Accurate prediction of neurologic changes in critically ill infants using pose AI.

Importance: Infant alertness and neurologic changes are assessed by exam, which can be intermittent and subjective. Reliable, continuous methods are needed. Objective: We hypothesized that our computer vision method to track movement, pose AI, could predict neurologic changes. Design: Retrospective observational study from 2021-2022. Setting: A level four urban neonatal intensive care unit (NICU). Participants: Infants with corrected age ≤1 year, comprising 115 patients with 4,705 hours of video data linked to electroencephalograms (EEG), including 46% female and 25.2% white non-Hispanic. Exposures: Pose AI prediction of anatomic landmark position and an XGBoost classifier trained on one-minute variance in pose. Main outcomes and measures: Outcomes were cerebral dysfunction, diagnosed from EEG readings by an epileptologist, and sedation, defined by the administration of sedative medications. Measures of algorithm performance were receiver operating characteristic-area under the curves (ROC-AUCs) on cross-validation and on two test datasets comprised of held-out infants and held-out video frames from infants used in training. Results: Infant pose was accurately predicted in cross-validation, held-out frames, and held-out infants (respective ROC-AUCs 0.94, 0.83, 0.89). Median movement increased with age and, after accounting for age, was lower with sedative medications and in infants with cerebral dysfunction (all P<5×10-3, 10,000 permutations). Sedation prediction had high performance on cross-validation, held-out frames, and held-out infants (ROC-AUCs 0.90, 0.91, 0.87), as did prediction of cerebral dysfunction (ROC-AUCs 0.91, 0.90, 0.76). Conclusions and Relevance: We used pose AI to predict sedation and cerebral dysfunction in 4,705 hours of video from a large, diverse cohort of infants. Pose AI may offer a scalable, minimally invasive method for neuro-telemetry in the NICU.

Comparative analysis of patients with new onset refractory status epilepticus preceded by fever (febrile infection-related epilepsy syndrome) versus without prior fever: An interim analysis.

Febrile infection-related epilepsy syndrome (FIRES) is a subset of new onset refractory status epilepticus (NORSE) that involves a febrile infection prior to the onset of the refractory status epilepticus. It is unclear whether FIRES and non-FIRES NORSE are distinct conditions. Here, we compare 34 patients with FIRES to 30 patients with non-FIRES NORSE for demographics, clinical features, neuroimaging, and outcomes. Because patients with FIRES were younger than patients with non-FIRES NORSE (median = 28 vs. 48 years old, p = .048) and more likely cryptogenic (odds ratio = 6.89), we next ran a regression analysis using age or etiology as a covariate. Respiratory and gastrointestinal prodromes occurred more frequently in FIRES patients, but no difference was found for non-infection-related prodromes. Status epilepticus subtype, cerebrospinal fluid (CSF) and magnetic resonance imaging findings, and outcomes were similar. However, FIRES cases were more frequently cryptogenic; had higher CSF interleukin 6, CSF macrophage inflammatory protein-1 alpha (MIP-1a), and serum chemokine ligand 2 (CCL2) levels; and received more antiseizure medications and immunotherapy. After controlling for age or etiology, no differences were observed in presenting symptoms and signs or inflammatory biomarkers, suggesting that FIRES and non-FIRES NORSE are very similar conditions.

Elevated phase amplitude coupling as a depression biomarker in epilepsy.

Depression is prevalent in epilepsy patients and their intracranial brain activity recordings can be used to determine the types of brain activity that are associated with comorbid depression. We performed case-control comparison of spectral power and phase amplitude coupling (PAC) in 34 invasively monitored drug resistant epilepsy patients' brain recordings. The values of spectral power and PAC for one-minute segments out of every hour in a patient's study were correlated with pre-operative assessment of depressive symptoms by Beck Depression Inventory-II (BDI). We identified an elevated PAC signal (theta-alpha-beta phase (5-25 Hz)/gamma frequency (80-100 Hz) band) that is present in high BDI scores but not low BDI scores adult epilepsy patients in brain regions implicated in primary depression, including anterior cingulate cortex, amygdala and orbitofrontal cortex. Our results showed the application of PAC as a network-specific, electrophysiologic biomarker candidate for comorbid depression and its potential as treatment target for neuromodulation.

Responsive Neurostimulation for People With Drug-Resistant Epilepsy and Autism Spectrum Disorder.

PURPOSE: Individuals with autism spectrum disorder (ASD) have comorbid epilepsy at much higher rates than the general population, and about 30% will be refractory to medication. Patients with drug-resistant epilepsy (DRE) should be referred for surgical evaluation, yet many with ASD and DRE are not resective surgical candidates. The aim of this study was to examine the response of this population to the responsive neurostimulator (RNS) System. METHODS: This multicenter study evaluated patients with ASD and DRE who underwent RNS System placement. Patients were included if they had the RNS System placed for 1 year or more. Seizure reduction and behavioral outcomes were reported. Descriptive statistics were used for analysis. RESULTS: Nineteen patients with ASD and DRE had the RNS System placed at 5 centers. Patients were between the ages of 11 and 29 (median 20) years. Fourteen patients were male, whereas five were female. The device was implanted from 1 to 5 years. Sixty-three percent of all patients experienced a >50% seizure reduction, with 21% of those patients being classified as super responders (seizure reduction >90%). For the super responders, two of the four patients had the device implanted for >2 years. The response rate was 70% for those in whom the device was implanted for >2 years. Improvements in behaviors as measured by the Clinical Global Impression Scale-Improvement scale were noted in 79%. No complications from the surgery were reported. CONCLUSIONS: Based on the authors' experience in this small cohort of patients, the RNS System seems to be a promising surgical option in people with ASD-DRE.

A multicenter retrospective study of patients treated in the thalamus with responsive neurostimulation.

Introduction: For drug resistant epilepsy patients who are either not candidates for resective surgery or have already failed resective surgery, neuromodulation is a promising option. Neuromodulatory approaches include responsive neurostimulation (RNS), deep brain stimulation (DBS), and vagal nerve stimulation (VNS). Thalamocortical circuits are involved in both generalized and focal onset seizures. This paper explores the use of RNS in the centromedian nucleus of the thalamus (CMN) and in the anterior thalamic nucleus (ANT) of patients with drug resistant epilepsy. Methods: This is a retrospective multicenter study from seven different epilepsy centers in the United States. Patients that had unilateral or bilateral thalamic RNS leads implanted in the CMN or ANT for at least 6 months were included. Primary objectives were to describe the implant location and determine changes in the frequency of disabling seizures at 6 months, 1 year, 2 years, and > 2 years. Secondary objectives included documenting seizure free periods, anti-seizure medication regimen changes, stimulation side effects, and serious adverse events. In addition, the global clinical impression scale was completed. Results: Twelve patients had at least one lead placed in the CMN, and 13 had at least one lead placed in the ANT. The median baseline seizure frequency was 15 per month. Overall, the median seizure reduction was 33% at 6 months, 55% at 1 year, 65% at 2 years, and 74% at >2 years. Seizure free intervals of at least 3 months occurred in nine patients. Most patients (60%, 15/25) did not have a change in anti-seizure medications post RNS placement. Two serious adverse events were recorded, one related to RNS implantation. Lastly, overall functioning seemed to improve with 88% showing improvement on the global clinical impression scale. Discussion: Meaningful seizure reduction was observed in patients who suffer from drug resistant epilepsy with unilateral or bilateral RNS in either the ANT or CMN of the thalamus. Most patients remained on their pre-operative anti-seizure medication regimen. The device was well tolerated with few side effects. There were rare serious adverse events. Most patients showed an improvement in global clinical impression scores.

Long-term outcomes after responsive neurostimulation for treatment of refractory epilepsy: a single-center experience of 100 cases.

OBJECTIVE: Despite antiepileptic drugs, more than 30% of people with epilepsy continue to have seizures. Patients with such drug-resistant epilepsy (DRE) may undergo invasive treatment such as resection, laser ablation of the epileptogenic focus, or vagus nerve stimulation, but many are not candidates for epilepsy surgery or fail to respond to such interventions. Responsive neurostimulation (RNS) provides a neuromodulatory option. In this study, the authors present a single-center experience with the use of RNS over the last 5 years to provide long-term control of seizures in patients with DRE with at least 1 year of follow-up. METHODS: The authors performed a retrospective analysis of a prospectively collected single-center database of consecutive DRE patients who underwent RNS system implantation from September 2015 to December 2020. Patients were followed-up postoperatively to evaluate seizure freedom and complications. RESULTS: One hundred patients underwent RNS placement. Seven patients developed infections: 2 responded to intravenous antibiotic therapy, 3 required partial removal and salvaging of the system, and 2 required complete removal of the RNS device. No postoperative tract hemorrhages, strokes, device migrations, or malfunctions were documented in this cohort. The average follow-up period was 26.3 months (range 1-5.2 years). In terms of seizure reduction, 8 patients had 0%-24% improvement, 14 had 25%-49% improvement, 29 experienced 50%-74% improvement, 30 had 75%-99% improvement, and 19 achieved seizure freedom. RNS showed significantly better outcomes over time: patients with more than 3 years of RNS therapy had 1.8 higher odds of achieving 75% or more seizure reduction (95% CI 1.07-3.09, p = 0.02). Also, patients who had undergone resective or ablative surgery prior to RNS implantation had 8.25 higher odds of experiencing 50% or more seizure reduction (95% CI 1.05-65.1, p = 0.04). CONCLUSIONS: Responsive neurostimulator implantation achieved 50% or more seizure reduction in approximately 80% of patients. Even in patients who did not achieve seizure freedom, significant improvement in seizure duration, severity, or postictal state was reported in more than 68% of cases. Infection (7%) was the most common complication. Patients with prior resective or ablative procedures and those who had been treated with RNS for more than 3 years achieved better outcomes.

Mood and Anxiety Disorders and Suicidality in Patients With Newly Diagnosed Focal Epilepsy: An Analysis of a Complex Comorbidity.

BACKGROUND AND OBJECTIVES: Mood, anxiety disorders, and suicidality are more frequent in people with epilepsy than in the general population. Yet, their prevalence and the types of mood and anxiety disorders associated with suicidality at the time of the epilepsy diagnosis are not established. We sought to answer these questions in patients with newly diagnosed focal epilepsy and to assess their association with suicidal ideation and attempts. METHODS: The data were derived from the Human Epilepsy Project study. A total of 347 consecutive adults aged 18-60 years with newly diagnosed focal epilepsy were enrolled within 4 months of starting treatment. The types of mood and anxiety disorders were identified with the Mini International Neuropsychiatric Interview, whereas suicidal ideation (lifetime, current, active, and passive) and suicidal attempts (lifetime and current) were established with the Columbia Suicidality Severity Rating Scale (CSSRS). Statistical analyses included the t test, χ2 statistics, and logistic regression analyses. RESULTS: A total of 151 (43.5%) patients had a psychiatric diagnosis; 134 (38.6%) met the criteria for a mood and/or anxiety disorder, and 75 (21.6%) reported suicidal ideation with or without attempts. Mood (23.6%) and anxiety (27.4%) disorders had comparable prevalence rates, whereas both disorders occurred together in 43 patients (12.4%). Major depressive disorders (MDDs) had a slightly higher prevalence than bipolar disorders (BPDs) (9.5% vs 6.9%, respectively). Explanatory variables of suicidality included MDD, BPD, panic disorders, and agoraphobia, with BPD and panic disorders being the strongest variables, particularly for active suicidal ideation and suicidal attempts. DISCUSSION: In patients with newly diagnosed focal epilepsy, the prevalence of mood, anxiety disorders, and suicidality is higher than in the general population and comparable to those of patients with established epilepsy. Their recognition at the time of the initial epilepsy evaluation is of the essence.

Analysis of Electrocorticography in Epileptic Patients With Responsive Neurostimulation Undergoing Scalp Electroencephalography Monitoring.

PURPOSE: To describe the relationship of electrocorticography events detected by a brain-responsive neurostimulation system (RNS) and their association with ictal and interictal activity detected on simultaneous scalp EEG. METHODS: We retrospectively identified patients with drug-resistant epilepsy implanted with RNS who subsequently underwent long-term scalp EEG monitoring. RNS detections were correlated to simultaneous activity recorded on scalp EEG to determine the characteristics of electrocorticography-stored long episodes associated with seizures or other findings on scalp EEG. RESULTS: Eleven patients were included with an average of 3.6 days of monitoring. Most RNS detections were of very brief duration (<10 seconds, 92.9%) and received one stimulation therapy (80.8%). A high proportion of long episodes (67.1%) were not identified as electrographic seizures on scalp EEG. Of those ictal-appearing (71.2%) long episodes, 68.2% had seizure correlates. Long episodes associated with seizures on scalp EEG had a longer median duration compared with those without (39.7 vs. 16.8 seconds, P < 0.002) and had broader spread pattern and were of higher amplitude on electrocorticography. Brief potentially ictal rhythmic discharges were the most common EEG findings associated with long episodes that did not have scalp EEG seizure correlates (100% for ictal- and 50% for non-ictal-appearing long episodes). CONCLUSIONS: Longer, broader spread and higher amplitude intracranial RNS detections are more likely to manifest as electrographic seizures on scalp EEG. Brief potentially ictal rhythmic discharges may serve as a scalp EEG biomarker of ictal intracranial episodes that are detected as long episodes by the RNS but not identified as electrographic seizures on scalp EEG.

Responsive Neurostimulation of the Thalamus for the Treatment of Refractory Epilepsy.

Introduction: One-third of patients with epilepsy continue to have seizures despite antiepileptic medications. Some of these refractory patients may not be candidates for surgical resection primarily because the seizure onset zones (SOZs) involve both hemispheres or are located in eloquent areas. The NeuroPace Responsive Neurostimulation System (RNS) is a closed-loop device that uses programmable detection and stimulation to tailor therapy to a patient's individual neurophysiology. Here, we present our single-center experience with the use of RNS in thalamic nuclei to provide long-term seizure control in patients with refractory epilepsy. Methods: We performed a prospective single-center study of consecutive refractory epilepsy patients who underwent RNS system implantation in the anterior (ANT) and centromedian (CM) thalamic nuclei from September 2015 to December 2020. Patients were followed postoperatively to evaluate seizure freedom and complications. Results: Twenty-three patients underwent placement of 36 RNS thalamic leads (CM = 27 leads, ANT = 9 leads). Mean age at implant was 18.8 ± 11.2 years (range 7.8-62 years-old). Two patients (8.7%) developed infections: 1 improved with antibiotic treatments alone, and 1 required removal with eventual replacement of the system to recover the therapeutic benefit. Mean time from RNS implantation to last follow-up was 22.3 months. Based on overall reduction of seizure frequency, 2 patients (8.7%) had no- to <25% improvement, 6 patients (26.1%) had 25-49% improvement, 14 patients (60.9%) had 50-99% improvement, and 1 patient (4.3%) became seizure-free. All patients reported significant improvement in seizure duration and severity, and 17 patients (74%) reported improved post-ictal state. There was a trend for subjects with SOZs located in the temporal lobe to achieve better outcomes after thalamic RNS compared to those with extratemporal SOZs. Of note, seizure etiology was syndromic in 12 cases (52.2%), and 7 patients (30.4%) had undergone resection/disconnection surgery prior to thalamic RNS therapy. Conclusion: Thalamic RNS achieved ≥50% seizure control in ~65% of patients. Infections were the most common complication. This therapeutic modality may be particularly useful for patients affected by aggressive epilepsy syndromes since a young age, those whose seizure foci are located in the mesial temporal lobe, and those who have failed prior surgical interventions.

Refractory Seizures Secondary to Vitamin B6 Deficiency in Parkinson Disease: The Role of Carbidopa-Levodopa.

Carbidopa-levodopa has been used for more than 50 years in the treatment of Parkinson disease (PD) and other movement disorders. Pyridoxal 5'-phosphate (PLP), an active form of vitamin B6 (pyridoxine), is involved in the decarboxylation of levodopa to dopamine; carbidopa, which is combined with levodopa to reduce peripheral levodopa conversion and minimize peripheral dopamine side effects, binds irreversibly with PLP. As a result, carbidopa-levodopa may cause vitamin B6 deficiency and associated sequelae, including seizures, especially in high doses. A 78-year-old gentleman with a 6-year history of PD on carbidopa-levodopa therapy and recent weight loss presented with new-onset myoclonus and focal to bilateral tonic-clonic seizures. Workup for vascular, infectious, malignant, metabolic, and autoimmune causes of seizure was unrevealing. The folate level was critically low at <2.20 ng/dL. Video EEG studies showed moderate cerebral dysfunction and seizures with diffuse onsets. Several anti-seizure medications (ASMs) were unsuccessfully tried, so empiric treatment with high-dose steroids was initiated eventually alongside intravenous vitamin B6 therapy. Following introduction of these interventions, the patient had no further epileptic events. The vitamin B6 level came back as undetectable at <1 µg/dL. The patient was discharged to a rehabilitation center for improved strength and function. At the time of writing, he remained on two ASMs as well as IV B6 supplementation. Vitamin B6 is a required cofactor in the decarboxylation of levodopa to dopamine, and high levodopa dosages may cause B6 deficiency; in addition, carbidopa binds B6 irreversibly. We recommend screening of vitamin B6 levels in PD patients, especially those requiring high or increasing doses of carbidopa-levodopa and those with poor nutrition.

Quantification of brain age using high-resolution 7 tesla MR imaging and implications for patients with epilepsy.

Purpose: Epilepsy patients exhibit morphological differences on neuroimaging compared to age-matched healthy controls, including cortical and sub-cortical volume loss and altered gray-white matter ratios. The objective was to develop a model of normal aging using the 7T MRIs of healthy controls. This model can then be used to determine if the changes in epilepsy patients resemble the changes seen in aging, and potentially give a marker for the severity of those changes. Methods: Sixty-nine healthy controls (24F/45M, mean age 36.5 ± 10.5 years) and forty-four epilepsy patients (24F/20M, 33.2 ± 9.9 years) non-lesional at 3T were scanned with volumetric T1-MPRAGE at 7T. These images were segmented and quantified using FreeSurfer. A linear regression-based model trained on healthy controls was developed to predict ages using derived imaging features among the epilepsy patient cohort. The model used 114 features with significant linear correlation with age. Results: The regression-based model estimated brain age with mean absolute error (MAE) of 6.6 years among controls. Comparable prediction accuracy of 6.9 years MAE was seen epilepsy patients. T-test of mean absolute error showed no difference in the prediction accuracy with controls and epilepsy patients (p = 0.68). However, average signed error showed elevated (+5.0 years, p = 0.0007) predicted age differences (PAD; brain-PAD=, predicted minus biological age) among epilepsy patients. Morphological metrics in the medial temporal lobe were major contributors to PAD. Additionally, patients with seizure frequency greater than once a week showed significantly elevated brain-PAD (+8.2 ± 5.3 years, n = 13) compared to patients with lower seizure frequency (3.7 ± 6.5 years, n = 31, p = 0.033). Major conclusions: Morphological patterns suggestive of premature aging were observed in non-lesional epilepsy patients vs. controls and in high seizure frequency patients vs. low frequency patients. Modeling brain age with 7T MRI may provide a sensitive imaging marker to assess the differential effects of the aging process in diseases such as epilepsy.

Electroencephalography at the height of a pandemic: EEG findings in patients with COVID-19.

OBJECTIVE: To characterize continuous video electroencephalogram (VEEG) findings of hospitalized COVID-19 patients. METHODS: We performed a retrospective chart review of patients admitted at three New York City hospitals who underwent VEEG at the peak of the COVID-19 pandemic. Demographics, comorbidities, neuroimaging, VEEG indications and findings, treatment, and outcomes were collected. RESULTS: Of 93 patients monitored, 77% had severe COVID-19 and 40% died. Acute ischemic or hemorrhagic stroke was present in 26% and 15%, respectively. Most common VEEG indications were encephalopathy/coma (60%) and seizure-like movements (38%). Most common VEEG findings were generalized slowing (97%), generalized attenuation (31%), generalized periodic discharges (17%) and generalized sharp waves (15%). Epileptiform abnormalities were present in 43% and seizures in 8% of patients, all of whom had seizure risk factors. Factors associated with an epileptiform VEEG included increasing age (OR 1.07, p = 0.001) and hepatic/renal failure (OR 2.99, p = 0.03). CONCLUSIONS: Most COVID-19 patients who underwent VEEG monitoring had severe COVID-19 and over one-third had acute cerebral injury (e.g., stroke, anoxia). Seizures were uncommon. VEEG findings were nonspecific. SIGNIFICANCE: VEEG findings in this cohort of hospitalized COVID-19 patients were those often seen in critical illness. Seizures were uncommon and occurred in the setting of common seizure risk factors.

The Sensitivity of Scalp EEG at Detecting Seizures-A Simultaneous Scalp and Stereo EEG Study.

PURPOSE: Compare the detection rate of seizures on scalp EEG with simultaneous intracranial stereo EEG (SEEG) recordings. METHODS: Twenty-seven drug-resistant epilepsy patients undergoing SEEG with simultaneous scalp EEG as part of their surgical work-up were included. A total of 172 seizures were captured. RESULTS: Of the 172 seizures detected on SEEG, 100 demonstrated scalp ictal patterns. Focal aware and subclinical seizures were less likely to be seen on scalp, with 33% of each observed when compared with focal impaired aware (97%) and focal to bilateral tonic-clonic seizures (100%) (P < 0.001). Of the 72 seizures without ictal scalp correlate, 32 demonstrated an abnormality during the SEEG seizure that was identical to an interictal abnormality. Seizures from patients with MRI lesions were statistically less likely to be seen on scalp than seizures from nonlesional patients (P = 0.0162). Stereo EEG seizures not seen on scalp were shorter in duration (49 seconds) compared with SEEG seizures seen on scalp (108.6 seconds) (P < 0.001). CONCLUSIONS: Scalp EEG is not a sensitive tool for the detection of focal aware and subclinical seizures but is highly sensitive for the detection of focal impaired aware and focal to bilateral tonic-clonic seizures. Longer duration of seizure and seizures from patients without MRI lesions were more likely to be apparent on scalp. Abnormalities seen interictally may at times represent an underlying seizure. The cognitive, affective, and behavioral long-term effects of ongoing difficult-to-detect seizures are not known.

Ictal head roll: a seizure semiology from the anterior prefrontal lobe.

Longstanding epilepsy can lead to modulation of cortical networks over time and unexpected seizure onset zones. Frontal lobe seizures, in particular, can have diverse semiologies and evolution patterns. We present a male patient with drug-resistant epilepsy secondary to severe traumatic brain injury who underwent bilateral stereo electroencephalography (SEEG) for surgical planning. SEEG localized an ictal circular head roll to the right anterior prefrontal region. This was followed by spread to the left orbitofrontal region and later the left amygdala and hippocampus, at which point a different semiology with behavioral arrest, lip smacking and oral automatisms began. This case, in which an ictal circular head roll was localized to the anterior prefrontal region, demonstrates the complexity of broad seizure networks that develop over time, leading to remote seizure spread.

Detection of Hippocampal Subfield Asymmetry at 7T With Automated Segmentation in Epilepsy Patients With Normal Clinical Strength MRIs.

While the etiology of hippocampal sclerosis (HS) in epilepsy patients remains unknown, distinct phenotypes of hippocampal subfield atrophy have been associated with different clinical presentations and surgical outcomes. The advent of novel techniques including ultra-high field 7T magnetic resonance imaging (MRI) and automated subfield volumetry have further enabled detection of hippocampal pathology in patients with epilepsy, however, studies combining both 7T MRI and automated segmentation in epilepsy patients with normal-appearing clinical MRI are limited. In this study, we present a novel application of the automated segmentation of hippocampal subfields (ASHS) software to determine subfield volumes of the CA1, CA2/3, CA4/DG, and the subiculum using ultra high-field 7T MRI scans, including T1-weighted MP2RAGE and T2-TSE sequences, in 27 patients with either mesial temporal lobe epilepsy (mTLE) or neocortical epilepsy (NE) compared to age and gender matched healthy controls. We found that 7T improved visualization of structural abnormalities not otherwise seen on clinical strength MRIs in patients with unilateral mTLE. Additionally, our automated segmentation algorithm was able to detect structural differences in volume and asymmetry across hippocampal subfields in unilateral mTLE patients compared to controls. Specifically, amongst unilateral mTLE patients with longer disease durations, volume loss was observed in the ipsilateral CA1 and CA2/3 subfields and contralateral CA1. There were no differences in subfield volumes in patients with NE compared to controls. We report the first application of 7T with automated segmentation to characterize the relationship between disease duration burden and asymmetry across specific hippocampal subfields in this population. Disease duration was found to have a statistically significant positive relationship with subfield asymmetry within the unilateral mTLE cohort. These findings highlight the ability of 7T MRI and automated segmentation to provide novel qualitative and quantitative information in epilepsy patients who are otherwise MRI-negative at clinical field strengths.

Deep Learning of Simultaneous Intracranial and Scalp EEG for Prediction, Detection, and Lateralization of Mesial Temporal Lobe Seizures.

In people with drug resistant epilepsy (DRE), seizures are unpredictable, often occurring with little or no warning. The unpredictability causes anxiety and much of the morbidity and mortality of seizures. In this work, 102 seizures of mesial temporal lobe onset were analyzed from 19 patients with DRE who had simultaneous intracranial EEG (iEEG) and scalp EEG as part of their surgical evaluation. The first aim of this paper was to develop machine learning models for seizure prediction and detection (i) using iEEG only, (ii) scalp EEG only and (iii) jointly analyzing both iEEG and scalp EEG. The second goal was to test if machine learning could detect a seizure on scalp EEG when that seizure was not detectable by the human eye (surface negative) but was seen in iEEG. The final question was to determine if the deep learning algorithm could correctly lateralize the seizure onset. The seizure detection and prediction problems were addressed jointly by training Deep Neural Networks (DNN) on 4 classes: non-seizure, pre-seizure, left mesial temporal onset seizure and right mesial temporal onset seizure. To address these aims, the classification accuracy was tested using two deep neural networks (DNN) against 3 different types of similarity graphs which used different time series of EEG data. The convolutional neural network (CNN) with the Waxman similarity graph yielded the highest accuracy across all EEG data (iEEG, scalp EEG and combined). Specifically, 1 second epochs of EEG were correctly assigned to their seizure, pre-seizure, or non-seizure category over 98% of the time. Importantly, the pre-seizure state was classified correctly in the vast majority of epochs (>97%). Detection from scalp EEG data alone of surface negative seizures and the seizures with the delayed scalp onset (the surface negative portion) was over 97%. In addition, the model accurately lateralized all of the seizures from scalp data, including the surface negative seizures. This work suggests that highly accurate seizure prediction and detection is feasible using either intracranial or scalp EEG data. Furthermore, surface negative seizures can be accurately predicted, detected and lateralized with machine learning even when they are not visible to the human eye.

Utility of 7 tesla MRI brain in 16 "MRI Negative" epilepsy patients and their surgical outcomes.

The objective is to quantitatively assess surgical outcomes in epilepsy patients who underwent scanning at 7T MRI whose lesions were undetectable at conventional field strengths (1.5T/3T). 16 patients who underwent an initial 1.5T/3T scan that was marked as non-lesional by a neuroradiologist and were candidates for epilepsy surgery were scanned at 7T. The 7T findings were evaluated by an expert neuroradiologist blinded to the suspected seizure onset zone (sSOZ). The relation of the neuroradiologist's findings compared with the sSOZ was classified as non-definite (no 7T lesion or lesion of no epileptogenic significance, or lesion of epileptogenic potential which localizes to the patient's sSOZ but is not the definitive cause), or definite (7T lesion of epileptogenic potential that highly localizes to the sSOZ and is confirmed through surgical intervention).. Each patient underwent neurosurgical intervention and postoperative Engel outcomes were obtained through retrospective chart review by an epileptologist. Of the 16 patients, 7 had imaging findings of definite epileptogenic potential at 7T while 9 had non-definite imaging findings. 15 out of 16 patients had Engel I, II, or III outcomes indicating worthwhile improvement. Patients with definite lesion status achieved Engel I surgical outcomes at higher rates (57.1%) than patients with non-definite lesion status (33.3%). Patients with normal clinical diagnostic scans at lower field strengths who have definite radiological findings on 7T corresponding to the sSOZ may experience worthwhile improvement from surgical intervention.

Brief potentially ictal rhythmic discharges and paroxysmal fast activity as scalp electroencephalographic biomarkers of seizure activity and seizure onset zone.

OBJECTIVE: The electroencephalographic (EEG) terms "brief potentially ictal rhythmic discharges" (BIRDs) and "paroxysmal fast activity" (PFA) are considered distinct entities; however, their definitions overlap, and they may have similar clinical significance. We investigated their clinical significance and their association with seizures and the seizure onset zone (SOZ). METHODS: We retrospectively identified an adult cohort (July 2015 to March 2018) whose long-term (>12 h) EEGs in any setting reported BIRDs (>4 Hz, lasting .5-10 s) and/or PFA. Different frequency cutoffs for PFA (>13 Hz or ≥8 Hz) were tested to compare their clinical significance. Patient demographics, clinical history, and EEG features were recorded. RESULTS: We identified 94 patients with BIRDs/PFA out of 3520 patients (3%); 36 were critically ill (12 with epilepsy), and 58 were noncritically ill (all with epilepsy). The frequency of BIRDs/PFA was largely dependent on EEG background: it tended to be slower (theta) in the absence of a posterior dominant rhythm or in the presence of continuous focal slowing in the same region (p = .01). Sixty-two of 94 patients (66%; 32/36 [89%] critically ill, 30/58 [52%] noncritically ill) had electrographic seizures during the recording. The scalp EEG SOZ colocalized with BIRDs/PFA in all cases. BIRDs with faster frequency (also qualifying as PFA by definition) had similar seizure risk to that of slower BIRDs (62%-71%), regardless of frequency cutoff used to define PFA. In addition, 30 of 30 (100%) patients with evolving BIRDs/PFA (which lasted a median of 6 s, range = 2-9.5 s) had electrographic seizures (>10 s), compared to 32 of 64 (50%) with nonevolving BIRDs (median = 1 s, range = .5-3.5 s; p < .01). SIGNIFICANCE: A high proportion of patients with BIRDs/PFA had seizures on EEG, regardless of their frequency (i.e., whether they also qualified as PFA), and their location colocalized with scalp SOZ in all cases. BIRDs appear to be a scalp EEG biomarker of uncontrolled seizure activity and a reliable localizing sign of the SOZ.

Psychiatric comorbidities, sexual health and neurodevelopmental outcomes of offspring in men with epilepsy.

BACKGROUND: There is a paucity of data in the literature specific to men with epilepsy on anti-seizure medication (ASM). The current study investigated the time to conception as well as the gestational and the neurodevelopmental outcomes of offspring of men with epilepsy on ASM compared to controls. Additionally, the prevalence of psychiatric comorbidities, and measures of sexual performance in males with and without epilepsy were analyzed. METHODS: A total of 450 male patients with and without epilepsy at one hospital were provided questionnaires to determine demographic characteristics, epilepsy history, type of ASM at the time of conception, comorbidities and sexual health. Time to conception, fertility methods and offspring birth and developmental history born to males with and without epilepsy was recorded. Survey data was evaluated using Student's t-test for continuous variables and Fisher's exact test for categorical variables. Odds ratio (OR) were calculated to determine associations between the measured data. RESULTS: After matching for age, we analyzed a total of 110 males with epilepsy and 110 without epilepsy. In the epilepsy group there was a higher rate of psychiatric comorbidities such as major depressive disorder, general anxiety disorder, bipolar disorder, and suicidal ideation when compared to the control group (N = 110; OR 3.39; 95% IC: 1.87-6.13, p < 0.001). Males with epilepsy also had a higher frequency of low erection scores when compared to males without epilepsy (N = 70 with epilepsy, N = 76 without epilepsy; OR 3.67; 95% IC: 1.44-9.39, p = 0.005). Of the 110 men with a diagnosis of epilepsy, 17 conceived children while using ASMs (38 total children). A total of 18.42% of children born to fathers on ASMs experienced developmental delays compared to 2.63% of controls, however this result was not statistically significant (p = 0.056). In addition, we did not find that offspring had significantly different birth weights or gestational ages in men on ASM compared to controls (p > 0.05). CONCLUSIONS: The present study suggests that men with epilepsy have an increased incidence of psychiatric comorbidities, and altered sexual performance, specifically erectile dysfunction, when compared with men without epilepsy. There was no statistically significant difference in the rates of developmental disorders and birth characteristics among those men with epilepsy on ASM at the time of conception and controls.