Abstract
Artificial intelligence (AI) is rapidly transforming health care, and its applications in epilepsy have increased exponentially over the past decade. Integration of AI into epilepsy management promises to revolutionize the diagnosis and treatment of this complex disorder. However, translation of AI into neurology clinical practice has not yet been successful, emphasizing the need to consider progress to date and assess challenges and limitations of AI. In this Review, we provide an overview of AI applications that have been developed in epilepsy using a variety of data modalities: neuroimaging, electroencephalography, electronic health records, medical devices and multimodal data integration. For each, we consider potential applications, including seizure detection and prediction, seizure lateralization, localization of the seizure-onset zone and assessment for surgical or neurostimulation interventions, and review the performance of AI tools developed to date. We also discuss methodological considerations and challenges that must be addressed to successfully integrate AI into clinical practice. Our goal is to provide an overview of the current state of the field and provide guidance for leveraging AI in future to improve management of epilepsy.
Key points
-
Artificial intelligence (AI) has the potential to maximize the value of data collected throughout the management of epilepsy, including neuroimaging and electroencephalography data, electronic medical records, and data from medical devices.
-
Machine learning dominated early applications of AI in epilepsy, but deep learning approaches have become increasingly popular.
-
Despite development of many AI tools with potential for the diagnosis and management of epilepsy, few have been implemented in clinical practice.
-
Collaborative efforts, including sharing of data and expertise, among researchers and clinicians are essential to realize the full potential of AI in epilepsy management.
-
Methodological and ethical considerations are pivotal for integrating AI into routine epilepsy management.
-
Future advancements in AI require robust clinical trials and ethical frameworks to ensure efficacy and patient safety in epilepsy treatment.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 print issues and online access
$209.00 per year
only $17.42 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
References
Kwan, P. & Brodie, M. J. Early identification of refractory epilepsy. N. Engl. J. Med. 342, 314–319 (2000).
Lerner, J. T. et al. Assessment and surgical outcomes for mild type I and severe type II cortical dysplasia: a critical review and the UCLA experience. Epilepsia 50, 1310–1335 (2009).
Hong, S. J. et al. Automated detection of cortical dysplasia type II in MRI-negative epilepsy. Neurology 83, 48–55 (2014).
Ahmed, B. et al. Cortical feature analysis and machine learning improves detection of “MRI-negative” focal cortical dysplasia. Epilepsy Behav. 48, 21–28 (2015).
El Azami, M. et al. Detection of lesions underlying intractable epilepsy on T1-weighted MRI as an outlier detection problem. PLoS ONE 11, e0161498 (2016).
Alaverdyan, Z., Jung, J., Bouet, R. & Lartizien, C. Regularized siamese neural network for unsupervised outlier detection on brain multiparametric magnetic resonance imaging: application to epilepsy lesion screening. Med. Image Anal. 60, 101618 (2020).
Snyder, K. et al. Distinguishing type II focal cortical dysplasias from normal cortex: a novel normative modeling approach. NeuroImage Clin. 30, 102565 (2021).
Chen, H. H. et al. Cognitive and epilepsy outcomes after epilepsy surgery caused by focal cortical dysplasia in children: early intervention maybe better. Childs Nerv. Syst. 30, 1885–1895 (2014).
Adler, S. et al. Novel surface features for automated detection of focal cortical dysplasias in paediatric epilepsy. NeuroImage Clin. 14, 18–27 (2016).
Spitzer, H. et al. Interpretable surface-based detection of focal cortical dysplasias: a Multi-centre Epilepsy Lesion Detection study. Brain J. Neurol. 145, 3859–3871 (2022).
Wagstyl, K. et al. Atlas of lesion locations and postsurgical seizure freedom in focal cortical dysplasia: a MELD study. Epilepsia 63, 61–74 (2022).
Jin, B. et al. Automated detection of focal cortical dysplasia type II with surface-based magnetic resonance imaging postprocessing and machine learning. Epilepsia 59, 982–992 (2018).
Wagstyl, K. et al. Planning stereoelectroencephalography using automated lesion detection: retrospective feasibility study. Epilepsia 61, 1406–1416 (2020).
Gill, R. S. et al. Multicenter validation of a deep learning detection algorithm for focal cortical dysplasia. Neurology 97, e1571–e1582 (2021).
Chen, S. et al. Voxel-based morphometry analysis and machine learning based classification in pediatric mesial temporal lobe epilepsy with hippocampal sclerosis. Brain Imaging Behav. 14, 1945–1954 (2020).
Park, Y. W. et al. Radiomics features of hippocampal regions in magnetic resonance imaging can differentiate medial temporal lobe epilepsy patients from healthy controls. Sci. Rep. 10, 19567 (2020).
Mo, J. et al. Automated detection of hippocampal sclerosis using clinically empirical and radiomics features. Epilepsia 60, 2519–2529 (2019).
Gleichgerrcht, E. et al. Radiological identification of temporal lobe epilepsy using artificial intelligence: a feasibility study. Brain Commun. 4, fcab284 (2022).
Chang, A. J. et al. MRI-based deep learning can discriminate between temporal lobe epilepsy, Alzheimer’s disease, and healthy controls. Commun. Med. 3, 33 (2023).
Hosseini, M. P., Nazem-Zadeh, M. R., Mahmoudi, F., Ying, H. & Soltanian-Zadeh, H. Support Vector Machine with nonlinear-kernel optimization for lateralization of epileptogenic hippocampus in MR images. Annu. Int. Conf. IEEE. Eng. Med. Biol. Soc. 2014, 1047–1050 (2014).
Beheshti, I. et al. FLAIR-wise machine-learning classification and lateralization of MRI-negative 18F-FDG PET-positive temporal lobe epilepsy. Front. Neurol. 11, 580713 (2020).
Keihaninejad, S. et al. Classification and lateralization of temporal lobe epilepsies with and without hippocampal atrophy based on whole-brain automatic MRI segmentation. PLoS ONE 7, e33096 (2012).
Bennett, O. F. et al. Learning to see the invisible: a data-driven approach to finding the underlying patterns of abnormality in visually normal brain magnetic resonance images in patients with temporal lobe epilepsy. Epilepsia 60, 2499–2507 (2019).
Mahmoudi, F. et al. Data mining MR image features of select structures for lateralization of mesial temporal lobe epilepsy. PLoS ONE 13, e0199137 (2018).
Caldairou, B. et al. MRI-based machine learning prediction framework to lateralize hippocampal sclerosis in patients with temporal lobe epilepsy. Neurology 97, e1583–e1593 (2021).
Louis, S. et al. Hippocampal sclerosis detection with NeuroQuant compared with neuroradiologists. AJNR Am. J. Neuroradiol. 41, 591–597 (2020).
Hadar, P. N. et al. Clinical validation of automated hippocampal segmentation in temporal lobe epilepsy. NeuroImage Clin. 20, 1139–1147 (2018).
Rebsamen, M. et al. A quantitative imaging biomarker supporting radiological assessment of hippocampal sclerosis derived from deep learning-based segmentation of T1w-MRI. Front. Neurol. 13, 812432 (2022).
Pardoe, H. R. et al. High resolution automated labeling of the hippocampus and amygdala using a 3D convolutional neural network trained on whole brain 700 μm isotropic 7T MP2RAGE MRI. Hum. Brain Mapp. 42, 2089–2098 (2021).
Rudie, J. D., Colby, J. B. & Salamon, N. Machine learning classification of mesial temporal sclerosis in epilepsy patients. Epilepsy Res. 117, 63–69 (2015).
Kim, D., Lee, J., Moon, J. & Moon, T. Interpretable deep learning-based hippocampal sclerosis classification. Epilepsia Open 7, 747–757 (2022).
Bernhardt, B. C., Hong, S. J., Bernasconi, A. & Bernasconi, N. Magnetic resonance imaging pattern learning in temporal lobe epilepsy: classification and prognostics. Ann. Neurol. 77, 436–446 (2015).
Hong, S. J., Bernhardt, B., Schrader, D. S., Bernasconi, N. & Bernasconi, A. Whole-brain MRI phenotyping in dysplasia-related frontal lobe epilepsy. Neurology 86, 643–650 (2016).
Mo, J. et al. Neuroimaging phenotyping and assessment of structural-metabolic-electrophysiological alterations in the temporal neocortex of focal cortical dysplasia IIIa. J. Magn. Reson. Imaging JMRI 54, 925–935 (2021).
Lee, H. M. et al. Unsupervised machine learning reveals lesional variability in focal cortical dysplasia at mesoscopic scale. NeuroImage Clin. 28, 102438 (2020).
Tustison, N. J. et al. The ANTsX ecosystem for quantitative biological and medical imaging. Sci. Rep. 11, 9068 (2021).
Henschel, L. et al. FastSurfer – a fast and accurate deep learning based neuroimaging pipeline. NeuroImage 219, 117012 (2020).
Lucas, A. et al. iEEG‐recon: a fast and scalable pipeline for accurate reconstruction of intracranial electrodes and implantable devices. Epilepsia 65, 817–829 (2024).
Li, K. et al. Optimizing trajectories for cranial laser interstitial thermal therapy using computer-assisted planning: a machine learning approach. Neurotherapeutics 16, 182–191 (2019).
Pérez-García, F. et al. Simulation of brain resection for cavity segmentation using self-supervised and semi-supervised learning. In Medical Image Computing and Computer Assisted Intervention – MICCAI 2020 (eds Martel, A. L. et al.) 115–125 (Springer, 2020).
Arnold, T. C. et al. Deep learning-based automated segmentation of resection cavities on postsurgical epilepsy MRI. NeuroImage Clin. 36, 103154 (2022).
Pérez-García, F. et al. A self-supervised learning strategy for postoperative brain cavity segmentation simulating resections. Int. J. Comput. Assist. Radiol. Surg. 16, 1653–1661 (2021).
Sarikaya, I. PET studies in epilepsy. Am. J. Nucl. Med. Mol. Imaging 5, 416–430 (2015).
Kerr, W. T. et al. Computer-aided diagnosis and localization of lateralized temporal lobe epilepsy using interictal FDG-PET. Front. Neurol. 4, 31 (2013).
Beheshti, I. et al. Pattern analysis of glucose metabolic brain data for lateralization of MRI-negative temporal lobe epilepsy. Epilepsy Res. 167, 106474 (2020).
Zhang, Q. et al. A deep learning framework for 18F-FDG PET imaging diagnosis in pediatric patients with temporal lobe epilepsy. Eur. J. Nucl. Med. Mol. Imaging 48, 2476–2485 (2021).
Kini, L. G. et al. Quantitative [18]FDG PET asymmetry features predict long-term seizure recurrence in refractory epilepsy. Epilepsy Behav. 116, 107714 (2021).
Sinclair, B. et al. Machine learning approaches for imaging-based prognostication of the outcome of surgery for mesial temporal lobe epilepsy. Epilepsia 63, 1081–1092 (2022).
Iglesias, J. E. et al. SynthSR: a public AI tool to turn heterogeneous clinical brain scans into high-resolution T1-weighted images for 3D morphometry. Sci. Adv. 9, eadd3607 (2023).
Lucas, A. et al. Multi-contrast high-field quality image synthesis for portable low-field MRI using generative adversarial networks and paired data. Preprint at medRxiv https://doi.org/10.1101/2023.12.28.23300409 (2023).
Flaus, A. et al. PET image enhancement using artificial intelligence for better characterization of epilepsy lesions. Front. Med. 9, 1042706 (2022).
Binder, J. R. FMRI is a valid noninvasive alternative to Wada testing. Epilepsy Behav. 20, 214–222 (2011).
Janecek, J. K. et al. Language lateralization by fMRI and Wada testing in 229 epilepsy patients: rates and predictors of discordance. Epilepsia 54, 314–322 (2013).
Gazit, T. et al. Probabilistic machine learning for the evaluation of presurgical language dominance. J. Neurosurg. 125, 481–493 (2016).
Torlay, L., Perrone-Bertolotti, M., Thomas, E. & Baciu, M. Machine learning – XGBoost analysis of language networks to classify patients with epilepsy. Brain Inform. 4, 159–169 (2017).
Kramer, M. A. & Cash, S. S. Epilepsy as a disorder of cortical network organization. Neuroscientist 18, 360–372 (2012).
Pedersen, M., Curwood, E. K., Archer, J. S., Abbott, D. F. & Jackson, G. D. Brain regions with abnormal network properties in severe epilepsy of Lennox-Gastaut phenotype: multivariate analysis of task-free fMRI. Epilepsia 56, 1767–1773 (2015).
Bharath, R. D. et al. Machine learning identifies “rsfMRI epilepsy networks” in temporal lobe epilepsy. Eur. Radiol. 29, 3496–3505 (2019).
Dai, X. J., Liu, H., Yang, Y., Wang, Y. & Wan, F. Brain network excitatory/inhibitory imbalance is a biomarker for drug-naive Rolandic epilepsy: a radiomics strategy. Epilepsia 62, 2426–2438 (2021).
Hwang, G. et al. Using low-frequency oscillations to detect temporal lobe epilepsy with machine learning. Brain Connect. 9, 184–193 (2019).
Mazrooyisebdani, M. et al. Graph theory analysis of functional connectivity combined with machine learning approaches demonstrates widespread network differences and predicts clinical variables in temporal lobe epilepsy. Brain Connect. 10, 39–50 (2020).
Gholipour, T. et al. Common functional connectivity alterations in focal epilepsies identified by machine learning. Epilepsia 63, 629–640 (2022).
Hao, S., Yang, C., Li, Z. & Ren, J. Distinguishing patients with temporal lobe epilepsy from normal controls with the directed graph measures of resting-state fMRI. Seizure 96, 25–33 (2022).
Nguyen, R. D. et al. Convolutional neural networks for pediatric refractory epilepsy classification using resting-state functional magnetic resonance imaging. World Neurosurg. 149, e1112–e1122 (2021).
Chiang, S., Levin, H. S. & Haneef, Z. Computer-automated focus lateralization of temporal lobe epilepsy using fMRI. J. Magn. Reson. Imaging 41, 1689–1694 (2015).
Yang, Z., Choupan, J., Reutens, D. & Hocking, J. Lateralization of temporal lobe epilepsy based on resting-state functional magnetic resonance imaging and machine learning. Front. Neurol. 6, 184 (2015).
Fallahi, A. et al. Dynamic functional connectivity in temporal lobe epilepsy: a graph theoretical and machine learning approach. Neurol. Sci. 42, 2379–2390 (2021).
Hunyadi, B. et al. A prospective fMRI-based technique for localising the epileptogenic zone in presurgical evaluation of epilepsy. NeuroImage 113, 329–339 (2015).
Nandakumar, N., Hsu, D., Ahmed, R. & Venkataraman, A. DeepEZ: a graph convolutional network for automated epileptogenic zone localization from resting-state fMRI connectivity. IEEE Trans. Biomed. Eng. 70, 216–227 (2023).
He, X. et al. Presurgical thalamic “hubness” predicts surgical outcome in temporal lobe epilepsy. Neurology 88, 2285–2293 (2017).
Wang, X. et al. Graph-theory based degree centrality combined with machine learning algorithms can predict response to treatment with antiepileptic medications in children with epilepsy. J. Clin. Neurosci. 91, 276–282 (2021).
Garner, R. et al. A machine learning model to predict seizure susceptibility from resting-state fMRI connectivity. In 2019 Spring Simulation Conference https://doi.org/10.23919/springsim.2019.8732859 (IEEE, 2019).
Amarreh, I., Meyerand, M. E., Stafstrom, C., Hermann, B. P. & Birn, R. M. Individual classification of children with epilepsy using support vector machine with multiple indices of diffusion tensor imaging. NeuroImage Clin. 4, 757–764 (2014).
Focke, N. K. et al. Automated MR image classification in temporal lobe epilepsy. NeuroImage 59, 356–362 (2012).
Del Gaizo, J. et al. Using machine learning to classify temporal lobe epilepsy based on diffusion MRI. Brain Behav. 7, e00801 (2017).
Fang, P. et al. Mapping the convergent temporal epileptic network in left and right temporal lobe epilepsy. Neurosci. Lett. 639, 179–184 (2017).
Huang, J., Xu, J., Kang, L. & Zhang, T. Identifying epilepsy based on deep learning using DKI images. Front. Hum. Neurosci. 14, 590815 (2020).
An, J. et al. Decreased white matter integrity in mesial temporal lobe epilepsy: a machine learning approach. NeuroReport 25, 788 (2014).
Kamiya, K. et al. Machine learning of DTI structural brain connectomes for lateralization of temporal lobe epilepsy. Magn. Reson. Med. Sci. 15, 121–129 (2016).
Munsell, B. C. et al. Evaluation of machine learning algorithms for treatment outcome prediction in patients with epilepsy based on structural connectome data. NeuroImage 118, 219–230 (2015).
Taylor, P. N. et al. The impact of epilepsy surgery on the structural connectome and its relation to outcome. NeuroImage Clin. 18, 202–214 (2018).
Sinha, N. et al. Structural brain network abnormalities and the probability of seizure recurrence after epilepsy surgery. Neurology 96, e758–e771 (2021).
Gleichgerrcht, E. et al. Deep learning applied to whole-brain connectome to determine seizure control after epilepsy surgery. Epilepsia 59, 1643–1654 (2018).
Gleichgerrcht, E. et al. Temporal lobe epilepsy surgical outcomes can be inferred based on structural connectome hubs: a machine learning study. Ann. Neurol. 88, 970–983 (2020).
Revell, A. Y. et al. A framework for brain atlases: lessons from seizure dynamics. NeuroImage 254, 118986 (2022).
Munsell, B. C. et al. Relationship between neuronal network architecture and naming performance in temporal lobe epilepsy: a connectome based approach using machine learning. Brain Lang. 193, 45–57 (2019).
Jeong, J. W., Lee, M. H., O’Hara, N., Juhász, C. & Asano, E. Prediction of baseline expressive and receptive language function in children with focal epilepsy using diffusion tractography-based deep learning network. Epilepsy Behav. 117, 107909 (2021).
Peter Binding, L. et al. The impact of temporal lobe epilepsy surgery on picture naming and its relationship to network metric change. NeuroImage Clin. 38, 103444 (2023).
Lee, M.-H. et al. Novel deep learning network analysis of electrical stimulation mapping-driven diffusion MRI tractography to improve preoperative evaluation of pediatric epilepsy. IEEE Trans. Biomed. Eng. 67, 3151–3162 (2020).
Cantor-Rivera, D., Khan, A. R., Goubran, M., Mirsattari, S. M. & Peters, T. M. Detection of temporal lobe epilepsy using support vector machines in multi-parametric quantitative MR imaging. Comput. Med. Imaging Graph. 41, 14–28 (2015).
Huang, J. et al. Coherent pattern in multi-layer brain networks: application to epilepsy identification. IEEE J. Biomed. Health Inform. 24, 2609–2620 (2020).
Zhou, B. et al. Machine learning for detecting mesial temporal lobe epilepsy by structural and functional neuroimaging. Front. Med. 14, 630–641 (2020).
Pustina, D. et al. Predicting the laterality of temporal lobe epilepsy from PET, MRI, and DTI: a multimodal study. NeuroImage Clin. 9, 20–31 (2015).
Sisodiya, S. M. et al. The ENIGMA-Epilepsy working group: mapping disease from large data sets. Hum. Brain Mapp. 43, 113–128 (2020).
Gleichgerrcht, E. et al. Artificial intelligence for classification of temporal lobe epilepsy with ROI-level MRI data: a worldwide ENIGMA-Epilepsy study. NeuroImage Clin. 31, 102765 (2021).
Tang, Y. et al. Machine learning-derived multimodal neuroimaging of presurgical target area to predict individual’s seizure outcomes after epilepsy surgery. Front. Cell Dev. Biol. 9, 669795 (2021).
Lee, H. M. et al. Decomposing MRI phenotypic heterogeneity in epilepsy: a step towards personalized classification. Brain J. Neurol. 145, 897–908 (2022).
Lucas, A. et al. Mapping hippocampal and thalamic atrophy in epilepsy: a 7-T magnetic resonance imaging study. Epilepsia 65, 1092–1106 (2024).
Rasheed, K. et al. Machine learning for predicting epileptic seizures using EEG signals: a review. IEEE Rev. Biomed. Eng. 14, 139–155 (2021).
Siddiqui, M. K., Morales-Menendez, R., Huang, X. & Hussain, N. A review of epileptic seizure detection using machine learning classifiers. Brain Inform. 7, 5 (2020).
Usman, S. M. et al. Using scalp EEG and intracranial EEG signals for predicting epileptic seizures: review of available methodologies. Seizure 71, 258–269 (2019).
Miltiadous, A. et al. Machine learning algorithms for epilepsy detection based on published EEG databases: a systematic review. IEEE Access. 11, 564–594 (2023).
Mercier, M. R. et al. Advances in human intracranial electroencephalography research, guidelines and good practices. NeuroImage 260, 119438 (2022).
Litt, B. & Echauz, J. Prediction of epileptic seizures. Lancet Neurol. 1, 22–30 (2002).
Andrzejak, R. G. et al. Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: dependence on recording region and brain state. Phys. Rev. E 64, 061907 (2001).
Andrzejak, R. G., Schindler, K. & Rummel, C. Nonrandomness, nonlinear dependence, and nonstationarity of electroencephalographic recordings from epilepsy patients. Phys. Rev. E 86, 046206 (2012).
Klatt, J. et al. The EPILEPSIAE database: an extensive electroencephalography database of epilepsy patients. Epilepsia 53, 1669–1676 (2012).
Brinkmann, B. H. et al. Crowdsourcing reproducible seizure forecasting in human and canine epilepsy. Brain 139, 1713–1722 (2016).
Kuhlmann, L. et al. Epilepsyecosystem.org: crowd-sourcing reproducible seizure prediction with long-term human intracranial EEG. Brain 141, 2619–2630 (2018).
Baldassano, S. N. et al. Crowdsourcing seizure detection: algorithm development and validation on human implanted device recordings. Brain 140, 1680–1691 (2017).
Wong, S. et al. EEG datasets for seizure detection and prediction – a review. Epilepsia Open. 8, 252–267 (2023).
Mirowski P. W., LeCun, Y., Madhavan, D. & Kuzniecky, R. In 2008 IEEE Workshop on Machine Learning for Signal Processing (eds Principe, J. C, Erdogmus, D. & Adali, T) 244–249 (IEEE, 2008).
Park, Y., Luo, L., Parhi, K. K. & Netoff, T. Seizure prediction with spectral power of EEG using cost-sensitive support vector machines. Epilepsia 52, 1761–1770 (2011).
Wang, N. & Lyu, M. R. Extracting and selecting distinctive EEG features for efficient epileptic seizure prediction. IEEE J. Biomed. Health Inform. 19, 1648–1659 (2015).
Richman, J. S. & Moorman, J. R. Physiological time-series analysis using approximate entropy and sample entropy. Am. J. Physiol. Heart Circ. Physiol. 278, H2039–H2049 (2000).
Song, Y. & Zhang, J. Discriminating preictal and interictal brain states in intracranial EEG by sample entropy and extreme learning machine. J. Neurosci. Methods 257, 45–54 (2016).
Truong, N. D. et al. Convolutional neural networks for seizure prediction using intracranial and scalp electroencephalogram. Neural Netw. 105, 104–111 (2018).
Kiral-Kornek, I. et al. Epileptic seizure prediction using big data and deep learning: toward a mobile system. EBioMedicine 27, 103–111 (2018).
Chung, Y. G. et al. Deep convolutional neural network based interictal-preictal electroencephalography prediction: application to focal cortical dysplasia type-II. Front. Neurol. 11, 594679 (2020).
Eberlein, M. et al. In 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) (eds Griol, D., Zheng, H. & Schmidt, H.) 2577–2582 (IEEE, 2018).
Yamamoto, S. et al. Data-driven electrophysiological feature based on deep learning to detect epileptic seizures. J. Neural Eng. 2021, 18, https://doi.org/10.1088/1741-2552/ac23bf (2021).
Sundararajan M, Taly A, Yan Q. Axiomatic attribution for deep networks. In International Conference on Machine Learning 3319–3328 (PMLR, 2017).
Bartolomei, F., Chauvel, P. & Wendling, F. Epileptogenicity of brain structures in human temporal lobe epilepsy: a quantified study from intracerebral EEG. Brain 131, 1818–1830 (2008).
Wang, G. et al. Seizure prediction using directed transfer function and convolution neural network on intracranial EEG. IEEE Trans. Neural Syst. Rehabil. Eng. Publ. IEEE Eng. 28, 2711–2720 (2020).
Peng, P., Xie, L. & Wei, H. A deep Fourier neural network for seizure prediction using convolutional neural network and ratios of spectral power. Int. J. Neural Syst. 31, 2150022 (2021).
Wu, X., Zhang, T., Zhang, L. & Qiao, L. Epileptic seizure prediction using successive variational mode decomposition and transformers deep learning network. Front. Neurosci. 16, 982541 (2022).
Yu, Z. et al. Epileptic seizure prediction using deep neural networks via transfer learning and multi-feature fusion. Int. J. Neural Syst. 32, 2250032 (2022).
Boonyakitanont, P., Lek-uthai, A., Chomtho, K. & Songsiri, J. A review of feature extraction and performance evaluation in epileptic seizure detection using EEG. Biomed. Signal. Process. Control. 57, 101702 (2020).
Liu, G., Xiao, R., Xu, L. & Cai, J. Minireview of epilepsy detection techniques based on electroencephalogram signals. Front. Syst. Neurosci. 15, 685387 (2021).
Acharya, U. R., Vinitha Sree, S., Swapna, G., Martis, R. J. & Suri, J. S. Automated EEG analysis of epilepsy: a review. Knowl. Based Syst. 45, 147–165 (2013).
Ghosh-Dastidar, S. & Adeli, H. A new supervised learning algorithm for multiple spiking neural networks with application in epilepsy and seizure detection. Neural Netw. 22, 1419–1431 (2009).
Faust, O., Acharya, U. R., Min, L. C. & Sputh, B. H. C. Automatic identification of epileptic and background EEG signals using frequency domain parameters. Int. J. Neural Syst. 20, 159–176 (2010).
Kharbouch, A., Shoeb, A., Guttag, J. & Cash, S. S. An algorithm for seizure onset detection using intracranial EEG. Epilepsy Behav. 22, S29–S35 (2011).
Liu, Y., Zhou, W., Yuan, Q. & Chen, S. Automatic seizure detection using wavelet transform and SVM in long-term intracranial EEG. IEEE Trans. Neural Syst. Rehabil. Eng. Publ. IEEE Eng. 20, 749–755 (2012).
Xiang, J. et al. The detection of epileptic seizure signals based on fuzzy entropy. J. Neurosci. Methods 243, 18–25 (2015).
Zheng, Y. X., Zhu, J. M., Qi, Y., Zheng, X. X. & Zhang, J. M. An automatic patient-specific seizure onset detection method using intracranial electroencephalography. Neuromodulation 18, 79–84 (2015).
Manzouri, F., Heller, S., Dümpelmann, M., Woias, P. & Schulze-Bonhage, A. A comparison of machine learning classifiers for energy-efficient implementation of seizure detection. Front. Syst. Neurosci. 12, 43 (2018).
Ehrens, D., Cervenka, M. C., Bergey, G. K. & Jouny, C. C. Dynamic training of a novelty classifier algorithm for real-time detection of early seizure onset. Clin. Neurophysiol. 135, 85–95 (2022).
Cook, M. J. et al. Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: a first-in-man study. Lancet Neurol. 12, 563–571 (2013).
Acharya, U. R., Oh, S. L., Hagiwara, Y., Tan, J. H. & Adeli, H. Deep convolutional neural network for the automated detection and diagnosis of seizure using EEG signals. Comput. Biol. Med. 100, 270–278 (2018).
Zhou, M. et al. Epileptic seizure detection based on EEG signals and CNN. Front. Neuroinform. 12, 95 (2018).
Gómez, C. et al. Automatic seizure detection based on imaged-EEG signals through fully convolutional networks. Sci. Rep. 10, 21833 (2020).
Li, Z. et al. Deep learning of simultaneous intracranial and scalp EEG for prediction, detection, and lateralization of mesial temporal lobe seizures. Front. Neurol. 12, 705119 (2021).
Caffarini, J. et al. Engineering nonlinear epileptic biomarkers using deep learning and Benford’s law. Sci. Rep. 12, 5397 (2022).
Zanetti, R., Pale, U., Teijeiro, T. & Atienza, D. Approximate zero-crossing: a new interpretable, highly discriminative and low-complexity feature for EEG and iEEG seizure detection. J. Neural Eng. 18, 066018 (2022).
Revell, A. Y. et al. A taxonomy of seizure spread patterns, speed of spread, and associations with structural connectivity. Preprint at bioRxiv https://doi.org/10.1101/2022.10.24.513577 (2022).
Pattnaik, A. R. et al. The seizure severity score: a quantitative tool for comparing seizures and their response to therapy. J. Neural Eng. 20, https://doi.org/10.1088/1741-2552/aceca1 (2003).
Revell, A.Y. et al. White matter signals reflect information transmission between brain regions during seizures. Preprint at bioRxiv https://doi.org/10.1101/2021.09.15.460549 (2022).
Chen, D., Wan, S. & Bao, F. S. Epileptic focus localization using discrete wavelet transform based on interictal intracranial EEG. IEEE Trans. Neural Syst. Rehabil. Eng. 25, 413–425 (2017).
Grinenko, O. et al. A fingerprint of the epileptogenic zone in human epilepsies. Brain J. Neurol. 141, 117–131 (2018).
Varatharajah, Y. et al. Integrating artificial intelligence with real-time intracranial EEG monitoring to automate interictal identification of seizure onset zones in focal epilepsy. J. Neural Eng. 15, 046035 (2018).
Cimbalnik, J. et al. Multi-feature localization of epileptic foci from interictal, intracranial EEG. Clin. Neurophysiol. 130, 1945–1953 (2019).
Klimes, P. et al. NREM sleep is the state of vigilance that best identifies the epileptogenic zone in the interictal electroencephalogram. Epilepsia 60, 2404–2415 (2019).
Conrad, E. C. et al. Spike patterns surrounding sleep and seizures localize the seizure-onset zone in focal epilepsy. Epilepsia 64, 754–768 (2023).
Zhao, X., Sole-Casals, J., Sugano, H. & Tanaka, T. Seizure onset zone classification based on imbalanced iEEG with data augmentation. J. Neural Eng. 19, 065001 (2022).
Rao, V. R. & Lowenstein, D. H. Epilepsy. Curr. Biol. 25, R742–R746 (2015).
Antoniades, A. et al. Detection of interictal discharges with convolutional neural networks using discrete ordered multichannel intracranial EEG. IEEE Trans. Neural Syst. Rehabil. Eng. 25, 2285–2294 (2017).
Abou Jaoude, M. et al. Detection of mesial temporal lobe epileptiform discharges on intracranial electrodes using deep learning. Clin. Neurophysiol. 131, 133–141 (2020).
Quon, R. J. et al. AiED: artificial intelligence for the detection of intracranial interictal epileptiform discharges. Clin. Neurophysiol. 133, 1–8 (2022).
Zhang, Y. et al. Refining epileptogenic high-frequency oscillations using deep learning: a reverse engineering approach. Brain Commun. 4, fcab267 (2022).
Zhang, Y. et al. Characterizing physiological high-frequency oscillations using deep learning. J. Neural. Eng. 19, 066027 (2022).
Medvedev, A. V., Agoureeva, G. I. & Murro, A. M. A long short-term memory neural network for the detection of epileptiform spikes and high frequency oscillations. Sci. Rep. 9, 19374 (2019).
Geng, D. et al. Deep learning for robust detection of interictal epileptiform discharges. J. Neural Eng. 18, 056015 (2021).
Baud, M. O. et al. Unsupervised learning of spatiotemporal interictal discharges in focal epilepsy. Neurosurgery 83, 683–691 (2018).
Charupanit, K., Sen-Gupta, I., Lin, J. J. & Lopour, B. A. Detection of anomalous high-frequency events in human intracranial EEG. Epilepsia Open. 5, 263–273 (2020).
Nejedly, P. et al. Utilization of temporal autoencoder for semi-supervised intracranial EEG clustering and classification. Sci. Rep. 13, 744 (2023).
Jeong, J. W. et al. Multi-scale deep learning of clinically acquired multi-modal MRI improves the localization of seizure onset zone in children with drug-resistant epilepsy. IEEE J. Biomed. Health Inform. 26, 5529–5539 (2022).
Mo, J. et al. Neuroimaging gradient alterations and epileptogenic prediction in focal cortical dysplasia IIIa. J. Neural Eng. 19, 025001 (2022).
Constantino, A. C. et al. Expert-level intracranial electroencephalogram ictal pattern detection by a deep learning neural network. Front. Neurol. 12, 603868 (2021).
Arcot Desai, S., Tcheng, T. & Morrell, M. Non-linear embedding methods for identifying similar brain activity in 1 million iEEG records captured from 256 RNS system patients. Front. Big Data 5, 840508 (2022).
Stirling, R. E. et al. Seizure forecasting using a novel sub-scalp ultra-long term EEG monitoring system. Front. Neurol. 12, 713794 (2021).
Ibrahim, G. M. et al. Presurgical thalamocortical connectivity is associated with response to vagus nerve stimulation in children with intractable epilepsy. NeuroImage Clin. 16, 634–642 (2017).
Mithani, K. et al. Connectomic profiling identifies responders to vagus nerve stimulation. Ann. Neurol. 86, 743–753 (2019).
Brinkmann, B. H. et al. Seizure diaries and forecasting with wearables: epilepsy monitoring outside the clinic. Front. Neurol. 12, 690404 (2021).
Meisel, C. et al. Machine learning from wristband sensor data for wearable, noninvasive seizure forecasting. Epilepsia 61, 2653–2666 (2020).
Nasseri, M. et al. Non-invasive wearable seizure detection using long-short-term memory networks with transfer learning. J. Neural Eng. 18, 056017 (2021).
Yew, A. N. J., Schraagen, M., Otte, W. M. & van Diessen, E. Transforming epilepsy research: a systematic review on natural language processing applications. Epilepsia 64, 292–305 (2023).
Savova, G. K. et al. Mayo clinical Text Analysis and Knowledge Extraction System (cTAKES): architecture, component evaluation and applications. J. Am. Med. Inform. Assoc. 17, 507–513 (2010).
Cui, L., Bozorgi, A., Lhatoo, S. D., Zhang, G. Q. & Sahoo, S. S. EpiDEA: extracting structured epilepsy and seizure information from patient discharge summaries for cohort identification. AMIA Annu. Symp. Proc. 2012, 1191–1200 (2012).
Guergana, K. S. et al. Mayo clinical text analysis and knowledge extraction system (cTAKES): architecture, component evaluation and applications. J. Am. Med. Inform. Assoc. 17, 507–513 (2010).
Garla, V. et al. The Yale cTAKES extensions for document classification: architecture and application. J. Am. Med. Inform. Assoc. 18, 614–620 (2011).
Hamid, H. et al. Validating a natural language processing tool to exclude psychogenic nonepileptic seizures in electronic medical record-based epilepsy research. Epilepsy Behav. 29, 578–580 (2013).
Beaulieu-Jones, B. K. et al. Predicting seizure recurrence after an initial seizure-like episode from routine clinical notes using large language models: a retrospective cohort study. Lancet Digit. Health 5, e882–e894 (2023).
Xie, K. et al. Extracting seizure frequency from epilepsy clinic notes: a machine reading approach to natural language processing. J. Am. Med. Inform. Assoc. 29, 873–881 (2022).
Xie, K., Litt, B., Roth, D. & Ellis, C. A. In Proceedings of the 21st Workshop on Biomedical Language Processing (eds Demner-Fushman, D., Cohen, K. B., Ananiadou, S. & Tsujii, J.) 369–375 (Association for Computational Linguistics, 2022).
Xie, K. et al. Long term epilepsy outcome dynamics revealed by natural language processing of clinic notes. Epilepsia 64, 1900–1909 (2023).
van Diessen, E., van Amerongen, R. A., Zijlmans, M. & Otte, W. M. Potential merits and flaws of large language models in epilepsy care: a critical review. Epilepsia 65, 873–886 (2024).
Ahmedt-Aristizabal, D. et al. Automated analysis of seizure semiology and brain electrical activity in presurgery evaluation of epilepsy: a focused survey. Epilepsia 58, 1817–1831 (2017).
Ahmedt-Aristizabal, D. et al. Deep learning approaches for seizure video analysis: a review. Epilepsy Behav. 154, 109735 (2024).
Peltola, J. et al. Semiautomated classification of nocturnal seizures using video recordings. Epilepsia 64, S65–S71 (2023).
Rai, P. et al. Automated analysis and detection of epileptic seizures in video recordings using artificial intelligence. Front. Neuroinform. https://doi.org/10.3389/fninf.2024.1324981 (2024).
Karácsony, T. et al. Novel 3D video action recognition deep learning approach for near real time epileptic seizure classification. Sci. Rep. 12, 19571 (2022).
Alim-Marvasti, A. et al. Machine learning for localizing epileptogenic-zone in the temporal lobe: quantifying the value of multimodal clinical-semiology and imaging concordance. Front. Digit. Health https://doi.org/10.3389/fdgth.2021.559103 (2021).
Martini, M. L. et al. Deep anomaly detection of seizures with paired stereoelectroencephalography and video recordings. Sci. Rep. 11, 7482 (2021).
Pérez-García, F. et al. Software tool for visualization of a probabilistic map of the epileptogenic zone from seizure semiologies. Front. Neuroinform. 16, 990859 (2022).
Alim-Marvasti, A. et al. Probabilistic landscape of seizure semiology localizing values. Brain Commun. 4, fcac130 (2022).
Pereira Choupina, H. M. et al. NeuroKinect 3.0: multi-bed 3Dvideo-EEG system for epilepsy clinical motion monitoring. Stud. Health Technol. Inform. 247, 46–50 (2018).
Jehi, L. et al. Development and validation of nomograms to provide individualised predictions of seizure outcomes after epilepsy surgery: a retrospective analysis. Lancet Neurol. 14, 283–290 (2015).
Alim-Marvasti, A., Vakharia, V. N. & Duncan, J. S. Multimodal prognostic features of seizure freedom in epilepsy surgery. J. Neurol. Neurosurg. Psychiatry 93, 499–508 (2022).
Acknowledgements
The authors acknowledge K. Xie (University of Pennsylvania) for his help with the section of the Review on electronic health records, and B. Scheid (University of Pennsylvania) and W. Ojemann (University of Pennsylvania) for their help with the section of the Review on medical devices. All provided useful references that enriched the content.
Author information
Authors and Affiliations
Contributions
All authors contributed to all aspects of the article.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Reviews Neurology thanks I. Heijink, A. Alim-Marvasti and M. Zijlmans for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Glossary
- Approximate zero-crossing
-
Point in a waveform where the signal crosses the zero line, useful in signal processing and analysis.
- Area under the receiver operating curve
-
A measure of how well a model can distinguish between different outcomes; higher values indicate better performance.
- Autoencoder
-
A neural network designed to learn embeddings of the input data through an encoding–decoding scheme.
- Bandpower
-
A measure of the power of a signal’s frequency components, often restricted to a specific range of frequencies.
- Betweenness centrality
-
A measure of how often a node appears on the shortest paths between other nodes in a network, indicating its role as a bridge or connector.
- Bidirectional encoder representations from transformers
-
A technique in natural language processing that helps computers to understand the context of words in sentences more effectively.
- Class imbalance
-
Occurs in data when one group of data (a class) is much larger than others, making it harder for machine learning models to generalize to datasets with a different distribution of classes.
- Convolutional neural network
-
A neural network that is especially good for analysing visual data, such as images, by recognizing patterns and structures within them.
- Cortical outlier detection
-
A technique that identifies unusual structures or patterns in brain imaging data; useful for studying brain health and disorders.
- Decision trees
-
Machine learning models that predict the value of a target variable by learning simple decision rules inferred from the data features.
- Deep learning
-
A type of machine learning that involves neural networks with many layers; excels in tasks such as image and speech recognition.
- Deep residual neural network
-
A type of advanced neural network that is especially good at learning from very large amounts of data, with a unique feature of learning from its own errors to improve accuracy.
- Degree centrality
-
A measure of how many connections or relationships a node has with others, indicating its importance or influence in the network.
- Elastic net regression
-
A statistical method that combines two techniques (least absolute shrinkage and selection operator regression and ridge regression) to predict or estimate outcomes; useful for overcoming difficulties associated with datasets with a large number of features and overfitting to limited data.
- Embedding
-
The process of mapping data onto a new space of higher or lower dimensionality, in which the data points are more clearly distinguishable or more easily manipulated.
- Extreme gradient boosting algorithm
-
A powerful and efficient algorithm used for structured data prediction, known for its speed and performance.
- False alarm rate
-
The rate at which a system incorrectly identifies a non-event as an event.
- Fast Fourier transformation
-
A method for analysing the frequency components of signals (such as audio) over time, represented visually in a spectrogram.
- Feature crafting
-
The process of selecting, manipulating and transforming raw data into features that better represent the underlying problem to predictive models.
- Features
-
Distinct, measurable attributes, properties or characteristics within a dataset that act as inputs for artificial intelligence models.
- Fractional anisotropy
-
A brain imaging measure that indicates how uniformly water flows in one direction, often used to assess the integrity of brain fibres.
- F-score
-
A statistical measure of the accuracy of a model, especially when there is an uneven class distribution; encompasses the precision (the proportion of positive identifications that were correct) and the recall (the proportion of positives that were identified correctly) of the model.
- Generative adversarial network
-
A framework for simultaneously training a generative model, which generates new data instances, and a discriminative model, which evaluates these data instances; the generative model learns to trick the discriminative model, creating data that looks like the inputs provided to the model.
- Graph convolutional neural networks
-
Neural networks that operate directly on data that are structured as a connected graph (nodes with connections between them), enabling the model to take advantage of the graphical structure of data.
- Graph theory metrics
-
Measures that derive from graph theory and are used to analyse the structures and connections within a set of data, such as social networks or brain connectivity networks.
- Ground truth labels
-
The true labels or classifications of data used to train and evaluate machine learning models.
- Held out dataset
-
A portion of data that is set aside and not used during training of a model and is later used to test and evaluate the performance of the model.
- High-dimensional
-
Including a large number of attributes or variables, often making analysis more complex.
- Hubness
-
The tendency of a node in a network to form a large number of connections with other nodes; hub nodes are central or highly influential within the network and strongly affect its structure and dynamics.
- Independent component analysis
-
A method to separate a mixed set of signals into individual, uncorrelated components, often used in audio or image processing.
- Integrated gradients
-
A method that helps to explain the decision-making of models, particularly in deep learning, by attributing the prediction to its input features.
- K-means clustering
-
A method for grouping data into a specified number of clusters, where each data point belongs to the cluster with the nearest mean value.
- Lagged global brain signal
-
A measure of time-delayed overall brain activity used in brain imaging analysis to understand brain network dynamics.
- Large language models
-
Artificial intelligence models that are designed to understand, generate and interpret large amounts of natural language text.
- Latent Dirichlet allocation
-
A generative statistical model traditionally used for topic modelling of documents; assumes each document includes a mixture of topics and iteratively refines its guesses about these topics and their associated words, revealing the main subjects discussed across a large set of documents.
- Leave-one-site-out cross-validation
-
A method used to test the effectiveness of a model, particularly when data originate from different locations, by systematically leaving out data from one site at a time.
- Linear discriminant analysis classifiers
-
Supervised learning methods that separate different groups or categories by identifying the features that most clearly distinguish them.
- Logistic regression classifiers
-
Methods that calculate the odds of something happening (for example, an event or a choice) on the basis of specific influencing factors.
- Long short-term memory recursive neural networks
-
A type of neural network that is especially good at processing sequences of data (such as text or time series) by remembering information for long periods.
- Machine learning
-
A field of artificial intelligence in which algorithms learn from data and make predictions or decisions.
- Mean diffusivity
-
A brain imaging measure that reflects how much water spreads out in tissue, helping to understand brain health and disease states.
- Mean kurtosis
-
A brain imaging measure of the degree of variability in water diffusion, providing insight into the complexity of tissue structure.
- Natural language processing
-
A branch of artificial intelligence that deals with training models to understand human natural language.
- Neural network
-
A computing system inspired by the structure, processing methods and learning ability of the human brain, consisting of layers of interconnected ‘neurons’, each of which can process information and transmit signals to others in the network, and often employing non-linearities in their structure that enable recognition of patterns in complex data that the network learns from.
- Non-negative matrix factorization
-
A technique used to break down a dataset into simpler parts, with the condition that none of these parts can have negative values; commonly used for data compression and pattern recognition.
- Normative modelling
-
A modelling approach that involves creating a standard or ‘normal’ model based on data from a healthy population, which can then be used to spot irregularities or deviations in individual cases.
- Quadratic discriminant analysis classifier
-
A technique that separates different groups or categories, similar to linear discriminant analysis classifiers but better suited to situations in which each group has unique characteristics.
- Random forest classifier
-
A machine learning model that consists of many individual decision trees that operate as an ensemble, in which each tree makes a class prediction and the class that gets the most votes becomes the model’s prediction.
- Recursive feature elimination
-
A technique for selecting important features in a dataset by repeatedly building a model and removing the least informative feature, as defined by the performance of the model trained with the set of features, each time.
- Recursive neural networks
-
A type of neural network architecture designed for processing structured data, such as data with a temporal component (time-series data), or data that follows a specific pattern (language).
- Resection mask
-
A mapped area, often in medical imaging, that shows the part of an organ or tissue planned for surgical removal.
- Self-supervised learning
-
A type of machine learning in which the system teaches itself using part of the data as a guide, often used when labelled data are scarce.
- Semi-supervised learning
-
A type of machine learning in which a small set of labelled data are combined with a large pool of unlabelled data to improve training efficiency and model performance.
- Spectral clustering
-
A technique in which similar data points are grouped together on the basis of their relationships to each other, often used for complex clustering tasks.
- Spiking neural networks
-
Neural networks that mimic the behaviour of biological neural networks, in which neurons fire in a discrete manner.
- Supervised learning
-
A type of machine learning in which known examples are used to train the model such that it can then make predictions about new data.
- Support vector classifier
-
A type of model used for classification tasks, which works by finding the best boundary to separate different categories of data.
- Support vector regressor
-
A type of model used for predicting continuous outcomes, based on the principles of finding the best line (or plane in higher dimensions) to fit the data.
- Topic modelling
-
The process of identifying topics present in a collection of documents.
- Transfer learning
-
A method to expedite learning in a new task by transferring knowledge from a related task that has already been learned; for example, a model that has been trained for seizure detection on the basis of scalp electroencephalography can be re-trained for seizure detection on the basis of intracranial electroencephalography through transfer learning with fewer data than might otherwise be required.
- Unsupervised learning
-
A type of machine learning in which hidden patterns in data are identified without known examples or labels guiding the model.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Lucas, A., Revell, A. & Davis, K.A. Artificial intelligence in epilepsy — applications and pathways to the clinic. Nat Rev Neurol (2024). https://doi.org/10.1038/s41582-024-00965-9
Accepted:
Published:
DOI: https://doi.org/10.1038/s41582-024-00965-9