This Pediatric Issue of the Journal of Korean Neurosurgical Society addresses the digital and robotic transformation of pediatric neurosurgery. It reviews the emerging applications of artificial intelligence and related technologies across diagnosis, monitoring, surgical practice, education, and communication, while also considering the unique scientific and ethical challenges of their use in children. Together, these articles offer a timely perspective on the future of safer, more precise, and more personalized pediatric neurosurgical care.

The pediatric brain represents a dynamic biological target characterized by rapid myelination and functional reorganization, which presents unique challenges for conventional, adult-centric artificial intelligence (AI) models. This review provides a structured overview of the evolution of AI applications in pediatric neuroimaging and neurosurgery, tracing the transition from early standardized pipelines and handcrafted imaging biomarkers to contemporary deep learning-based approaches for segmentation, prediction, and anomaly detection. Recent advances indicate a paradigm shift from static image interpretation toward dynamic and interactive intelligence, in which AI systems actively support clinical decision-making during surgery rather than functioning solely as diagnostic tools. This new paradigm is supported by four technological domains : brain foundation models designed to capture age-aware neurodevelopmental representations; spatial computing technologies for three-dimensional, context-aware-visualization; physical AI systems integrating robotic safety constraints; and multimodal AI agents that act as cognitive surgical copilots by synthesizing imaging, physiological, and intraoperative data in real time. By shifting the role of AI from preoperative assessment to intraoperative guidance, this paradigm offers new opportunities to enhance surgical precision, safety, and workflow efficiency in pediatric neurosurgery. This review aims to provide neurosurgeons with a conceptual framework for understanding and adopting next-generation AI technologies that align with the dynamic nature of the developing brain and the clinical demands of pediatric neurosurgical care.

Pediatric neuro-oncology is a critical field of neurosurgery, representing the leading cause of disease-related mortality in children. Despite its rarity, it encompasses over 100 diverse disease entities, which significantly complicates preoperative differential diagnosis and surgical planning. This review examines how artificial intelligence (AI) can address these unmet clinical challenges throughout the perioperative period. Preoperatively, AI-driven radiogenomic models extract pixel-level features to enable non-invasive molecular subtyping, such as predicting B-Raf proto-oncogene alteration status in pediatric low-grade gliomas (pLGGs). Such insights are vital for determining the extent of resection (EOR) with consideration of availability of targeted therapies. Furthermore, AI facilitates automated tumor segmentation, allowing for meticulous surgical planning and more accurate assessment of surgical risks. Intraoperatively, AI significantly accelerates diagnostic turnaround times, which is essential for real-time decision-making. Emerging technologies, including Oxford Nanopore sequencing with neural network classifiers or stimulated Raman histology, allow for the rapid identification of tumor characteristics in operation time window. These tools directly inform the optimal EOR, particularly in cases like medulloblastoma where molecular subgroups dictate surgical aggressiveness. Additionally, AI integration into intraoperative neurophysiological monitoring enhances the preservation of critical neurological functions. Postoperatively, multimodal deep learning models integrate clinical, imaging, and genomic data to improve prognostic accuracy and standardize response assessment via AI integration. While challenges such as data scarcity and the "black box" nature of algorithms persist, innovative strategies offer potential solutions to AI application. AI serves as a transformative tool for personalized precision management, potentially bridging diagnostic disparities and optimizing clinical outcomes for children with central nervous system tumors.

Extended reality (XR), encompassing virtual reality, augmented reality (AR), and mixed reality, has emerged as a transformative technology in neurosurgery. This narrative review examines the current applications of XR technologies across four major domains : surgical planning and visualization, intraoperative navigation, medical education and training, and patient/caregiver communication. The integration of artificial intelligence-based lesion segmentation with XR platforms has enabled precise three-dimensional visualization of complex intracranial pathologies. Modern AR navigation systems, utilizing both optical see-through and video see-through head-mounted displays, have demonstrated submillimeter accuracy in recent clinical studies. Educational applications of XR have shown significant improvements in anatomical understanding and surgical skill acquisition among neurosurgery trainees. Furthermore, XR-enhanced patient communication tools have improved comprehension of complex neurosurgical procedures and informed consent quality. Despite current limitations including hardware constraints, workflow integration challenges, and the need for larger validation studies, XR technologies hold substantial promise for advancing pediatric neurosurgical care. This review provides a critical analysis of current evidence, discusses the advantages and limitations of different XR modalities, and offers perspectives on future developments in this rapidly evolving field.

Pediatric neurodevelopmental disorders (NDDs) as defined by the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision : intellectual disabilities, communication disorders, attention-deficit/hyperactivity disorder (ADHD), autism spectrum disorder (ASD), specific learning disorder, and motor disorders together with closely associated pediatric neurological conditions such as cerebral palsy (CP) and epilepsy, present with symptom fluctuations and variable performance, making longitudinal assessment challenging. Conventional clinical scales, questionnaires, and caregiver diaries are largely subjective and offer limited temporal coverage. This review examines wearable technologies applicable to these conditions, focusing on measurable digital biomarkers and real-world clinical application. A narrative review was performed. Wearable technologies were categorized into motion trackers (accelerometers/gyroscopes, inertial measurement units [IMUs]), physiological monitors (photoplethysmography-derived heart rate and heart rate variability, and electrodermal activity), wearable electroencephalography (EEG)/brain sensors, smart glasses/cameras, and textile-based wearables. IMU and accelerometer based systems achieved posture and gait classification accuracies of approximately 85%, distinguished children with ADHD from controls with 80-90% accuracy, and detected stereotyped behaviors in children with ASD with sensitivities exceeding 90%. Wearable EEG demonstrated high performance for ASD classification (accuracy, 96%; sensitivity, 100%) and for automated sleep staging in epilepsy(accuracy, 80.8-83.5%), and also enabled seizure-related signal monitoring through detection of ictal fast activity. For tonic-clonic seizures, wearable devices achieved sensitivities of 89-94%, whereas tonic-seizure detection performance remains limited. Smartwatch vibrotactile stimulation reduced stereotyped behaviors and improved task performance, and wrist-worn biosignals enabled prediction of aggressive behavior approximately three minutes before onset. Augmented reality based smart glasses supported social-communication coaching in children with ASD. In children with CP, smart-shoe systems classified daily activities with high accuracy and quantified gait asymmetry. Wearables enable objective and continuous assessment of symptoms and behaviors in children with NDDs and related pediatric neurological conditions across real-world environments and are expanding toward personalized intervention. Remaining challenges include integration into clinical workflows (electronic medical record [EMR] linkage, clinician-friendly data summarisation, and reimbursement pathways), ethical and privacy considerations, and standardisation of validation protocols.

Pediatric drug-resistant epilepsy remains a significant neurological challenge, with up to one-third of affected children failing to achieve seizure control with pharmacologic therapy. Surgical intervention offers the greatest likelihood of seizure reduction or freedom; however, precise localization and treatment of epileptogenic networks in pediatric patients are often complicated by multilobar onset patterns, evolving neurodevelopmental anatomy, and limitations of conventional invasive monitoring. Over the past two decades, stereoelectroencephalography (SEEG) has emerged as a preferred invasive diagnostic modality due to its minimally invasive approach, three-dimensional sampling capability, and favorable safety profile. Parallel advances in robotic stereotaxy have further enhanced the precision, efficiency, and reproducibility of electrode implantation and stereotactic interventions. Robotic assistance has expanded beyond diagnostic SEEG implantation to include stereotactic ablative therapies, such as radiofrequency thermocoagulation and magnetic resonance-guided laser interstitial thermal therapy, as well as neuromodulation procedures including deep brain stimulation and responsive neurostimulation. These technologies enable precise targeting of deep and distributed epileptogenic networks while minimizing surgical morbidity. Pediatric clinical studies demonstrate that robotic-assisted SEEG achieves high diagnostic yield, excellent safety profiles, and accuracy comparable to or exceeding conventional stereotactic techniques. Similarly, robot-assisted ablative and neuromodulation procedures show promising efficacy in carefully selected pediatric populations, although complete seizure freedom remains less common in neuromodulatory approaches. Despite increasing adoption, the pediatric robotic epilepsy surgery literature remains predominantly retrospective and heterogeneous, with limited prospective comparative data. This review synthesizes current evidence regarding robotic applications in pediatric epilepsy surgery, including SEEG implantation, stereotactic ablation, and neuromodulation. We highlight technical considerations, clinical outcomes, and safety profiles, while identifying critical gaps in evidence related to long-term outcomes, and the direct clinical impact of robotic accuracy. Continued prospective and multicenter investigations are essential to define the optimal role of robotic assistance in improving outcomes for pediatric patients with drug-resistant epilepsy.

Neurocritical care relies on continuous assessment of neurological function and physiology under time pressure, yet bedside teams must interpret high-frequency, multimodal data that include hemodynamic waveforms, intracranial pressure (ICP), electroencephalography (EEG), near-infrared spectroscopy, neuroimaging, and electronic health record (EHR) context. Artificial intelligence (AI) and machine learning are increasingly used to fuse these biosignals, reduce interobserver variability, and generate dynamic risk estimates that can support monitoring, early warning, and prognostication. In pediatric and neonatal populations, where developmental physiology and smaller case volumes amplify uncertainty, AI-enabled tools have shown particularly strong performance in selected domains. Examples include prediction of impending intracranial hypertension using features from arterial blood pressure and ICP waveforms, automated EEG trend analysis for neurodevelopmental outcome prediction after perinatal asphyxia, magnetic resonance imaging-based neuroprognostication in neonatal hypoxic-ischemic encephalopathy, and interoperable EHR-based models for mortality and new morbidity risk stratification in the intensive care unit. However, many studies remain retrospective, and generalizability across institutions can be limited by differences in data capture, clinical practice, and outcome definitions. This review summarizes clinically relevant applications of multimodal biosignal analysis in neurocritical care, highlights validation and implementation considerations, and proposes priorities for future pediatric translation, including multisite evaluation, calibration, explainability, and prospective impact studies.

Anesthesia for pediatric neurosurgery represents a highly complex and challenging field, characterized by age-dependent physiological variability, heterogeneous patient populations, and the critical need to protect the developing central nervous system. Conventional clinical approaches often rely on adult-derived data, which may inadequately reflect the distinct neurophysiological and hemodynamic characteristics of neonates, infants and children. Recent advances in artificial intelligence (AI) have enabled the integration of multimodal perioperative data, including physiologic signals and neurophysiologic monitoring, with the aim of supporting clinical decision-making in complex surgical settings. This review summarizes the current landscape of AI applications relevant to pediatric anesthesia, with particular attention to preoperative risk assessment, airway management, and real-time prediction of intraoperative adverse events such as hypoxemia and hemodynamic instability. Although AI-based approaches have demonstrated encouraging results in adult populations, their application to pediatric neuroanesthesia remains limited. The integration of AI into this field faces several distinct challenges, including the scarcity of high-quality datasets for rare neurosurgical conditions, substantial heterogeneity across developmental stages, and difficulties in aligning model outputs with clinically interpretable physiologic mechanisms. Addressing these limitations will require the development of explainable, physiology-informed AI frameworks and disease-specific models tailored to conditions such as moyamoya disease or complex craniofacial reconstruction. Ultimately, AI should be positioned as an adjunctive decision-support tool that complements, rather than replaces, anesthesiologists' expertise. Through multidisciplinary collaboration and human-centered implementation, AI may contribute to improved perioperative safety and long-term neurodevelopmental outcomes in vulnerable pediatric neurosurgical patients.

Objective : This study aims to evaluate the effects of poly(3-hydroxybutyrate)/alginate (PHBA/ALG) tubular graft containing cytidine 5'-diphosphocholine (citicoline) on axonal regeneration in an experimental rat sciatic nerve injury model and to compare the results of this method with autologous nerve grafting. Methods : A total of 24 Wistar-Hannover albino rats were randomly divided into three groups (n=8). In all animals, a 10 mm segment was removed from the right sciatic nerve under microscope. The nerve defect was repaired using autologous nerve graft (group O), PHBA graft only (group T), and PHBA graft containing citicoline (group S). Sixteen weeks later, all animals were euthanized and sciatic nerve samples were taken for histopathological examination. Sections were stained with hematoxylin-eosin and Masson Trichrome. Results : The groups were compared statistically in terms of regenerated axon diameter, number of blood vessels, nerve sheath degeneration, edema, inflammatory cell infiltration, and necrosis. Histopathological evaluation showed that regenerative axons were formed in all three groups. PHBA and citicoline-containing PHBA grafts successfully bridged the damaged area. While the best results were obtained in the autograft group, the results of the PHBA group containing citicoline were statistically significantly better than the PHBA group. In general, the distribution of histopathological parameters showed that citicoline administration showed protective effects on parameters such as inflammation and cell death. Axon diameter : group S had significantly larger axon diameters than group T (p<0.05). Number of blood vessels : the highest values were in the autograft group and the difference between groups O and T was significant (p<0.05). Conclusion : PHBA-based tubular graft with oriented nanofiber structure containing citicoline supports peripheral nerve regeneration. Although our results are not better than autograft, it may be a promising alternative for cases where autologous graft is not available.

Objective : Alzheimer's disease is a progressive neurodegenerative disorder characterized by amyloid-beta (Aβ) peptide aggregation, representing a major therapeutic target. Emerging evidence suggests certain chemotherapeutic agents may attenuate Aβ pathology. Methods : This study investigated the effects of imatinib, a tyrosine kinase inhibitor with limited blood-brain barrier (BBB) penetration, and nilotinib, with enhanced BBB permeability, in an intracerebroventricular streptozotocin (ICV-STZ) rat model of Alzheimer's disease. Outcomes included behavioral assessments (learning latency), hippocampal CA1 and CA3 neuronal counts, and brain concentrations of tumor necrosis factor (TNF)-α, nuclear factor kappa B (NF-κB), brain-derived neurotrophic factor (BDNF), and neuregulin-1 (NRG-1). Results : ICV-STZ administration significantly elevated TNF-α and NF-κB levels and reduced BDNF and NRG-1 expression. Both imatinib and nilotinib mitigated these alterations, with imatinib demonstrating greater efficacy despite its limited BBB permeability. Imatinib and nilotinib reduced TNF-α and NF-κB levels, increased BDNF and NRG-1 expression, and significantly improved cognitive performance, with latency periods extending from 69.8 seconds in the disease model to 193.5 and 183.1 seconds, respectively. Conclusion : Imatinib and nilotinib ameliorated neuroinflammation, restored neurotrophic support, and improved cognitive deficits in a preclinical Alzheimer's disease model. These findings highlight the therapeutic potential of tyrosine kinase inhibitors, warranting further translational research in human studies.

Objective : To evaluate the efficacy of individualized dual antiplatelet therapy (DAPT) using VerifyNow in reducing periprocedural ischemic complications following carotid artery stenting (CAS), particularly in patients with unstable plaques. Methods : This retrospective, single-center study included 187 patients who underwent elective CAS between January 2018 and September 2024. Patients were divided into two groups : a conventional DAPT (c-DAPT) group treated with aspirin and clopidogrel (n=89), and an individualized DAPT (i-DAPT) group (n=98) where antiplatelet regimens were adjusted based on preoperative platelet reactivity measured by VerifyNow. The primary outcome was the incidence of ischemic lesions on diffusion-weighted imaging (DWI) following CAS. Risk factors and the impact of DAPT personalization on ischemic changes were analyzed using multivariate logistic regression. Results : DWI-positive ischemic lesions were observed in 62.9% of the c-DAPT group and 51% of the i-DAPT group (p=0.07). Although the difference was not statistically significant, there was a trend toward reduced ischemic changes with individualized therapy. Multivariate analysis identified hypertension, diabetes mellitus, and unstable plaque as independent predictors of post-CAS ischemic changes. Subgroup analysis showed that VerifyNow-guided i-DAPT significantly reduced ischemic events in patients with unstable plaques (odds ratio, 0.49; p=0.03), whereas no significant effect was observed in other subgroups. Conclusion : VerifyNow-guided personalization of antiplatelet therapy is a safe and may have the potential to reduce periprocedural ischemic complications in CAS, particularly in patients with unstable plaques. Further multicenter prospective studies are needed to validate these findings and optimize individualized antiplatelet regimens.

Objective : Recurrent lumbar disc herniation (RLDH) is a significant challenge following lumbar discectomy, with recurrence rates of 5-15%. Established risk factors include male gender, diabetes mellitus, smoking, and obesity, but the role of paraspinal muscles in recurrence is unclear. This study was conducted to identify key risk factors for RLDH, including the volume of paraspinal muscles with machine learning. Methods : We used data from 126 patients who underwent lumbar discectomy between January 2003 and September 2023 and had follow-up outpatient visits for more than 6 months at a single institution. Variables selected for the model, comprising demographic and clinical variables, medical history, LDH operation-related variables, and magnetic resonance imaging (MRI) measurements for RLDH. Based on clinical symptoms and radiologic results, the patients were classified into RLDH and non-RLDH groups, and RLDH was defined at the same surgical level on follow-up MRI. Results : Totally, 38 patients were included in the RLDH group and 88 in the non-RLDH group. The volume of quadratus lumborum was identified as a risk factor for RLDH (odds ratio, 7.894; p=0.001). Among the five different machine learning algorithms, extreme gradient boosting achieved the best result with an accuracy of 0.794 and area under the curve of 0.811. In terms of Shapley Additive exPlanations value analysis, the weight, volume of quadratus lumborum, psoas major, and vertebra were key features for predicting RLDH. Conclusion : The prediction model would be of great assistance for surgeons to make surgical decisions or establish observation intervals.

Objective : Prevertebral soft tissue swelling (PSTS) is a significant complication of anterior cervical spine surgery (ACSS) that causes dysphagia, dysphonia, and possibly life-threatening airway obstruction. This study aims to develop and internally validate an interpretable machine learning model to predict significant PSTS after single-level ACSS using a small, single-center dataset. Methods : We retrospectively analyzed data from 62 patients who underwent single-level ACSS in our center, from January 2014 to December 2022. Postoperative swelling over 7.0 mm (above 75th percentile) was defined as significant PSTS. We developed an elastic net regularized logistic regression model over 1000 iterations with nested 5-fold cross validation for hyperparameter tuning and bootstrap validation. The model's interpretability was assessed using SHapley Additive exPlanations (SHAP) approach, and its clinical utility by decision curve analysis. Results : A total of 16 (25.8%) out of 62 patients developed significant PSTS. Our model achieved a bootstrap-validated area under the ROC curve value of 0.84 (95% confidence interval, 0.73-0.92) with good calibration (Hosmer-Lemeshow p=0.42). According to our results, important predictors of PSTS included 1) low preoperative serum albumin (SHAP importance, 0.42), 2) upper level surgery above C5 (0.38), and 3) male sex (0.31). Decision curve analysis demonstrated net benefit across probability thresholds of 15-45%. Conclusion : Our machine learning model effectively predicted risk of significant PSTS following single-level ACSS despite small sample size. Such findings offer new opportunities for risk stratification and prevention strategies. This study is a proof of concept for artificial intelligence (AI)-based risk stratification strategy; further external validation could improve its relevance in the clinical setting.

Objective : Ventriculoperitoneal shunt (VPS) treats hydrocephalus effectively, but manual ventricular catheter placement has a high misplacement rate. This study assesses the feasibility, accuracy, and safety of robot-assisted VPS. Methods : A retrospective analysis was performed on the clinical data of 87 patients from 2022 to 2024, with 42 receiving robot-assisted VPS and 45 receiving conventional manual VPS. Logistic regression analysis was performed to identify risk factors affecting placement accuracy. Results : The robot-assisted cohort showed smaller bilateral frontal horn diameters and a lower Evans index (p<0.05). It achieved higher placement accuracy, required fewer puncture attempts, and had a lower postoperative complication rate (p<0.05). No significant difference was found in operation time (p>0.05). Logistic regression analysis confirmed robot-assistance as a key factor for accurate catheter placement (odds ratio, 6.52; p<0.001). Conclusion : Robot-assisted VPS achieves precise placement, reduces puncture attempts and complications, and does not increase time costs, confirming its clinical feasibility, accuracy, and safety.

Objective : The aim of this study was to report the early clinical experience with stereotactic biopsy using the Medtronic Stealth Autoguide Cranial Robotic Guidance Platform, representing the first relevant clinical study in the Republic of Korea (ROK). We evaluated potential advantages and limitations related to workflow integration, instrument setup, procedural efficiency, accuracy, and safety. Methods : A retrospective case series was conducted across three centers in the ROK. From April to July 2025, 17 consecutive patients underwent frameless stereotactic biopsy using the Stealth Autoguide system. Data on demographics, operative time, and complications were collected, and descriptive analyses were performed. Results : Biopsy using the Stealth Autoguide System was successfully completed in 16 of 17 patients. The median operative time was 48 minutes (interquartile range, 33-64), and the median target alignment error was 0.4 mm (interquartile range, 0.3-0.6). One case required conversion to the conventional manual method due to targeting inaccuracy. Three cases of asymptomatic intracerebral hemorrhage occurred, including one at the tissue-harvest site and two with cortical hemorrhage, all of which were managed conservatively. Conclusion : The Stealth Autoguide system facilitates a more efficient frameless stereotactic biopsy than conventional methods by reducing operative time, enhancing accuracy, and allowing for a smaller, muscle-sparing incision. However, some concerns remain, and further clinical validation is needed to define its optimal indications and safety profile.

Objective : The 2024-2025 medical crisis in the Republic of Korea, triggered by mass trainee departures, caused severe staffing shortages in tertiary hospitals. The impact of this disruption on neurosurgeons' workload, lifestyle, and mental health has not been evaluated. In this study, we aimed to determine the prevalence of burnout among neurosurgeons during the crisis, assess associated changes in workload and lifestyle, and identify key factors linked to burnout severity. Methods : We conducted a nationwide, questionnaire-based observational study from April 29 to June 4, 2025, targeting board-certified neurosurgeons in university hospitals across the Republic of Korea. Participants were recruited through departmental professional contacts and completed an online survey capturing baseline characteristics, paired pre-post crisis values for 14 workload and lifestyle variables, and three current-status items. Burnout was assessed using the Copenhagen Burnout Inventory (CBI) as the primary tool and the Maslach Burnout Inventory-Human Services Survey (MBI-HSS) as a supplementary measure. Results : Of 61 respondents, 36 (59.0%) met high-burnout criteria. After the crisis onset, all workload and lifestyle factors shifted unfavorably, with the largest increases in weekend/holiday duty (86.9%), night duty (82.0%), and sleep disturbance call (68.9%), and the largest declines in number of colleagues (-50.8%), leisure time (-47.5%) and family time (-44.3%). High-burnout surgeons had greater workloads, more off-hour duties, and less recovery time than low-burnout peers. Burnout severity correlated with staffing levels, working hours, night duty, time spent with family and leisure activities, job satisfaction, intention to leave, and all MBI subscales. Conclusion : Burnout among Korean neurosurgeons is widespread after the national medical crisis, with substantial deterioration in workload and lifestyle. Coordinated efforts at both hospital and national levels are essential to protect surgeon well-being and maintain the quality of neurosurgical care.