en Computational Neuroscience Computational Neuroscience Original Original <div style="text-align: justify;"><span style="font-size:14px;"><span style="font-family:Tahoma;"><span style="line-height:2;"><b>Introduction:</b> Epileptic seizure prediction is an essential clinical goal, as unpredictable seizures severely affect patients&rsquo; safety and quality of life. Although the electroencephalogram (EEG) has traditionally been used for seizure prediction, unimodal approaches often suffer from high false-positive rates. Recent evidence suggests that electrocardiogram (ECG) features may provide complementary information. This study aimed to design and evaluate a multimodal deep learning framework that integrates EEG and ECG signals for improved seizure prediction.<br> <b>Methods:</b> Thirty patients with drug-resistant epilepsy underwent simultaneous EEG and ECG monitoring. Signals were preprocessed using band-pass filtering, independent component analysis, and normalization. Statistical, spectral, and nonlinear features were extracted, and a hybrid Convolutional Neural Network&ndash;Long Short-Term Memory (CNN&ndash;LSTM) model was trained using stratified 4-fold cross-validation. The classification task was defined as binary discrimination between preictal (seizure-predictive) and interictal (non-seizure or baseline) states, allowing the model to identify physiological changes preceding seizure onset. Performance metrics included accuracy, sensitivity, specificity, false-positive rate, and area under the curve (AUC).<br> <b>Results:</b> The multimodal CNN&ndash;LSTM system achieved 92.6 &plusmn; 0.5% accuracy, 90.4 &plusmn; 0.5% sensitivity, 94.1 &plusmn; 0.3% specificity, and an AUC of 0.95 &plusmn; 0.01, with a false-positive rate of 7.2 &plusmn; 0.8%. These results significantly outperformed unimodal EEG-only and ECG-only models, which demonstrated accuracies of 84.3% and 77.5%, respectively. Statistical analysis confirmed the superiority of the multimodal approach, F(2, 6) = 8.73, p = .006.<br> <b>Conclusion:</b> Integrating EEG and ECG features in a multimodal CNN&ndash;LSTM framework enhances seizure prediction accuracy and reduces false alarms compared with unimodal models. The findings underscore the translational potential of multimodal deep learning for real-time early-warning systems and wearable applications in epilepsy management.</span></span></span></div> CNN–LSTM model, EEG–ECG integration, multimodal deep learning, seizure prediction, preictal state 0 0 http://bcn.iums.ac.ir/browse.php?a_code=A-10-8362-1&slc_lang=en&sid=1 Jabbar Khazaal Jabbar Al-Bkhaitawi gabarman579@gmail.com 13700319475328460057940 13700319475328460057940 No Department of Biomedical Engineering, SR. C., Islamic Azad University, Tehran, Iran Keivan Maghooli Keivan_maghooli@iau.ac.ir 13700319475328460057941 13700319475328460057941 Yes Department of Biomedical Engineering, SR. C., Islamic Azad University, Tehran, Iran Ali Sheikhani sheikhani al 81@srbiau.ac.ir 13700319475328460057942 13700319475328460057942 No Department of Biomedical Engineering, SR. C., Islamic Azad University, Tehran, Iran Nader Jafarnia Dabanloo jafarnia@srbiau.ac.ir 13700319475328460057943 13700319475328460057943 No Department of Biomedical Engineering, SR. C., Islamic Azad University, Tehran, Iran