Background/Objectives: Artificial intelligence (AI) has rapidly expanded across medical imaging with proposed applications in diagnosis, prognostication, and surgical planning. Concerns remain regarding methodological robustness and clinical readiness for many published models. This systematic review aimed to conduct a methodological audit of AI imaging studies relevant to contemporary neurosurgical practice—including intracranial, cerebrovascular, spinal, and connectomics-based applications—published in 2025. Methods: Following PRISMA guidelines and PROSPERO registration (CRD420261284068), PubMed was searched for studies published in 2025 evaluating machine learning or deep learning applications in MRI- or CT-based imaging. Three reviewers independently extracted data on validation strategy, data leakage risk, human comparator use, calibration reporting, and CLAIM/TRIPOD-AI adherence. Risk of bias was assessed using PROBAST+AI. Results: Of 1776 screened records, 91 studies met the inclusion criteria. China led contributions (54.9%), oncology was the most common domain (37.4%), and MRI was the predominant modality (67.0%). External validation was reported in 75.8% of studies, and 66.0% used multicenter cohorts. Data leakage risk was low in 93.4%. However, only 18.7% included human comparators, calibration was reported in 30.8%, and none achieved full CLAIM/TRIPOD-AI compliance. Conclusions: AI imaging studies published in 2025 demonstrate encouraging progress in multicenter design and external validation. However, persistent gaps in human benchmarking, calibration, and reporting suggest further methodological development is needed.

Background: Data extraction for systematic reviews is highly resource-intensive. This study evaluated four frontier large language models (LLMs) on complex structured metadata extraction from specialized neuroimaging artificial intelligence (AI) literature to determine their performance in automated evidence synthesis. Methods: We compared Google Gemini 3 Pro Preview, Anthropic Claude Opus 4.5, Perplexity Sonar Pro, and OpenAI GPT 5.2. Using a standardized prompt, each model extracted 22 variables from 91 peer-reviewed neuroimaging AI articles. The variables were stratified into low-, medium-, and high-complexity tiers. The performance was measured via the exact-match accuracy against a consensus-based expert ground truth. Results: The overall exact-match accuracy was moderate. Gemini 3 Pro Preview achieved the highest overall rate (56.4%), followed by Sonar Pro (52.1%), Claude Opus 4.5 (51.3%), and GPT 5.2 (46.5%). Gemini significantly outperformed all other models (p < 0.001). The performance declined dramatically as the variable complexity increased. Across models, the accuracy was 88.9–92.9% for low-complexity categorical fields, 47.0–63.3% for medium-complexity text extraction, and 2.7–15.5% for high-complexity variables requiring clinical judgment or multi-section synthesis. The most common type of error was misclassification. All four models scored 0% on the main performance metric, but this reflected a representational mismatch with the ground truth rather than extraction failure, indicating that the exact-match accuracy underestimates the true semantic performance. Conclusions: Frontier LLMs can effectively automate the retrieval of simple categorical data, but have serious difficulties with methodological variables that are complex. Although extraction can be fully automated for low-complexity fields, human review remains essential for context-dependent variables that require clinical judgment.

Background/Objectives: Carotid body paragangliomas, commonly referred to as Carotid Body Tumors (CBTs), are rare, highly vascular paragangliomas arising at the carotid bifurcation and pose significant surgical challenges due to their proximity to critical neurovascular structures. Optimal management remains debated, particularly for large or complex lesions. This study aims to present a structured neurosurgical operative workflow as an educational and practical resource to help young surgeons understand operative decision-making and technical execution from a neurosurgical perspective. Methods: We retrospectively reviewed patients diagnosed with CBTs and identified three cases that underwent microsurgical resection by a single neurosurgeon. Clinical presentation, radiographic findings, operative strategies, intraoperative microsurgical techniques, and postoperative outcomes were analyzed. Surgical procedures for all three cases are further illustrated with technical video demonstrations highlighting meticulous microsurgical techniques performed by a single neurosurgeon. Results: All three patients presented with either incidental or slowly progressive neck masses, with imaging demonstrating classic splaying of the internal and external carotid arteries. One patient exhibited elevated catecholamine metabolites, while another had a familial history of paragangliomas. Preoperative embolization was successfully performed in all three cases. Complete tumor resection was achieved in each patient. One patient developed post-embolization embolic ischemic changes with transient neurological deficits that were resolved within several hours. No permanent cranial nerve deficits, vascular injuries, or tumor recurrences were observed. Pathology confirmed paraganglioma in all cases. Conclusions: Surgical resection remains an effective treatment for CBTs, which are commonly managed by vascular or head and neck surgeons. This case series illustrates the technical feasibility of CBT resection using a comprehensive neurosurgical strategy that integrates endovascular preparation, cerebral perfusion assessment, and meticulous microsurgical technique. Rather than proposing novel surgical innovation, this report aims to provide a structured operative framework and detailed video-based illustration of complex carotid bifurcation management from a neurosurgical perspective.

Intracranial epidermoid tumors (ETs) are rare, benign lesions that often arise within the cerebellopontine angle and are typically slow-growing with favorable outcomes. Malignant transformation is exceedingly rare and poorly understood, with limited data on imaging findings, medical and operative management, and overall survival. We conducted a PRISMA-guided systematic review and pooled analysis of 68 patients from 58 studies with pathologically confirmed malignant transformation of intracranial ETs. Clinical presentation, tumor location, radiologic evolution, management, histopathology, molecular findings, and survival were analyzed. The Kaplan-Meier method was used for survival analysis. Mean age at diagnosis was 49 years with a slight male predominance. Most tumors arose in the cerebellopontine angle (97%). Malignant transformation was associated with contrast enhancement (98%) and rapid growth (57%). Longitudinal imaging revealed heterogeneous T2 signal changes (56%) and loss of diffusion restriction on diffusion-weighted imaging (43%), termed "Loss of Fluidity" and "Flip Sign." Gross total resection was achieved in 24%; subtotal resection in 66%, often supplemented with radiotherapy (46%) and chemotherapy (21%). Median overall survival was 23 months, with 58% mortality within six months and recurrence in 63%. Immunohistochemistry confirmed squamous differentiation with high Ki-67 (mean 41%). Malignant transformation of intracranial ETs is rare but highly aggressive. Early recognition of clinical deterioration and identifying characteristic imaging changes may enable prompt detection. Maximal safe resection remains the cornerstone of management, with multimodal approaches potentially improving outcomes. Long-term surveillance and molecular studies are essential to guide prognosis and targeted treatment.

Background: The reliable integration of large language models (LLMs) into neuroimaging data extraction workflows remains unresolved. Prior benchmarking shows that exact-match accuracy underestimates LLM extraction performance, but whether inter-model consensus and variable complexity can guide automation remains unclear. We evaluated whether inter-model consensus can serve as a confidence signal for human–artificial intelligence (AI) extraction and can guide complexity-stratified workflow triage. Methods: Four frontier LLMs were queried via OpenRouter with an identical zero-shot structured prompt to extract 22 predefined variables from 91 peer-reviewed neuroimaging AI articles, yielding 2002 article–variable items per model. Variables were stratified a priori into low- (n = 7), medium- (n = 8), and high-complexity (n = 7). Performance was compared with an expert reference using exact-match and semantic-equivalence accuracy. Item-level consensus and five triage strategies characterized the efficiency–accuracy trade-off. Results: Semantic-equivalence accuracy converged to 80.5–83.4% across models despite approximately ten percentage-point exact-match differences. Unanimous 4/4 consensus occurred in 45.6% (910/1994) of items, with exact-match accuracy of 85.8%, rising to 95.3% after semantic normalization; however, 14.2% still failed to match the reference. Reliability was complexity-dependent: 96.6% for low-complexity variables, 73.2% for medium-complexity variables, and 38.1% for high-complexity variables. A hybrid strategy auto-accepting 4/4 items and routing 3/4 items to rapid verification reduced estimated review effort by approximately 59%. Conclusions: Inter-model consensus is useful, but it is incomplete and depends on variable complexity. We show that LLM-assisted extraction in neuroimaging AI is a complexity-stratified workflow design problem: low-complexity neuroimaging variables may be selectively automated, while medium-complexity variables require rapid verification, and high-complexity methodological variables should remain human-led.