EEG-FuseFormer: A Transformer-Driven Feature Fusion Framework for Seizure Onset Prediction
2026-06-01 • Machine Learning
Machine Learning
AI summaryⓘ
The authors developed a new method called EEG-FuseFormer to predict when seizures will start in epilepsy patients by analyzing brainwave data (EEG). Their approach combines features from two neural network models, one that reads raw EEG signals and another that looks at transformed EEG data, then merges these using a transformer encoder for better predictions. Tested on a well-known dataset, their model showed very high accuracy and worked well even when applied to different patients, especially after some fine-tuning. They also examined how fast their model runs on different hardware to understand its practical use.
EpilepsySeizure predictionEEG (Electroencephalogram)CNN-LSTMResNet-18Transformer encoderShort-Time Fourier Transform (STFT)Cross-patient validationFeature fusionComputational complexity
Authors
Vigneshwar Hariharan, Chithra Reghuvaran, Arlene John, Nhat Pham, Omer Rana, Deepu John, Ganesh Neelakanta Iyer
Abstract
Epilepsy is one of the most common neurological disorders globally, characterized by recurring seizures and significantly impacting the quality of life. Despite advancements in diagnostic techniques, the mitigation of risks faced by epilepsy patients remains challenging due to the unpredictability of seizure events. An accurate forecast of seizure onset helps to reduce risks in epilepsy patients. In this paper, we propose EEG-FuseFormer, a transformer-based feature fusion framework for seizure-onset prediction that combines intermediate features extracted from Convolutional Neural Networks-Long Short-Term Memory (CNN-LSTM) and ResNet-18 networks. The CNN-LSTM architecture captures both spatial and temporal features directly from the raw signal, whereas the ResNet-18 extracts features from the Short-Time Fourier Transform (STFT) representation of the EEG signals. Fusion is carried out using a transformer encoder, and the final prediction is generated using fully connected dense layers. The CHB-MIT dataset was used to validate the proposed model. The results show that the proposed model achieves a mean recall of 98.85% and outperforms most of the state-of-the-art methods. This study evaluates the ability of the proposed feature fusion model to generalize in cross-patient testing scenarios. Fine-tuning pre-trained models on limited target patient data (target adaptation) within the cross-patient validation framework results in higher recall, precision, and F1-score metrics in comparison to the conventional cross-patient validation approach. Finally, the runtime-based computational complexity of the model is assessed across diverse hardware platforms to highlight the performance-complexity trade-off.