Traditional artificial neural networks consist of nodes with nonoscillatory dynamics. Biological neural networks, on the other hand, consist of oscillatory components embedded in an oscillatory environment. Motivated by this feature of biological neurons, we describe a neural network framework with explicit damped, oscillatory node dynamics. We express the oscillatory dynamics using two history-dependent terms to connect these dynamics with standard recurrent neural network formulations, and apply physical constraints from observed brain dynamics to choose the oscillatory frequencies and stationary constraints to reduce the number of free parameters. We then optimize and illustrate network performance by classifying hand-written digits and simulated neuronal spike train activity and show that these oscillatory network elements support accurate classification with few trainable parameters. Choosing oscillator frequencies according to a proposed theory for brain rhythms improves classification accuracy compared to alternative frequency configurations and compared to standard recurrent neural network frameworks with comparable numbers of parameters. Compared to existing approaches, the proposed resonant recurrent network (RRN) utilizes oscillatory dynamics expressed as a straightforward extension of standard recurrent neural networks, produces interpretable features for classification, and performs well with few parameters when oscillator frequencies follow a configuration observed in vivo. We propose that RRNs may serve as efficient, biologically inspired building blocks to achieve complex goals in biological and artificial neural networks.

Thalamic neuromodulation has emerged as a promising therapy to reduce seizures in patients with drug-resistant epilepsy who are not candidates for resective or ablative surgical treatments. This advance creates therapeutic opportunities for patients with seizure foci that are generalized, poorly localized, multifocal, or in eloquent cortex. However, optimal treatment requires improved understanding of the thalamocortical circuits and cellular mechanisms that support seizure initiation, propagation, and termination. With its dense, nucleus-specific reciprocal connectivity with cortical and limbic regions and intrinsic oscillatory properties, the thalamus is well-constructed and well-positioned to influence epileptic activity. Importantly, distinct thalamic nuclei exhibit differing engagement depending on seizure type and propagation patterns. Here, we review current knowledge of thalamocortical anatomy and function, cellular mechanisms of ictal propagation, and the role of the thalamus in generalized and focal seizures, with emphasis on human studies. We further examine how these anatomical and mechanistic insights inform neuromodulatory interventions aimed at improving seizure control.

SUMMARY:Healthy sleep is electrographically defined by oscillations that drive alternating up- and down-states that regulate neuronal excitability. Slow oscillations-the hallmark rhythm of deep nonrapid eye movement sleep, temporally organize faster, more focal rhythms, including thalamocortical sleep spindles and hippocampal sharp-wave ripples. These cascading rhythms have been implicated as a critical activity-dependent mechanism that binds coordinated neuronal activity to support systems-level consolidation of previous experience into long-term memory. Epilepsy is a disease characterized by sporadic pathologic neuronal activity, including interictal epileptiform discharges and seizures. Epileptic activity is frequently potentiated during sleep, but the interactions with specific graphoelements that comprise sleep microarchitecture are complex. Critically, the diverse ways in which epileptic activity interdigitates with sleep microstructure imply that epileptiform activity can distort and disrupt the same sleep rhythms that normally support memory consolidation. Cognitive comorbidities are common in epilepsy, and impairments in memory consolidation are increasingly recognized. In this review, we synthesize leading evidence on the complex interactions between sleep microarchitecture and interictal epileptiform discharges. We first summarize electrophysiologic data on how slow oscillations interact with spindles and interictal epileptiform discharges and then assemble these findings into a unifying framework for interictal epileptiform discharge-slow oscillation-spindle dynamics. Finally, we review evidence on how disruption of these sleep oscillations can contribute to cognitive dysfunction in epilepsy and highlight implications for developmental and epileptic encephalopathies.

The cross-regional interplay of slow oscillations, spindles, and ripples during sleep is believed to support systems memory consolidation but remains understudied in humans. Using a validated behavioral task and simultaneous intracranial neural recordings from the orbitofrontal cortex, thalamus, and hippocampus in 19 patients with epilepsy, we examined the cross-regional interplay of sleep oscillations (slow oscillations, spindles, and ripples), alongside epileptic spikes, and their role in motor memory consolidation. Orbitofrontal slow oscillations robustly modulate spindle and ripple oscillations within and across regions during sleep. Although most combinations of oscillation rates positively predicted overnight performance change in a motor task, hippocampal ripple rate and coupled hippocampal-orbitofrontal ripple rates were the most reliable predictors across subjects. In contrast, rates of most sleep oscillations coupled to epileptic spikes were negative predictors of overnight motor performance change, with the rate of slow oscillations co-occurring with epileptic spikes the most reliable predictors of negative change across subjects. These findings provide direct evidence of a hierarchical cascade of sleep oscillations in human motor memory processing and reveal that epileptic spikes coupled to sleep oscillations interfere with this process in patients with epilepsy.

The widespread use of calcium imaging has produced large-scale datasets capturing neuronal population activity across diverse experimental contexts, posing challenges for analyzing complex, high-dimensional data. Dimensionality reduction (DR) methods have been pivotal in addressing these challenges by simplifying data into interpretable, low-dimensional structures, while capturing essential network dynamics. Among DR methods, Nonnegative Matrix Factorization (NMF) can produce biologically meaningful representations through its nonnegativity constraint and parts-based decomposition, making it especially suited for analyzing neuronal calcium signals. To enhance accessibility and standardization in the analysis of state-dependent neuronal dynamics, we introduce Calcium Network dynamiCs (CaNetiCs), an open-source toolbox centered on NMF, integrating standardized DR methods (PCA, ICA, UMAP), geometric low-dimensional component space analyses, and neuronal network simulation modules. We validate our toolbox by applying it to two diverse experimental datasets that describe responses to graded anesthesia: whole-ganglion cellular calcium imaging of C. elegans and two-photon imaging of murine somatosensory cortex. Our analyses recapitulate previously observed trends, such as network suppression and decorrelation with anesthesia, while uncovering novel insights into neuronal activity under differing contexts. CaNetiCs provides an accessible, modular, and interpretable framework, facilitating broader adoption of standardized dimensionality reduction methodologies for deeper exploration of neuronal network dynamics across experimental paradigms. The open-source code, along with documentation, is available at https://github.com/dannycarbonero/CaNetiCs.

Stochastic Actor-Oriented Models (SAOMs) were designed in the social network setting to capture network dynamics representing a variety of influences on network change. The standard framework assumes the observed networks are free of false positive and false negative edges, which may be an unrealistic assumption. We propose a hidden Markov model (HMM) extension to these models, consisting of two components: 1) a latent model, which assumes that the unobserved, true networks evolve according to a Markov process as they do in the SAOM framework; and 2) a measurement model, which describes the conditional distribution of the observed networks given the true networks. An expectation-maximization algorithm is developed for parameter estimation. We address the computational challenge posed by a massive discrete state space, of a size exponentially increasing in the number of vertices, through the use of the missing information principle and particle filtering. We present results from a simulation study, demonstrating our approach offers improvement in accuracy of estimation, in contrast to the standard SAOM, when the underlying networks are observed with noise. We apply our method to functional brain networks inferred from electroencephalogram data, revealing larger effect sizes when compared to the naive approach of fitting the standard SAOM.

Background:Sleep-dependent memory consolidation is supported by sleep spindles during stages 2 and 3 non-rapid eye movement sleep. Sleep spindles and sleep-dependent memory consolidation are both decreased in Rolandic epilepsy (RE). Non-invasive auditory stimulation evokes SOs and SO-spindle complexes in healthy adults but the impact on memory consolidation has been inconsistent. Objective:We investigated the effects of auditory stimulation during sleep on SOs, SO-spindle complexes, and sleep-dependent memory consolidation in children with RE and controls. Methods:A prospective cross-over study was conducted in children with RE and control. Children completed two nap visits with auditory or sham stimulation. SOs and SO-spindle complexes rates were measured offline using validated detectors. Sleep-dependent memory consolidation was assessed using the motor sequence typing task. Results:Auditory stimulation evoked SOs and SO-spindle complexes broadly with maximal effect over frontal electrodes. Compared to sham, stimulation delivered during background activity evoked SOs (29.8% increase, p<0.001) and SO-spindle complexes (16.8% increase, p<0.001) and stimulations delivered near the peak of an ongoing SO upstate maximally evoked SOs (51.3% increase, p<0.001) and SO-spindle complexes (32.3% increase, p<0.001). Changes in frontal SO (1.9% improvement per increase in SO/min; p<0.001) and SO-spindle complexes (9.5% improvement per increase in SO-spindle/min) event rates due to auditory stimulation positively predicted changes in sleep-dependent memory consolidation. Conclusion:Auditory stimulation reliably modulates sleep oscillations when delivered on background activity and during the upstate of SOs. As increased event rates improve memory consolidation, stimulation paradigms to increase SO and SO-spindle complex rates are required to enhance memory.

OBJECTIVE:We investigated the effects of auditory stimulation during sleep on slow oscillations (SOs), SO-spindle complexes, and sleep-dependent memory consolidation in children with Rolandic Epilepsy (RE) and controls. METHODS:Participants completed two nap visits with auditory or sham stimulation. SOs and SO-spindle complexes rates were measured offline using validated detectors. Sleep-dependent memory consolidation was assessed using the motor sequence typing task. RESULTS:Auditory stimulation evoked SOs and SO-spindle complexes with maximal effect over frontal electrodes. Compared to sham, stimulation of background activity increased SOs (29.8 %, p < 0.001) and SO-spindle complexes (16.8 %, p < 0.001); stimulation of an ongoing SO upstate maximally evoked SOs (51.3 % increase, p < 0.001) and SO-spindle complexes (32.3 % increase, p < 0.001). Changes in frontal SO (1.9 % improvement per increase in SO/min; p < 0.001) and SO-spindle complexes (9.5 % improvement per increase in SO-spindle/min; p = 0.007) event rates due to auditory stimulation positively predicted changes in sleep-dependent memory consolidation. CONCLUSION:Auditory stimulation reliably modulates sleep oscillations when delivered on background activity and during the upstate of SOs. As increased event rates improve memory consolidation, stimulation paradigms to increase SO and SO-spindle complex rates are required to enhance memory. SIGNIFICANCE:Auditory stimulation paradigms that increase SO and SO-spindle complex rates may enhance sleep-dependent memory consolidation.

Interictal epileptiform spikes, high-frequency ripple oscillations, and their co-occurrence - spike ripples - in human scalp or intracranial voltage recordings are well-established epileptic biomarkers independent of etiology. While clinically significant, the neural mechanisms generating these electrographic events remain unclear. Stroke is a well-established risk factor for epilepsy with an estimated 11% of patients developing epilepsy within 5 years post-stroke. Cortical stroke induced via photothrombosis results in epilepsy in rats. As mice allow for cell-type specific analysis, we sought to determine whether focal cortical stroke in mice produces characteristic human epileptic biomarkers specific to the lesioned tissue. We induced unilateral focal stroke in the motor cortex using photothrombosis in the broadly used C57BL/6 mice and obtained intermittent bilateral local field potential recordings over many weeks. We observed spike, ripple, and spike ripple biomarkers in the peri-stroke mouse cortex that shared consistent morphology as humans with epilepsy. In addition, similar to humans, we found spike ripples detected using automated procedures developed from human intracranial recordings classified the pathological hemisphere with the highest sensitivity and specificity among the three biomarkers. Expert validation of the automatically detected biomarkers confirmed these observations. These results demonstrate a translational cortical stroke mouse model with a defined injury zone that produces localized electrographic biomarkers as in human epilepsy, enabling the investigation of the cellular and circuit mechanisms of pathologic interictal activity.

The cross-regional interplay of slow oscillations, sleep spindles, and ripples during sleep is believed to support systems memory consolidation but is understudied in humans. Using a validated behavioral task and intracranial neural recordings from orbitofrontal cortex, thalamus, and hippocampus in 19 epilepsy patients, we examined the cross-regional interplay of sleep-oscillations and their role in memory consolidation. Orbitofrontal slow oscillations robustly modulate sleep rhythms both within and across regions. Most combinations of oscillation rates predict overnight memory consolidation, but hippocampal ripple rate and coupled hippocampal-orbitofrontal ripples were the strongest positive predictors of memory consolidation. In contrast, epileptic spikes coupled to sleep oscillations strongly predicted reduced memory consolidation, with the strongest negative effect observed when epileptic spikes were coupled to slow oscillations. These findings provide direct evidence of the hierarchical cascade of sleep oscillations in human memory processing and reveal how epileptic spikes disrupt this process in patients with epilepsy.

Epilepsy is a major neurological disorder characterized by recurrent, spontaneous seizures. For patients with drug-resistant epilepsy, treatments include neurostimulation or surgical removal of the epileptogenic zone (EZ), the brain region responsible for seizure generation. Precise targeting of the EZ requires reliable biomarkers. Spike ripples - high-frequency oscillations that co-occur with large amplitude epileptic discharges - have gained prominence as a candidate biomarker. However, spike ripple detection remains a challenge. The gold-standard approach requires an expert manually visualize and interpret brain voltage recordings, which limits reproducibility and high-throughput analysis. Addressing these limitations requires more objective, efficient, and automated methods for spike ripple detection, including approaches that utilize deep neural networks. Despite advancements, dataset heterogeneity and scarcity severely limit machine learning performance. Our study explores long-short term memory (LSTM) neural network architectures for spike ripple detection, leveraging data augmentation to improve classifier performance. We highlight the potential of combining training on augmented and in vivo data for enhanced spike ripple detection and ultimately improving diagnostic accuracy in epilepsy treatment.

Background and ObjectivesRolandic epilepsy (RE), the most common childhood focal epilepsy syndrome, is characterized by a transient period of sleep-activated epileptiform activity in the centrotemporal regions and variable cognitive deficits. Sleep spindles are prominent thalamocortical brain oscillations during sleep that have been mechanistically linked to sleep-dependent memory consolidation in animal models and healthy controls. Sleep spindles are decreased in RE and related sleep-activated epileptic encephalopathies. To further evaluate the association between this electrographic biomarker and cognitive dysfunction in this common disease, we investigate whether children with RE have deficient sleep-dependent memory consolidation and whether impaired memory consolidation is associated with reduced sleep spindles in the centrotemporal regions.MethodsIn this prospective case-control study, children were trained and tested on a validated probe of memory consolidation, the motor sequence task (MST). Sleep spindles were measured from high-density EEG during a 90-minute nap opportunity between MST training and testing using an automated sleep spindle detector validated for use in children with and without epilepsy.ResultsTwenty-three children with RE (9 with active disease, 5F, age 6.9-12.8 years; 14 with resolved disease, 8F, age 8.8-17.8 years) and 19 age-matched and sex-matched controls (8F, age 6.9-18.7 years) were enrolled. Children with active epilepsy had decreased memory consolidation compared with control children (p = 0.001, mean percentage reduction 25.7%, 95% CI 10.3%-41.2%) and compared with children with resolved epilepsy (p = 0.007, mean percentage reduction 21.9%, 95% CI 6.2%-37.6%). Children with active epilepsy had decreased sleep spindle rates in the centrotemporal region compared with controls (p = 0.008, mean decrease 2.5 spindles per minute, 95% CI 0.7-4.4 spindles per minute). Spindle rate, but not spike rate or spike-wave index, correlated with sleep-dependent memory consolidation (p = 0.004, mean MST improvement of 3.9%, 95% CI 1.3%-6.4%, for each unit increase in spindles per minute).DiscussionChildren with RE have impaired sleep-dependent memory consolidation during the active period of disease that correlates with a deficit in the sleep spindle rate. This finding identifies a noninvasive biomarker to aid diagnosis and a potential etiologic mechanism to guide therapeutic discovery of cognitive dysfunction in RE and related sleep-activated epilepsy syndromes.

In severe epileptic encephalopathies, epileptic activity contributes to progressive cognitive dysfunction. Epileptic encephalopathies share the trait of spike-wave activation during non-REM sleep (EE-SWAS), a sleep stage dominated by sleep spindles, which are brain oscillations known to coordinate offline memory consolidation. Epileptic activity has been proposed to hijack the circuits driving these thalamocortical oscillations, thereby contributing to cognitive impairment. Using a unique dataset of simultaneous human thalamic and cortical recordings in subjects with and without EE-SWAS, we provide evidence for epileptic spike interference of thalamic sleep spindle production in patients with EE-SWAS. First, we show that epileptic spikes and sleep spindles are both predicted by slow oscillations during stage two sleep (N2), but at different phases of the slow oscillation. Next, we demonstrate that sleep-activated cortical epileptic spikes propagate to the thalamus (thalamic spike rate increases after a cortical spike, P approximate to 0). We then show that epileptic spikes in the thalamus increase the thalamic spindle refractory period (P approximate to 0). Finally, we show that in three patients with EE-SWAS, there is a downregulation of sleep spindles for 30 s after each thalamic spike (P < 0.01). These direct human thalamocortical observations support a proposed mechanism for epileptiform activity to impact cognitive function, wherein epileptic spikes inhibit thalamic sleep spindles in epileptic encephalopathy with spike and wave activation during sleep.

Objective:Interictal epileptiform spikes, high-frequency ripple oscillations, and their co-occurrence (spike ripples) in human scalp or intracranial voltage recordings are well-established epileptic biomarkers. While clinically significant, the neural mechanisms generating these electrographic biomarkers remain unclear. To reduce this knowledge gap, we introduce a novel photothrombotic stroke model in mice that reproduces focal interictal electrographic biomarkers observed in human epilepsy. Methods:We induced a stroke in the motor cortex of C57BL/6 mice unilaterally (N=7) using a photothrombotic procedure previously established in rats. We then implanted intracranial electrodes (2 ipsilateral and 2 contralateral) and obtained intermittent local field potential (LFP) recordings over several weeks in awake, behaving mice. We evaluated the LFP for focal slowing and epileptic biomarkers - spikes, ripples, and spike ripples - using both automated and semi-automated procedures. Results:Delta power (1-4 Hz) was higher in the stroke hemisphere than the non-stroke hemisphere in all mice ( p <0.001). Automated detection procedures indicated that compared to the non-stroke hemisphere, the stroke hemisphere had an increased spike ripple ( p =0.006) and spike rates ( p =0.039), but no change in ripple rate ( p =0.98). Expert validation confirmed the observation of elevated spike ripple rates ( p =0.008) and a trend of elevated spike rate ( p =0.055) in the stroke hemisphere. Interestingly, the validated ripple rate in the stroke hemisphere was higher than the non-stroke hemisphere ( p =0.031), highlighting the difficulty of automatically detecting ripples. Finally, using optimal performance thresholds, automatically detected spike ripples classified the stroke hemisphere with the best accuracy (sensitivity 0.94, specificity 0.94). Significance:Cortical photothrombosis-induced stroke in commonly used C57BL/6 mice produces electrographic biomarkers as observed in human epilepsy. This model represents a new translational cortical epilepsy model with a defined irritative zone, which can be broadly applied in transgenic mice for cell type specific analysis of the cellular and circuit mechanisms of pathologic interictal activity. Key Points:Cortical photothrombosis in mice produces stroke with characteristic intermittent focal delta slowing.Cortical photothrombosis stroke in mice produces the epileptic biomarkers spikes, ripples, and spike ripples.All biomarkers share morphological features with the corresponding human correlate.Spike ripples better lateralize to the lesional cortex than spikes or ripples.This cortical model can be applied in transgenic mice for mechanistic studies.

We evaluated whether spike ripples, the combination of epileptiform spikes and ripples, provide a reliable and improved biomarker for the epileptogenic zone compared with other leading interictal biomarkers in a multicentre, international study. We first validated an automated spike ripple detector on intracranial EEG recordings. We then applied this detector to subjects from four centres who subsequently underwent surgical resection with known 1-year outcomes. We evaluated the spike ripple rate in subjects cured after resection [International League Against Epilepsy Class 1 outcome (ILAE 1)] and those with persistent seizures (ILAE 2-6) across sites and recording types. We also evaluated available interictal biomarkers: spike, spike-gamma, wideband high frequency oscillation (HFO, 80-500 Hz), ripple (80-250 Hz) and fast ripple (250-500 Hz) rates using previously validated automated detectors. The proportion of resected events was computed and compared across subject outcomes and biomarkers. Overall, 109 subjects were included. Most spike ripples were removed in subjects with ILAE 1 outcome (P < 0.001), and this was qualitatively observed across all sites and for depth and subdural electrodes (P < 0.001 and P < 0.001, respectively). Among ILAE 1 subjects, the mean spike ripple rate was higher in the resected volume (0.66/min) than in the non-removed tissue (0.08/min, P < 0.001). A higher proportion of spike ripples were removed in subjects with ILAE 1 outcomes compared with ILAE 2-6 outcomes (P = 0.06). Among ILAE 1 subjects, the proportion of spike ripples removed was higher than the proportion of spikes (P < 0.001), spike-gamma (P < 0.001), wideband HFOs (P < 0.001), ripples (P = 0.009) and fast ripples (P = 0.009) removed. At the individual level, more subjects with ILAE 1 outcomes had the majority of spike ripples removed (79%, 38/48) than spikes (69%, P = 0.12), spike-gamma (69%, P = 0.12), wideband HFOs (63%, P = 0.03), ripples (45%, P = 0.01) or fast ripples (36%, P < 0.001) removed. Thus, in this large, multicentre cohort, when surgical resection was successful, the majority of spike ripples were removed. Furthermore, automatically detected spike ripples localize the epileptogenic tissue better than spikes, spike-gamma, wideband HFOs, ripples and fast ripples.

Calcium imaging allows recording from hundreds of neurons in vivo with the ability to resolve single cell activity. Evaluating and analyzing neuronal responses, while also considering all dimensions of the data set to make specific conclusions, is extremely difficult. Often, descriptive statistics are used to analyze these forms of data. These analyses, however, remove variance by averaging the responses of single neurons across recording sessions, or across combinations of neurons, to create single quantitative metrics, losing the temporal dynamics of neuronal activity, and their responses relative to each other. Dimensionally Reduction (DR) methods serve as a good foundation for these analyses because they reduce the dimensions of the data into components, while still maintaining the variance. Non-negative Matrix Factorization (NMF) is an especially promising DR analysis method for analyzing activity recorded in calcium imaging because of its mathematical constraints, which include positivity and linearity. We adapt NMF for our analyses and compare its performance to alternative dimensionality reduction methods on both artificial and in vivo data. We find that NMF is well-suited for analyzing calcium imaging recordings, accurately capturing the underlying dynamics of the data, and outperforming alternative methods in common use.

Journal Article Accepted manuscript Reply: The challenge of assessing invasive biomarkers for epilepsy surgery and To plan efficacious epilepsy surgery Get access Wen Shi, Wen Shi Department of Neurology, Massachusetts General Hospital, Boston, MA 02114, USAHarvard Medical School, Boston, MA 02115, USA Search for other works by this author on: Oxford Academic PubMed Google Scholar Uri Eden, Uri Eden Center for Systems Neuroscience, Boston University, Boston, MA 02215, USADepartment of Mathematics and Statistics, Boston University, Boston, MA 02215, USA Search for other works by this author on: Oxford Academic PubMed Google Scholar Mark A Kramer, Mark A Kramer Center for Systems Neuroscience, Boston University, Boston, MA 02215, USADepartment of Mathematics and Statistics, Boston University, Boston, MA 02215, USA Search for other works by this author on: Oxford Academic PubMed Google Scholar Catherine J Chu Catherine J Chu Department of Neurology, Massachusetts General Hospital, Boston, MA 02114, USAHarvard Medical School, Boston, MA 02115, USA Correspondence to: Catherine J. Chu, 100 Cambridge Street, 20th floor, Boston, MA 02114, USA. E-mail: cjchu@mgh.harvard.edu https://orcid.org/0000-0001-7670-9313 Search for other works by this author on: Oxford Academic PubMed Google Scholar Brain, awae165, https://doi.org/10.1093/brain/awae165 Published: 16 May 2024 Article history Received: 07 May 2024 Accepted: 13 May 2024 Published: 16 May 2024

Consistent observations across recording modalities, experiments, and neural systems find neural field spectra with 1/f-like scaling, eliciting many alternative theories to explain this universal phenomenon. We show that a general dynamical system with stochastic drive and minimal assumptions generates 1/f-like spectra consistent with the range of values observed in vivo without requiring a specific biological mechanism or collective critical behavior.

OBJECTIVE:Early identification of infants at risk of cerebral palsy (CP) enables interventions to optimize outcomes. Central sleep spindles reflect thalamocortical sensorimotor circuit function. We hypothesized that abnormal infant central spindle activity would predict later contralateral CP. METHODS:We trained and validated an automated detector to measure spindle rate, duration, and percentage from central electroencephalogram (EEG) channels in high-risk infants (n = 35) and age-matched controls (n = 42). Neonatal magnetic resonance imaging (MRI) findings, infant motor exam, and CP outcomes were obtained from chart review. Using univariable and multivariable logistic regression models, we examined whether spindle activity, MRI abnormalities, and/or motor exam predicted future contralateral CP. RESULTS:The detector had excellent performance (F1 = 0.50). Spindle rate (p = 0.005, p = 0.0004), duration (p < 0.001, p < 0.001), and percentage (p < 0.001, p < 0.001) were decreased in hemispheres corresponding to future CP compared to those without. In this cohort, PLIC abnormality (p = 0.004) and any MRI abnormality (p = 0.004) also predicted subsequent CP. After controlling for MRI findings, spindle features remained significant predictors and improved model fit (p < 0.001, all tests). Using both spindle duration and MRI findings had highest accuracy to classify hemispheres corresponding to future CP (F1 = 0.98, AUC 0.999). CONCLUSION:Decreased central spindle activity improves the prediction of future CP in high-risk infants beyond early MRI or clinical exam alone. SIGNIFICANCE:Decreased central spindle activity provides an early biomarker for CP.