Search Results (62)

The educational pathway for expertise in intraoperative neurophysiological monitoring (IONM) is complex and lengthy, requiring a solid foundation in neuroscience, neurophysiology, and neuroanatomy. It also demands direct familiarity with a broad range of neurosurgical scenarios, including supratentorial, infratentorial, and spinal procedures, gained through exposure to at least ten distinct surgical approaches. Intraoperative neurophysiology must be tailored to each patient’s preoperative assessments. It relies on a variety of methods to collect, analyze, and report neurophysiological signals that are relevant to the surgical procedure. Despite its importance, there remains a substantial shortage of training tools designed to support realistic practice and skill development. To address this gap, we developed a comprehensive framework (ION-Sim) that integrates all laboratory testing modalities and adapts them to the operating room environment. ION_sim supports the simulation and analysis of spontaneous EEG and EMG activity, a wide range of evoked potentials, and intraoperative stimulus–response testing protocols. The framework provides a unified environment for practicing, testing, and validating the core neurophysiological procedures employed during neurosurgical interventions. In addition, it incorporates a robust data-management architecture, maintaining a database with system setups, user profiles, educational performance metrics, and automatically generating reports. This structure enables the longitudinal tracking of objective skill acquisition and facilitates standardized assessments of trainee progress. ION_Sim is distributed both as a ready-to-use application, suitable for direct integration into teaching and training programs, and as a modular scientific library. Through its dedicated APIs, users can design customized configurations, create novel simulation scenarios, and extend the platform to support additional research or educational objectives. It is available upon request for educational purposes and is open-source and released under the GNU General Public License, ensuring transparency, reproducibility, and long-term accessibility for the scientific and clinical communities. Full article

Computed tomography angiography (CTA)-based segmentation of the superficial temporal arteries (STAs) and facial vessels (FVs) is important for neurosurgical and reconstructive planning. Nevertheless, segmentation of STAs and FVs remains challenging because of their small caliber, tortuous courses, and proximity to high-intensity bony structures. This study aims to develop a deep learning framework for accurate automated segmentation of these craniofacial vessels. A single-input 3D nnU-Net v2 model was trained using raw CTA volumes, while a Fusion-based Vesselness Map (FVM) was constructed from multiscale vessel-enhancement filters to emphasize small vascular structures and suppress irrelevant regions such as the skull and skin. Instead of being used as an additional input channel, the FVM was incorporated into the loss function as a spatial prior to guide the network toward vessel boundaries and distal branches. In 72 clinical cases, the FVM-guided model improved segmentation accuracy compared with a baseline model trained with Dice Focal Loss, particularly in boundary delineation. For the STAs, the Average Symmetric Surface Distance decreased from 6.543 mm to 2.941 mm. Qualitative evaluation further showed reduced segmentation noise and fewer false positives near bone and distal branches. These findings suggest that integrating classical vessel enhancement into deep learning supervision can improve morphologically consistent craniofacial vessel segmentation and support preoperative surgical planning. Full article

Introduction: Neurosurgery has evolved from an anatomy-driven analog discipline into a digitally augmented field supported by multimodal imaging, neuronavigation, intraoperative imaging, neurophysiological monitoring, robotics, augmented reality, and artificial intelligence. Objective: To examine how this transition has altered professional responsibility, informed consent, training, and medico-legal accountability in neurosurgical practice. Methods: We performed a structured narrative review of the literature on digital neurosurgery and its ethical and professional implications, focusing on publications from 1990 onward and supplemented by landmark historical papers. Sources were selected for relevance to cranial, spinal, skull base, stereotactic, and neuro-oncological neurosurgery, and then synthesized into thematic domains including brain shift, eloquent cortex preservation, stereotactic accuracy, intraoperative neurophysiology, workflow integration, equity, and liability. Results: Digital systems improve lesion localization, function-preserving surgery, stereotactic precision, documentation, and training, but they also introduce new vulnerabilities related to registration error, brain shift, platform dependence, data overload, cost, cybersecurity, deskilling, and diffuse accountability. Conclusions: Digital augmentation expands rather than diminishes the neurosurgeon’s responsibility. The neurosurgeon remains accountable for surgical indication, interpretation of technology-generated information, intraoperative override, and communication of technology-specific risks. The central ethical challenge is to integrate digital tools without weakening patient-centered judgment. Full article

Eye-tracking has been utilized to characterize visual behavior in medical image visualization and interpretation, yet neurosurgeons remain underrepresented. Characterizing neurosurgery-specific visual expertise is important for understanding expert search strategies, informing training, and developing computational models. This study examined gaze behavior in naïve observers (N_p = 29), neurosurgery registrars (N_p = 16), and consultant neurosurgeons (N_p = 24), viewing normal (N_p = 20) and pathological (N_p = 19) brain MR images under a free-viewing paradigm. To capture expertise-related characteristics, we analyzed two features at each fixation location: (i) fixation duration, reflecting temporal allocation of visual attention, and (ii) three-dimensional fractal dimension (3DFD) around each fixation location, quantifying local structural complexity. To assess pathological-type effects, we grouped similar pathologies into five stimulus groups. Linear mixed-effects modelling revealed systematic expertise-related differences, with experts exhibiting longer fixation durations in pathological stimulus groups and pathology-type-dependent complexity sampling. Combined fixation duration and 3DFD features captured complementary aspects of visual expertise, improving Random Forest classifier’s accuracy (>93%) compared to individual features, for all five stimulus groups. These findings highlight neurosurgery-specific markers of visual expertise and demonstrate that combining behavioral and image-derived features could underpin computational models and training tools that emulate expert-level strategies in neurosurgical image interpretation. Future work should evaluate its applicability to other medical domains. Full article

Background/Objectives: Accurate automatic segmentation of multimodal magnetic resonance imaging (MRI) is essential for neurosurgical planning and image-guided procedures. However, existing three-dimensional segmentation models often struggle with low lesion-to-tissue contrast, ambiguous tumor boundaries, small enhancing tumor regions, and performance degradation caused by missing imaging modalities. This study aimed to develop a robust segmentation framework that improves cross-modal representation learning, boundary recovery, and segmentation performance under incomplete-input conditions. Methods: We propose PF-CMNet, a Progressive Frequency-Aware Cross-Modal Network with Missing-Modality Distillation for three-dimensional brain tumor segmentation. The network introduces a Cross-Modal Selective Frequency Attention module in the early encoder stage to model modality-specific frequency responses and spatially adaptive cross-modal correlations. A Progressive Cross-Scale Detail Fusion decoder is further employed to aggregate multilevel semantic features and refine high-resolution boundary details. To enhance robustness under missing-modality conditions, a teacher–student distillation strategy transfers full-modality predictions and shallow feature knowledge to a student network trained with random modality dropout. Results: On the MSD Task01_BrainTumour dataset, PF-CMNet achieved an average Dice score of 84.3%, with Dice scores of 79.6%, 82.8%, and 90.4% for enhancing tumor, tumor core, and whole tumor, respectively. On the BraTS2021 dataset, the model achieved an average Dice score of 88.2% and the lowest average 95th percentile Hausdorff distance among the compared methods. In predefined complete-modality absence stress tests, where unavailable MRI sequences were zero-masked to model the absence of input modalities rather than partial image degradation, the distilled model maintained average Dice scores of 78.64%, 82.58%, 58.39%, 82.03%, and 79.29% when FLAIR, T1, T1ce, T2, and T1 + T2 were unavailable, respectively. Conclusions: PF-CMNet provides a unified framework for multimodal brain tumor segmentation, improving full-modality segmentation accuracy, boundary consistency, and robustness to incomplete MRI inputs while maintaining a favorable accuracy–efficiency trade-off. Full article

Background/Objectives: Transsphenoidal surgery has become the gold standard for the treatment of sellar and parasellar lesions, but it remains associated with significant anatomical challenges and the risk of intraoperative complications. The limitations of conventional imaging in depicting the complex three-dimensional anatomy of the skull base have led to a growing interest in virtual (VR) and augmented reality (AR) technologies, which offer enhanced spatial visualization, preoperative simulation, and image-guided intraoperative navigation. This systematic review aims to evaluate the current evidence on the role of virtual and augmented reality in transsphenoidal surgical interventions, with a focus on their impact on preoperative planning, intraoperative orientation, surgical outcomes, and neurosurgical training. Methods: A systematic literature search was conducted in accordance with PRISMA 2020 guidelines across PubMed, Scopus, and Web of Science for the period 2015–2025. MeSH terms and free-text keywords related to transsphenoidal surgery, sphenoid sinus anatomy, and VR/AR technologies were combined using Boolean operators. Risk of bias was assessed using RoB 2.0 for RCTs; methodological quality was assessed using the Newcastle–Ottawa Scale for observational studies and AMSTAR 2 for systematic reviews. Clinical, morphometric, and experimental studies evaluating VR/AR applications were included. Data were extracted using a standardized protocol and synthesized through qualitative analysis, with subgroup analysis by technology type (VR vs. AR) and clinical application domain. Results: A total of 218 publications were identified, of which 52 met the inclusion criteria (clinical studies n = 12, simulation and technology studies n = 30, morphological studies n = 10). VR-based three-dimensional reconstructions were consistently associated with improved preoperative spatial orientation and anatomical landmark recognition. AR systems demonstrated a meaningful contribution to intraoperative navigation, with reported reductions in time to target and improved visualization of critical neurovascular structures. VR platforms showed high effectiveness in surgical training, with shorter learning curves and improved technical performance. However, the majority of included studies were small observational cohorts, simulation studies, or expert overviews, with substantial heterogeneity in methodology, technology platforms, and outcome measures, precluding quantitative meta-analysis. Conclusions: Virtual and augmented reality represent clinically promising adjuncts to transsphenoidal surgery, with demonstrated benefits in preoperative planning, intraoperative navigation, and surgical training. These conclusions should be interpreted in the context of a predominantly early-phase and heterogeneous evidence base. Standardized protocols, larger prospective studies, and randomized trials are needed before the integration of VR/AR with navigation systems and artificial intelligence can be established as a routine component of personalized transsphenoidal surgery. Full article

Background/Objectives: Glioblastoma is the most common and aggressive primary malignant brain tumor in adults, characterized by infiltrative growth and poor prognosis. Achieving maximal resection without inducing neurological deficits remains a challenge in glioblastoma surgery. While 5-aminolevulinic acid-based fluorescence-guided surgery supports intraoperative tumor visualization, its reliability is limited by patient variability and weak fluorescence signals. This study proposes a machine learning framework to enhance fluorescence-guided surgery sensitivity by analyzing surgical microscope images at the pixel level. Methods: Fluorescence-mode neurosurgical microscope images of synthetic samples with known Protoporphyrin IX (PPIX) concentrations were used to train three classifiers (Support Vector Machine, Naïve Bayes, Neural Network) for pixel-wise fluorescence detection. In parallel, three contrastive-learning-based Variational Autoencoders (VAE, $\beta$ = 1, 2, 3) were evaluated for detecting weak fluorescence beyond visual perception. Additionally, a regression model was trained to relate pixel features to PPIX concentration. The best-performing VAE ($\beta$ = 1) was subsequently trained on real intraoperative data, and its detection sensitivity was compared to annotations from four experienced surgeons. Results: The proposed model achieved the highest detection rates on synthetic test data when calibrated for 99% specificity. Applied to real intraoperative images, the model revealed fluorescent areas substantially larger than those marked by experienced surgeons. In non-5-ALA control cases, minimal false positives were observed, indicating a specificity exceeding 99.9%. The regression model reliably quantified PPIX concentration in synthetic samples ($R^2=0.92$). Conclusions: By enabling more sensitive and objective fluorescence detection, this approach offers a valuable tool for improving surgical decision-making and facilitating safer, more extensive tumor resections. Full article

Background/Objectives: Delayed cerebral ischemia (DCI) is a major cause of poor outcome after aneurysmal subarachnoid hemorrhage (aSAH). Beyond large-vessel vasospasm, DCI reflects a systemic, multifactorial process involving inflammation, hematologic dysregulation, and organ dysfunction. Stroke-associated pneumonia (SAP), a frequent aSAH complication linked to stroke-induced immunodepression, may aggravate secondary ischemic injury. Unlike prior studies focusing on classical predictors alone, we included pneumonia and longitudinal respiratory parameters alongside inflammatory, hematologic, and renal markers. Using machine learning, this study aimed to identify predictors of DCI and functional outcome from routinely collected intensive care data. Methods: In this retrospective single-center study, 182 aSAH patients treated in a neurosurgical intensive care unit were included. Clinical data, SAP status, and longitudinal inflammatory, hematologic, renal, and respiratory parameters were extracted. DCI and functional outcome were assessed. Continuous variables were summarized as minimum, maximum, and mean values. Supervised machine learning models combining 12 feature selection methods and 12 classifiers were trained using five-fold cross-validation and evaluated by accuracy, F1-score, and AUC. Results: DCI occurred in 22% of patients, and SAP in 27%. The machine learning models achieved a mean accuracy of 59.7% (F1-score 58.8%, AUC 59.7%) for DCI prediction. No single dominant feature emerged; predictive patterns included leukocyte counts, CRP, erythrocyte indices, platelet variability, renal function, and oxygenation metrics. Functional outcome prediction performed moderately better (mean AUC 65.7%) and shared overlapping predictors. Conclusions: DCI reflects systemic instability in aSAH, with longitudinal inflammatory and respiratory variability outperforming static thresholds. Dynamic risk stratification may enable earlier detection of deterioration, supporting future time-series modeling and external validation. Full article

Cerebral aneurysm clipping remains a key surgical approach despite advancements in endovascular techniques. However, training for this procedure is complex due to the variable and fragile nature of aneurysmal tissues. This study evaluates the mechanical behaviour of human basilar arteries during clipping and compares them to 3D-printed models used for neurosurgical training. Mechanical tests were performed on ten cadaveric basilar arteries, distinguishing between healthy and plaque-affected segments. Plaque-affected regions required significantly higher clipping force (1.73 ± 0.22 N) compared to healthy segments (0.45 ± 0.19 N), confirming that atherosclerosis markedly increases arterial stiffness. Six 3D-printed phantom materials were evaluated; none accurately replicated the biomechanical response of real arteries. The Flex Anatomical material showed the highest stiffness (44.51 ± 0.98 N), while Silicone 40A was the most compliant (1.05 ± 0.12 N), yet both deviated substantially from biological performance. These findings underscore the current limitations of anatomical models that lack realistic biomechanical properties. Full article

Background: Three-dimensional (3D) printing technology has rapidly emerged as a transformative tool in medicine, enabling the conversion of two-dimensional scans into highly accurate 3D models. This technology, especially when combined with artificial intelligence (AI) and advanced materials, offers numerous applications in surgical planning, simulation-based training, and patient-specific care. Methods: This review examines current literature and case studies on the use of 3D printing technology in various fields of medicine, especially in surgical specialties. Key applications include surgical planning, mock surgeries, biopsy guide creation, and customized implant fabrication across various surgical fields. Results: 3D printing is transforming surgery by enabling precise visualization of tumors and critical structures, significantly enhancing preoperative planning for conditions such as bone, soft tissue (e.g., neuroblastomas), renal, and maxillofacial tumors. In reconstruction surgeries, patient-specific 3D-printed implants ensure better anatomical compatibility, particularly in maxillofacial, neurosurgical, and vascular applications. Puncture guides improve procedural accuracy in interventions like percutaneous nephrolithotripsy. Detailed anatomical models aid in simulation-based training, increasing preparedness for complex procedures. Additionally, patient-specific implants and AI-integrated decision support systems are paving the way for more personalized and efficient surgical care. Conclusions: 3D printing technology, especially when combined with AI, is reshaping modern surgery by improving both accuracy, safety, and personalized healthcare. Its applications extend across multiple specialties, offering new possibilities in surgical planning, training, and patient-specific treatments. As AI and bioprinting continue to evolve, the potential for real-time applications, such as live-printed tissue implants and enhanced decision support, could drive the next phase of innovation in various fields. Full article

Purpose: Cadaveric dissection is a cornerstone of neurosurgical education, providing trainees with a realistic 3D understanding of anatomy and a safe environment to practice surgical approaches. A preservation technique was developed that merges the advantages of fresh-frozen and embalmed cadavers, maintaining tissue realism while enhancing durability. This approach preserves flexibility and natural color, improves anatomical detail, and creates a safe, long-lasting model ideal for neurosurgical training. Methods: Four specimens were thawed, cannulated, and irrigated before implementing a protocol consisting of low concentration formaldehyde with glycerol and ethanol for extended preservation. The specimens were prepared for both neurosurgery training and educational purposes, and their condition was evaluated with a semi-quantitative scale. Each specimen was evaluated independently by two raters, blinded to the time-point, using a semi-quantitative scale anchored to predefined criteria (0–3 per domain). Inter-rater reliability was calculated using the intraclass correlation coefficient (ICC [2,k]) for continuous scores and Cohen’s κ for categorical agreement. Results: Over nine years of intermittent use, the specimens remained in good condition: tissues retained sufficient softness for dissection, injected vessels stayed vivid in color, and no foul odor or microbial growth was observed. The evaluation employed a semi-quantitative scale, with results ranging from 11/14 to 14/14. The mean values demonstrate stable tissue quality over time, with only minor variations in color and perfusion. The inter-rater reliability was high (ICC = 0.91; κ = 0.88). Conclusions: The preservation method leverages the strengths of both fresh-frozen and embalmed models. The results suggest feasibility of long-term reuse, although further quantitative validation is needed. Full article

Background: Cognitive and neuropsychological effects of cancer and its treatments have gained increasing attention over the past decade, with growing evidence of persistent deficits across multiple cancer types. While numerous studies have examined these effects, the literature remains fragmented, and no comprehensive bibliometric synthesis has been conducted to map the field’s intellectual structure and emerging trends. Methods: A bibliometric and science mapping analysis was performed using the Scopus database to identify peer-reviewed articles published between 2015 and 2025 on neuropsychological or cognitive outcomes in adult cancer populations. Data from 179 eligible publications were analyzed with VOSviewer and Microsoft Power BI, applying performance metrics and network mapping techniques, including co-authorship, bibliographic coupling, co-citation, and keyword co-occurrence analyses. Results: Publication output increased steadily over the decade, with leading contributions from the Journal of Neuro-Oncology, Psycho-Oncology, and Brain Imaging and Behavior. Co-citation analysis identified three core intellectual pillars: (i) clinical characterization of cancer-related cognitive impairment, (ii) mechanistic and neuroimaging-based investigations, and (iii) neurosurgical and neuropathological research in brain tumors. Keyword mapping revealed emerging themes in sleep and circadian rhythm research, biological contributors to cognitive decline, and scalable rehabilitation strategies such as web-based cognitive training. Collaborative networks, while showing dense local clusters, remained moderately fragmented across disciplines. Conclusions: This review provides the first quantitative, decade-spanning map of cognitive oncology research, highlighting both consolidated knowledge areas and underexplored domains. Future efforts should prioritize methodological standardization, cross-disciplinary collaboration, and integration of cognitive endpoints into survivorship care, with the ultimate aim of improving functional outcomes and quality of life for cancer survivors. Full article

Background/Objectives: Minimally invasive spine surgery (MISS) is complex and requires proficiency with a variety of technological and robotic modalities. Acquiring these skills is a long and involved process, often with a steep learning curve. This paper seeks to characterize the state of MISS training in neurosurgical and orthopedic residency programs, focusing on their effectiveness at minimizing substantial learning curves in the field, as well as highlighting potential areas for future growth. Methods: We conducted a scoping review of the PubMed, Scopus, and Embase databases utilizing the PRISMA extension for scoping reviews. Results: Of the 100 studies initially identified, 16 were included in our final analysis. MISS training types could be broadly grouped into four categories: virtual simulation (including AR and VR), physical models, hybrid didactic and simulation, and mentored training. Training with these modalities led to improvements in resident performance across multiple different MISS techniques, including percutaneous pedicle screw fixation, MIS dural repair, MIS-TLIF, MIS-LLIF, MIS-ULBD, microscopic discectomy/disk herniation repair, percutaneous needle placement, and surgical navigation. Specific improvements included reduced error rate, operation time, and fluoroscopy exposure, as well as increased procedural knowledge, accuracy, and confidence. Conclusions: The incorporation of MISS training modalities in spine surgery residency leads to increases in simulated performance and could serve as a means of overcoming significant learning curves in the field. Full article

Mesial temporal lobe epilepsy (mTLE) surgery, particularly selective amygdalohippocampectomy (sAHE), is a recognized treatment for pharmacoresistant temporal lobe epilepsy (TLE). Accurate intraoperative orientation is crucial for complete resection while maintaining functional integrity. This study evaluated the usability and effectiveness of multimodal neuronavigation and microscope-based augmented reality (AR) with intraoperative computed tomography (iCT) and navigated intraoperative ultrasound (iUS) in 28 patients undergoing resective surgery. Automatic iCT-based registration provided high initial navigation accuracy. Navigated iUS was utilized to verify navigational accuracy and assess the extent of resection during the procedure. AR support was successfully implemented in all cases, enhancing surgical orientation, surgeon comfort, and patient safety, while also aiding training and education. At one-year follow-up, 60.7% of patients achieved complete seizure freedom (ILAE Class 1), rising to 67.9% at the latest follow-up (median 4.6 years). Surgical complications were present in three cases (10.7%), but none resulted in permanent deficits. The integration of microscope-based AR with iCT and navigated iUS provides a precise and safe approach to resection in TLE surgery, additionally serving as valuable tool for neurosurgical training and education. Full article

Background: Acute Kidney Injury (AKI) is a pivotal concern in neurocritical care, impacting patient survival and quality of life. This study harnesses machine learning (ML) techniques to predict the occurrence of AKI in patients receiving hyperosmolar therapy, aiming to optimize patient outcomes in neurocritical settings. Methods: We conducted a retrospective cohort study of 4886 patients who underwent hyperosmolar therapy in the neurosurgical intensive care unit (ICU). Comparative predictive analyses were carried out using advanced ML algorithms—eXtreme Gradient Boosting (XGBoost), Gradient Boosting Machine (GBM), Random Forest (RF)—against standard multivariate logistic regression. Predictive performance was assessed using an 8:2 training-testing data split, with model fine-tuning through cross-validation. Results: The RF with KNN imputation showed slightly better performance than other approaches in predicting AKI. When applied to an independent test set, it achieved a sensitivity of 79% (95% CI: 70–87%) and specificity of 85% (95% CI: 82–88%), with an overall accuracy of 84% (95% CI: 81–87%) and AUROC of 0.86 (95% CI: 0.82–0.91). The multivariate logistic regression analysis, while informative, showed less predictive strength compared to the ML models. Delta chloride levels and serum osmolality proved to be the most influential predictors, with additional significant variables including pH, age, bicarbonate, and the osmolar gap. Conclusions: The prominence of delta chloride and serum osmolality among the predictive variables underscores its potential as a biomarker for AKI risk in this patient population. Full article