EEG signal Detection

1. Electroencephalogram Data Analysed Through
the Lens of Machine Learning to Detect Signs
of Epilepsy
J.Rajalakshmi
Research scholar
Department of CSE
Sethu Institute of Technology
Dr.S.Siva Ranjani,
Professor,
Department of CSE,
Sethu Institute of Technology.
Dr G. Sugitha,
Professor,
Department CSE,
Muthayammal Engineering College
S C Prabanand,
Department of AI & DS,
Bannari Amman Institute of
Technology,

2. OBJECTVE
We create a multi-variate technique employing EEG data to
comprehend the functional connectivity of the brain at the sensor
level.
The temporal, periodic, & time-frequency domains were probed
using five different measurements to investigate eight distinct
connection properties.
The K-Nearest Neighbor method was used to evaluate the
solution
Both EG and HC groups demonstrated high levels of
classification accuracy (89%), although EG showed significant
spatial-temporal abnormalities in the front central regions at the
beta frequency band compared to HC.

3. OBJECTVE
Classification accuracy for EG and NEAD was only
approximately 79% because of the well-documented
comorbidity of NEAD and epileptic episodes.
This study suggests that seizure-free EEG data may be
used to reliably differentiate between those with HC and
those with specialized epilepsy.
Although more research is needed to develop a
diagnostic tool that might aid in treatment.

4. REVIEWS
Autocorrelation in the electroencephalogram was utilised
to calculate a measure of pacemaker cells by the authors
.Frequency-response seizure diagnosis relies on
distinguishing differences between normal and transient
EEGs in the frequency domain .
Time series complexity may be quantified in nonlinear ways
using metrics like correlation dimension (CD), maximal
Lyapunov exponent (LLE), and a posteriori error .
EEGs may usefully be characterised by this feature. Using
CD to characterise interictal EEG for seizure forecasting,
researchers found that CD scores obtained from epileptic
EEG data are substantially lower again for lesion than other
areas of the brain .
The past twenty years have seen a growth in the usage of
artificial neural networks (ANN) for the classification of
EEG data .

5. REVIEWS
To properly record seizures, a key indication of the severity
of epilepsy, seizure detection may be useful for people with
epilepsy.
This in turn may aid in differentiating between epileptic
seizures and other occurrences, such as psychogenic non-
epileptic seizures .
If seizures are identified and alarms are set off, SUDEP
(sudden unexpected death in epilepsy) may be avoided.
Automatic seizure detection is especially useful when dealing
with long-term digital data, such as subcutaneously implanted
electroencephalography (EEG), since it is impractical to manually
analyse such data.

6. Proposed Method
Pre-processing of Data
The objective of this research is to provide a graphical
representation of the functional connectivity of the brain between two
EEG signals x i & y i, and to do so throughout the time, frequency, and
time-frequency domains via the use of five unique methods.
These dimensions and their derivatives are then used to derive
classification features.
Mutual Information
Mutual information quantifies the degree to which two signals are
interdependent by the quantity of information sent from one to the
other (MI).
𝐺 𝑥 = 𝑖=1
𝑗
𝑝(𝑥𝑖) 𝑙𝑜𝑔 𝑝(𝑥𝑖)

7. Correlation
How two signals are related when one of them is moved through
time relative to the other.
In Eq. below, we find the formula used to calculate the cross-
correlation coefficients used to this study.
𝐻𝑥𝑦 𝑛 =
𝑥 𝑙 + 𝑦[𝑙 + 1]
∞
𝑗=−∞
𝑥 𝑙
∞
𝑗 =−∞
2
∗ 𝑦 𝑙 + 1
∞
𝑗 =−∞
2
Method
The time-frequency-power graphs used in the visualization were
generated using MATLAB's spectrum analyzer tool and a short-
time Fourier transform.
When looking at the time-domain representation of the EEG
without segmentation, we utilized a finite impulse response band
pass filter with a 1-70 Hz frequency range and a stop filter with a
50 Hz frequency range.

8. Proposed Method

9. 0
0.2
0.4
0.6
0.8
1
1.2
comorbid diseases and possible causes
The frequency with which patients with seizures
visit the ED, together with any other diseases.

10. Epilepsy patients' rates of ED
visits and hospitalizations.

11. Patient 18 had 11 focal tonic seizures
in a 19-hour period

12. ROC signal of the personalised
automated seizure detection system.

13. An SVM-based automated seizure identification process was
presented, and its performance was compared to that of a uni-modal
EEG method.
The proposed method used a number of multi-modal combinations
within ACCM, ECG, and EEG. By providing a framework for seizure
suspect epochs within the previously recorded data, features
generated from the electroencephalogram (EEG) and
electrocardiogram (ECG) may supplement human data interpretation
of recordings.
Conclusion

14. REFERENCES
[1] M. Pievani, et al., Functional network disruption in the degenerative dementias, Lancet Neurol. 10 (9)
(2011) 829–843, https://doi.org/10.1016/S1474-4422(11) 70158-2.
[2] F. Varela, et al., The brainweb: phase synchronization and large-scale integration, Nat. Rev. Neurosci. 2 (4)
(2001) 229–239, https://doi.org/10.1038/35067550.
[3] V. Brodbeck, et al., EEG microstates of wakefulness and NREM sleep, NeuroImage 62 (3) (2012) 2129–2139,
https://doi.org/10.1016/j.neuroimage.2012.05.060.
[4] G. Lioi, et al., Directional connectivity in the EEG is able to discriminate wakefulness from NREM sleep,
Physiol. Meas. 38 (9) (2017) 1802–1820, https://doi.org/10.1088/1361-6579/aa81b5.
[5] G.K. Dash, et al., Interictal regional paroxysmal fast activity on scalp EEG is common in patients with
underlying gliosis, Clin. Neurophysiol. 129 (5) (2018) 946–951, https://doi.org/10.1016/j.clinph.2018.02.007.
[6] R. Renzel, et al., Persistent generalized periodic discharges: a specific marker of fatal outcome in cerebral
hypoxia, Clin. Neurophysiol. 128 (1) (2017) 147–152, https://doi.org/10.1016/j.clinph.2016.10.091.
[7] H. Watanabe, et al., Effect of hyperventilation on seizures and EEG findings during routine EEG, Clin.
Neurophysiol. 129 (5) (2018) e38, https://doi.org/10.1016/j.clinph.2018.02.097.
[8]Arends J, Thijs RD, Gutter T, Ungureanu C, Cluitmans P, Van Dijk J, et al. Multimodal nocturnal seizure
detection in a residential care setting: A long-term prospective trial. Neurology 2018;91(21):e2010–9.

15. [9] Beck M, Simony C, Zibrandtsen I, Kjaer TW. Readiness among people with epilepsy to carry body-worn monitor
devices in everyday life: A qualitative study. Epilepsy Behav 2020;112:107390.
https://doi.org/10.1016/j.yebeh.2020.107390
.
[10]Beniczky S, Jeppesen J. Non-electroencephalography-based seizure detection. Curr Opin Neurol
2019;32(2):198–204.
[11]Beniczky S, Ryvlin P. Standards for testing and clinical validation of seizure detection devices. Epilepsia
2018;59:9–13.
[12]Beniczky S, Wiebe S, Jeppesen J, Tatum WO, Brazdil M, Wang Y, et al. Automated seizure detection using
wearable devices: A clinical practice guideline of the International League Against Epilepsy and the International
Federation of Clinical Neurophysiology. Epilepsia 2021;62(3):632–46.
[13]Boonyakitanont P, Lek-uthai A, Chomtho K, Songsiri J. A review of feature extraction and performance
evaluation in epileptic seizure detection using EEG. Biomed Signal Process Control 2020;57:101702.
https://doi.org/10.1016/j.bspc.2019.101702.
[14]Bouchequet P. rsleep: Analysis of Sleep Data. R package version 1.0.3. 2020. Available from: https://CRAN.R-
project.org/package=rsleep.
[15] Smys, S., and C. Vijesh Joe. "Big data business analytics as a strategic asset for health care industry." Journal of
ISMAC 1, no. 02 (2019): 92-100.
[16] Chen, J. I. Z., & Yeh, L. T. (2020). Data Forwarding in Wireless Body Area Networks. Journal of Electronics, 2(02),
80-87.
[17]K. R. Devi, S. Suganyadevi, S. Karthik and N. Ilayaraja, "Securing Medical Big data through Blockchain
technology," 2022 8th International Conference on Advanced Computing and Communication Systems (ICACCS),
2022, pp. 1602-1607, doi: 10.1109/ICACCS54159.2022.9785125.

16. [18] S. Suganyadevi, K. Renukadevi, K. Balasamy and P. Jeevitha, "Diabetic Retinopathy Detection Using Deep
Learning Methods," 2022 First International Conference on Electrical, Electronics, Information and
Communication Technologies (ICEEICT), 2022, pp. 1-6. https://doi.org/10.1109/ICEEICT53079.2022.9768544.
[19] Suganyadevi, S., Seethalakshmi, V. & Balasamy, K. A review on deep learning in medical image analysis. Int J
Multimed Info Retr (2021). https://doi.org/10.1007/s13735-021-00218-1
[20] Balasamy K, Suganyadevi S (2021) “A fuzzy based ROI selection for encryption and watermarking in medical
image using DWT and SVD” Multimed Tools Appl 80, 7167–7186, https://doi.org/10.1007/s11042-020-09981-5.
[21] Suganyadevi S, Shamia D, Balasamy K (2021) An IoT-based diet monitoring healthcare system for women.
Smart Healthc Syst Des Secur Priv Asp. https://doi.org/10.1002/9781119792253.ch8
[22] Balasamy, K., Krishnaraj, N. & Vijayalakshmi, K. Improving the security of medical image through neuro-fuzzy
based ROI selection for reliable transmission. Multimed Tools Appl 81, 14321–14337 (2022).
https://doi.org/10.1007/s11042-022-12367-4
[23] Balasamy, K., Krishnaraj, N. & Vijayalakshmi, K. An Adaptive Neuro-Fuzzy Based Region Selection and
Authenticating Medical Image Through Watermarking for Secure Communication. Wireless Pers Commun (2021).
https://doi.org/10.1007/s11277-021-09031-9
[24] Balasamy, K., Krishnaraj, N., Ramprasath, J., & Ramprakash, P. (2021). A secure framework for protecting
clinical data in medical IoT environment. Smart Healthcare System Design: Security and Privacy Aspects.
https://doi.org/10.1002/9781119792253.ch9

17. Thank You