Singh, Rohin

Artificial intelligence (AI) is reshaping neurosurgery, offering unprecedented opportunities to enhance diagnostics, personalize treatment, and predict outcomes. At the heart of this transformation is the ability to effectively harness big data (BD) within the electronic medical record. Understanding these data structures is essential for making sense of the vast volumes of information generated in modern neurosurgical practice. Equally important are the machine learning (ML) models driving these advancements. From supervised learning and convolutional neural networks to generative AI, these tools are already making a mark in areas such as brain tumor segmentation and spine surgery outcome predictions. Their versatility highlights the potential of ML to complement clinical expertise and streamline decision-making in neurosurgery. However, adopting BD and ML also brings ethical challenges that cannot be ignored. Bias in algorithms threatens to reinforce health disparities, whereas concerns about data privacy demand vigilance in handling sensitive patient information. In addition, the question of liability looms large as ML increasingly influences clinical decisions. The aim of the study was to provide a roadmap for neurosurgeons navigating the evolving intersection of BD, ML, and ethical responsibility in the AI era.

AbstractObjectiveNeuromodulation for pediatric refractory epilepsy (PRE) is preferred when the seizure onset zone is not amenable to surgical resection, due to multifocal onset or involvement of eloquent cortex. Given its rapidly evolving landscape, we describe our institutional experience with neurostimulation therapies including deep brain stimulation (DBS) of the anterior nucleus (DBS‐ANT), the centromedian nucleus (DBS‐CM), responsive neurostimulation (RNS), and chronic subthreshold stimulation (CSS) to treat PRE.MethodsA retrospective chart review was conducted to identify pediatric patients (≤18 years of age) who presented to our institution with PRE. Patients were included who had at least 1 year of follow‐up after neurostimulation was started. Patients with ≥50% seizure frequency reduction at last follow‐up compared to baseline were classified as responders.ResultsA total of 35 patients (21 females) were included in the series. Nine patients underwent DBS‐ANT, 9 underwent DBS‐CM (+ ANT in 7 patients), 5 underwent RNS, and 12 underwent CSS. The median age at seizure onset and surgery was 8 and 16 years, respectively. Structural lesions were the most common etiology of epilepsy (54%). Eighteen patients had prior interventions and 13 received invasive EEG. Post‐operative complications following neurostimulation implantation were noted in three patients (9%). However, no permanent deficits were noted. Median follow‐up time was 46 months (range 13–162 months). Twenty‐three patients (59%) were responders (3/9 DBS‐ANT; 7/9 DBS‐CM; 4/5 RNS; 9/12 CSS). Further, median seizure reduction was 25% (DBS‐ANT), 89% (DBS‐CM), 87% (RNS), and 88% (CSS) across the cohort. Six patients (17%) had improved subjective neurocognitive outcomes while the other 29 had stable neurocognitive outcomes.SignificanceDBS, RNS, and CSS are safe and effective options for PRE. An understanding of treatment approaches can optimize results by tailoring therapy to individual patients.Plain Language SummaryThis study investigates the use of various neurostimulation therapies for treating pediatric refractory epilepsy (PRE) including deep brain stimulation (DBS), responsive neurostimulation (RNS), and chronic subthreshold stimulation (CSS). Overall, neuromodulation was found to be effective in reducing seizures in most patients, with 59% of patients showing at least a 50% reduction in seizure frequency. Additionally, the treatments were generally safe, with few complications and no instances of permanent deficits.