A Review of Machine Learning Approaches for Epileptic Seizure Prediction

Selim, Sahar; Elhinamy, Ethar; Othman, Hisham; Abouelsaadat, Wael; Salem, Mohammed A.-M.

doi:10.1109/icces48960.2019.9068190

Cited by 18 publications

(7 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The interictal period is the duration belonging to neither the preictal nor the ictal period in the sequence of EEG signals. To date, epileptic seizure prediction research works have employed CNN and RNN models [ 44 , 45 ] for classifying the high-dimensional preictal and interictal EEG patterns in the spatial and temporal domains; however, the conversion of EEG signals into a Euclidean grid structure causes the results to suffer from a loss of adjacent spatial information. Hence, the proposed approach exploits the Graph Convolutional Neural Network (GCNN), following the use of a spiking encoder, which consumes minimal computational and storage resource across the channels after feature extraction.…”

Section: Proposed Epileptic Seizure Prediction Methodologymentioning

confidence: 99%

Patient-Specific Preictal Pattern-Aware Epileptic Seizure Prediction with Federated Learning

Saemaldahr

Ilyas

2023

Sensors

View full text Add to dashboard Cite

Electroencephalography (EEG) signals are the primary source for discriminating the preictal from the interictal stage, enabling early warnings before the seizure onset. Epileptic siezure prediction systems face significant challenges due to data scarcity, diversity, and privacy. This paper proposes a three-tier architecture for epileptic seizure prediction associated with the Federated Learning (FL) model, which is able to achieve enhanced capability by utilizing a significant number of seizure patterns from globally distributed patients while maintaining data privacy. The determination of the preictal state is influenced by global and local model-assisted decision making by modeling the two-level edge layer. The Spiking Encoder (SE), integrated with the Graph Convolutional Neural Network (Spiking-GCNN), works as the local model trained using a bi-timescale approach. Each local model utilizes the aggregated seizure knowledge obtained from the different medical centers through FL and determines the preictal probability in the coarse-grained personalization. The Adaptive Neuro-Fuzzy Inference System (ANFIS) is utilized in fine-grained personalization to recognize epileptic seizure patients by examining the outcomes of the FL model, heart rate variability features, and patient-specific clinical features. Thus, the proposed approach achieved 96.33% sensitivity and 96.14% specificity when tested on the CHB-MIT EEG dataset when modeling was performed using the bi-timescale approach and Spiking-GCNN-based epileptic pattern learning. Moreover, the adoption of federated learning greatly assists the proposed system, yielding a 96.28% higher accuracy as a result of addressing data scarcity.

show abstract

Section: Proposed Epileptic Seizure Prediction Methodologymentioning

confidence: 99%

Patient-Specific Preictal Pattern-Aware Epileptic Seizure Prediction with Federated Learning

Saemaldahr

Ilyas

2023

Sensors

View full text Add to dashboard Cite

show abstract

“…SVM [15][16][17][18], k-nearest neighbour (k-NN) [19], random forest (RF) [20] and linear discriminant analysis (LDA) [21,22] are the most widely used methods. These machine learning techniques were discovered to offer decent classification accuracy for seizure prediction [23]. Using the knowledge, they had learned during their medical training or professional experience, they automatically pulled features from their brains.…”

Section: Related Workmentioning

confidence: 99%

Epileptic seizure prediction based on Convolutional neural networks and optimization techniques

shanmugam,

NALLASAMY

2024

Preprint

View full text Add to dashboard Cite

According to the World Health Organization, fifty million individuals worldwide suffer from epilepsy, a neurological condition. While electroencephalography (EEG) is crucial for diagnosing epilepsy and monitoring the brain activity of epilepsy patients, it requires a specialist to examine all EEG recordings to find epileptic activity. The automatic recognition of epilepsy is becoming a crucial concern since manual detection is a tedious and time-consuming process that imposes a tremendous burden on neurologists. To overcome this issue, a seizure prediction technique is proposed based on convolutional neural networks (CNN) for feature extraction and an extreme gradient boosting (XGBoost) algorithm for seizure classification. CHB-MIT dataset is used for training and testing. Moreover , the proposed framework is optimized using a meta-heuristic algorithm called Aquila optimization for efficient and robust performance. The proposed method shows qualitative results achieving 99.02% of accuracy, 98.67% of specificity and 98.12% of sensitivity and the proposed method is compared with state-of-the-art methods.

show abstract

“…It includes extracting only some ruling EEG frequencies and analyzing only those to detect seizures. Other techniques used are: algorithm based on Short Time Fourier Transform [16], random-forest classification algorithm [17], Long short-term memory (LSTM) [18]and CNN [1], [8], [18], Naïve Bayes [15], [19], Support Vector Machine (SVM) [15], [19], Linear Discriminant Analysis (LDA) [19], variational modal decomposition (VMD) [20] and a deep forest (DF) [20] model, pyramidal one-dimensional convolutional neural network (P-1D-CNN) [21], 1D CNN-LSTM [22], decision tree (DT) [9], shallow artificial neural network (ANN) [9], [15], K-Nearest Neighbors (KNN) [15], and convolutional neural networks [1], [8].…”

Section: Figure 1phases Of Seizure and Their Description Is Given In ...mentioning

confidence: 99%

Machine Learning Model to Detect Seizures Using EEG Signals

Chandel

Kaur

Grover

et al. 2023

Preprint

View full text Add to dashboard Cite

Epilepsy is a disease of grave concern these days due to the negligence in its treatment in many parts of the world. Its detection and diagnose requires high skill, large amount of time and money. Thus, due to lack of treatment, epilepsy which can be diagnosed with simple epileptic drugs turn refractory. This can be avoided if it is detected at an early stage. Also, the data received after a patient undergo EEG is quite complex. Visualizing that data in an effective way and knowing important timestamps in a recorded EEG signal can help one save time and increase accuracy of detection. An automated system utilizing conventional machine learning is thus proposed in this study that uses features extracted from EEG signals. We have used a seizure detection model and visualized data and the result using various python libraries. Seizure detection is a model which is able to identify the presence of abnormal activities in the brain. Seizure prediction is a model which is able to predict in advance if he/she is going to face seizures in coming time by just studying the EEG signals of present state of that patient. Supervised Machine learning (random forest classifier) was employed to analyze recorded EEG signals for epilepsy detection. Data in the datasets was visualized using matplotlib. Classifier was visualized using Graphviz and pydot. Random forest model predicted epilepsy with a good accuracy of 96.87%, Sensitivity came out to be 98.4% and Specificity was 90.7%.

show abstract

A Review of Machine Learning Approaches for Epileptic Seizure Prediction

Cited by 18 publications

References 32 publications

Patient-Specific Preictal Pattern-Aware Epileptic Seizure Prediction with Federated Learning

Patient-Specific Preictal Pattern-Aware Epileptic Seizure Prediction with Federated Learning

Epileptic seizure prediction based on Convolutional neural networks and optimization techniques

Machine Learning Model to Detect Seizures Using EEG Signals

Contact Info

Product

Resources

About