using electrocardiogram (ECG) is one of the inertial measurement unit and muscular activity
best ways for detecting epilepsy seizures, because acquisition sensors and these signals were
it is comparatively easier compared to EEG. The transmitted from the used device.
results showed that the proposed algorithm Helmy & Helmy, (2015) [32] created the first
detected epileptic seizures efficiently and mobile application which was called Seizario,
compared to other models, it has been found to be This app offered a set of useful properties that
more effective. help to provide assistance to epilepsy patients.
Seizario app. are introduced a new accelerometer-
6.2. Researches based on ACM based learning algorithms with elaborate finite-
state-machines for the automatic detecting of
Whereas, Nijsen et al. (2005) [25] used the grand mal seizures and harmful falls. Seizario’s
visual analysis of 3-D accelerometer (ACM) and detecting algorithms were designed to operate
video/EEG recordings to predict epilepsy. The efficiently on mobiles and showed great potential
study results showed good results, 91% of the with more than 95% detection rate for each of
seizure with motor phenomena have been detected seizures and falls, with a minimum number of
by using recorded data with these sensors and false alarms.
considered feasible for detecting seizures. Whilst Ribeiro et al. (2016) [33] used the
Cuppens et al. (2009) [2] suggested another wearable device from ACM and embedded
detection systems applied with accelerometers sensors by applying machine learning algorithms
ACM that are attached to wrists and ankles. This obtained KNN 99% and C4.5 & PART 98% and
system used the development of an automatic there was no valuable variation at the results
detection algorithm. A detection system can solve between KNN, PART, and C4.5… An experiment
the unsuitable use problem of EEG in the daily showed that the user of wearable accelerometer
lives of patients, by using the development of an sensor was found very efficiently with a limited
automatic detection algorithm. This resulted in an number of false alarms (Tonpe et al., 2017) [34].
algorithm of sensitivity of up to 91.67% and Also Kusmakar et al. (2017) [35] used wearable
accuracy of up to83.92%. ACM sensors and their results showed that the
Another study used ACM with a different sensitivity was 23% with 0:72=24h.
algorithm based on a Bayesian approach and
found that the patient in standing up status 6.3. Researches based on mixed sensors
promotes a good detection close to 90% of (e.g. EDA, ACM, ECG, EMG, SpO2
seizures and 25% of false alarms [26]. video, audio skin, temperature and
Schulc et al. (2011) [27] used a Wii Remote BR)
(R) which encased an accumulator and Bluetooth
antenna and by using this device achieved 100% Poh M . (2011) [36] proposed an approach to
temporary sensitivity and specificity of over 88% monitor sympathetic nervous system activity
while a positive predictive value ≥75% of false during epileptic seizures with the use of a
alarm ratio. wearable sensor measuring electro-dermal activity
Another study used wireless ACM sensor with (EDA) and ACM. The sensitivity of the method
some physiological parameters, the false alarms reached 94% tonic-clonic seizures with low rate
were reduced with realizing a high level of of false alarms (<;1 per 24h).
detection [28]. Another study used a method to distinguish
Borujeny et al.(2013) [29] revealed the benefit between ordinary movement and hyper motor
of Wireless Sensor Networks (WSN) in the seizures [37]. The ACM was applied in both wrist
application of health monitoring to detect epilepsy and ankle to get the dataset with synchronized
seizures. This application can be used in a clinical video, audio, EEG, ECG, upper arm EMG and
environment or at the patient’s home. They used ACM data. The results were achieved 100%
ACM in the portable device. This study obtained sensitivity and Positive Predictive Values PPV
85% sensitivity with 3FP, while another study got which ranged from 34.91% to 100%.
a high sensitivity of up to 90% and a low rate of Other researchers monitored the human health
false alarms (0.2/day) which used the same device by observing some physiological signals such as
with a different algorithm [30]. electrocardiogram (ECG), respiration and
The study of Gheryani et al. (2017) [31] electromyography (EMG) signals. (Altini et al.,
examined the Wireless Body Area Network 2011) [38] promised to use these findings to
(WBAN) to detect and identify seizures by the
4
Alla Fikrat Alwindawi Al-Khwarizmi Engineering Journal, Vol. 16, No. 2, P.P. 1- 13 (2020)
monitor the epilepsy seizure. Becq et al. (2011) et al., 2015) [45]. A hierarchical classification
[39] used motogrogram to detect different types of method has been implemented for the detection of
epilepsy seizures. by collecting data from triaxis various types of epileptic seizures utilizing data
accelerometers and magnetometers. from wearable sensors (EDA and ACM). They
Conradsen et al. (2012) [30] applied an obtained a general sensitivity of up to 89.1% and
automatic multi-modal intelligent seizure a general accuracy of up to 93.1% was obtained
acquisition (MISA) system to detect motor with motor activity By using a k-nearest neighbor
seizures from electromyography data. Their (KNN). Whereas with no motor activity the
results showed that the MISA system has highly sensitivity reached a percentage of 97.1% and the
sensitive short detection latency and a low rate of accuracy reached 92.9%
false detection. Moreover, the results have shown Single seizure detecting methods have the
that the multimodal detecting system is superior problem of high false positive rate Cogan et
when compared with a uni-modal system. The al,.2015) [46]. A group of signals that may be
proposed system has a potential to detect seizures easily monitored by a wrist -worn device have
according to multi-modal data. been found and which produced a distinctive
A group of researchers developed a small pattern during the time of seizures for patients in
medallion which is attached to the patient's neck epilepsy monitoring unit (EMU) A group of five
with the use of a necklace type grip that consists bio-signals have been chosen, which may be
of combining the spectral analysis method of monitored at the wrist and are known or believed
anomalous currents with modern processors and to be affected by seizures they are: heart rate
mobile communications technology such as GPS (HR), arterial oxygenation (SpO2), accelero-
and Bluetooth [40]. The device disassembles and metry (ACC), electrodermal activity (EDA) and
amplifies only the primary brain currents, the ones temperature (TEMP). Several researchers have
of epileptic seizures, and warns the patient of notices great HR changes at the beginning of
possible danger The voice warning is used to some kinds of seizures; they needed to collect
connect the mobile to the patient with the use of a more data from epileptic patients in a day by day
specified program, and to send an emergency text basis, in addition to EMU settings prior to being
that the seizure reports are imminent and the confident of having found a seizure detection
current location with the use of GPS system for method beneficial for a wide range of people.
the emergency service, the doctor, and the Data collection is continuing, so we will have
National Center for Epilepsy Monitoring and opportunity to further evaluate our findings. We
Communication in advance. expect that more sophisticated methodologies
In the recent years, cloud computing in the mentioned in Subsection.
area of health-care started gaining importance. Sasikala et al. 2016 [47] proposed the design,
Pandey et al. (2012) [41] presented a model for development and validation of a wrist-based
on-line monitoring of the patient’s health with the wearable device that can not only detect seizures
use of cloud computing technology. Also, Internet efficiently, but over time can also predict a seizure
of Healthcare Things (IoHT) became very before the actual onset. The results concluded that
essential in field healthcare. Abdulaziz A. seizure detection would primarily involve the
Albesher (2019) [42] provides an overview of measurement of the following parameters: (i)
different aspects of IoT in healthcare. motion , (ii) electro-dermal activity (EDA), (iii)
Forkan et al. (2014) [43] presented a model skin temperature, and (iv) heart rate.
which is based on a service-oriented model A new multi criteria decision system (MCDS)
enabling real-time aided living services. It offers a was proposed by Ahmed et al. (2017) [48], they
flexible middle-ware layer hiding the complexity obtained a precision of 96% with 90% recall over
in managing sensor data from various types of the synthetic data. In this study, they used the
sensors and contextual information as well. parameters of Electro-cardiograph (ECG),
Fortino et al. (2014) [44] presented a model electrodermal activity (EDA), body motion and
which is based on combining the utilization of breathing rate (BR).
BSNs and cloud computing. It performs Different physical factors were applied to
monitoring of assisted living by wearable sensors transmit the signals of seizures to a cell phone
which transmit data to the cloud using a mobile. device Ramirez-Alaminos et al. (2017) [49], these
A study presented a method which considered physical factors are temperature, heart rate, and
the first step in the direction of a reliable extreme motion. The authors examined different
ambulatory monitoring system for epileptic temperature values, heartbeats and tested the
seizures with or without motor activities (Helderg battery life of the used system.
5
Alla Fikrat Alwindawi Al-Khwarizmi Engineering Journal, Vol. 16, No. 2, P.P. 1- 13 (2020)
Audio identification was applied on a group of way for inferring seizures and usual arbitrary
epilepsy patient in the Netherlands (De Bruijine at motions. The patient is not required to wear any
al., 2009) [50]. The used system was comprised of specific clothes, markers or sensors, therefore it is
three stages; this system depended on a theory of entirely non-intrusive.
Bayesian to classify the characteristics vector. Saleh M. Gaber et al. (2017) [52] depended on
A proposal of a non-intrusive video analysis the EEG Network on Chip (NoC) concept to
system for the patient’s body parts movement display and predict unstable brain waves. This
analysis in Epilepsy Monitoring Unit is present by prevents the risks of seizure during the driving
(Mandal et al.2012) [51]. The system utilized skin and childhood stage.
colour modeling, head/face posture template Table 1 provided a comprehensive summary of
matching and facial detection for the analysis and the state-of-the-art literature on epileptic seizure
quantification of the head motions. Epileptic detection including the methods and algorithms
patients’ heads have been analyzed in a holistic used and the results obtained.
Table 1.1,
The review of different method, Algorithm and results for the detection of seizures. ECG= Electrocardiography ,
EEG= Electroencephalography , HRV=Heart Rate Viability, ACM= Accelerometer , KNN= K-Near Neighbors,
EDA= Electro-dermal activity, EMG= Electromyography, SpO2=Arterial Oxygenation , TEMP= Temperature,
NoC=Network on Chip , FAR=False Alarm Rate , FDR= False Discovery Rate, FP= False Positive, PPV=
Positive Predictive Value, SNE= Sensitivity, SVDD=support vector data description , SVDD=support vector data
description, PPV= postivie predictive value , FDR= false detection rate , H= hours , Avg= Average, FPR= false
6
Alla Fikrat Alwindawi Al-Khwarizmi Engineering Journal, Vol. 16, No. 2, P.P. 1- 13 (2020)
7
Alla Fikrat Alwindawi Al-Khwarizmi Engineering Journal, Vol. 16, No. 2, P.P. 1- 13 (2020)
Fürbass et al. EEG ,ECG and Obtained 86% sensitivity with Avg
35 Automatic multimodal detection
(2017)[55] EMG of false detection 12.8 in 24h
Sopic et al. E-Glass using 4 EEG electrodes The Sensitivity 93.80% and 93.37%
36 ECG
(2018)[56] specificity
Results showed that sensitivity 77.6
Elger et al. Low Complex Novelty classifier
37 Heart Rate % and Avg of FAR 2.56 false alarms
(2018)[57] detection
per night
Obtained 97% sensitivity with
Geertsema et al. Analyzing of Oscillatory motion
38 Video median of false alarms 0.78 per
(2018)[58] pattern and set threshold
night
There are different technologies developed for capture seizure data. The table 1.2 shows the
tracking seizure information. The developed commercial device with their details.
devices and mobile applications will becomes a
tool to help patients, families, and clinicians
8
Alla Fikrat Alwindawi Al-Khwarizmi Engineering Journal, Vol. 16, No. 2, P.P. 1- 13 (2020)
Table 1.2
Review of different commercial devices
Commercial device Used apparatus Detection Method Website
Advance Brain
EEG with different NO. Wireless EEG www.advancebrainmonitring.com
Monitoring
of channels
ACM, bed motion
Emfit Wireless transmission www.emfit.com
sensor
Bio-sensor for measuring EDA www.affectiva.com
Affectiva Wrist-band
,ACM and body TEMP
Mobile Application Wireless transmission : sending
www.epdetect.com
EpDetect (base on ACM of SMS ,movement detection and
mobile) GPS position location
Respiration sensor and EDA
Smart belt Chest-belt ,wearable .
sensor
Raw signal of EDA , ACM www.movisens.com
Movisens Electrodes
TEMP and air pressure
Wireless alarm bell, wired
ACM , bed motion
Epi Watcher version integrated and message www.vahlkamp.nl
sensor
to responsible person
Mio Alpha
Watch Heart rate monitor www.alphaheartrate.com
Streapless
ADLX330 Wrist-band ACM www.sparkfun.com
Pulseguard Wrist-band Heart rate changing www.pulseguard.org
SensAlert200 Bed motion sensor Detecting unusual movement www.sensorium.co.uk
Detecting unusual movement
SAMI-sleep activity Video-Camera
and send alarms and video www.samialert.com
monitor Night vision monitoring
records
Video-Camera
www.babyping.com
Baby-Ping Night vision monitoring Wi-Fi connection
with voice
7. Discussion 8. References
The epilepsy seizure detection and alert system [1] Kurada, A., Srinivasan, T., Hammond, S.,
is very important and urgent for saving the Ulate-Campos, A. and Bidwell, J., 2019.
patient’s life. Many social organizations Seizure detection devices for use in antiseizure
encourage researchers to design such a system to medication clinical trials: A systematic review.
make a penitent’s life more safer and prevent case Seizure, 66, pp.61-69.
of Sudden Unexpected Death in Epilepsy [2] Ceulemans, B., Cuppens, K., Lagae, L., Van
(SUDEP). Huffel, S. and Vanrumste, B., 2009. P152
This review paper gives a useful overview of Detection of nocturnal frontal lobe seizures in
the spectrum of non-EEG body signals which are pediatric patients by means of accelerometers:
related to seizures. The review shows that these preliminary results. European Journal of
body signals (such as Electromyography (EMG), Paediatric Neurology, 13, p.S67.
Heart rate (HR), oxygen level (SpO2), [3] S. M. Klim and S. M. Klim, “Big-data
accelerometer (ACM) and eta) individually fail to Management using Map Reduce on Cloud:
detect seizures, so the effective seizure system is Case study, EEG Images Data,” Al-Khwarizmi
based on multi-body signals. Engineering Journal, vol. 13, no. 1, pp. 129–
In spite of if seizure alerts happen or not, these 137, 2017.
systems enhances the patient monitoring process [4] Fisher, R., Boas, W., Blume, W., Elger, C.,
by collecting the statistical date about the number Genton, P., Lee, P. and Engel, J., 2020.
of seizures during the period of time and the time Epileptic Seizures And Epilepsy: Definitions
between seizures. This information is very Proposed By The International League Against
important for the diagnosti purpose. Finally, the Epilepsy (ILAE) And The International Bureau
paper gives an up to date list of some available For Epilepsy (IBE).
seizures detection commercial devices.
9
Alla Fikrat Alwindawi Al-Khwarizmi Engineering Journal, Vol. 16, No. 2, P.P. 1- 13 (2020)
[5] Vergara, P., de la Cal, E., Villar, J., González, Cardiac arrhythmias in focal epilepsy: a
V. and Sedano, J., 2017. An IoT Platform for prospective long-term study. Lancet
Epilepsy Monitoring and Supervising. Journal (London, England). 18 December 2004.
of Sensors, 2017, pp.1-18. DOI:https://doi.org/10.1016/S01406736(04)1
[6] Pitkänen, A. and Sutula, T., 2002. Is epilepsy a 7594-6
progressive disorder? Prospects for new [17] Noor K Al-Qazzaz, Israa F. Abdulazez,
therapeutic approaches in temporal-lobe Salma A. Ridha, Simulation Recoding of an
epilepsy. The Lancet Neurology, 1(3), pp.173- ECG and PCG for Feature Extraction. Al-
181. Khwarizmi Engineering Journal Vol.10,
[7] Benbadis, S., 2001. Epileptic seizures and No.4, P.P. 81-91, 2014.
syndromes. Neurologic Clinics, 19(2), pp.251- [18] Van Elmpt, W. J. C., Nijsen, T. M. E., Griep,
270. P. A. M., & Arends, J. B. A. M., A model of
[8] Hashimoto, H., Fujiwara, K., Suzuki, Y., heart rate changes to detect seizures in severe
Miyajima, M., Yamakawa, T., Kano, M., epilepsy. Seizure, 15(6), pp.366–375, 2006.
Maehara, T., Ohta, K., Sasano, T., Matsuura, [19] Malarvili, M. B., & Mesbah, M., Newborn
M. and Matsushima, E., 2013. Epileptic Seizure Detection Based on Heart Rate
Seizure Monitoring by Using Multivariate Variability. IEEE Transactions on
Statistical Process Control. IFAC Proceedings Biomedical Engineering, 56(11), pp.2594–
Volumes, 46(31), pp.249-254. (pp. 1–4), IEEE, 2603, 2009.
2013. [20] F. Massé, M. Bussel, A. Serteyn, J. Arends,
[9] Engel, J., 2006. ILAE classification of epilepsy & J. Penders, “Miniaturized wireless ECG
syndromes. Epilepsy Research, 70, pp.5-10. monitor for real-time detection of epileptic
[10] Fisher RS, e., 2020. Instruction Manual For seizures,” ACM Transactions on Embedded
The ILAE 2017 Operational Classification Computing Systems, 12(4), pp. 111-117,
Of Seizure Types. - Pubmed - NCBI. [online] 2013.
Ncbi.nlm.nih.gov. Available at: DOI: 10.1145/1921081.1921095
<https://www.ncbi.nlm.nih.gov/pubmed/282 [21] I. Osorio, “Automated Seizure Detection
76064> [Accessed 18 May 2020].. Using Ekg,” International Journal of Neural
[11] Page, A., Shea, C., & Mohsenin, T. Systems, 24(02), 1450001, 2014.
,Wearable seizure detection using https://doi.org/10.1142/S0129065714500014
convolutional neural networks with transfer [22] S. Behbahani, N. J. Dabanloo, A. M.
learning. In 2016 IEEE International Nasrabadi, & A. Dourado, “Prediction of
Symposium on Circuits and Systems epileptic seizures based on heart rate
(ISCAS), pp. 1086–1089, IEEE, 2016 variability,” Technology and Health Care :
[12] Islam, T., Advanced Interfacing Techniques Official Journal of the European Society for
for the Capacitive Sensors ,pp. 73–109, Engineering and Medicine. 14;24(6), pp.795-
Springer, Cham, 2017. 810, Nov 2016.
[13] [Cogan, D., Nourani, M., Harvey, J., & [23] G. Shamim, Y. U. Khan, M. Sarfraz, & O.
Nagaraddi, V., Epileptic seizure detection Farooq, “Epileptic seizure detection using
using wristworn biosensors. In 2015 37th heart rate variability,”. In 2016 International
Annual International Conference of the IEEE Conference on Signal Processing and
Engineering in Medicine and Biology Communication (ICSC). 26-28 Dec. 2016.
Society (EMBC), pp. 5086–5089, IEEE, DOI:10.1109/ICSPCom.2016.7980585
2015. [24] T.M.E Nijsen, J. B.A. M. Arends, P. A.M.
[14] Pantelopoulos, A., & Bourbakis, N. G., A Griep, & P.J. M. Cluitmans, “The potential
Survey on Wearable Sensor-Based Systems value of three-dimensional accelerometry for
for Health Monitoring and Prognosis. IEEE detection of motor seizures in severe
Transactions on Systems, Man, and epilepsy,” Epilepsy and Behavior,7(1),
Cybernetics, Part C (Applications and pp.74-84, Aug 2005.
Reviews), 40(1), pp.1–12, 2010.. [25] P. Jallon, “ A Bayesian approach for epileptic
[15] Baker GA, Jacoby A, Gorry J, et al. Quality seizures detection with 3D accelerometers
of life of people with epilepsy in Iran, the sensors,”. In 2010 Annual International
Gulf, and Near East. Epilepsia, 46(1), Conference of the IEEE Engineering in
pp.132–140, 2005. Medicine and Biology pp.6325–6328. IEEE,
[16] Rugg-Gunn, F. J., Simister, R. J., Squirrell, 2010
M., Holdright, D. R., & Duncan, J. S.,
10
Alla Fikrat Alwindawi Al-Khwarizmi Engineering Journal, Vol. 16, No. 2, P.P. 1- 13 (2020)
12
(2020) 13 -1 ! ،2 ا د،16 ا ا ارز ا اوي ت ا ا ء
!, ا
عFR8ت اAJ;X مAqX oSpDZ .تQA`8 اn<J Lm ةAm;8 اj8دي إiZ 5W وLe;اfg وث5d b>VJ FطAOS8AJ ] @`_;فX^J عFR8 ا0J;X فF<Z
.nZFS8أس اFJ EEG بA?W أsJ رUDZ x1d ، nZFS28 0>V68AJ vgw@ F@ وھ; أ، (EEG) غA@58 اs1?OG / ;Z51m ام5OD4AJ 0>WاFS28 ي512rD8ا
€4• وا1 طFm;DZ .o1SSRS8 واo1~dA>8ه اA>DX_| ا2Z LD8{ت اYfS8 ا5d أL وھ0J;68وث ا5d z>W ]71>6G وnZFS8ة ا5gAV@ ورةFp8 اo@و
EEG. رةAƒ إF1„ راتAƒ إj8 إA7S1SRG Lm 56DVG LD8;اق وا4‚ اLm 0dADS8 ا08;S`S8ت اAJ;68 اog •fY8 ا0SqX أo@
og •fY8ز اA7B U1SRG لAˆ@ Lm AZApr8 اo@ 5Z5<8 اL?IG LD8ة اF1‡‚ت اQArS8اض اF<D4Q 04 دراF1m;G ; ھ08ArS8 ھ…ه اo@ ف578ا
.ً AZرAˆG حAD@ 51B •f3 مAqX و، 01@;ارزO8 اb18A4 وأ، A74A1W oYSZ LD8 اUVˆ8رات اAƒ إz~@ 08;S`S8 اLr1r`8| اW;8 اLm 0J;X
56DVG LD8 واAاؤھ5G ارoYSZ تAJ;X فAfD3 ا0SqX أL?IG LD8 ا01~`>8ت اQArS8 اo@ 5Z5<8ك اA6اض أن ھF<D4‚ ا012Sg Œ;ﺿG ، ‹8…8 0ˆ1DX
.0WA?8 ا0p_O6@ ، 0`ZF@ ، مAVB‚دة ا5<D@ راتAƒم إ5ODVZ …ي8م اAq68 اL ھ018A<m فAfD3Q وا0>WاFS8 ا0SqX أF~3 أ.UVˆ8رات اAƒ إj8إ
13