Article Text

Download PDFPDF
Evaluation of ChatGPT in knowledge of newly evolving neurosurgery: middle meningeal artery embolization for subdural hematoma management
  1. Stefan W Koester,
  2. Joshua S Catapano,
  3. Joelle N Hartke,
  4. Robert F Rudy,
  5. Tyler S Cole,
  6. Anant Naik,
  7. Ethan A Winkler,
  8. Christopher S Graffeo,
  9. Visish M Srinivasan,
  10. Ashutosh P Jadhav,
  11. Andrew F Ducruet,
  12. Michael T Lawton,
  13. Felipe C Albuquerque
  1. Department of Neurosurgery, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, Arizona, USA
  1. Correspondence to Dr Felipe C Albuquerque, c/o Neuroscience Publications Barrow Neurological Institute, Phoenix, Arizona, USA; neuropub{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Chat-Generative Pre-Trained Transfer (ChatGPT) is a readily accessible artificial intelligence (AI)-enabled chatbot recently developed by the company OpenAI (San Francisco, California, USA). The software incorporates a large language model with more than 175 billion parameters, with the capability of providing rapid answers to user-generated questions.1 The usefulness of ChatGPT in the healthcare domain is of increasing interest given its ability to obtain, digest, and divulge medically correct and informative text on demand. Using the United States Medical Licensing Examination (step 1, step 2 CK, and step 3) as a test of physician competency, Kung et al2 assessed the ability of ChatGPT to obtain basic medical information and make clinical decisions. It passed these examinations with approximately 60% accuracy.2 In a study by Hopkins et al,3 ChatGPT generated a correct response for 60.2% of neurosurgical board-style questions. Given the increasingly rapid pace of new research and innovation in neurosurgery, and particularly within the endovascular realm, the ability of ChatGPT to recognize and discuss new information within neurosurgery is of particular importance and interest to neurosurgeons. If able to efficiently update and maintain a contemporary summary of recent research, ChatGPT could be an enabling technology for physicians, but it poses risks if it provides a dated or skewed representation. To assess and rate its performance and ability to understand new and evolving fields within neurosurgery, we asked ChatGPT to create a review of middle meningeal artery embolization (MMAE) for chronic subdural hematoma.


ChatGPT (version 4.0) was tasked with writing a 2700-word review regarding the technique of MMAE for chronic subdural hematoma as well as its indications, evidence of clinical utility, limitations, and future directions. The precise prompt used to query ChatGPT was, ‘GPT-4, middle meningeal artery embolization is a procedure performed for subdural hematomas. Write a 2700-word review …

View Full Text


  • X @Graffeo

  • Contributors Conceptualization: JSC, SWK, AN. Data curation: JSC, SWK. Formal analysis: SWK. Investigation: JSC, SWK, AN. Methodology: JSC, SWK, AN. Project administration: JSC, FCA, JNH. Resources: JSC, SWK, AN. Software: AN, SWK. Supervision: JSC, SWK. Validation: JSC, JNH, CSG, VMS. Visualization: AN, JSC, SWK. Writing - original draft: JSC, SWK. Writing - review and editing: SWK, JSC, AN, JNH, RFR, TSC, EAW, CSG, VMS, APJ, AFD, MTL, FCA.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Disclaimer FCA and AFD serve on the editorial board of the Journal of NeuroInterventional Surgery. The other authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this manuscript.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

Linked Articles