Artificial intelligence models for clinical prediction are proliferating across neurosurgery, but their aggregate discriminative performance, distribution along the care pathway and methodological quality are unclear. We systematically reviewed and meta-analysed these studies in neurosurgery and its subspecialities along with their risk of bias appraisal. We performed a PRISMA search for peer-reviewed original studies using AI to predict a clinical outcome in neurosurgical populations. Metrics were pooled using random effects modelling. Each study was classified as diagnostic, prognostic or forecasting and was assigned an evidence level. Clinical prediction task sub-stratification analysis was performed. Risk-of-bias was assessed using PROBAST. 1063 studies were included. Across 451 studies reporting AUC, pooled discrimination was 0.86, ranging from 0.84 (spine) to 0.87 (neuro-oncology), with extreme heterogeneity. Performance varied by clinical pathway stage, highest for diagnosis, grading and molecular tasks and lowest for post-operative tasks. Only 51 studies were using prospective data with external validation, whereas 619 rested on internal validation alone or no validation. 89 percent studies were at high risk-of-bias, driven by the analysis domain. AI models report high performance across all the neurosurgical subspecialities, concordant across AUC, accuracy, sensitivity and specificity and robust to risk-of-bias stratification. Performance is highest for imaging-anchored diagnostic tasks and lowest at decision-facing post-operative stage. However, the literature is dominated by optimism-prone, internally-validated and poorly calibrated models with extreme heterogeneity. External validation, calibration analysis and reporting guidelines adherence remain prerequisites for clinical adoption.
PURPOSE Contemporary neurosurgical training faces challenges due to reduced operative exposure and the limited, inconsistent feedback provided to trainees. Wearable technologies may help address these issues by offering objective data and real-time support. This scoping review systematically evaluates the evidence of wearable technology in surgical training, analysing device types, their applications, and the supporting evidence for their effectiveness. MATERIALS AND METHODS A systematic literature search was conducted in 10 pre-clinical and clinical databases from inception to May 2024, searching for keywords "wearable technology" AND "surgery" AND "training". The search yielded 25,806 results. After removing duplicates, papers were screened by title and abstract (n = 12,306), followed by full-text screening (n = 482). RESULTS Forty-one studies, including 985 participants, were included in the review. Head-mounted displays were predominantly used for augmented and mixed reality applications, while smart glasses facilitated telementoring and procedure recording. Other wearables enabled biomechanical analysis and ergonomic assessments. Reported outcomes varied widely, with clinician-centred metrics, such as procedural accuracy, error rates, completion times, and satisfaction, being the most frequently assessed. Devices including HTC VIVE and HoloLens were associated with improved surgical performance. Clinicians often reported enhanced visualisation and anatomical orientation, though barriers such as learning curves, technical issues, and discomfort were noted. CONCLUSIONS This review provides a comprehensive evaluation of wearable technologies in surgical training, highlighting their potential to improve performance metrics and support a range of educational uses for neurosurgical trainees. However, key challenges must be addressed to enable broader adoption and integration into surgical education.
