An automated algorithm for stereoelectroencephalography electrode localization and labelling.

PURPOSE: Stereoelectroencephalography (sEEG) is increasingly utilized for localization of seizure foci, functional mapping, and neurocognitive research due to its ability to target deep and difficult to reach anatomical locations and to study in vivo brain function with a high signal-to-noise ratio. The research potential of sEEG is constrained by the need for accurate localization of the implanted electrodes in a common template space for group analyses. METHODS: We present an algorithm to automate the grouping of sEEG electrodes by trajectories, labelled by target and insertion point. This algorithm forms the core of a pipeline that fully automates the entire process of electrode localization in standard space, using raw CT and MRI images to produce atlas labelled MNI coordinates. RESULTS: Across 196 trajectories from 20 patients, the pipeline successfully processed 190 trajectories with localizations within 0.25±0.55 mm of the manual annotation by two reviewers. Six electrode trajectories were not directly identified due to metal artifacts and locations were interpolated based on the first and last contact location and the number of contacts in that electrode as listed in the surgical record. CONCLUSION: We introduce our algorithm and pipeline for automatically localizing, grouping, and classifying sEEG electrodes from raw CT and MRI. Our algorithm adds to existing pipelines and toolboxes for electrode localization by automating the manual step of marking and grouping electrodes, thereby expedites the analyses of sEEG data, particularly in large datasets.

Longitudinal point-of-care assessment of psychomotor vigilance in children in the epilepsy monitoring unit.

The epilepsy monitoring unit (EMU) is a complex and dynamic operational environment, where the cognitive and behavioural consequences of medical and environmental changes often go unnoticed. The psychomotor vigilance task (PVT) has been used to detect changes in cognition and behaviour in numerous contexts, including among astronauts on spaceflight missions, pilots, and commercial drivers. Here, we piloted serial point-of-care administration of the PVT in children undergoing invasive monitoring in the EMU. Seven children completed the PVT throughout their hospital admission and their performance was associated with daily seizure counts, interictal epileptiform discharges, number of antiseizure medications (ASMs) administered, and sleep quality metrics. Using mixed-effects models, we found that PVT reaction time and accuracy were adversely affected by greater number of ASMs and interictal epileptiform activity. We show that serial point-of-care PVT is simple and feasible in the EMU and may enable greater understanding of individual patient responses to medical and environmental alterations, inform clinical decision-making, and support quality-improvement and research initiatives.

The impact of simultaneous intracranial recordings on scalp EEG: A finite element analysis.

BACKGROUND: In this study, we examined the utility of simultaneous scalp and stereotactic intracranial electroencephalography (SSIEEG) in epilepsy patients. Although SSIEEG offers valuable insights into epilepsy and cognitive function, its routine use is uncommon. Challenges include interpreting post-craniotomy scalp EEG due to surgically implanted electrodes. NEW METHOD: We describe our methodology for conducting SSIEEG recordings. To simulate the potential impact on EEG interpretation, we computed the leadfield of scalp electrodes with and without burrholes using Finite Element Analysis to compare the resulting sensitivity volume and waveforms of simulated intracranial signals between skulls with and without burrholes. RESULTS: The presence of burr holes in the skull layer of the leadfield models did not discernibly modify simulated waveforms or scalp EEG topology. Using realistic SEEG burr hole diameter, the difference in the average leadfield of scalp electrodes was 0.12% relative to the effect of switching two nearby electrodes, characterized by the cosine similarity difference. No patients experienced adverse events related to SSIEEG. COMPARISON WITH EXISTING METHODS: Although there is increasing acceptance and interest in SSIEEG, few studies have characterized the technical feasibility. Here, we demonstrate through modelling that scalp recordings from SSIEEG are comparable to that through an intact skull. CONCLUSION: The placement and simultaneous acquisition of scalp EEG during invasive monitoring through stereotactically inserted EEG electrodes is routinely performed at the Hospital for Sick Children. Scalp EEG recordings may assist with clinical interpretation. Burr holes in the skull layer did not discernibly alter EEG waveforms or topology.

Dissociable default-mode subnetworks subserve childhood attention and cognitive flexibility: Evidence from deep learning and stereotactic electroencephalography.

Cognitive flexibility encompasses the ability to efficiently shift focus and forms a critical component of goal-directed attention. The neural substrates of this process are incompletely understood in part due to difficulties in sampling the involved circuitry. We leverage stereotactic intracranial recordings to directly resolve local-field potentials from otherwise inaccessible structures to study moment-to-moment attentional activity in children with epilepsy performing a flexible attentional task. On an individual subject level, we employed deep learning to decode neural features predictive of task performance indexed by single-trial reaction time. These models were subsequently aggregated across participants to identify predictive brain regions based on AAL atlas and FIND functional network parcellations. Through this approach, we show that fluctuations in beta (12-30 Hz) and gamma (30-80 Hz) power reflective of increased top-down attentional control and local neuronal processing within relevant large-scale networks can accurately predict single-trial task performance. We next performed connectomic profiling of these highly predictive nodes to examine task-related engagement of distributed functional networks, revealing exclusive recruitment of the dorsal default mode network during shifts in attention. The identification of distinct substreams within the default mode system supports a key role for this network in cognitive flexibility and attention in children. Furthermore, convergence of our results onto consistent functional networks despite significant inter-subject variability in electrode implantations supports a broader role for deep learning applied to intracranial electrodes in the study of human attention.

Which is more deleterious to cognitive performance? Interictal epileptiform discharges vs anti-seizure medication.

Children with epilepsy commonly have comorbid neurocognitive impairments that severely affect their psychosocial well-being, education, and future career prospects. Although the provenance of these deficits is multifactorial, the effects of interictal epileptiform discharges (IEDs) and anti-seizure medications (ASMs) are thought to be particularly severe. Although certain ASMs can be leveraged to inhibit IED occurrence, it remains unclear whether epileptiform discharges or the medications themselves are most deleterious to cognition. To examine this question, 25 children undergoing invasive monitoring for refractory focal epilepsy performed one or more sessions of a cognitive flexibility task. Electrophysiological data were recorded to detect IEDs. Between repeated sessions, prescribed ASMs were either continued or titrated to <50% of the baseline dose. Hierarchical mixed-effects modeling assessed the relationship between task reaction time (RT), IED occurrence, ASM type, and dose while controlling for seizure frequency. Both presence (ß ± SE = 49.91 ± 16.55 ms, p = .003) and number of IEDs (ß ± SE = 49.84 ± 12.51 ms, p < .001) were associated with slowed task RT. Higher dose oxcarbazepine significantly reduced IED frequency (p = .009) and improved task performance (ß ± SE = -107.43 ± 39.54 ms, p = .007). These results emphasize the neurocognitive consequences of IEDs independent of seizure effects. Furthermore, we demonstrate that inhibition of IEDs following treatment with select ASMs is associated with improved neurocognitive function.

Electrocorticography-Guided Resection Enhances Postoperative Seizure Freedom in Low-Grade Tumor-Associated Epilepsy: A Systematic Review and Meta-Analysis.

BACKGROUND: Low-grade cerebral neoplasms are commonly associated with medically intractable epilepsy. Despite increasing evidence that epileptogenic brain regions commonly extend beyond visible tumor margins, the utility of extended surgical resections leveraging intraoperative electrocorticography (ECoG) remains unclear. OBJECTIVE: To determine whether ECoG-guided surgery is associated with improved postoperative seizure control. METHODS: We performed a systematic review and meta-analysis encompassing both adult and pediatric populations. The primary outcome measure was postoperative seizure freedom as defined by Engel class I outcome. Class I/II outcome served as a secondary measure. Relevant clinical and operative data were recorded. A random-effects meta-analysis based on the pooled odds ratio (OR) of seizure freedom was performed on studies that reported comparative data between ECoG-guided surgery and lesionectomy. RESULTS: A total of 31 studies encompassing 1115 patients with medically refractory epilepsy met inclusion criteria. Seven studies reported comparative data between ECoG-guided surgery and lesionectomy for meta-analysis. Tumor resection guided by ECoG was associated with significantly greater postoperative seizure freedom (OR 3.95, 95% CI 2.32-6.72, P < .0001) and class I/II outcome (OR 5.10, 95% CI 1.97-13.18, P = .0008) compared with lesionectomy. Postoperative adverse events were rare in both groups. CONCLUSION: These findings provide support for the utilization of ECoG-guided surgery to improve postoperative seizure freedom in cases of refractory epilepsy associated with low-grade neoplasms. However, this effect may be attenuated in the presence of concomitant cortical dysplasia, highlighting a need for improved presurgical and intraoperative monitoring for these most challenging cases of localization-related epilepsy.

Vagus Nerve Stimulation Modulates Phase-Amplitude Coupling in Thalamic Local Field Potentials.

OBJECTIVE: The antiseizure effects of vagus nerve stimulation (VNS) are thought to be mediated by the modulation of afferent thalamocortical circuitry. Cross-frequency phase-amplitude coupling (PAC) is a mechanism of hierarchical network coordination across multiple spatiotemporal scales. In this study, we leverage local field potential (LFP) recordings from the centromedian (CM) (n = 3) and anterior (ATN) (n = 2) nuclei in five patients with tandem thalamic deep brain stimulation and VNS to study neurophysiological changes in the thalamus in response to VNS. MATERIALS AND METHODS: Bipolar LFP data were recorded from contact pairs spanning target nuclei in VNS "on" and "off" states. RESULTS: Active VNS was associated with increased PAC between theta, alpha, and beta phase and gamma amplitude in CM (q < 0.05). Within the ATN, PAC changes also were observed, although these were less robust. In both nuclei, active VNS also modulated interhemispheric bithalamic functional connectivity. CONCLUSIONS: We report that VNS is associated with enhanced PAC and coordinated interhemispheric interactions within and between thalamic nuclei, respectively. These findings advance understanding of putative neurophysiological effects of acute VNS and contextualize previous animal and human studies showing distributed cortical synchronization after VNS.

Epilepsy disrupts hippocampal phase precision and impairs working memory.

OBJECTIVE: Working memory deficits are prevalent in childhood epilepsy. Working memory processing is thought to be supported by the phase of hippocampal neural oscillations. Disruptions in working memory have previously been linked to the occurrence of transient epileptic activity. This study aimed to resolve the associations between oscillatory neural activity, transient epileptiform events, and working memory in children with epilepsy. METHODS: Intracranial recordings were acquired from stereotactically implanted electrodes in the hippocampi, epileptogenic zones, and working memory-related networks of children with drug-resistant epilepsy during a 1-back working memory task. Interictal epileptic activity was captured using automated detectors. Hippocampal phase and interregional connectivity within working memory networks were indexed by Rayleigh Z and the phase difference derivative, respectively. Trials with and without transient epileptiform events were compared. RESULTS: Twelve children (mean age = 14.3 ± 2.8 years) with drug-resistant epilepsy were included in the study. In the absence of transient epileptic activity, significant delta and theta hippocampal phase resetting occurred in response to working memory stimulus presentation (Rayleigh z-score = 9, Rayleigh z-score = 8). Retrieval trials that were in phase with the preferred phase angle were associated with faster reaction times (p = .01, p = .03). Concurrently, delta and theta coordinated interactions between the hippocampi and working memory-related networks were enhanced (phase difference derivative [PDD] z-scores = 6-11). During retrieval trials with pre-encoding or pre-retrieval transient epileptic activity, phase resetting was attenuated (Rayleigh z-score = 5, Rayleigh z-score = 1), interregional connectivity was altered (PDD z-scores = 1-3), and reaction times were prolonged (p = .01, p = .03). SIGNIFICANCE: This work highlights the role of hippocampal phase in working memory. We observe poststimulus hippocampal phase resetting coincident with enhanced interregional connectivity. The precision of hippocampal phase predicts optimal working memory processing, and transient epileptic activity prolongs working memory processing. These findings can help guide future treatments aimed at restoring memory function in this patient population.

Interictal discharges delay target-directed eye movements and impair attentional set-shifting in children with epilepsy.

OBJECTIVE: The theory of transient cognitive impairment in epilepsy posits that lapses in attention result from ephemeral disruption of attentional circuitry by interictal events. Eye movements are intimately associated with human attention and can be monitored in real time using eye-tracking technologies. Here, we sought to characterize the associations between interictal epileptiform discharges (IEDs), gaze, and attentional behavior in children with epilepsy. METHODS: Eleven consecutive children undergoing invasive monitoring with stereotactic electrodes for localization-related epilepsy performed an attentional set-shifting task while tandem intracranial electroencephalographic signals and eye-tracking data were recorded. Using an established algorithm, IEDs were detected across all intracranial electrodes on a trial-by-trial basis. Hierarchical mixed-effects modeling was performed to delineate associations between trial reaction time (RT), eye movements, and IEDs. RESULTS: Hierarchical mixed-effects modeling revealed that both the presence of an IED (ß ± SE = 72.74 ± 24.21 ms, p = .003) and the frequency of epileptiform events (ß ± SE = 67.54 ± 17.30 ms, p < .001) were associated with prolonged RT on the attentional set-shifting task. IED occurrence at the time of stimulus presentation was associated with delays in gaze initiation toward the visual targets (p = .017). SIGNIFICANCE: The occurrence of epileptiform activity in close temporal association with stimulus presentation is associated with delays in target-directed gaze and prolonged response time, hallmarks of momentary lapses in attention. These findings provide novel insights into the mechanisms of transient impairments in children and support the use of visual tracking as a correlate of higher order attentional behavior.

Decoding Intracranial EEG With Machine Learning: A Systematic Review.

Advances in intracranial electroencephalography (iEEG) and neurophysiology have enabled the study of previously inaccessible brain regions with high fidelity temporal and spatial resolution. Studies of iEEG have revealed a rich neural code subserving healthy brain function and which fails in disease states. Machine learning (ML), a form of artificial intelligence, is a modern tool that may be able to better decode complex neural signals and enhance interpretation of these data. To date, a number of publications have applied ML to iEEG, but clinician awareness of these techniques and their relevance to neurosurgery, has been limited. The present work presents a review of existing applications of ML techniques in iEEG data, discusses the relative merits and limitations of the various approaches, and examines potential avenues for clinical translation in neurosurgery. One-hundred-seven articles examining artificial intelligence applications to iEEG were identified from 3 databases. Clinical applications of ML from these articles were categorized into 4 domains: i) seizure analysis, ii) motor tasks, iii) cognitive assessment, and iv) sleep staging. The review revealed that supervised algorithms were most commonly used across studies and often leveraged publicly available timeseries datasets. We conclude with recommendations for future work and potential clinical applications.

Atypical Functional Connectivity During Unfamiliar Music Listening in Children With Autism.

Background: Atypical processing of unfamiliar, but less so familiar, stimuli has been described in Autism Spectrum Disorder (ASD), in particular in relation to face processing. We examined the construct of familiarity in ASD using familiar and unfamiliar songs, to investigate the link between familiarity and autism symptoms, such as repetitive behavior. Methods: Forty-eight children, 24 with ASD (21 males, mean age = 9.96 years ± 1.54) and 24 typically developing (TD) controls (21 males, mean age = 10.17 ± 1.90) completed a music familiarity task using individually identified familiar compared to unfamiliar songs, while magnetoencephalography (MEG) was recorded. Each song was presented for 30 s. We used both amplitude envelope correlation (AEC) and the weighted phase lag index (wPLI) to assess functional connectivity between specific regions of interest (ROI) and non-ROI parcels, as well as at the whole brain level, to understand what is preserved and what is impaired in familiar music listening in this population. Results: Increased wPLI synchronization for familiar vs. unfamiliar music was found for typically developing children in the gamma frequency. There were no significant differences within the ASD group for this comparison. During the processing of unfamiliar music, we demonstrated left lateralized increased theta and beta band connectivity in children with ASD compared to controls. An interaction effect found greater alpha band connectivity in the TD group compared to ASD to unfamiliar music only, anchored in the left insula. Conclusion: Our results revealed atypical processing of unfamiliar songs in children with ASD, consistent with previous studies in other modalities reporting that processing novelty is a challenge for ASD. Relatively typical processing of familiar stimuli may represent a strength and may be of interest to strength-based intervention planning.

The anterior and centromedian thalamus: Anatomy, function, and dysfunction in epilepsy.

The thalamus is a densely connected collection of nuclei that play a critical role in gating information flow across the neocortex. Through diffuse reciprocal cortico-thalamo-cortical connectivity, the anterior and centromedian nuclei exert remarkable control over cortically expressed activity. Consequently, mounting evidence implicates these thalamic centres in both the genesis and propagation of aberrant epileptiform activity across the brain. The present work reviews existing literature with regards to the anatomy, function, and dysfunction of the anterior and centromedian thalamic nuclei as they relate to epileptogenesis and ictal dynamics in humans. A confluence of electrophysiological, anatomical, and neuromodulatory evidence links these thalamic hubs to a variety of epilepsy syndromes. These data are discussed as they relate to targeted thalamic neuromodulation.

Connectomic Profiles and Cognitive Trajectories After Epilepsy Surgery in Children.

BACKGROUND AND OBJECTIVES: Neurocognitive outcomes after surgery for temporal lobe epilepsy in childhood are variable. Postoperative changes are not directly predicted by seizure freedom, and associations between epilepsy, neuropsychological function, and developing neural networks are poorly understood. Here, we leveraged whole-brain connectomic profiling in magnetoencephalography (MEG) to retrospectively study associations between brain connectivity and neuropsychological function in children with temporal lobe epilepsy undergoing resective surgery. METHODS: Clinical and MEG data were retrospectively analyzed for children who underwent temporal lobe epilepsy surgery at the Hospital for Sick Children from 2000 to 2021. Resting-state connectomes were constructed from neuromagnetic oscillations via the weighted-phase lag index. Using a partial least-squares (PLS) approach, we assessed multidimensional associations between patient connectomes, neuropsychological scores, and clinical covariates. Bootstrap resampling statistics were performed to assess statistical significance. RESULTS: A total of 133 medical records were reviewed, and 5 PLS analyses were performed. Each PLS analysis probed a particular neuropsychological domain and the associations between its baseline and postoperative scores and the connectomic data. In each PLS analysis, a significant latent variable was identified, representing a specific percentage of the variance in the data and relating neural networks to clinical covariates, which included changes in rote verbal memory (n = 41, p = 0.01, σ2 = 0.38), narrative/verbal memory (n = 57, p = 0.00, σ2 = 0.52), visual memory (n = 51, p = 0.00, σ2 = 0.43), working memory (n = 44, p = 0.00, σ2 = 0.52), and overall intellectual function (n = 59, p = 0.00, σ2 = 0.55). Children with more diffuse, bilateral intrinsic connectivity across several frequency bands showed lower scores on all neuropsychological assessments but demonstrated a greater propensity for gains after resective surgery. DISCUSSION: Here, we report that connectomes characterized by diffuse connectivity, reminiscent of developmentally immature networks, are associated with lower preoperative cognition and postoperative cognitive improvement. These findings provide a potential means to understand neurocognitive function in children with temporal lobe epilepsy and expected changes postoperatively.

Very preterm brain at rest: longitudinal social-cognitive network connectivity during childhood.

Very preterm (VPT: ≤32 weeks of gestational age) birth poses an increased risk for social and cognitive morbidities that persist throughout life. Resting-state functional network connectivity studies provide information about the intrinsic capacity for cognitive processing. We studied the following four social-cognitive resting-state networks: the default mode, salience, frontal-parietal and language networks. We examined functional connectivity using magnetoencephalography with individual head localization using each participant's MRI at 6 (n = 40) and 8 (n = 40) years of age compared to age- and sex-matched full-term (FT) born children (n = 38 at 6 years and n = 43 at 8 years). VPT children showed increased connectivity compared to FT children in the gamma band (30-80 Hz) at 6 years within the default mode network (DMN), and between the DMN and the salience, frontal-parietal and language networks, pointing to more diffuse, less segregated processing across networks at this age. At 8 years, VPT children had more social and academic difficulties. Increased DMN connectivity at 6 years was associated with social and working memory difficulties at 8 years. Therefore, we suggest that increased DMN connectivity contributes to the observed emerging social and cognitive morbidities in school age.

Spectral changes following resective epilepsy surgery and neurocognitive function in children with epilepsy.

Decelerated resting cortical oscillations, high-frequency activity, and enhanced cross-frequency interactions are features of focal epilepsy. The association between electrophysiological signal properties and neurocognitive function, particularly following resective surgery, is, however, unclear. In the current report, we studied intraoperative recordings from intracranial electrodes implanted in seven children with focal epilepsy and analyzed the spectral dynamics both before and after surgical resection of the hypothesized seizure focus. The associations between electrophysiological spectral signatures and each child's neurocognitive profiles were characterized using a partial least squares analysis. We find that extent of spectral alteration at the periphery of surgical resection, as indexed by slowed resting frequency and its acceleration following surgery, is associated with baseline cognitive deficits in children. The current report provides evidence supporting the relationship between altered spectral properties in focal epilepsy and neuropsychological deficits in children. In particular, these findings suggest a critical role of disrupted thalamocortical rhythms, which are believed to underlie the spectral alterations we describe, in both epileptogenicity and neurocognitive function.NEW & NOTEWORTHY Spectral alterations marked by decelerated resting oscillations and ectopic high-frequency activity have been noted in focal epilepsy. We leveraged intraoperative recordings from chronically implanted electrodes pre- and postresection to understand the association between these electrophysiological phenomena and neuropsychological function. We find that the extent of spectral alteration, indexed by slowed resting frequency and its acceleration following resection, is associated with baseline cognitive deficits. These findings provide novel insights into neurocognitive impairments in focal epilepsy.

Phase Resetting in the Anterior Cingulate Cortex Subserves Childhood Attention and Is Impaired by Epilepsy.

The neural mechanisms that underlie selective attention in children are poorly understood. By administering a set-shifting task to children with intracranial electrodes stereotactically implanted within anterior cingulate cortex (ACC) for epilepsy monitoring, we demonstrate that selective attention in a set-shifting task is dependent upon theta-band phase resetting immediately following stimulus onset and that the preferred theta phase angle is predictive of reaction time during attentional shift. We also observe selective enhancement of oscillatory coupling between the ACC and the dorsal attention network and decoupling with the default mode network during task performance. When transient focal epileptic activity occurs around the time of stimulus onset, phase resetting is impaired, connectivity changes with attentional and default mode networks are abolished, and reaction times are prolonged. The results of the present work highlight the fundamental mechanistic role of oscillatory phase in ACC in supporting attentional circuitry and present novel opportunities to remediate attention deficits in children with epilepsy.

A review of magnetoencephalography use in pediatric epilepsy: an update on best practice.

Introduction: Magnetoencephalography (MEG) is a noninvasive technique that is used for presurgical evaluation of children with drug-resistant epilepsy (DRE).Areas covered: The contributions of MEG for localizing the epileptogenic zone are discussed, in particular in extra-temporal lobe epilepsy and focal cortical dysplasia, which are common in children, as well as in difficult to localize epilepsy such as operculo-insular epilepsy. Further, the authors review current evidence on MEG for mapping eloquent cortex, its performance, application in clinical practice, and potential challenges.Expert opinion: MEG could change the clinical management of children with DRE by directing placement of intracranial electrodes thereby enhancing their yield. With improved identification of a circumscribed epileptogenic zone, MEG could render more patients as suitable candidates for epilepsy surgery and increase utilization of surgery.

Detection of high-frequency oscillations in electroencephalography: A scoping review and an adaptable open-source framework.

PURPOSE: High frequency oscillations (HFOs) are putative biomarkers of epileptogenicity. These electrophysiological phenomena can be effectively detected in electroencephalography using automated methods. Nonetheless, the implementation of these methods into clinical practice remains challenging as significant variability exists between algorithms and their characterizations of HFOs. Here, we perform a scoping review of the literature pertaining to automated HFO detection methods. In addition, we propose a framework for defining and detecting HFOs based on a simplified single-stage time-frequency based detection algorithm with clinically-familiar parameters. METHODS: Several databases (OVID Medline, Web of Science, PubMed) were searched for articles presenting novel, automated HFO detection methods. Details related to the algorithm and various stages of data acquisition, pre-processing, and analysis were abstracted from included studies. RESULTS: From the 261 records screened, 57 articles presented novel, automated HFO detection methods and were included in the scoping review. These algorithms were categorized into 3 groups based on their most salient features: energy thresholding, time-frequency analysis, and data mining/machine learning. Algorithms were optimized for specific datasets and suffered from low specificity. A framework for user-constrained inputs is proposed to circumvent some of the weaknesses of highly performant detectors. CONCLUSIONS: Further efforts are required to optimize and validate existing automated HFO detection methods for clinical utility. The proposed framework may be applied to understand and standardize the variations in HFO definitions across institutions.

Predicting PTSD severity using longitudinal magnetoencephalography with a multi-step learning framework.

Objective. The present study explores the effectiveness of incorporating temporal information in predicting post-traumatic stress disorder (PTSD) severity using magnetoencephalography (MEG) imaging data. The main objective was to assess the relationship between longitudinal MEG functional connectome data, measured across a variety of neural oscillatory frequencies and collected at two timepoints (Phase I and II), against PTSD severity captured at the later time point.Approach. We used an in-house developed informatics solution, featuring a two-step process featuring pre-learn feature selection (CV-SVR-rRF-FS, cross-validation with support vector regression (SVR) and recursive random forest feature selection) and deep learning (long-short term memory recurrent neural network, LSTM-RNN) techniques.Main results. The pre-learn step selected a small number of functional connections (or edges) from Phase I MEG data associated with Phase II PTSD severity, indexed using the PTSD CheckList (PCL) score. This strategy identified the functional edges affected by traumatic exposure and indexed disease severity, either permanently or evolving dynamically over time, for optimal predictive performance. Using the selected functional edges, LSTM modelling was used to incorporate the Phase II MEG data into longitudinal regression models. Single timepoint (Phase I and Phase II MEG data) SVR models were generated for comparison. Assessed with holdout test data, alpha and high gamma bands showed enhanced predictive performance with the longitudinal models comparing to the Phase I single timepoint models. The best predictive performance was observed for lower frequency ranges compared to the higher frequencies (low gamma), for both model types.Significance. This study identified the neural oscillatory signatures that benefited from additional temporal information when estimating the outcome of PTSD severity using MEG functional connectome data. Crucially, this approach can similarly be applied to any other mental health challenge, using this effective informatics foundation for longitudinal tracking of pathological brain states and predicting outcome with a MEG-based neurophysiology imaging system.