Comparing Additive and Multiplicative Attention in CNN–LSTM for Ictal Detection
Abstract views: 25
/ 
PDF downloads: 30
Keywords:
EEG, Ictal Detection, Seizure, LSTM, CNN-LSTM, Bahdanau Attention, Luong AttentionAbstract
This study comparatively analyzes the performance of six deep learning-based models in
epileptic seizure detection. In particular, the effects of additive (Bahdanau) and multiplicative (Luong)
attention mechanisms, integrated into the CNN-LSTM backbone, are evaluated in detail. EEG signals
were divided into 256 sample-length windows with a stride value of 256, and the training, validation, and
test sets were separated by 70, 15, and 15 percent, respectively. Z-score standardization was calculated
only with the training statistics and applied to the other subsets. The Bahdanau attention mechanism
added to the CNN-LSTM structure achieved the highest performance, with an area under the curve
(AUPRC) value of 0.9884, receiver operating characteristic (AUROC) value of 0.9876, and F1 score of
0.9436. The CNN-LSTM model, utilizing the Luong attention mechanism, achieved a similarly high level
of success with an AUPRC value of 0.9857 and an AUROC value of 0.9849. The results reveal that
attentional mechanisms enhance the model's ability to focus on critical temporal patterns in EEG signals,
thereby improving both sensitivity and overall accuracy in seizure detection. These findings suggest that
the addition of an attention mechanism to the LSTM and CNN-LSTM architecture offers a practical and
reliable improvement in terms of sensitivity-specificity trade-off, training time, and clinical applicability.
Downloads
References
N. Jiwani, K. Gupta, M. H. U. Sharif, N. Adhikari, and N. Afreen, “A LSTM-CNN Model for Epileptic Seizures Detection using EEG Signal,” in 2022 2nd International Conference on Emerging Smart Technologies and Applications (eSmarTA), IEEE, Oct. 2022, pp. 1–5. doi: 10.1109/eSmarTA56775.2022.9935403.
Z. Wang and X. Zhou, “Prediction of Epileptic Seizures Based on CNN-LSTM Network,” in 2022 2nd International Conference on Frontiers of Electronics, Information and Computation Technologies (ICFEICT), IEEE, Aug. 2022, pp. 131–135. doi: 10.1109/ICFEICT57213.2022.00031.
X. Cao, S. Zheng, J. Zhang, W. Chen, and G. Du, “A hybrid CNN-Bi-LSTM model with feature fusion for accurate epilepsy seizure detection,” BMC Med Inform Decis Mak, vol. 25, no. 1, p. 6, Jan. 2025, doi: 10.1186/s12911-024-02845-0.
H. Kashefi Amiri, M. Zarei, and M. R. Daliri, “Epileptic seizure detection from electroencephalogram signals based on 1D CNN-LSTM deep learning model using discrete wavelet transform,” Sci Rep, vol. 15, no. 1, p. 32820, Sep. 2025, doi: 10.1038/s41598-025-18479-9.
J. Su et al., “A model for epileptic EEG detection and recognition based on Multi-Attention mechanism and Spatiotemporal,” Sci Rep, vol. 15, no. 1, p. 31993, Aug. 2025, doi: 10.1038/s41598-025-17256-y.
T. Zhang, J. Chen, and K. Polat, “Enhanced epileptic seizure detection using CNNs with convolutional block attention and short-term memory networks,” Behavioural Brain Research, vol. 496, p. 115831, Jan. 2026, doi: 10.1016/j.bbr.2025.115831.
J. Li, Y. Hou, W. Li, R. Wu, and M. Liu, “Improved Arrhythmia Detection via CNN-LSTM with Attention Mechanism in Electrocardiogram Signals,” 2025, pp. 351–361. doi: 10.1007/978-981-96-2528-4_28.
F. Liu et al., “An Attention-based Hybrid LSTM-CNN Model for Arrhythmias Classification,” in 2019 International Joint Conference on Neural Networks (IJCNN), IEEE, Jul. 2019, pp. 1–8. doi: 10.1109/IJCNN.2019.8852037.
Y. Heryadi and H. L. H. S. Warnars, “Learning temporal representation of transaction amount for fraudulent transaction recognition using CNN, Stacked LSTM, and CNN-LSTM,” in 2017 IEEE International Conference on Cybernetics and Computational Intelligence (CyberneticsCom), IEEE, Nov. 2017, pp. 84–89. doi: 10.1109/CYBERNETICSCOM.2017.8311689.
E. Ali, M. Angelova, and C. Karmakar, “Epileptic seizure detection using CHB-MIT dataset: The overlooked perspectives,” R Soc Open Sci, vol. 11, no. 5, May 2024, doi: 10.1098/rsos.230601.
A. L. Goldberger et al., “PhysioBank, PhysioToolkit, and PhysioNet: Components of a new research resource for complex physiologic signals.,” Circulation, vol. 101, no. 23, Jun. 2000, doi: 10.1161/01.CIR.101.23.e215.
J. Guttag, “CHB-MIT Scalp EEG Database (version 1.0.0). PhysioNet. RRID:SCR_007345. https://doi.org/10.13026/C2K01R.0 Ali Shoeb. Application of Machine Learning to Epileptic Seizure Onset Detection and Treatment. PhD Thesis, Massachusetts Institute of Technology, September 2009.,” https://physionet.org/content/chbmit/1.0.0/, 2010.
S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Comput, vol. 9, no. 8, pp. 1735–1780, Nov. 1997, doi: 10.1162/neco.1997.9.8.1735.
D. Bahdanau, K. Cho, and Y. Bengio, “Neural Machine Translation by Jointly Learning to Align and Translate,” May 2016, [Online]. Available: http://arxiv.org/abs/1409.0473
M.-T. Luong, H. Pham, and C. D. Manning, “Effective Approaches to Attention-based Neural Machine Translation,” Sep. 2015, [Online]. Available: http://arxiv.org/abs/1508.04025
W. Qiao, X. Bi, L. Han, and Y. Zhang, “Epilepsy Prediction and Detection Using Attention-CssCDBN with Dual-Task Learning,” Sensors, vol. 25, no. 1, p. 51, Dec. 2024, doi: 10.3390/s25010051.
K. Yan et al., “Automated seizure detection in epilepsy using a novel dynamic temporal-spatial graph attention network,” Sci Rep, vol. 15, no. 1, p. 16392, May 2025, doi: 10.1038/s41598-025-01015-0.
W. Zhao et al., “Residual and bidirectional LSTM for epileptic seizure detection,” Front Comput Neurosci, vol. 18, Jun. 2024, doi: 10.3389/fncom.2024.1415967.
H. Torkey, S. Hashish, S. Souissi, E. E.-D. Hemdan, and A. Sayed, “Seizure Detection in Medical IoT: Hybrid CNN-LSTM-GRU Model with Data Balancing and XAI Integration,” Algorithms, vol. 18, no. 2, p. 77, Feb. 2025, doi: 10.3390/a18020077.
