Follow the papers below.

Acharya, U. R., Oh, S. L., Hagiwara, Y., Tan, J. H., & Adeli, H. (2018). Deep convolutional neural network for the automated detection and diagnosis of seizure using EEG signals. Computers in biology and medicine, 100, 270-278.

Jukić, S., & Kevrić, J. (2018). PARALLEL PROCESSING OF EEG SIGNALS OF PEDIATRIC PATIENTS FOR EPILEPTIC SEIZURE PREDICTION. INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH, 7(5).

Tsiouris, Κ. Μ., Pezoulas, V. C., Zervakis, M., Konitsiotis, S., Koutsouris, D. D., & Fotiadis, D. I. (2018). A Long Short-Term Memory deep learning network for the prediction of epileptic seizures using EEG signals. Computers in biology and medicine, 99, 24-37.

Measuring EEGs as a complex system using entropies is a very fruitful approach. It reveals hidden properties of the observed system without knowing it's internal structure and topology.

It can discriminate walking state, sleeping state, epilepsy and even deepness of narcosis.

It is a very useful approach used even in cardiology and ECGs.

1) Entropies for detection of epilepsy in EEG - ScienceDirect

https://www.sciencedirect.com › pii

https://www.sciencedirect.com/science/article/pii/S0169260705001525

2) Detection of epileptic seizure based on entropy analysis of short-term EEG - NCBI - NIH

https://www.ncbi.nlm.nih.gov › articles

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5854404/

3)

Entropies for detection of epilepsy in EEG - ACM Digital Library - Association for Computing Machinery

https://dl.acm.org › citation

https://dl.acm.org/citation.cfm?id=1647790

4) Detection of epileptic seizure based on entropy analysis of short-term EEG - PLOS

https://journals.plos.org › plosone › article

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0193691

You can choose Non-Cooperative and Deceptive based emotion detection.

Or you can think about this topic

Epileptic Seizures Prediction Using Machine Learning Methods

Dear Alhassan Alharbi ,

Without any exaggeration seizure prediction is one of the most challenging scope in biological signal processing which is very hot to publish. For the first time it published two times in Nature in current year that illuminates its importance. One of them is a valuable review article published just last month and I can send it for you if you want. One of the most challenging in this problem mentioned is to predict the exact time of the seizure onset instead of determine the horizon. You can see more detail here:

https://www.researchgate.net/project/Detection-of-Focal-Epileptic-Seizure-Using-NIRS-Signal-Based-on-Discrete-Wavelet-Transform

It will be a priceless project,if you could do it. In addition, I have a rich database for it and have I lot with experience and some practical idea. So if you want to start your research in this field, it would be my honer to collaborate you.

If you have any question feel free to ask me.

Good luck