Cortical beta oscillations (13–30 Hz) reflect sensorimotor processing, but are not well understood in balance recovery. We hypothesized that sensorimotor cortical activity would increase under challenging balance conditions. We predicted greater beta power when balance was challenged, either by more difficult perturbations or by lower balance ability. In 19 young adults, we measured beta power over motor cortical areas (electroencephalography, Cz electrode) during three magnitudes of backward support -surface translations. Peak beta power was measured during early (50–150 ms), late (150–250 ms), and overall (0–400 ms) time bins, and wavelet-based analyses quantified the time course of evoked beta power. An ANOVA was used to compare peak beta power across perturbation magnitudes in each time bin. We further tested the association between perturbation-evoked beta power and individual balance ability measured in a challenging beam walking task. Beta power increased ~50 ms after perturbation, and to a greater extent in larger perturbations. Lower individual balance ability was associated with greater beta power in only the late (150–250 ms) time bin. These findings demonstrate greater sensorimotor cortical engagement under more challenging balance conditions, which may provide a biomarker for reduced automaticity in balance control that could be used in populations with neurological impairments.

Objective Seizure frequency and seizure freedom are among the most important outcome measures for patients with epilepsy. In this study, we aimed to automatically extract this clinical information from unstructured text in clinical notes. If successful, this could improve clinical decision-making in epilepsy patients and allow for rapid, large-scale retrospective research. Materials and Methods We developed a finetuning pipeline for pretrained neural models to classify patients as being seizure-free and to extract text containing their seizure frequency and date of last seizure from clinical notes. We annotated 1000 notes for use as training and testing data and determined how well 3 pretrained neural models, BERT, RoBERTa, and Bio_ClinicalBERT, could identify and extract the desired information after finetuning. Results The finetuned models (BERTFT, Bio_ClinicalBERTFT, and RoBERTaFT) achieved near-human performance when classifying patients as seizure free, with BERTFT and Bio_ClinicalBERTFT achieving accuracy scores over 80%. All 3 models also achieved human performance when extracting seizure frequency and date of last seizure, with overall F1 scores over 0.80. The best combination of models was Bio_ClinicalBERTFT for classification, and RoBERTaFT for text extraction. Most of the gains in performance due to finetuning required roughly 70 annotated notes. Discussion and Conclusion Our novel machine reading approach to extracting important clinical outcomes performed at or near human performance on several tasks. This approach opens new possibilities to support clinical practice and conduct large-scale retrospective clinical research. Future studies can use our finetuning pipeline with minimal training annotations to answer new clinical questions.

I.AbstractCortical beta oscillations (13-30 Hz) reflect sensorimotor cortical activity, but have not been fully investigated in balance recovery behavior. We hypothesized that more challenging balance conditions would lead to greater recruitment of cortical sensorimotor brain regions for balance recovery. We predicted that beta power would be enhanced when balance recovery is more challenging, either due to more difficult perturbations or due to lower intrinsic balance ability. In 19 young adults, we measured beta power evoked over motor cortical areas (Cz electrode) during 3 magnitudes of backward support-surface translational perturbations using electroencephalography. Peak beta power was measured during early (50-150 ms), late (150-250 ms), and overall (0-400 ms) time bins, and wavelet-based analyses quantified the time course of evoked beta power and agonist and antagonist ankle muscle activity. We further assessed the relationship between individual balance ability measured in a challenging beam walking task and perturbation-evoked beta power within each time bin. In balance perturbations, cortical beta power increased ∼50 ms after perturbation onset, demonstrating greater increases with increasing perturbation magnitude. Balance ability was negatively associated with peak beta power in only the late (150-250 ms) time bin, with higher beta power in individuals who performed worse in the beam walking task. Additionally, the time course of cortical beta power followed a similar waveform as the evoked muscle activity, suggesting these evoked responses may be initially evoked by shared underlying mechanisms. These findings support the active role of sensorimotor cortex in balance recovery behavior, with greater recruitment of cortical resources under more challenging balance conditions. Cortical beta power may therefore provide a biomarker for engagement of sensorimotor cortical resources during reactive balance recovery and reflect the individual level of balance challenge.

Objective: To determine the effect of epilepsy on intracranial electroencephalography (EEG) functional connectivity, and the ability of functional connectivity to localize the seizure onset zone, controlling for spatial biases. Approach: We analyzed intracranial EEG data from patients with drug-resistant epilepsy admitted for pre-surgical planning. We calculated intracranial EEG functional networks and determined whether changes in functional connectivity lateralized the seizure onset zone using a spatial subsampling method to control for spatial bias. We developed a “spatial null model” to localize the seizure onset zone electrode using only spatial sampling information, ignoring EEG data. We compared the performance of this spatial null model against models incorporating EEG functional connectivity and interictal spike rates. Main results: 110 patients were included in the study, although the number of patients differed across analyses. Controlling for spatial sampling, the average connectivity was lower in the seizure onset zone region relative to the same anatomic region in the contralateral hemisphere. A model using intra-hemispheric connectivity accurately lateralized the seizure onset zone (average accuracy 75.5%). A spatial null model incorporating spatial sampling information alone achieved moderate accuracy in classifying seizure onset zone electrodes (mean AUC = 0.70, 95% CI 0.63-0.77). A model incorporating intracranial EEG functional connectivity and spike rate data further outperformed this spatial null model (AUC 0.78, p = 0.002 compared to spatial null model). However, a model incorporating functional connectivity without spike rate data did not significantly outperform the null model (AUC 0.72, p = 0.38). Significance: Intracranial EEG functional connectivity is reduced in the seizure onset zone region, and interictal data predict seizure onset zone electrode localization and laterality, however a predictive model incorporating functional connectivity without interictal spike rates did not significantly outperform a spatial null model. We propose constructing a spatial null model to provide an estimate of the pre-implant hypothesis of the seizure onset zone, and to serve as a benchmark for further machine learning algorithms in order to avoid overestimating model performance because of electrode sampling alone.

Objective: Evaluating patients with drug-resistant epilepsy often requires inducing seizures by tapering antiseizure medications (ASMs) in the epilepsy monitoring unit (EMU). The relationship between ASM taper strategy, seizure timing, and severity remains unclear. In this study, we developed and validated a pharmacokinetic model of total ASM load and tested its association with seizure occurrence and severity in the EMU. Methods:We studied 80 patients who underwent intracranial electroencephalographic recording for epilepsy surgery planning. We developed a first order pharmacokinetic model of the ASMs administered in the EMU to generate a continuous metric of overall ASM load. We then related modeled ASM load to seizure likelihood and severity. We determined the association between the rate of ASM load reduction, the length of hospital stay, and the probability of having a severe seizure. Finally, we used modeled ASM load to predict oncoming seizures.Results: Seizures occurred in the bottom 50th percentile of sampled ASM loads across the cohort (p < .0001, Wilcoxon signed-rank test), and seizures requiring rescue therapy occurred at lower ASM loads than seizures that did not require rescue therapy (logistic regression mixed effects model, odds ratio = .27, p = .01).Greater ASM decrease early in the EMU was not associated with an increased likelihood of having a severe seizure, nor with a shorter length of stay.