Aug 2026· Brain Topography· Vol 39· 0 citations· 34 references
Medicine
TL;DR
The hypothesis is that the sudden transition to an epileptic seizure and the abrupt qualitative change at seizure offset occur through nonlinear processes, while the seizure period is mainly governed by linear autocorrelations, which are responsible for spreading epileptic activity and maintaining ictal dynamics.
Abstract
Abrupt qualitative changes of the dynamics of complex systems that consist of a large number of dynamical units like the human brain are governed by nonlinear processes. A typical example of such a phenomenon is the peri-ictal transition of focal onset seizures. However, the reliable extraction of nonlinear features from short data segments is a highly nontrivial task. We apply a recently introduced method of nonlinear signal analysis that requires neither any kind of phase space reconstruction, necessary e.g. for the estimation of the Lyapunov exponent or the correlation dimension, nor the estimation of probability densities, which is mandatory for measures based on information theory like e.g. mutual information. The method is completely data driven and detects with high sensitivity signatures of determinism, distinguishing between regular, chaotic and stochastic dynamics and stands out for its very high temporal resolution. Using a double window approach our method provides a time-frequency resolution, visualizing the time evolution of nonlinear processes along the frequency spectrum without band-pass filtering beforehand. We apply this index to 100 intracranial EEG recordings of 37 patients with temporal lobe epilepsy provided by different institutions. Our hypothesis is that the sudden transition to an epileptic seizure and the abrupt qualitative change at seizure offset occur through nonlinear processes, while the seizure period is mainly governed by linear autocorrelations, which are responsible for spreading epileptic activity and maintaining ictal dynamics. When selecting electrodes near or within the seizure onset zone (SOZ), we observe pronounced features of non-linear determinism in all cases a few seconds before the seizure became visually detectable by specialists. Seizure offset as well as the immediate post seizure period are also characterized by pronounced nonlinear features, while almost no significant signatures of nonlinear features could be detected during the seizure period. The present work reveals nonlinear dynamics leading to epileptic seizures and opens novel perspective for the analysis of the comparison between inter and pre-ictal intervals, seizure offset and the post-seizure period.
Epilepsy is a neurological disorder characterized by unpredictable seizures that can severely affect patients’ quality of life. Diagnosis primarily relies on the analysis of the electroencephalogram (EEG), a complex, noisy, and nonstationary signal representing the summation of various brain rhythms (delta, theta, alpha, beta, and gamma). Since manual EEG analysis is difficult and time-consuming, automatic approaches are necessary to efficiently detect seizures and diagnose the disease. In this work, we propose a novel framework that integrates wavelets for brain rhythms estimation with Chaos-Embedding and the Higuchi fractal dimension to extract highly discriminative features. These features are then used in a robust representation space and classified using K-Nearest Neighbors is employed to differentiate between healthy patients, epileptic patients, and epileptic patients during seizures. Experimental results show that the proposed approach achieves near-perfect performance, outperforming existing methods and providing a promising tool for the automatic diagnosis of epilepsy based on brain rhythms.
Zayneb Brari, Ines Bouzouita, S. Belghith· International Conference on...· 0 citations
It is demonstrated that the squared Pearson correlation coefficient provides a simple quantitative criterion for distinguishing chaos from noise directly from observed time-series data.
Epileptic seizures arise from the abnormal, sustained recruitment of neuronal populations, yet the temporal organisation of activity within individual seizures remains poorly understood. We hypothesised that seizures follow structured trajectories through a restricted state space rather than random patterns of activity. To test this idea, we analysed continuous long-term EEG recordings from six mice with pilocarpine-induced temporal lobe epilepsy and developed a symbolic framework that transforms seizure activity into sequences of discrete burst states. Across more than one thousand spontaneous seizures, we found that seizures are organised by sparse, animal-specific transition rules linking a small repertoire of recurrent burst motifs. These rules evolve over the course of epilepsy: early seizures display substantial variability in burst composition, and, to a smaller but statistically significant degree, in transition structure, whereas later seizures become more stereotyped in which burst types they use, a pattern suggestive of consolidation of the epileptic network into a stable dynamical regime. Seizure microstructure was further modulated by circadian phase, indicating that biological rhythms influence not only when seizures occur but also how they unfold. Moreover, burst sequences exhibited higher-order temporal dependencies that could not be explained by first-order Markov statistics alone. Together, these results reveal an evolving seizure grammar that links ictal dynamics to disease progression and circadian regulation, establishing symbolic dynamics as a powerful framework for quantifying the internal organisation of seizures. significance statement Why seizures differ from one another, and how those differences relate to the evolution of epilepsy in each subject, remain poorly understood. We demonstrate that seizures are structured dynamical trajectories rather than uniform episodes of abnormal activity. Using symbolic and information-theoretic analyses of long-term recordings in epileptic mice, we identify a limited repertoire of recurrent activity patterns whose organisation evolves during chronic epilepsy and is modulated by circadian phase. These findings suggest that the internal structure of seizures encodes information about network reorganisation over time, providing a class of biomarkers for tracking disease progression.
Vinicius Lima, A. Ghestem, V. Jirsa et al.· bioRxiv· 0 citations
Complex dynamical systems often undergo transitions from periodic to chaotic behaviour as bifurcation parameters vary, making timely detection of these changes essential. Conventional approaches based on the maximal Lyapunov exponent (MLE) generally require either knowledge of the governing equations or sufficiently long, uniformly sampled time series. Their performance degrades when the available data are scanty or contain missing observations, making reliable phase-space reconstruction difficult. We propose a methodology that combines Topological Data Analysis (TDA), specifically 0-D sublevel persistence, with Machine Learning (ML) classifiers to distinguish periodic and chaotic regimes directly from time series. Sublevel persistence extracts topological features by analysing the evolution of minima and maxima, revealing repeating signatures for periodic dynamics and more scattered patterns for chaotic dynamics. These features are used to train logistic regression, support vector machine, and k-nearest neighbour classifiers. Hyperparameters are validated using K-fold cross-validation, yielding average classification accuracies exceeding 90%. The trained classifiers provide binary predictions, identifying periodic and chaotic behaviour in previously unseen data. The proposed methodology is evaluated on the Duffing, R\"ossler, and Lorenz systems, where the detected transitions closely agree with those identified using the MLE, demonstrating the reliability of the approach. It is further applied to real-world ECG signals to classify normal and abnormal heartbeats, producing encouraging performance across standard statistical metrics. The proposed framework provides an effective alternative for analysing sparse or incomplete time series and is particularly useful in experimental settings where conventional nonlinear time-series methods are limited.
Reliable anomaly detection in time series is crucial yet challenging owing to the chaotic nature of signals from sources such as financial markets or natural systems. However, most state-of-the-art approaches rely heavily on extensive parameter tuning and large labeled datasets, which are rarely available in production, making them unsuitable for highly dynamic environments. Because field data are predominantly unlabeled, fully unsupervised anomaly detection has become an operational necessity for practical industrial deployment. We present an enhanced, fully autonomous implementation of the Brain-Inspired Cortical Coding (BIC) framework that effectively eliminates the requirement for manual parameter tuning and labeling. The BIC framework combines adaptive clustering with entropy-oriented cortical coding. Unlike most advanced methods, BIC learns directly from unlabeled data and provides stable performance under varying operational conditions. The evaluation was conducted using synthetic Lorenz attractor signals and real-world artificial intelligence for IT operations (AIOps) key performance indicator (KPI) data characterized by rapidly evolving patterns. The comparative experiments included long short-term memory (LSTM), extreme learning machines (ELM), variational autoencoders (VAE), generative adversarial network (GAN)-based models, Isolation Forest, and statistical baselines. Results show BIC outperforms state-of-the-art models, achieving F1-scores of 0.84 and 0.75 under low- and high-noise Lorenz conditions, and an F1-score of approximately 0.80 on the AIOps KPI dataset across both short and long training regimes. Furthermore, under extended evaluation, BIC maintains stable detection performance, whereas competing models show significant deterioration or require additional retraining. These results demonstrate that BIC provides a robust and scalable solution for anomaly detection in noisy, chaotic, and rapidly evolving time-series environments.
In patients with epilepsy, seizures are associated with pathological neural synchronization. However, the preictal period preceding a seizure often exhibits reduced spatial synchronization compared to normal cognition. This observation aligns with the concept of the brain as a complex dynamical system, where a reduction in dimensionality and resilience can precede a phase transition. The Critical Brain Hypothesis suggests a connection between the loss of healthy scale-free behavior and various disorders, including epilepsy. Our study investigates preictal changes by utilizing network features, such as mean node degree and mean clustering coefficient, derived from thresholded correlation matrices of patient intracranial electrocorticographic electrode data. We observed a suppression of intermittent high-synchronization periods within the feature space during the minutes leading up to seizure onset. This constriction of the explored hypervolume in the preictal state indicates a breakdown in the brain’s ability to maintain normal coherence. We use these preictal changes to predict the probability of seizure onset using a Support Vector Machine algorithm. These discrete predictions can then be combined into real-time continuous seizure risk forecasts via Bayesian updating. This innovative and computationally lightweight approach has the potential to significantly improve upon static predictions, providing opportunities for more adaptable, quantitative, and interpretable tools for managing seizures.
Jeremy Goetz, John M. Beggs, Robert M. Worth et al.· bioRxiv· 0 citations
