About

Douglas Kondziolka, MD, is a professor and vice chair of clinical research in the Department of Neurosurgery at NYU Langone Health, where he joined the faculty in 2012. He is also the director of the Center for Advanced Radiosurgery, where he treats patients with Gamma Knife radiosurgery and is recognized as the most experienced active neurosurgeon in this area worldwide. His clinical practice involves treating neurological conditions such as metastatic brain tumors, acoustic neuroma, meningioma, primary brain tumors, brain vascular malformations, movement and behavioral disorders, and trigeminal neuralgia. Prior to NYU Langone, he served as chief of stereotactic and functional neurosurgery and director of the Center for Brain Function and Behavior at the University of Pittsburgh, and worked with the Pittsburgh Penguins as their neurosurgeon from 2002 to 2012. Dr. Kondziolka's research spans many conditions, including benign and malignant tumors, vascular malformations, and functional disorders, with a focus on stereotactic radiosurgery. His work involves collecting patient outcomes and developing innovative artificial intelligence tools in medicine. He launched the world’s largest brain tumor dataset for research in 2022. His research also includes the informed consent process for surgery, neurostimulation for major depression, and clinical trials for stroke and neuroprogenitor cell implantation. He has published extensively, with 685 medical journal articles, 280 book chapters, and has edited 8 books. As editor-in-chief of the journals Neurosurgery, Operative Neurosurgery, and Neurosurgery Practice, he has contributed significantly to the field. Dr. Kondziolka has received numerous awards, including the Mahaley Clinical Award, the Lars Leksell Award, and the Jacob I. Fabrikant Award, and has served as president of several major neurosurgical societies. His leadership roles include being a past president of the American Society for Stereotactic and Functional Neurosurgery, the International Stereotactic Radiosurgery Society, and the Congress of Neurological Surgeons.

Research topics

• Medicine
• Internal medicine
• Surgery
• Computer Science
• Radiology
• Oncology
• Psychology
• Artificial Intelligence
• Machine Learning
• Data science
• Neuroscience
• Medical emergency
• Software engineering
• Medical physics
• Engineering
• General surgery
• Nuclear medicine

Selected publications

• AI-Powered Pipeline Transforms Neurosurgical Articles Into High-Quality Graphical Abstracts

  Neurosurgery Open · 2026-04-17

  articleOpen access

  BACKGROUND AND OBJECTIVES: articles into graphical abstracts using Cascade Styling Sheets (CSS) templates and iterative prompting of a frontier vision language model and to conduct a human evaluation of this pipeline. METHODS: We developed an automated pipeline to convert extracted manuscript content into standardized graphical abstracts. The pipeline implements a custom CSS profile designed to match existing journal standards. Using Claude Sonnet-3.5, we generated structured hypertext markup language summaries organized into 6 sections: Objectives, Background, Methods, Results, Discussion, and Conclusion. The model selected up to 2 representative figures per manuscript based on caption analysis. We evaluated performance using 100 randomly selected articles published between 2020 and 2024 (95 from Neurosurgery, 4 from Operative Neurosurgery, 1 from Neurosurgery Practice). Three Editorial Review Board members independently assessed abstracts using 3 binary criteria: (1) proper formatting, (2) factual accuracy, and (3) visual appeal. RESULTS: Generated graphical abstracts achieved proper formatting in 85% of cases (95% CI: 76.7%-90.7%), factual accuracy in 99% (95% CI: 94.4%-99.9%), and visual appropriateness in 82% (95% CI: 73.3%-88.3%). Overall, 70% of abstracts (95% CI: 60.5%-78.1%) met all 3 criteria and were deemed "publication ready" without manual intervention. Error analysis revealed poor figure selection (40.0%) as the most common failure mode, followed by title replacement errors from PDF extraction (26.7%). CONCLUSION: Our artificial intelligence-CSS pipeline demonstrates the feasibility of automating graphical abstract generation for neurosurgical manuscripts, achieving publication-ready quality in 70% of cases with 99% factual accuracy. This technology offers a scalable augmentation tool that can reduce the design burden for authors, enhancing visual scientific communication in neurosurgical publishing while complementing human expertise.

  PublisherDOI

• A Multi-AI Agent Framework for Interactive Neurosurgical Education and Evaluation: From Vignettes to Virtual Conversations

  Neurosurgery Open · 2026-03-13

  articleOpen access

  BACKGROUND AND OBJECTIVES: Traditional medical board examinations present clinical information in static vignettes with multiple-choices (MC), fundamentally different from how physicians gather and integrate data in practice. Recent advances in large language models (LLMs) offer promising approaches to creating more realistic clinical interactive conversations. However, these approaches are limited in neurosurgery, where patient communication capacity varies significantly and diagnosis heavily relies on objective data such as imaging and neurological examinations. We aimed to develop and evaluate a multi–artificial intelligence (AI) agent conversation framework for neurosurgical case assessment that enables realistic clinical interactions through simulated patients and structured access to objective clinical data. METHODS: We developed a framework to convert 608 Self-Assessment in Neurological Surgery first-order diagnosis questions into conversation sessions using 3 specialized AI agents: patient AI for subjective information, system AI for objective data, and clinical AI for diagnostic reasoning. We evaluated generative pretrained transformer 4o's (GPT-4o's) diagnostic accuracy across traditional vignettes, patient-only conversations, and patient + system AI interactions, with human benchmark testing from 10 neurosurgery residents. RESULTS: GPT-4o showed significant performance drops from traditional vignettes to conversational formats in both MC (89.0%-60.9%, P < .0001) and free-response scenarios (78.4%-30.3%, P < .0001). Adding access to objective data through system AI improved performance (to 67.4%, P = .0015; and 61.8%, P < .0001, respectively). Questions requiring image interpretation showed similar patterns but lower accuracy. Residents outperformed GPT-4o in free-response conversations (70.0% vs 28.3%, P = .0030) using fewer interactions and reported high educational value of the interactive format. CONCLUSION: This multi-AI agent framework provides both a more challenging evaluation method for LLMs and an engaging educational tool for neurosurgical training. The significant performance drops in conversational formats suggest that traditional MC testing may overestimate LLMs' clinical reasoning capabilities, while the framework's interactive nature offers promising applications for enhancing medical education.

  PublisherDOI

• Risk-Weighted Impact: Reframing Risk Analyses for Medical Decisions

  Neurosurgery · 2026-04-23

  articleSenior author

  BACKGROUND AND OBJECTIVES: Developing precise metrics for clinical use is vital to translating outcomes data to practice. Risk probabilities, such as the complication rate of surgery or the lifetime risk of an aneurysm rupture, are widely used for counseling patients, but their comparison may be misleading if risks are spread over different time horizons. This study evaluates a new risk-assessment approach called risk-weighted impact (RWI) that applies event probabilities to estimate the average number of years of life impacted by event occurrence. METHODS: Decision-making policies based on RWI and cumulative lifetime event risk were applied to determine management in a simplified model of incidental cerebral aneurysms through Monte Carlo simulation (1000 iterations of 10 000 synthetic patients). In addition, a web-based application was created to simplify risk-assessment calculations and comparisons. RESULTS: When treatment of incidental cerebral aneurysms was simulated using both risk assessment methods, there was disagreement in 25.2% (95% CI: 24.4%-26.1%) of cases, with the RWI policy preferring observation, while event-risk policy preferred intervention. In these patients, the number of poor outcomes was nearly the same, 110 (95% CI: 91-129) in RWI policy and 111 (95% CI: 90-132) in event-risk policy, but the RWI policy resulted in 874.8 fewer quality-adjusted life year lost (95% CI: 299.5-1466.6) due to adverse events occurring an average of 11.3 years later (95% CI: 8.2-14.1 years). CONCLUSION: Only using cumulative lifetime event risks may understate the impact of an up-front treatment given that a larger proportion of risk is assumed at an earlier age, when more years of life are in jeopardy. RWI offers an alternative approach to thinking about risk, using the same inputs (event probabilities and life expectancy) to compare estimated patient impact. RWI is more aligned with clinical objectives and is a valuable metric for risk assessment and decision making.

  PublisherDOI

• P11.14.A STEREOTACTIC RADIOSURGERY FOR OVARIAN CANCER BRAIN METASTASES: AN INTERNATIONAL RADIOSURGERY RESEARCH FOUNDATION RETROSPECTIVE STUDY

  Neuro-Oncology · 2025-10-01

  articleOpen access

  Abstract BACKGROUND Ovarian cancer rarely metastasises to the brain, thus studies reporting the outcomes of affected patients are lacking. Stereotactic radiosurgery (SRS) is now the mainstay of management for patients with brain metastases (BM) from most primary sites, but little evidence of its efficacy in ovarian cancer is available. The current study was undertaken to provide guidance for SRS management using pooled data from multiple institutions. MATERIAL AND METHODS Centers participating in the IRRF (International Radiosurgery Research Foundation) were asked to provide outcome data for patients who had SRS for ovarian cancer brain metastases between 2020 and 2024 and at least one clinical and imaging follow-up after the procedure. Primary endpoints included survival from SRS, local tumor response according to RANO-BM criteria and occurrence of adverse radiation effects (ARE). Cox regression analyses were performed to identify variables impact each endpoint. RESULTS 128 patients had SRS for a total of 532 BM treated. Epithelial histology was the most common (91%). Median age at SRS was 62 years (IQR 56-70). Median KPS was 80% (IQR 80-90%) and 71.9% of patients had neurological symptoms at presentation. Other active systemic metastases were present in 41.4%. The median number of treated BM was 2 (IQR1-3) and the median cumulative treatment volume was 6 cc (IQR 2-12.9). The median margin dose was 18 Gy (IQR 16-20). At last follow-up, 19.5% of patients were still alive. The median overall survival (OS) after SRS was 27 months, and 6-, 12- and 24-month OS was 83.8%, 74.2% and 52.2%, respectively. Multivariate analyses revealed that increasing age at SRS (HR 1.02, p=0.05), active systemic metastases (HR 2, p=0.005) and increasing number of brain metastases (HR 1.05, p=0.04) were associated with worse survival, while repeating SRS (HR 0.4, p=0.002) led to improved survival. Local failure occurred in 12.8% of treated BM. Actuarial progression-free survival (PFS) at 6, 12 and 24 months was 92.3%, 86.6% and 70.3%, respectively. Only prior WBRT (HR 4.36, p=0.03) led to worse local control on multivariate analyses. New remote BM appeared in 49.5% of patients and 14.5% suffered from leptomeningeal dissemination. ARE were seen in 12.1% of BM but were symptomatic in only 3.2%. On multivariate analyses, prior SRS (HR 3.13, p=0.002) was associated with increased risk of ARE. CONCLUSION Ovarian cancer brain metastases can be safely and effectively managed with SRS as the primary treatment modality in most patients.

  PublisherOA PDFDOI

• Neurosurgery Practice: A Journal Comes of Age

  Neurosurgery · 2025-09-15

  article1st authorCorresponding
  PublisherDOI

• Radiosurgery for Sporadic Facial Nerve Schwannoma: An International Multi-institutional Study of 60 Cases

  Otology & Neurotology · 2025-11-13

  article

  OBJECTIVE: To characterize patient outcomes after primary stereotactic radiosurgery (SRS) for the management of sporadic facial nerve schwannoma. STUDY DESIGN: Retrospective cohort study. SETTING: Six tertiary referral centers across the United States and United Kingdom. PATIENTS: Adults undergoing SRS from 2000 through 2023 for sporadic facial nerve schwannoma along any segment of the facial nerve were included. Patients with NF2-related schwannomatosis were excluded. INTERVENTION: Stereotactic radiosurgery. MAIN OUTCOME MEASURE: Long-term tumor control. RESULTS: Among 60 patients meeting inclusion, the median age at SRS was 52 years (IQR: 41 to 64) with a median tumor size of 19.5 mm (IQR: 14.7 to 22.8). Tumors commonly involved the internal auditory canal (73%), cisternal (49%), geniculate/labyrinthine (47%), and tympanic segments (22%). Two patients experienced SRS failure and underwent salvage treatment; salvage-free survival rates (95% CI; number still at risk) at 1, 3, 5, and 10 years after SRS were 100% (100 to 100; 55), 100% (100 to 100; 36), 100% (100 to 100; 18), and 87% (72 to 100; 9), respectively. Among 31 (52%) patients with House-Brackmann (HB) grade I facial function at presentation, only 6 demonstrated worse facial function at a median of 3.2 years (IQR: 1.7 to 6.6) after SRS. Of 18 patients with serviceable hearing (AAO-HNS class A/B) at SRS, 13 maintained serviceable hearing at a median of 1.0 years (IQR: 0.5 to 4.9) of post-SRS audiometric follow-up. CONCLUSIONS: Durable tumor control after primary SRS for sporadic facial nerve schwannoma is achieved in most patients. Among those with HB grade I facial function at presentation, treatment with SRS harbors limited additional risk of facial paresis beyond observation alone.

  PublisherDOI

• Evaluating the Performance and Fragility of Large Language Models on the Self-Assessment for Neurological Surgeons

  Neurosurgery · 2025-12-08

  articleOpen access

  BACKGROUND AND OBJECTIVES: The Congress of Neurological Surgeons Self-Assessment for Neurological Surgeons questions are widely used by neurosurgical residents to prepare for written board examinations. Recently, these questions have also served as benchmarks for evaluating large language models' (LLMs) neurosurgical knowledge. LLMs show significant promise for transforming neurosurgical practice; however, they are susceptible to in-text distractions and confounding factors. Given the increasing use of generative artificial intelligence and ambient dictation technologies, clinical text is at a larger risk for the inclusion of extraneous details. The aim of this study was to assess the performance of state-of-the-art LLMs on neurosurgery board-like questions and to evaluate their robustness to the inclusion of distractor statements. METHODS: A comprehensive evaluation was conducted using 28 state-of-the-art LLMs. These models were tested on 2904 neurosurgery board examination questions derived from the Congress of Neurological Surgeons Self-Assessment for Neurological Surgeons. In addition, the study introduced a distraction framework to assess the fragility of these models. The framework incorporated simple, irrelevant distractor statements containing polysemous words with clinical meanings used in nonclinical contexts to determine the extent to which such distractions degrade model performance on standard medical benchmarks. RESULTS: Six of the 28 tested LLMs achieved board-passing outcomes, with the top-performing models scoring over 15.7% above the passing threshold. When exposed to distractions, accuracy across various model architectures was significantly reduced-by as much as 20.4%-with 1 model failing that had previously passed. Both general-purpose and medical open-source models experienced greater performance declines compared with proprietary variants when subjected to the added distractors. CONCLUSION: While current LLMs demonstrate an impressive ability to answer neurosurgery board-like examination questions, their performance is markedly vulnerable to extraneous, distracting information. These findings underscore the critical need for developing novel mitigation strategies aimed at bolstering LLM resilience against in-text distractions, particularly for safe and effective clinical deployment.

  PublisherDOI

• Hypofractionation of Gamma Knife Radiosurgery for Intracranial Meningiomas: A Retrospective Multicenter Study and Systematic Review of Literature

  Neurosurgery · 2025-10-27 · 1 citations

  articleSenior author

  BACKGROUND AND OBJECTIVES: Hypofractionated Gamma Knife radiosurgery (hfGKRS) is increasingly considered for treating large or near-critical structure meningiomas because of potential safety advantages. However, data on optimal fractionation and long-term outcomes remain limited. This study evaluated the longer-term tumor control and toxicity after hfGKRS for intracranial meningiomas at 2 large centers, supplemented by a systematic review and meta-analysis of existing literature. METHODS: The analysis included 34 patients (site 1 = 25, site 2 = 9, median age 62.6 years) with 40 tumors (median volume 11.2 cm 3 ). 62% was low-grade (World Health Organization grade 0-1) and 38% was high-grade (World Health Organization grade 2-3). The most common fractionation schemes were 20 Gy in 5 fractions for low-grade and 21 Gy in 3 fractions for high-grade tumors. The mean follow-up was 28.8 months. RESULTS: Only 6 of 34 patients did not have any previous treatment including surgery and/or radiotherapy. 82% of patient patients had neurological deficits before stereotactic radiosurgery. The estimated rate of 5-year tumor progression for low-grade and high-grade tumors was 7.7% (95% CI 0.41%-30%) and 36% (95% CI 12%-62%). Symptoms improved in 12 patients (35%) and worsened in 6 patients (16%), with 1 case attributed to tumor progression and no significant visual deterioration in 16 tumors within 3 mm of the optic apparatus. There was no statistically significant association between fractionation (3 vs 5) scheme and tumor control ( P = .07) or survival ( P = .12). Karnofsky Performance Status performance was a significant predictor of death (HR 0.89, P = .012) and tumor progression (HR 0.93, P = .048). The combined meta-analysis revealed a 5-year tumor control rate of 91.6% for low-grade and 37.9% for high-grade meningiomas. CONCLUSION: hfGKRS demonstrates durable control and acceptable safety for low-grade intracranial meningiomas. High-grade tumors showed less favorable outcomes comparable with single-session Gamma Knife radiosurgery historical data. Further prospective data are needed to confirm these findings and optimize fractionation strategies.

  PublisherDOI

• Automating the Referral of Bone Metastases Patients With and Without the Use of Large Language Models

  Neurosurgery · 2025-08-15 · 2 citations

  article

  BACKGROUND AND OBJECTIVES: Bone metastases, affecting more than 4.8% of patients with cancer annually, and particularly spinal metastases require urgent intervention to prevent neurological complications. However, the current process of manually reviewing radiological reports leads to potential delays in specialist referrals. We hypothesized that natural language processing (NLP) review of routine radiology reports could automate the referral process for timely multidisciplinary care of spinal metastases. METHODS: We assessed 3 NLP models-a rule-based regular expression (RegEx) model, GPT-4, and a specialized Bidirectional Encoder Representations from Transformers (BERT) model (NYUTron)-for automated detection and referral of bone metastases. Study inclusion criteria targeted patients with active cancer diagnoses who underwent advanced imaging (computed tomography, MRI, or positron emission tomography) without previous specialist referral. We defined 2 separate tasks: task of identifying clinically significant bone metastatic terms (lexical detection), and identifying cases needing a specialist follow-up (clinical referral). Models were developed using 3754 hand-labeled advanced imaging studies in 2 phases: phase 1 focused on spine metastases, and phase 2 generalized to bone metastases. Standard McRae's line performance metrics were evaluated and compared across all stages and tasks. RESULTS: In the lexical detection, a simple RegEx achieved the highest performance (sensitivity 98.4%, specificity 97.6%, F1 = 0.965), followed by NYUTron (sensitivity 96.8%, specificity 89.9%, and F1 = 0.787). For the clinical referral task, RegEx also demonstrated superior performance (sensitivity 92.3%, specificity 87.5%, and F1 = 0.936), followed by a fine-tuned NYUTron model (sensitivity 90.0%, specificity 66.7%, and F1 = 0.750). CONCLUSION: An NLP-based automated referral system can accurately identify patients with bone metastases requiring specialist evaluation. A simple RegEx model excels in syntax-based identification and expert-informed rule generation for efficient referral patient recommendation in comparison with advanced NLP models. This system could significantly reduce missed follow-ups and enhance timely intervention for patients with bone metastases.

  PublisherDOI

• Most Roads Lead to Cushing: Mapping Neurosurgical Training Lineages in the United States

  World Neurosurgery · 2025-09-05

  article
  PublisherDOI

Recent grants

• NIH Grant K08NS001723

  NIH · $440k · 1999

Frequent coauthors

• L. Dade Lunsford

  University of Pittsburgh Medical Center

  2309 shared

• John C. Flíckinger

  University of Pittsburgh Medical Center

  1579 shared

• Ajay Niranjan

  University of Pittsburgh Medical Center

  811 shared

• Hideyuki Kano

  681 shared

• Jason P. Sheehan

  University of Virginia Health System

  582 shared

• David Mathieu

  Brown University

  415 shared

• Inga S. Grills

  Beaumont Health

  290 shared

• Gene H. Barnett

  Cleveland Clinic

  208 shared

Awards & honors

• Mahaley Clinical Award for brain tumor clinical research fro…
• Lars Leksell Award from the World Federation of Neurosurgica…
• Penfield Lecturer of the Canadian Neurosurgical Society
• Robert Florin Award from the American Association of Neurolo…
• Integra Foundation Award from the American Association of Ne…